Two New Targets for Type 2 Diabetes Care

Pulmonary Function and Sleep Breathing:Two New Targets for Type 2 Diabetes Care

Albert Lecube,1,2 Rafael Simo,2,3 Maria Pallayova,4,5 Naresh M. Punjabi,6,7

Carolina Lopez-Cano,1 Cecilia Turino,8 Cristina Hernandez,2,3 and Ferran Barbe8,9

1Endocrinology and Nutrition Department, Hospital Universitari Arnau de Vilanova, Institut de RecercaBiomedica de Lleida, Universitat de Lleida, Lleida, Catalonia 25198, Spain; 2Centro de Investigacion Biomedica enRed de Diabetes y Enfermedades Metabolicas Asociadas, Instituto de Salud Carlos III, Madrid 28029, Spain;3Endocrinology and Nutrition Department, Hospital Universitari Vall d’Hebron, Diabetes and MetabolismResearch Unit, Vall d’Hebron Institut de Recerca, Universitat Autonoma de Barcelona, Barcelona, Catalonia 08035,Spain; 4Department of Medicine, Weill Cornell Medicine, New York, New York 10065; 5Department of HumanPhysiology and Sleep Laboratory, Faculty of Medicine, Pavol Jozef Safarik University, Kosice 04011, Slovak Republic;6Division of Pulmonary and Critical CareMedicine, Department ofMedicine, Johns Hopkins University, Baltimore,Maryland 21224; 7Department of Epidemiology, Bloomberg School of Public Health, Johns Hopkins University,Baltimore, Maryland 21224; 8Respiratory Department, Hospital Universitari Arnau de Vilanova-Santa Marıa,Institut de Recerca Biomedica de Lleida, Universitat de Lleida, Lleida, Catalonia 25198, Spain; and 9Centro deInvestigacion Biomedica en Red de Enfermedades Respiratorias, Instituto de Salud Carlos III, Madrid 28029, Spain

ABSTRACT Population-based studies showing the negative impact of type 2 diabetes (T2D) on lung function are overviewed. Among the well-

recognized pathophysiological mechanisms, the metabolic pathways related to insulin resistance (IR), low-grade chronic inflammation, leptin

resistance, microvascular damage, and autonomic neuropathy are emphasized. Histopathological changes are exposed, and findings reported

from experimental models are clearly differentiated from those described in humans. The accelerated decline in pulmonary function that

appears in patients with cystic fibrosis (CF) with related abnormalities of glucose tolerance and diabetes is considered as an example to further

investigate the relationship between T2D and the lung. Furthermore, a possible causal link between antihyperglycemic therapies and pulmonary

function is examined. T2D similarly affects breathing during sleep, becoming an independent risk factor for higher rates of sleep apnea, leading to

nocturnal hypoxemia and daytime sleepiness. Therefore, the impact of T2D on sleep breathing and its influence on sleep architecture is

analyzed. Finally, the effect of improving some pathophysiological mechanisms, primarily IR and inflammation, as well as the optimization of

blood glucose control on sleep breathing is evaluated. In summary, the lung should be considered by those providing care for people with

diabetes and raise the central issue of whether the normalization of glucose levels can improve pulmonary function and ameliorate sleep-

disordered breathing. Therefore, patients with T2D should be considered a vulnerable group for pulmonary dysfunction. However, further

research aimed at elucidating how to screen for the lung impairment in the population with diabetes in a cost-effective manner is

needed. (Endocrine Reviews 38: 550 – 573, 2017)

T ype diabetes (TD) is a complex disease thataffects % of the middle-aged population ().

Its global prevalence is expanding dramatically, drivenby an obesogenic environment that promotes easieraccess to high energy–dense food along with sedentarybehavior, acting on susceptible genotypes. In addition,the increase in the ageing population is a noteworthycontributing factor.

TD is associated with devastating health conse-quences, affecting both large and small vessels, and

leading to retinopathy and visual loss, nephropathyand impaired kidney function, and peripheral neu-ropathy and amputations, as well as life-threateningmacrovascular disease (). Consequently, it has be-come one of the main threats to public healthworldwide. The problem increases when we take intoaccount that up to one-half of all sufferers are unawarethey have the disease, as people can develop TDwithout symptoms (). However, subjects with undi-agnosed TD share the same risk factors for micro- and

ISSN Print: 0163-769X

ISSN Online: 1945-7189

Printed: in USA

Copyright © 2017

Endocrine Society

Received: 26 June 2017

Accepted: 29 August 2017

First Published Online:

4 September 2017

550 https://academic.oup.com/edrv doi: 10.1210/er.2017-00173

REVIEWD

ownloaded from

https://academic.oup.com

/edrv/article/38/6/550/4103361 by guest on 21 July 2022

https://academic.oup.com/edrv

http://dx.doi.org/10.1210/er.2017-00173

macrovascular disease, and diabetic complications in-crease over time ().

Although the lung has not classically been includedin the list of organs that could be affected by diabetes,its great vascularization and abundant collagen andelastin fibers make the lung parenchyma a potentialtarget for chronic hyperglycemia (). In fact, thereare good reasons to believe that the same histo-logical and physiologic disturbances that account forcomplications in other organ systems may also affectpulmonary function (, ). In this regard, it should benoted that the increased glycemic exposure is asso-ciated with reduced pulmonary function in TD, and

the % decrease in forced expiratory volume in second (FEV) is an independent predictor of all-cause mortality in this population (). Basic researchand epidemiological and clinical data also support thenotion that TD has a deleterious effect on sleepbreathing, becoming an independent risk factor forsevere nocturnal hypoxemia ().

Although this review mainly focuses on the clinicaland epidemiological evidence for an association betweenTD and pulmonary function, the impact of TD onsleep breathing and sleep architecture is also updated. Inaddition, an overview of the underlying pathophysio-logical mechanisms involved in this association is given.

Clinical and Epidemiological Evidence

Baseline spirometric values are lower in patientswith T2DCross-sectional studies conducted during the last twodecades have shown that adults with TD have lowerforced vital capacity (FVC) and FEV values thanadults without TD (, –) (Table ; Fig. ). Amajority of the available reports have examined rel-atively large cohorts and have taken into accountseveral cofounding factors such as age, sex, body massindex (BMI), smoking history, and diabetes duration.In one of the earliest published works, the Cardio-vascular Health Study, in which the spirometric ref-erence values for a healthy population aged to years were determined, it was observed that TDwas significantly associated with a reduction in FEV(). Three years later, in , Barrett-Connor andFrette noted that in the Rancho Bernardo Study, in-volving men and women aged from to years, the FEV and FVC values were independentlylower in men with TD than in men without diabetes,particularly in those patients with diabetes duration ofat least years (). Interestingly, no associations weredetected between lung function and the prevalence of

TD or fasting plasma glucose (FPG) levels in women,probably because of the small number of women in-cluded in the study. Since then, several large epide-miological studies including the Fremantle DiabetesStudy, Framingham Heart Study, British Women’sHeart and Health Study, Normative Aging Study,Copenhagen City Heart Study, Atherosclerosis Risk inCommunities (ARIC) Study, and Strong Heart Studyhave consistently found an association between TDand several respiratory parameters such as FEV, FVC,and the FEV/FVC ratio (–, , , , ). Inaddition, a meta-analysis of studies on lung functionin , patients with diabetes and , healthycontrols and a systematic review of studies bothsupported the epidemiological evidence that TD isassociated with mild, but substantial, ventilation ab-normalities, more resembling a restrictive pattern (, ).

Although most of the empirical data are fromNorth America and Europe, there is corroboratingevidence that TD is also associated with impairedlung function in Asians (, , , , ). For ex-ample, in a cross-sectional study comprising ,Koreans over years of age, individuals with TDand those with impaired fasting glucose showed anincreased odds ratio [. (.–.)] for presenting

ESSENTIAL POINTS

· Type 2 diabetes exerts a deleterious effect on pulmonary function

· The reduced lung function is negatively associated with fasting plasma glucose, glycated hemoglobin, diabetes duration,and its severity

· Insulin and leptin resistance, low-grade chronic inflammation, microvascular damage, and autonomic neuropathy play anessential role in the lung damage associated with type 2 diabetes

· The histopathological findings include the thickening of the alveolar epithelia and pulmonary capillary basal lamina, thereduction of the alveolar space, higher degrees of fibrosis and microangiopathy, and modifications in mucus secretion

· Type 2 diabetes affects breathing during sleep, is an independent risk factor for higher rates of sleep apnea, and leads tonocturnal hypoxemia and daytime sleepiness

· The role of some therapeutic options such as insulin sensitizers and incretin-based therapies on lung function deservefurther research

· Patients with type 2 diabetes are recommended to be considered a vulnerable group for pulmonary dysfunction

551doi: 10.1210/er.2017-00173 https://academic.oup.com/edrv

REVIEWD

ownloaded from



http://dx.doi.org/10.1210/er.2017-00173


a restrictive ventilatory pattern, which remainedconsiderable after controlling for smoking, BMI, waistcircumference, sex, and age (). Another studyconducted in a Japanese population demonstrated thatvital capacity (VC) was independent and negativelyassociated with the presence of TD in both sexes ().A recent study involving data from , subjectsfrom the Hispanic and Latin American communityhas also revealed that those with TD had lower meanFEV and FVC values and also a higher punctuationin a dyspnea index than those without diabetes, withhigher impairment when the analysis was performedseparately for those with a known lung disease ().Finally, when, in the same clinical setting, the role ofdifferent races regarding the negative impact of TDon lung function was compared, whites seemed toshow more global impairment than African Ameri-cans and Hispanics ().

Apart from spirometry, there are also data on theassociation between impaired diffusing capacity forcarbon monoxide (DLCO) corrected for alveolarvolume and TD (, , ). Despite a small studysample, Chance et al. reported that DLCO at rest wassimilar among control subjects, nonobese subjectswith TD, and obese patients with TD (). Withexercise, however, patients with TD had a % to %reduction in DLCO compared with those withoutdiabetes. Although the decrease in DLCO with exercise

remained even after several covariates had been takeninto consideration, the noteworthy differences persistedonly in obese subjects with diabetes. These data suggestthat alterations in the lung parenchyma of obese in-dividuals with diabetes may impair alveolar-capillaryrecruitment during exercise due to connective tissuedeposition (). Reduced lung diffusing capacity in thelungs of subjects with TD could also result from theobserved aberrations in lung parenchyma. The decreasein the microvascular surface area and the increase inperfusion heterogeneity appear to be correlated withdisease duration and the presence of diabetes-relatedextrapulmonary complications (, ). For example,the reduction in DLCO in patients with TD has beenfound to be more common in those with signs ofdiabetic microangiopathy, as indicated by a notablepositive correlation with the -hour protein excretionrate ().

Other spirometric measures that have been de-scribed as being reduced in patients with TD includepeak expiratory flow (PEF), maximum voluntaryventilation (MVV), total lung capacity (TLC), andmaximum expired pressure (, , , , , ).

The association between metabolic data andspirometric valuesThe reduced lung function in patients with TD isassociated with an increase in fasting plasma glucose,

Table 1. Cross-Sectional Studies Evaluating the Relationship Between T2D and Impaired Lung Function

T2D vs Without T2D

Author Year Sample N FEV1 FVC DLCO

Barrett-Connor and Frette (12) 1996 Rancho Bernardo Study 1239 ↓a ↓a —

Davis et al. (13) 2000 Fremantle Diabetes Study 495 ↓ ↓ —

Lange et al. (14) 2002 Copenhagen City Heart Study 12,062 ↓ ↓ —

Walter et al. (15) 2003 Framingham Heart Study 3254 ↓ ↓ —

Lawlor et al. (16) 2004 British Women’s Heart and Health Study 3911 ↓ ↓ —

Litonjua et al. (17) 2005 Normative Aging Study 704 ↓b ↓b —

Chance et al. (18) 2008 Patients with T2D and nonsmokers 114 ↔ ↓ ↓c

Yeh et al. (19) 2008 ARIC 11,262 ↓ ↓ —

Oda et al. (20) 2009 Japanese men and women 2608 — ↓ —

Oda and Kawai (21) 2009 Japanese men and women 1353 — ↓ —

Yeh et al. (22) 2011 Strong Heart Study 2396 ↓ ↓ —

Klein et al. (23) 2012 Clinical setting study 4164 ↓ ↓ ↓

Huang et al. (24) 2014 Clinical setting study 584 ↓ ↓ —

Klein et al. (25) 2016 Hispanic Community Health Study/Study of Latinos 14,455 ↓ ↓ —

a

Association only noted in those with a diabetes duration of .10 years.b

FEV1 and FVC lower in the regular smokers but not in nonsmokers.c

Decrease in DLCO only noted with exercise and not at rest.

552 Lecube et al Type 2 Diabetes, Pulmonary Function, and Sleep Endocrine Reviews, December 2017, 38(6):550–573

REVIEWD

ownloaded from



glycated hemoglobin (HbAc), diabetes duration, andits severity (, , , , ,). In the ARIC study,a graded inverse association was observed amongFVC, FEV, and fasting glucose and diabetes severity(stratified by duration of diabetes and the types ofantidiabetic medications used) (). Data from theFremantle Diabetes Study showed that TD duration,more than glycemic control, was associated with re-duced pulmonary function, mainly FEV and PEF(). In contrast, data from the Hispanic CommunityHealth Study/Study of Latinos has shown that thepresence of macroalbuminuria in patients with TD isrelated to a % lower FVC and a % lower FEV thanthose with normoalbuminuria ().

The inverse relation between metabolic data andspirometric values has also been described in subjectswithout diabetes. In the Rancho Bernardo Study,higher FPG values were correlated with lower FEVand FVC in men without diabetes, suggesting that thenegative effect of glycemia on pulmonary functionmight precede the diagnosis of diabetes (). Also,among the , participants not known to have TDassessed in the cross-sectional Copenhagen City HeartStudy, there was a notable association between ele-vated plasma glucose and a reduction in lung function(). In the Normative Aging Study, in which onlymale subjects were included, those who developedTD had lower FEV and FVC than subjects withoutdiabetes (). In addition, differences in lung functionbetween groups were observed both prior to the di-agnosis of diabetes and at the moment when thediagnostic criteria for diabetes were met, even afteradjustment for age, height, weight, and smokingstatus (). In a Japanese population, a lower VC wasindependently associated with FPG in men, but therelationship was not conclusive in women ().

Is the decline in spirometric values accelerated insubjects with T2D?Although an independent association exists betweenimpaired lung function and TD, inferences regardingcausal effects cannot be made due to the cross-sectional nature of some of the available studies. Al-though longitudinal studies are available, data oncausal effects are equivocal with a comparable numberof positive and negative studies on the differences inthe longitudinal change in FEV and FVC betweenthose with and without TD.

The ARIC Study and the Fremantle Diabetes Studyrevealed a greater decline in FVC in subjects with TDthan in the control population, after and years offollow-up, respectively (, , ). In the prospectiveARIC Study conducted on a TD population (n =,) with years of follow-up, a considerable de-crease in FVC in patients with diabetes was observedcompared with the control group of , healthymiddle-aged subjects ( vs mL/year; P = .) ().In addition, TD severity, as indicated by the use ofantidiabetic medications, showed a faster and gradedrelationship with absolute FVC and FVC decline ().In the Fremantle Diabetes Study, a forced spirometrywas performed in a group of Australian non-smoking patients with TD at the beginning of thestudy (). After years of observation, a notableannual FEV decline of mL was revealed incomparison with a decline of to mL in thehealthy, nonsmoking controls. More importantly, thestudy revealed a sizeable positive correlation betweenFVC decline and a higher baseline HbAc: for every %increase in the HbAc level, a FVC decline of %predicted value was observed ().

By contrast, when lung function was longitudi-nally analyzed in the Danish patients with TD

11 13

29 30

85

7

16 21 22

2527

28 2315

2417

26

14

201918

Barcelona (7)Seattle-Washington (11)California (13)Australia (14)Massachusetts (15)

United Kingdom (16)United States (17)United States (18)Arizona-Oklahoma-Dakota (19)Chicago (20)

Copenhagen (21) (22)Japan (23) (24) (26)Bijapur (25)Fuzhou (27)

Seoul (28)United States (29)Dallas-Texas (30)Colombia (85)

© 2017 Endocrine Reviews ENDOCRINE SOCIETY

Figure 1. Worlddistribution of cross-sectional studies evaluatingthe relationship betweenT2D and impaired lungfunction.


REVIEWD

ownloaded from



http://dx.doi.org/10.1210/er.2017-00173


included in the Copenhagen City Heart Study over a-year period, the decline of FVC and FEV wassimilar to that observed in their counterparts withoutdiabetes (). Similar results were reported by the samegroup when patients with diabetes were followedover a -year period. These findings suggest that thereduced pulmonary function described in patientswith TD seems not to be progressive in the long term(). However, it should be noted that the recruitedpatients with diabetes were the sum of insulin-dependent and non–insulin-dependent subjectswith diabetes mellitus and that the definition of diabeteswas self-reported or based on a single measurement ofnonfasting blood glucose $ mg/dL ().

Differing results were also observed when lungfunction was evaluated in healthy subjects according towhether they developed TD during the follow-upperiod. In the longitudinal analysis of , subjectsparticipating in the Copenhagen City Heart Study,after adjusting for confounders, the participantsnewly diagnosed with TD lost mL of FVC and mL of FEV annually over a -year observationperiod, almost twice as much as control subjects ().By contrast, in the Normative Aging Study, there wereno differences in the rates of FEV or FVC change overtime between men who developed diabetes andsubjects without diabetes ().

Collectively, the cross-sectional and longitudinaldata indicate that, although patients with TD mayhave lower FEV, FVC, and DLCO values thansubjects without diabetes, the causal relationship be-tween TD and lung function is not well defined.Nonetheless, although the loss in lung function inpatients with TD is of moderate magnitude and evensubclinical, it may be of clinical significance, par-ticularly given the high prevalence of comorbid-associated conditions such as cardiac and renalfailure in TD.

What comes first, lung dysfunction or T2D?The vast majority of studies that have evaluated thisquestion have considered how often the spirometricparameters are capable of identifying healthy subjectswho will develop TD. In the Fremantle DiabetesStudy, the decline in lung function began ~ yearsbefore the diagnosis of TD (). Previously, in theCopenhagen City Heart Study, an association betweenthe newly diagnosed diabetes and impaired pulmonaryfunction was also observed (). Likewise, in a Swed-ish cohort, Engstrom and Janzon () and Engstromet al. () reported that low FVC and FEV predictedincident diabetes independent of adiposity. Also,data from the National Health and Nutrition Exam-ination reported that FVC and FEV, togetherwith the presence of a restrictive ventilatory pat-tern, were significantly and inversely associated withthe incidence of diabetes in the United States ().More recently, in the Strong Heart Study of Native

Americans, an impaired lung function was detectedbefore the development of metabolic syndrome orTD (). In this regard, the risk of incident TDincreased by % for every % decrease in predictedFVC percentage during the -year follow up ().

Overall, these findings suggest that the etiopa-thogenic mechanisms that precede the development ofTD, such as IR and low-grade chronic inflammation,also play an essential role in the lung damage andrespiratory abnormalities associated with TD.

Pathophysiological Mechanisms UnderlyingPulmonary Dysfunction in T2D

The association between pulmonary dysfunction andTD has been increasingly documented. However, theunderlying causes of its initiation and aggravation havenot been fully elucidated. Few pathophysiologicalmechanisms have been well documented to participatein the development of the diabetic lung, but the po-tential impact of other suggested underlying mecha-nisms on the lung function still requires furtherevidence before being formally recognized (Fig. ).Although described separately, each explains a part ofthe whole picture, and many of them act simulta-neously in patients with TD (Fig. ).

Well-recognized pathogenic mechanisms

IRIR is one of the main pathophysiological mechanismsleading to TD, and its role in initiating lung ab-normalities merits attention (). The relationshipbetween ventilatory function and indirect measures ofIR was first assessed with cross-sectional data from male participants without diabetes in the NormativeAging Study (). Males in the highest tertile of fastinginsulin levels showed a considerable decrease in theirmean unadjusted FVC of mL in comparison withmales in the lowest tertile. After adjusting for potentialconfounders including age, height, BMI, waist-to-hipcircumference ratio, physical activity, alcohol intake,and smoking in separate multiple linear regressionmodels, fasting insulin and the fasting IR index werenegatively associated not only with FVC but also withFEV. In this study, the extent to which variation infasting insulin levels was associated with variability inFVC was comparable in magnitude to the effect oflifetime cigarette smoking ().

Similar results were extended to women when thebaseline data from women without diabetesbetween and years of age participating in theBritish Women’s Heart and Health Study showeda linear inverse association of IR with lung functionmeasurements such as FEV and FVC (). With fulladjustment for all potential confounding factors, forevery standard deviation increase in log FEV and log


REVIEWD

ownloaded from



FVC, the homeostatic model assessment of IR(HOMA-IR) score decreased by % and %, re-spectively (). The authors suggested that the asso-ciation was, at least in part, explained by early lifefactors that affected lung growth and programmed IR.In a study of morbidly obese women without di-abetes, IR was also recognized as an independentdeterminant of pulmonary dysfunction (). In thisregard, patients with HOMA-IR $. showed alteredlower airway resistance, with lower values of FEV andmaximum midexpiratory flow in comparison withwomen with lower IR. In addition, IR contributedindependently to FEV, maximum midexpiratoryflow, and FVC variance among all subjects (). Inanother study, nonsmoking subjects (.% fe-males) without lung disease were divided accordingto HOMA-IR $.. A logistic regression analysisrevealed an increase in the likelihood of having IR of. times with every one-point decrease in the percentFEV/FVC ratio for predicted values ().

More recently, among a total of Hispanic andAfrican American adolescents aged to years,those with HOMA-IR in the top quartile had a lowerpercent FEV/FVC ratio in addition to a lower re-sidual volume, residual volume/TLC ratio, expiratory

reserve volume, and functional residual capacity ().After adjusting for general and truncal adiposity,HOMA-IR remained a major predictor of expiratoryreserve volume and the FEV/FVC ratio in multi-variate analysis.

Mechanisms linking pulmonary function impair-ment in insulin-resistant populations remain to beelucidated. One of the advocated mechanisms isa reduction in mitochondrial fitness in the skeletalmuscle, provoking an impairment of muscle strength(, ). In this regard, in the Normative Aging Studyof men, the handgrip dynamometry, a measure-ment of skeletal muscle strength, was negatively as-sociated with fasting insulin levels after adjustment forpotential confounders ().

The systemic inflammation mediated by IR mayalso be a contributing factor. In this way, when therelationship between pulmonary function and sys-temic inflammation was assessed among Hispanic andAfrican American obese adolescents with asthma,Rastogi et al. () showed that T helper (Th) cellresponse was associated with IR. In addition, whendata from nonsmoking Korean male subjectswithout a history of malignancy, asthma, chronic lungdisease, pulmonary tuberculosis, liver disease, or severe

Figure 2. Well-recognized pathogenic mechanisms underlying pulmonary dysfunction in T2D. ASM, airway smooth muscle; NE,neutrophil elastase.

Insulin resistanceLeptin resistance and

leptin-induced inflammationLung microangiopathy

Autonomic

neuropathy

Low-grade chronic

inflammation

Increased ASM

responsiveness

Systemic

inflammation

Impaired

muscle strength Focal lung inflammationSurfactant defects Increased

parasympathetic

tone and NE activity

Type 2 Diabetes

© 2017 Endocrine Reviews ENDOCRINE SOCIETY


REVIEWD

ownloaded from



http://dx.doi.org/10.1210/er.2017-00173


cardiovascular disease were divided into quartilesbased on FVC or FEV (percent predicted), HOMA-IR significantly increased as the FVC or FEV (percentpredicted) decreased, whereas the individuals in thelowest FVC or FEV quartile had the highest highsensitive C-reactive protein (CRP) level (). Stepwisemultiple logistic regression analyses showed thathigher high-sensitive CRP and HOMA-IR, togetherwith abdominal obesity, were independent predictorsof the lowest FVC and FEV (percent predicted)quartiles.

The relationship between IR and inflammation alsooffers a biological explanation for its link with incidentasthma-like symptoms among adults (, –). Theconsequences of hyperinsulinemia include the ability toincrease the proliferation of primary human airwaysmooth muscle cells and also its hyperresponsiveness andcontractility upon insulin exposure (, ). Additionally,the administration of intranasal insulin in mice increasedcollagen deposition in the lungs and positively regulatedthe activation of b-catenin, a positive regulator ofepithelial–mesenchymal transition and fibrosis, throughthe mediation of phosphatidylinositol -kinase/Akt ().In a prospective study with a total of subjects fromthe general population in Denmark who participated ina health examination between and , baseline IRwas associated with an increased risk of developingwheezing and asthma-like symptoms after a -yearfollow-up period (). This effect was independent of sexand stronger than that of obesity. Similarly, in a repre-sentative sample of adolescents from the United Stateswith and without asthma in the to NationalHealth and Nutrition Examination Survey, IR wasnegatively associated with FEV and FVC ().However, this finding was notable only in over-weight or obese adolescents and was more pro-nounced for FEV among adolescents with asthma.For example, an increment in HOMA-IR from theth percentile to the th percentile was associatedwith ~-mL lower FEV and ~-mL lower FVC.Similarly, in a retrospective study performed with, Korean subjects who visited for a routinehealth checkup, HOMA-IR showed a noteworthycorrelation with FEV in men, and in binary logisticregression analysis, HOMA-IR was independentlyassociated with airway hyperresponsiveness in bothsexes ().

A final explanation accounting for the deleteriouseffects of IR on lung function may be related to insulinreceptors that have been isolated in type II alveolar(AT-II) epithelial cells (, ). These receptors play animportant role by mediating the cellular uptake ofglucose, which is a major substrate for the biosynthesisof surfactant phospholipids (). Recently, in a pop-ulation-based random sample, Fernandez-Real et al.() described how increased circulating levels ofsurfactant protein A, the major surfactant-associatedprotein, were associated with altered glucose tolerance

and IR. Therefore, surfactant defects in insulin re-sistant subjects may also lead to an increase in airwayresistance. Altogether, the metabolic pathways relatedto IR could be crucial in initiating lung abnormalitiesin patients with TD and point to the lung as a newcomponent of the metabolic syndrome.

Leptin resistance and leptin-induced inflammationLeptin is a proinflammatory cytokine derived fromadipose tissue that mediates the balance between foodintake and energy expenditure (). However, leptin isalso involved in fetal lung development, protects AT-IIepithelial cells from hypoxia-induced apoptosis, andincreases surfactant protein B in animal models (,). In human lung, leptin is produced in bronchialepithelial cells and alveolar macrophages (), andfully functional receptors for leptin have been iden-tified in bronchial and alveolar epithelial cells andbronchial smooth muscle cells (, ). Therefore,leptin may play a role in the regulation of airwaydiameter and the pathogenesis of respiratory diseases(, ). In addition, leptin also leads to bronchodi-lation by signaling through its receptor in the braincholinergic neurons, which decreases parasympathetictone in smooth muscle cells ().

The presumed negative effect of leptin on airwaycaliber is mediated in two different ways: () its in-flammatory role inducing focal epithelial injury, smallairway disappearance, and centrilobular emphysema,and () the development of leptin resistance, leading toincreased parasympathetic tone, which in turn causesbronchoconstriction (). Leptin resistance has alsobeen related to an imbalance between alpha- anti-trypsin and the neutrophil elastase activity, resulting inan overactivity of neutrophil elastase that may degradelung tissue proteins, thus leading to pulmonary dis-orders (). Both conditions, inflammation and leptinresistance, are common disturbances in patients withTD ().

In cross-sectional studies, serum leptin was relatedto impaired lung function. Data from a representativesample of nonobese subjects participating in theThird National Health, Nutrition, and ExaminationSurvey showed that, independently of other risk fac-tors, circulating leptin levels were inversely related toFEV (). Additionally, there was a noteworthy as-sociation between serum leptin and inflammatorymarkers such as CRP, leukocytes, and fibrinogen ().Similarly, in African American men and womenin the Jackson Heart Study, serum leptin was stronglyinversely and independently associated with FEV andFVC (percent predicted) values, especially in men andobese women (). Moreover, in a group of healthy -year-old children, an association betweenhigher leptin concentrations and lower lung function,FEV, and FVC was shown, independent of adiposity(). Finally, in a nested case-control study of rescueworkers during the terrorist attacks on September


REVIEWD

ownloaded from



in New York City, elevated leptin was associatedwith more than twofold increased chances of abnor-mal FEV after adjustment for BMI ().

More related with leptin resistance, a major negativecorrelation between leptin and FEV and MVV wasobserved in a small group of obese children and ad-olescents, reinforcing the idea that leptin status is a keyfactor in developing a restrictive lung pattern (). It isworth mentioning that in a subset of subjects aged to years from the Global Asthma and AllergyNetwork of Excellence clinical follow-up survey, serumleptin was positively associated with the severity ofasthma, especially in females (). In contrast, themagnitude of the reduction in leptin levels was a pre-dictor of improvements in FVC, FEV, and PEF after year of interdisciplinary intervention to achieve weightloss in postpuberal obese adolescents ().

Nevertheless, to the best of our knowledge, thereare no studies assessing serum leptin concentrationsand pulmonary function in subjects with TD thatcould unequivocally confirm this association.

Low-grade chronic inflammation stateLow-grade chronic inflammation is inherent in TDpathophysiology (). Therefore, persistent inadequateglucose control over time may alter the regulation ofinflammatory pathways that are involved in the im-pairment of pulmonary function (–). In thisregard, the relevance of tumor necrosis factor (TNF)-ain the inflammatory processes of the lung has beendemonstrated in a mouse model of acute lung in-flammation (). In this model, pharmacologic (eta-nercept) and genetic TNF-a inhibition (TNF-areceptor knockout) significantly reduced inflam-matory cell infiltration, proapoptotic pathway activa-tion, and tight junction abnormalities in the lung ().In humans, the administration of etanercept in pa-tients with refractory asthma significantly improvedtheir respiratory function and, in particular, post-bronchodilator FEV when compared with placebo(). Moreover, in obese nonsmoking women withoutprior respiratory disease, the soluble TNF-a receptor (TNF-R) contributed independently to FEV andFVC impairments (). It should be noted that TNF-R is mainly expressed on epithelial lung cells, whereasTNF-R is primarily found on the surface of cells ofmyeloid origin ().

Cross-sectional studies also indicate that systemicinflammation may be linked to early perturbations ofpulmonary function. Among , subjects withoutknown pulmonary disease, there was a strong inverseassociation between CRP levels and quartiles of FEVthat remained significantly high after adjustment forage, sex, components of metabolic syndrome, andfitness level (). More recently, the British RegionalHeart Study of subjects aged to years andwith no history of cardiovascular disease or dia-betes demonstrated considerable inverse associations

of baseline FVC, FEV, and the FEV/FVC ratio withblood glucose levels and markers of inflammation(CRP, interleukin-, and leukocyte count), even afteradjustment for a wide range of variables including age,BMI, and smoking status (). Finally, among a total of subjects with TD from Colombia, those withHbAc .% (n = ) had significantly lower residualvalues (observed minus predicted) for FVC and FEVin comparison with patients with adequate metaboliccontrol (). These differences were not explained bythe usual determinants of lung function such as age,height, sex, cigarette smoking, or other lifetime ex-posures. More interestingly, there was a trend towardlower lung function, especially for FVC, and increasedlevels of inflammatory mediators, especially CRP, asHbAc levels increased (). These results reinforce thepotential association between inflammatory markersand decreased lung function in patients with TD.

It should be noted that although significantly reduced,the lung parameters in the previously-mentioned studiesgenerally remained within the normal range.

Microvascular damage and lung microangiopathyThe lung alveolar-capillary network constitutes thelargest microvascular bed in humans, making it ex-tremely susceptible to be a target for TD-associatedmicroangiopathy (). However, it has been suggestedthat owing to its large reserves, symptoms and dis-ability occur in the lung later than in smaller mi-crovasculature of the kidney or retina, despite a similarseverity of anatomic involvement (). Hence, diabeticlung microangiopathy causes pulmonary dysfunctionsthat may remain undiagnosed for a prolonged period.

The loss of lung microvascular reserve can bequantified noninvasively from DLCO and its ratio tothe alveolar ventilation. In fact, lung volume andDLCO can decline by ~% without dyspnea at rest. Itshould be noted that, although subclinical, criticaldecrements in pulmonary DLCO have been reportedin patients with TD (, , ). More importantly,the reduction of DLCO significantly correlated withthe degree of diabetic retinopathy and the -hourprotein excretion rate, suggesting that the processesrelated to microangiopathy in TD are also involved inlung damage (, ).

The clinical studies in this field have been per-formed with limited sample size. However, consistentresults have been reported showing a decreased DLCOwhen adjusted to alveolar volume in patients withdiabetes, in particular in those with diabetic reti-nopathy or microalbuminuria (, , , ). Re-cently, the Hispanic Community Health Study/Studyof Latinos comprising , subjects showed thatpulmonary function impairment mirrored kidneymicroangiopathy, as those with TD and macro-albuminuria showed significantly lower lung function(FVC and FEV) and higher dyspnea score than thosewithout TD and with normoalbuminuria ().

“The clinical studies in this fieldhave been performed withlimited sample size.”


REVIEWD

ownloaded from



http://dx.doi.org/10.1210/er.2017-00173


The permeability of lung epithelium was also assessedin nonsmoking patients with TD with no history ofcardiorespiratory disease, aged to years, who un-derwent a technetium m-diethyltriaminepentaaceticacidaerosol scintigraphy: % of them showed a delayedmTc-diethyltriaminepentaaceticacid clearance, mainlythose with late complications and diabetes duration. years ().

Physiologically, one part of pulmonary microcir-culation is inactive. However, closed vessels aregradually recruited in case of higher demand foroxygen (e.g., during physical exercise) that normallyleads to an increase of DLCO (). In a cross-sectionalstudy involving nonsmoking patients with TDwithout overt cardiopulmonary disease and healthynonsmoking subjects without diabetes, Weir et al. ()quantified pulmonary microvascular reserves. Patientswith TD showed a % to % lower pulmonaryblood flow, pulmonary capillary blood volume, andDLCO corrected by pulmonary blood flow at exercisethan control subjects, with diminished recruitmentof alveolar capillaries. Lung microvascular indexeswere significantly related to HbAc, extrapulmonarymicroangiopathy, including retinopathy, abnor-malities in both motor and sensory nerves, andmicroalbuminuria (). More recently, similar resultshave been observed using a postural variation of DLCOas an early marker of impaired alveolar-capillary re-cruitment ().

Autonomic neuropathyAlthough more involved in sleep-breathing disordersand pulmonary dysfunction in type diabetes (TD),diabetic autonomic neuropathy (DAN) has also beenassociated with an increased prevalence of lungfunction impairment in TD (, ). DAN may affectup to % of the population with diabetes, but there islittle information regarding its involvement in theabnormalities of pulmonary function.When the kinetics ofnorepinephrine were studied in patients with DAN, thelung clearance of iodine--meta-iodobenzylguanidinewas significantly decreased compared with patientswithout sympathetic nervous dysfunction ().

In patients with DAN, both abnormalities in thebronchomotor tone and specific airway responsivenessto different stimuli have been described, indicatinga defective control of mechanisms regulating bronchialcaliber (). In this regard, in five patients with severesymptomatic DAN, bronchial provocation testing withcold air failed to show a noteworthy reduction inspecific airways conductance in comparison with fivepatients without DAN (). Similarly, a group of subjects with DAN presented a significantly raisedcough reflex threshold in contrast to nonneuropathicpatients with TD, supporting the vagal denervation ofthe respiratory tract (). Another group of six patientswith DAN showed an increased bronchial reactivityafter the nebulization of histamine, reducing their

FEV by at least % compared with a matchedcontrol group of six subjects with no evidence of DAN(). Moreover, patients with DAN presenteda significantly lower capacity of bronchodilation afterinhalation of the atropine-like drug ipratropiumbromide than a control group of patients withoutDAN ().

Peculiar changes in breathing pattern and greaterventilatory requirements have been observed duringincremental exercise in subjects with DAN, alsosuggesting an altered control of breathing in stressfulconditions (). For example, subjects with DANresponded to heavy exercise by excessively increasingtheir respiratory rate and alveolar ventilation incomparison with patients without DAN ().Likewise, autonomic neuropathy involving re-spiratory muscles may also impair thoracic mobility(). Subsequently, patients with DAN showedreduced respiratory muscle strength (measured asmaximum static inspiratory pressure) than controls without DAN, with a negative correlationbetween muscle strength and both resting heart rateand diabetes duration (). Finally, when assessedby the neuropathy disability score, diabetic poly-neuropathy was associated with substantial respi-ratory muscle impairment in males with TDwithout pulmonary disease (, ). Altogether,reduced airway vagal tone seems to play a nontrivialrole in the relationship between TD and impairedlung function.

Other potential pathogenic mechanisms

Decreased muscle strengthDecreases in muscle mass and functional muscle ca-pacity, two key factors in pulmonary function, havebeen reported in TD (). In fact, the prevalence ofinspiratory muscle weakness in a sample of pa-tients with TD from Brazil was % (). Data from subjects from the cross-sectional Normative AgingStudy showed that handgrip strength was negativelyassociated with fasting insulin after adjustment forpotential confounders including age, BMI, and usualphysical activity level (). More recently, a cleardeterioration of lean mass and muscle functions wasobserved among adult patients with TD of up to years old (). Impairment in muscular strengthmay be due to lower levels of physical activity, buthyperglycemia and IR can also impair muscle con-tractile function and muscle force generation (, ).In addition, patients with TD also present lower levelsof plasma carnitine (an amino acid stored in skeletalmuscles) than healthy controls, whereas carnitinelevels in patients with TD positively correlated withFVC, FEV, FEV/FVC, and inspiratory musclestrength (). Therefore, it seems reasonable topostulate that this reduced functional capacity ofskeletal muscle, particularly the diaphragm but also the


REVIEWD

ownloaded from



accessory skeletal muscles of respiration, can havenegative consequences for lung function ().

Muscle strength and muscle endurance were ex-plored through the maximum static inspiratorypressure and MVV in nonsmoking patients withTD without lung disease and compared with that of healthy subjects (). Muscle strength was sig-nificantly reduced and related to lung volumes andHbAc level, whereas impaired endurance of re-spiratory muscles prevailed in patients with micro-vascular complications (). More recently, in theBerlin Aging Study II involving subjects witha mean age of 6 years, muscle mass and ab-dominal obesity proved to be the most importantfactors influencing lung function in TD, and themeasurement of grip strength was recommended fora better interpretation of pulmonary function tests inthis population (). Finally, weeks of inspiratorymuscle training reversed the loss of inspiratory musclestrength in a small group of subjects with TD(). However, this improvement in inspiratorymuscle strength was not accompanied by changes inFEV, FVC, or MVV ().

Nonenzymatic glycosylation of lung proteinsIt has been suggested that nonenzymatic glycosylationof pulmonary connective tissue proteins, especiallycollagen but also elastin and fibronectin, may favor theobserved restrictive pattern described in patients withTD as a consequence of a less compliant pulmonaryparenchyma (, ). It should be noted that gly-cosylated collagen has a low physiological turnoverrate, more stiffness, and more resistance to digestionby collagenase and pepsin than nondiabetic collagen(). Therefore, the deleterious effect on collagenstructure in pulmonary parenchyma and cartilages ofthe chest wall will limit chest mobility (, ). Ina study involving patients with insulin-dependentdiabetes mellitus, those with limited joint mobilityshowed a notable decrease in TLC, FVC, and FEVcompared with those without joint limitation, sug-gesting that altered respiratory mechanics were an-other manifestation of a generalized disturbance incollagen metabolism (). It is also important to notethat AT-II epithelial cells specifically express themRNA of the receptors for advanced glycation endproducts (RAGE) (). The AGE–RAGE ligationactivates pathophysiological cascades, leading to pul-monary endothelial cell dysfunction, proinflammatoryeffects, and cell apoptosis (, ). In this regard, itshould be noted that RAGE plays a role in asthmairrespective of the identity of the allergens involved inanimal models ().

Defects in the bronchiolar surfactant layer and thedeficit in glucagon-like peptide 1 concentrationsDefects in the bronchiolar surfactant layer that isinvolved in maintaining airway stability and diameter

may likewise be considered a contributing factor to theimpairment of airway caliber regulation in patientswith TD. When the alveolocapillary barrier isdamaged, surfactant proteins escape from the alveolarspace into the vascular compartment. Therefore, se-rum concentration of surfactant protein D has beenevaluated in a large variety of pulmonary diseases andproposed as a potential systemic biomarker for lungdiseases (, ). In addition, it is worth mentioningthat elevated serum levels of surfactant protein Dhave also been found in nonpulmonary diseases thatadversely affect the lungs as part of their systemicrepercussion (, ). A population-based randomsample study described how increased circulatinglevels of surfactant protein A, the major surfactant-associated protein, were associated with alteredglucose tolerance and IR (). Therefore, we suggestthat serum surfactant proteins should be contem-plated as a serum biomarker of lung impairment, asa first step toward identifying those patients withTD for whom a respiratory function study would berecommendable.

Glucagon-like peptide (GLP-) receptor has beenfound abundantly in the lung, where it seems to beimplicated in the regulation of the lipid fraction ofsurfactants (). In addition, experimental studieshave shown that GLP- plays a role in the stimulationof surfactant production by pneumocytes (, ).In fact, in a rat model of lung hypoplasia, surfactantproduction was improved by the administration of theGLP- receptor agonist liraglutide (). In addition,in a rat model with streptozotocin (STZ)–induceddiabetes, reduced lung levels of surfactant A and Bwere restored after this treatment was administered(). Therefore, the underlying deficit in GLP-concentrations that exists in TD could also en-hance the airway resistance observed in these pa-tients. Furthermore, treatment with GLP- analogscould result in an improvement in ventilatorypatterns. A clinical trial specifically designed toaddress this question is ongoing (ClinicalTrials.gov:NCT).

The Histopathological Evidence

The histopathological changes induced by thepathophysiological mechanisms previously describedinclude the thickening of the alveolar epithelia andpulmonary capillary basal lamina (BL), the reductionof the alveolar space, a higher degree of fibrosis, thepresence of centrolobular emphysema and pulmo-nary microangiopathy, and modifications in mucussecretion (, ). These abnormalities are ho-mogeneous throughout the whole lung parenchyma.Findings reported from experimental models will beclearly differentiated from those reported in humans(Fig. ).

“As amylin receptors havebeen detected in lungmembrane, these datasuggest a physiological rolefor this molecule in lungchanges found in T2D.”


REVIEWD

ownloaded from



http://dx.doi.org/10.1210/er.2017-00173


Experimental evidence: animal experiments

Thickness of basement membranes and expansionof interstitiumExperimental studies on mice and hamsters in whichdiabetes was induced by STZ injection have shownthat hyperglycemia affects the structure and functionalproperties of the alveolar capillary endothelial cells(). Compared with age-matched animals, diabeticanimals exhibited a thickened BL provided with focalnodules, similar in density and distribution to thosefound in the glomerular BL in diabetes, with ~% oflung capillaries showing a narrowed or collapsed lu-men (, ). Moreover, reduced capillary bloodvolume, impaired alveolar microvascular perfusion,and reduced alveolar–capillary recruitment have beenobserved in model rats with TD (). A reducedbreakdown of the connective tissue proteins con-tributing to the increased thickness of both the alveolarand endothelial BL has been found in STZ rats withdiabetes in comparison with nondiabetic rats ().The interstitial space between the alveolar and theendothelial BL is also expanded in animal modelsexposed to chronic hyperglycemia. This is due to anincrease of matrix proteins, mainly type I and IIIcollagen and elastin fibers, whereas a simultaneousreduction of sulfate proteoglycan concentrations oc-curs (). The nonenzymatic glycation of collagenincreases matrix stiffness, challenges its digestion byproteases, and slows down its turnover (, ). Inaddition, AGE may also upregulate the tissue ex-pression of profibroting cytokines (, ). In thisregard, the abnormal connective tissue metabolism aswell as collagen crosslinking in thoracic and lung tissuemight limit chest mobility and lung function ().This expansion of the interstitium results in fibrosisand induration of the lung, which in turn leads toa reduction in lung volume and its compliance in theseanimals. In addition, emphysema has been observed in

lungs from rabbits with alloxan-induced diabetes().

Changes in alveolar epithelial cells andalveolus morphologyNo alterations related to chronic hyperglycemia havebeen reported in the structure of the alveolar epithelialtype I cell. These are large and squamous cells re-sponsible for gas exchange and ion transport andconstitute % of the alveolar epithelial layer. How-ever, an altered morphology of the smaller cuboidalAT-II cells, which are involved in surfactant pro-duction, has been observed in the Zucker diabetic fattyrat, an animal model that reproduces TD (). Themost prevalent abnormalities described in AT-II aredilated endoplasmic reticulum, an age-related increasein the volume of lamellar bodies, and impairedsurfactant production (). The altered surfactantproduction, in combination with the increased ex-tracellular matrix, may induce a partial alveolus col-lapse and, consequently, a reduction of the alveolarspace. In fact, microscopic observation of lungs fromSTZ rats showed smaller alveoli, but increased innumber in comparison with nondiabetic lung samples().

In addition, structural alterations have also beendescribed in lung endothelial cells, including an ex-panded endoplasmic reticulum, more multivesicularbodies and lysosomes, and the presence of fusedplasmalemmal vesicles ().

Changes in muscle and airway mucus secretionThe alterations detected in the respiratory musclesconsist of structural changes with muscle atrophy,augmented lipid deposition, decreased mitochondria,as well as muscle fiber transformation (). Thesechanges are related to a reduced cellular glucose up-take and fatty acid oxidation because of IR.

Three studies have shown that amylin, a pancreatichormone cosecreted with insulin, could stimulatemucus secretion at the submucosal side of isolated rattrachea (–). As amylin receptors have beendetected in lung membrane, these data suggesta physiological role for this molecule in lung changesfound in TD, in which hyperinsulinemia is the rule.However, so far, no studies have been conducted indiabetic models with TD.

Human evidence: postmortem examinations

Thickness of basement membranes and expansionof interstitiumLong-standing TD is characterized by widespreadalterations to the BL. An early report based on autopsyobservations from patients with TD confirmed thepresence of microangiopathy in the lung, involvingalveolar septa and pleura, and appearing not only asa thickening of the basal membrane, but also as

Figure 3. Human lung histopathological evidence inducedby T2D.

Increasedmucus

secretion

Thickness ofbasement

membranes

Expansionof interstitium

Increasedinterstitial

matrix proteins

Structuralchanges inrespiratorymuscles Altered

alveolartype II cells

Nodularfibrosis

Microangiopathy

© 2017 Endocrine ReviewsENDOCRINE SOCIETY


REVIEWD

ownloaded from



hyalinosis, plasmorrhagia, and insudation (). In-terestingly, similar to the retina, vascular leakage alsooccurs in the lung of patients with TD. In this regard,perfusion chest computed tomographies have shownincreased vascular permeability in TD (). Simi-larly, the epithelial and capillary BL of alveoli weresignificantly thicker in patients with diabetes thanthey were in age-matched control subjects (), with astrong correlation with the thickness of the BL of renaltubules and muscle capillaries, suggesting a similarpathologic mechanism (, ). Furthermore, ina histopathological study of the lungs of autopsiedpatients with TD and controls without diabetes,the thickness of alveolar capillary walls, pulmonaryarteriolar walls, and alveolar walls had significantlyincreased in the former group (). From electronmicrograph images obtained from autopsy mate-rial, the BL separating the capillary from the al-veolar space in six patients with TD was ~%thicker compared with six controls without di-abetes ( 6 vs 6 nm; P , .) ().Therefore, the thickening of the capillary BL andthe accumulation of connective tissue in the in-terstitial space could induce a narrowing of thecapillary wall and a partial collapse of capillarylumen (, ).

Overall, the thickening of the walls of pulmo-nary alveoli caused by increased amounts of inter-stitial matrix proteins and the thickening of the basalmembrane of alveoli leads to a decrease in pulmonaryparenchyma elasticity (, , ).

Changes in alveolar epithelial cells andalveolus morphologyWhen histopathological changes in the lung obtainedfrom the autopsies of patients with TD werecompared with findings in the lung of subjectswithout diabetes, the former showed a major increasein the amount of collagen in the alveolar walls, themain airways, and the walls of the great vessels and alsopulmonary hemorrhages (). In addition, a specifictype of fibrosis, a nodular deposition of collagen in theinterstitium of the alveolar walls, was described in thelung of subjects with TD ().

The Paradigm of CF–Related Diabetes

CF is a systemic disease involving pulmonary functionand abnormalities of glucose tolerance and diabetes,which allows us to investigate the deleterious impact ofchronic hyperglycemia on lung function.

CF and CF-Related DiabetesCF is the most common autosomal-recessive condi-tion in the white population, affecting ~ out of every live births (). CF is caused by defects in anepithelial chloride channel, the CF transmembrane

regulator, which is expressed across tissues, includingairways, intestine, and pancreas (, ). The in-cidence of diabetes in patients with CF is ~ timesgreater than in the general population and is considereda clinical entity distinct from that of TD and TD(). CF-related diabetes (CFRD) may reach a preva-lence of up to % in adults with CF, with a higherincidence in females. It is associated with impaired lungfunction and an increased chance of treatment failure inpulmonary exacerbations ().

The hallmark of CFRD is an early insulin de-ficiency due to reduced beta cell mass (). However,the destruction of beta cells is incomplete, and othermechanisms such as IR, genetic predisposition, theimpairment of the enteroinsular axis, inflammation,and increased oxidative stress have also been describedas playing a contributory role (–). IR is usuallymild and related to infection status and steroid therapy().

Using longitudinal data from to in individuals with diabetes from the U.K. Cystic FibrosisRegistry, an HbAc value of$.% was associated witha threefold increased risk of death, independently ofother potential confounders ().

The influence of glucose intolerance and diabeteson pulmonary function in CFIn a cross-sectional analysis of patients with CFenrolled in the European Epidemiologic Registry ofCystic Fibrosis, the predicted percentage of FEV waslower in people with diabetes than in those withoutdiabetes across all ages (). Thus, the mean value ofpredicted percentage of FEV in the group of yearsor older was % in the patients without diabetes, ascompared with % in the group with diabetes. Inaddition, the predicted FVC value was also lower inpeople with CFRD than in those without diabetes in allage groups. Similarly, according to data from the U.K.Cystic Fibrosis Database, CFRD in female subjectswithout chronic Pseudomona aeruginosa infection wasassociated with a % lower predicted FEV comparedwith subjects with CF with normal glucose tolerance(NGT) ().

According to recent data in individuals with CF,the decline in FEV is driven primarily by the episodicpulmonary exacerbations, and only % of them fullyrecover their -month baseline lung function afterintravenous antibiotic treatment (). In addition, thedecline in lung function is accelerated after the de-velopment of CFRD. In this regard, in a prospectivemulticenter study following patients with CF for years, both FEV and FVC declined significantly inthe CFRD, whereas patients without diabetes did notshow any notable change ().

It is worth mentioning that pulmonary functionbegins to decline as early as to years prior to thediagnosis of CFRD, implying that impaired glucosetolerance (IGT) may also be detrimental. Lavie et al.


REVIEWD

ownloaded from



http://dx.doi.org/10.1210/er.2017-00173


() showed that patients with CF with IGT pre-sented a lower mean predicted FEV than patientswith NGT ( 6 vs 6 %; P , .).

In a prospective study enrolling patients withCF, the rate of lung function decline over a -yearobservation period was directly related to the baselineoral glucose tolerance test (OGTT) classification:patients with IGT had a greater loss of lung functionthan those with NGT, and those with CFRD withoutfasting hyperglycemia had the greatest loss (). Thefunctional pulmonary parameters (FEV and FVC)and the BMI at baseline were comparable, and theresults were independent of sex, age, BMI, respiratorymicrobiology, and use of corticosteroids. In addition,patients in the lowest quartile for insulin productionat baseline experienced the highest rates of pulmo-nary function decline over time, suggesting a linearinverse relationship between insulin production andlung deterioration (). Similarly, in a retrospectivestudy involving patients with CFRD and pa-tients with CF without diabetes, matched by age, sex,and chronic P. aeruginosa lung infection, statisticallyimportant differences in FEV and FVC betweengroups emerged . and years prior to the di-agnosis of diabetes, respectively (). More recently,a retrospective study comprising patients with CFshowed marked decreases in FVC, FEV, and forcedexpiratory flow at % of expired FVC in CFRD withthe onset of year prior to the diagnosis of diabetes().

Similar results appeared when other measures ofglucose intolerance apart from OGTT were used inpatients with CF without CFRD. In this regard, pa-tients with normal OGTT but with a maximum con-tinuous glucose monitoring value $ mmol/Lexhibited a considerable impairment in lung function(FEV: .6 .% vs .6 %, P = .; FVC: .6 .% vs . 6 .%, P = .) independent ofage, age at diagnosis, sex, genotype, exocrine pancreaticinsufficiency, BMI, and P. aeruginosa colonization ().

Why does CFRD accelerate lung dysfunction?The mechanisms leading to impaired pulmonaryfunction in CFRD are unclear, but the combinedeffects of hyperglycemia and insulin deficiency wouldseem to be the main contributing factor. Lungfunction in subjects with CF is critically dependenton maintaining normal weight and lean body mass.Therefore, insulin deficiency might affect pulmonaryfunction by creating a catabolic state with excessiveprotein and fat breakdown (, ). In contrast,hyperglycemia could cause a decline in pulmonaryfunction directly by reducing the defense capacityagainst infection (). In this way, patients with TDshow a significant decrease in the percentage ofactivated macrophages when compared with subjectswithout diabetes that is partially normalized aftermetabolic improvement (, ). In addition, when

blood glucose levels are modestly elevated (.mg/dL or . mmol/L), the presence of glucose in airwayfluid contributes to the progression of respiratorydisease (). Furthermore, hyperglycemia has beennoted to be associated with increased oxidative stressin CF, contributing to lung disease by creatinga proinflammatory environment in the airways (,). The Th pathway and the production ofinterleukin- could provide a better understandingof the relation between beta-cell dysfunction andlung impairment in CF. In fact, Th has been in-volved in beta-cell destruction occurring in TD,TD, and, interestingly, in lung inflammation as well(, ).

Two mechanisms previously related to the negativeimpact of TD on lung function could also affect theprogression toward CFRD with CF, namely AGEs andthe decreased levels of GLP-. Plasma AGE andneutrophil-derived human SA levels are signif-icantly elevated in CFRD and correlate negatively withFEV (). After a standardized breakfast, patientswith CF and CFRD were found to have significantlydecreased GLP- compared with healthy controls(). Finally, based on chest computed tomography,CFRD has also been associated with structural lungdisease that occurs prior to changes in lung functionsuch as FEV ().

Insulin treatment and lung functionTo date, there is no convincing proof that long-acting insulins, short-acting insulins, insulin ana-logs, or oral antidiabetic agents have a distinctiveadvantage over one another in controlling hyper-glycemia or other clinical outcomes associated withCFRD (). However, the Cystic Fibrosis Foun-dation clinical practice guidelines support the ad-ministration of insulin as the mainstay therapy,and this remains the most widely used treatmentmethod ().

Some uncontrolled trials suggest that a once-dailyinjection of intermediate or long-acting insulin im-proves weight and lung function. Following years ofinsulin administration in patients with CF olderthan years with abnormal OGTT in spite of normalfasting glycemia, FEV stabilized in insulin-treatedinsulinopenic subjects (. 6 .% vs . 6.%), but decreased in NGT patients who remainedwithout insulin treatment (. 6 .% vs . 6.%; P = .) (). However, two trials with datafor the comparison of insulin vs placebo in CFRDdid not report any critical differences between thegroups regarding the primary outcomes of bloodglucose level, lung function, and nutritional status(, ). Whether insulin may be of greaterbenefit to respiratory function when given prior tothe diagnosis of CFRD, after which structural lungdisease may be irreversible, is a subject for futureinvestigation ().


REVIEWD

ownloaded from



The Deleterious Effect of T2D on Sleep-Breathing Disorders and Sleep Architecture

In recent years, increasing evidence has alsoappeared suggesting an association between sleepapnea syndrome (SAS) and TD (). The availabledata suggest that long-term exposure to intermittenthypoxemia and sleep fragmentation increasessympathetic nerve activity that favors the disordersof glucose metabolism (). Alternatively, somestudies propose the idea that TD might contributeto the development of sleep-breathing disorders(Fig. ).

Nighttime awakenings, sleep quality, and daytimesleepiness in T2DDisruptions of sleep occur more frequently in patientswith TD than in the population without diabetes.Depending on age, diabetes control and duration,diabetes treatment, and associated complications andcomorbidities, nocturnal awakenings in TD mightbe due to nocturia, thirst, symptomatic nocturnalhypoglycemia, pain due to peripheral neuropathy,associated restless legs syndrome, or the need forself-monitoring of blood glucose during the night(–). These frequent nighttime awakeningscontribute to sleep quality reduction, and according tothe Pittsburgh Sleep Quality Index, almost two out ofevery three patients with TD are classified as poorsleepers (Pittsburgh Sleep Quality Index globalscore .) (). In addition, sleep quality reductionhas been found to be associated with basal hyper-glycemia but not with HbAc, thus strengthening thepotential relationship between nocturnal hyperglyce-mia and sleep quality ().

TD is also related to excessive daytime sleepinessindependently of other classic risk factors such asSAS (, ). In consecutive, obese subjectswaiting for bariatric surgery, the Epworth SleepinessScale global score was higher in men, older patients,and those with TD, whereas no relationship be-tween daytime somnolence and polysomnographymeasurements, including the Apnea-Hypopnea In-dex (AHI), were reported (). Similarly, in a case-control study involving individuals, the exces-sive daytime sleepiness (Epworth Sleepiness Scalescore .) was almost double among patients withTD in comparison with control subjects (.% vs.%), and those with HbAc$.% and FPG$.mmol/L showed a higher risk of excessive daytimesleepiness than subjects with HbAc ,.% [oddsratio . (% confidence interval .–.); P =.] ().

Pathophysiological mechanisms underlyingsleep-breathing disorders in T2DBreathing control is critical for respiratory responsesto changes in the internal and external environment,including but not limited to sleep, physical activity,stress, acute and chronic diseases, and other condi-tions (, ). In adults with TD, previous re-search has demonstrated abnormal and bluntedventilatory responses to isocapnic hypoxia with analtered breathing pattern, whereas the ventilatoryresponse to hypercapnia is well preserved (Table )(, ). In fact, although the respiratory minutevolume was similar between groups, the increase inthe amount of air that was breathed during minutein response to hypoxia was almost four times lessin patients with TD in comparison with control

Higher index of microarousalsand sleep fragmentation Changes in

sleep architecture

Excessive daytime sleepinessand poor sleep quality

Nocturnal hypoxia

ApneaHypopneaT2D

Control ApneaHypopnea

Stage 2

Stage 1

REM

SWSstage

For the same apnea-hypopnea index

Type 2 Diabetes

Non-REM sleepREM sleepO2

O2

© 2017 Endocrine ReviewsENDOCRINE SOCIETY

Figure 4. The deleteriouseffect of T2D on sleep-breathing disorders andsleep architecture. SWS,slow-wave sleep. EEGarousal image reproducedfrom Sleep Heart HealthStudy, courtesy of NationalSleep Research Resource,https://sleepdata.org/datasets/shhs/pages/mop/6-623-mop-eeg-arousal.md.


REVIEWD

ownloaded from



https://sleepdata.org/datasets/shhs/pages/mop/6-623-mop-eeg-arousal.md



http://dx.doi.org/10.1210/er.2017-00173


subjects (). Therefore, patients with TD seem tobe unable to respond in a homeostatic mannerto hypoxic challenge. The main pathophysiologicalmechanisms involved in the sleep-breathing disor-ders that occur in patients with TD are summarizedin Table .

Although the underlying pathophysiology re-mains poorly understood, the findings suggest thatthe chemoreflex, an integrated response involvingchemoreceptor detection, the activity of respiratorycenters in the medulla, afferent and efferent neuralconduction, and respiratory muscle function arealtered in TD, which may impact the regula-tion of blood gases and oxygen delivery withimportant clinical implications (–). Othernonchemoreflex influences may also contributeto reduced ventilatory responses to hypoxia inTD. Specifically, DAN can lead to alterations inboth afferent and efferent motor pathway func-tions (, ). Consequently, impaired respiratorymuscle function with decreased respiratory musclestrength, impaired chest wall function, and/or air-flow limitation may result in a blunted ventilatoryresponse to hypoxemia in TD. Among the humoralventilatory control mechanisms, leptin plays animportant role in the respiratory system, rangingfrom its actions in the lung to its involvement inmany common disorders of the respiratory system(, ). As previously mentioned, the metabolic

pathways related to leptin resistance in diabetescould therefore contribute to deficiencies in centralrespiratory control mechanisms and the defectiveresponses of peripheral and central chemoreceptors(, ). Finally, the breathing pattern generatorand the ventilatory rhythm generator responsiblefor the central control of ventilation could be af-fected in some persons with TD who developvascular lesions and/or brain atrophy (–).

The bidirectional relation between T2D andsleep-breathing disordersAs mentioned previously, TD, breathing, and sleepare closely interrelated. During the last two decades,increasing evidence has appeared suggesting an as-sociation between SAS and TD, two common dis-orders that occur with increased frequency in theobese population (). The available data suggest thatlong-term exposure to intermittent hypoxemia andsleep fragmentation increases sympathetic nerve ac-tivity, contributing to glucose metabolism disorders(). In this regard, a high prevalence of fastinghyperglycemia, IR, and TD has been found amongpatients with SAS in comparison with healthy subjects().

Alternatively, both obesity-hypoventilation syn-drome and SAS are common comorbidities of TDthat have been linked to lung dysfunction and variousadverse metabolic outcomes (, –). Thus,some studies propose that IR and chronic hyper-glycemia might contribute to the development ofSAS. Vgontzas et al. () found that women withpolycystic ovary syndrome, a condition associatedwith IR, were much more likely to have SAS anddaytime sleepiness than controls, suggesting thatIR is a mediator of SAS in humans. Addition-ally, the analysis of cross-sectional data from theSleep Heart Health Study found that subjectswith clinically diagnosed TD had more episodesof periodic breathing, an abnormality of the cen-tral control of ventilation, and more severe sleephypoxemia; after correcting for the main factorsinvolved in the development of SAS, the differencein sleep hypoxemia was eliminated (). Recently,the Sweet Sleep Study has described how thepresence of TD produces a more severe pattern ofsleep breathing characterized by higher rates ofsleep apneas for the same AHI (). This dele-terious effect of TD on sleep breathing leads toexcessive daytime sleepiness, a feature that affectsalmost one-fourth of the general population withTD ().

In addition, the established changes in chemoreflexsensitivity, along with sympathetic activation andhypothalamic-pituitary-adrenal axis hyperactivity,present in TD significantly increase the risk of theexacerbation of other clinical abnormalities in re-sponse to hypoxia/hypoxemia (e.g., in states of obesity,

Table 2. Pathophysiological Mechanisms Contributing tothe Deleterious Effect of T2D on Sleep Breathing

Changes in Chemoreflex Sensitivity

Abnormal threshold in chemoreceptor activation

Afferent and efferent neural condition

Respiratory muscle function

Changes in nonchemoreflex influences

Decreased respiratory muscle strength

DAN

Situations that produce airflow limitation

Humoral ventilatory control mechanisms

IR

Leptin resistance

Changes in breathing pattern and ventilatory rhythmgenerator

Cerebrovascular disease

Brain atrophy

Sympathetic hyperactivity

Hypothalamus pituitary axis hyperactivity


REVIEWD

ownloaded from



sleep-disordered breathing, coronary artery disease,etc.).

Effects of T2D on sleep architectureand microarousalsSleep is an active state with its own characteristic stagesthat recurs throughout the night. The succession ofcycles and the length of each stage and cycle composethe sleep architecture of each person (). Besides thenon–stage-specific disruptions of sleep, stage-specific disturbances can also occur in diabetes.However, the results do not always exhibit the sameconcordance. For example, Palloyva et al. ()reported a significantly increased percentage time inrapid eye movement (REM) sleep in subjects withTD compared with controls without TD. Morerecently, the small reduction in the amount of timethat patients with TD spent in stage of non-REMdisappeared when patients and control subjectswere matched by the AHI (). In a similar way,baseline differences for non-REM sleep stages re-ported between participants with and without TDin the Sleep Heart Health Study disappeared in themultivariate regression analyses when the means forage, sex, BMI, race, and neck circumference wereadjusted ().

In contrast, patients with TD showed higher sleepfragmentation through higher rates of microarousalsduring sleep than control subjects, and these differ-ences were mainly observed during the non-REMsleep (). Microarousals, characterized by centralnervous system reactivity producing electroencepha-lographic activation and sympathetic overactivity,were independently associated with fasting plasmaglucose (). Interestingly, data from the CoronaryArtery Risk Development in Young Adults sleep studyshowed that a % rise in sleep fragmentation wasassociated with increments in fasting insulin andfasting glucose by % and %, respectively ().Therefore, it seems that a relationship between sleepfragmentation and TD exists. In addition, as sleepfragmentation is involved with increased levels oflipids and blood pressure, this enlarged rate in themicroarousal index may also be implicated in theprogression of cardiovascular disease in TD (,).

T2D and nocturnal hypoxemiaTD is an independent risk factor for severenocturnal hypoxemia in obese patients. In fact, theduration of sleeping time with oxygen satura-tion ,% was three- to fourfold higher in pa-tients with TD in comparison with subjectswithout diabetes matched according to their BMIand waist circumference (). In addition, in-termittent hypoxemia is favored during the REMsleep in patients with TD (). This is an im-portant finding because sleep-related hypoxemia in

patients with SAS is known to be a major stimulusleading to oxidative stress and endothelial dys-function and might contribute to the increased riskof cardiovascular events observed in the populationwith diabetes ().

In addition, findings from previous studiessuggest that when TD occurs in parallel with SAS,it is associated with a specific breathing patternduring sleep that cannot be explained by age, sex,BMI, or SAS severity (, , ). In this way, thecomposition of AHI in patients with TD ischaracterized by an increase in apnea events, withno differences or even a reduction in hypopneaepisodes (). This result suggests that at the samelevel of AHI, patients with TD have more severesleep breathing disorders than subjects withoutdiabetes.

Finally, laboratory-based investigations have im-plicated DAN in this relationship, so that patients withTD with DAN are more likely to have not onlycentral, but also obstructive sleep apnea, than thosewithout it (). Therefore, an impairment of theupper airway reflexes, possibly due to alterations to theautonomic nervous fibers involved in their regulation,could lead to an inability of individuals with diabetesto respond appropriately to nocturnal airflow re-duction episodes, leading to more severe intermittenthypoxemic episodes.

Therapeutic Implications

The previous data raise the central issue of whether thenormalization of blood glucose levels can significantlyimprove pulmonary function and ameliorate sleep-breathing disorders in TD. However, only experi-mental data and pilot interventional studies aimed atexploring the impact on respiratory parameters ofsome therapeutic options such as the effect of im-proving glycemic control on sleep breathing arecurrently available. Therefore, further research on thisissue is warranted.

Data from experimental modelsThe beneficial effects of both metformin, a drugcurrently used to raise insulin sensitivity, and insulinon pulmonary development, sleep breathing, and lungfunction have been observed in some experimentalmodels (–).

Ramadan et al. () showed direct evidence for thecontribution of IR to the development of apneic ep-isodes and how oral treatment with metformin notonly prevented but also reversed apnea episodes inSprague-Dawley rats. More recently, Chen et al. ()have demonstrated that the subcutaneous adminis-tration of metformin ( and mg/kg) in neonatalWistar rats with experimental bronchopulmonarydysplasia was accompanied by a sizeable reduction in


REVIEWD

ownloaded from



http://dx.doi.org/10.1210/er.2017-00173


lung inflammation, alveolar septum thickness, andfibrosis, as well as improvements in vascularizationand survival.

Regarding the effect of insulin, Hein et al. ()demonstrated, in STZ-induced diabetic rats, that thedecreased ventilatory responses to hypercapnic andhypoxic challenges that appeared by the third andfourth week were prevented with insulin treatment.However, it is difficult to ascertain whether this effectwas attributed to the single reduction in blood glucose,to the insulin signaling itself, or a combination of botheffects.

Data from human studies: insulin sensitizersand insulinTwo studies have evaluated the effect of drugs usedto raise insulin sensitivity in subjects with chronicobstructive pulmonary disease and TD. In a ret-rospective one involving Korean patients, treat-ment with insulin sensitizers was independentlyassociated with improvements in FVC comparedwith insulin therapy (). More recently, in anopen-label prospective study that evaluated pa-tients with chronic obstructive pulmonary diseaseand TD or IGT, treatment with metformin for months improved the St. George’s RespiratoryQuestionnaire score by a median of five points andthe transition dyspnea index by two points (). Inaddition, inspiratory mouth pressure increased by. cm HO. Similarly, in a cross-sectional studyinvestigating pulmonary function among pa-tients with TD from Colombia without knownlung diseases, the metformin group showed signif-icantly lower differences from the expected values ofFVC compared with those treated with secreta-gogues after adjustment for metabolic control andthe duration of the disease ().

In contrast, patients on insulin therapy presentlower spirometric values. In this regard, data from subjects with TD participating in the cross-sectional Copenhagen City Heart Study demon-strated that the impairment of pulmonary functionwas more pronounced in patients treated with in-sulin compared with those only treated with oralagents or diet; the average FEV and FVC were and mL lower, respectively, in insulin-treatedpatients than in those treated with oral agents ().However, these negative data on lung functionassociated with insulin therapy may be related morewith the severity and duration of diabetes than withthe insulin itself. Another confounding factor notconsidered in these studies is the weight gain as-sociated with insulin therapy, a considerable clinicaleffect that could result in a negative impact on lungfunction.

The influence of insulin on alveolar-capillary mem-brane conductance was also explored in asymp-tomatic nonsmoking patients with TD receiving diet

and hypoglycemic drugs (). After the infusion of IU of regular insulin for days, a notable improve-ment in DLCO and alveolar capillary membraneconductance in comparison with saline infusion wasfound, independently of the metabolic effects. As novariations in VC, FEV, pulmonary wedge pressure, orpulmonary arteriolar resistance were observed, itseemed that the effect of insulin improving gas ex-change was mediated through the facilitation ofalveolar-capillary interface conductance ().

Data from humans: glucose optimization andimprovement of inflammationMore recently, a case-control study comprising patients with TD and controls without diabetesprovided evidence that the increased number ofnocturnal oxygen desaturations described in di-abetes exhibited a considerable reversibility aftera -day period of blood glucose optimization ().This positive effect on the repetitive nocturnalhypoxemia and reoxygenation cycle might be at-tributable not only to the normalization of thediabetic milieu, but also to the effect of exogenousinsulin. However, there was not a noteworthy in-crease in total insulin dosage, pointing to glycemicoptimization rather than the effect of insulin per seas the main factor accounting for the improvementof nocturnal oximetry.

The improvement of inflammation has also beenreported as having beneficial effects on pulmonaryfunction and sleep breathing (, ). However,most of the patients recruited for these studies hadno TD. The outcome of the neutralization ofTNF-a by etanercept (a fusion protein that binds toTNF-a molecules and blocks their interaction withthe cell surface) was explored in eight obese malepatients with SAS. In this placebo-controlled double-blind study, the anti-inflammatory action of the drugwas associated with a sizeable decrease in sleepinessand also with a marked reduction in the sum ofapneas and hypopnea events per hour comparedwith placebo (). Treatment with etanercept inpatients with refractory asthma has also improvedthe respiratory function and, in particular, post-bronchodilator FEV, in comparison with subjectswho received placebo, raising the possibility thatanti–TNF-a agents could be useful in the preventionand/or management of the impairment of lungfunction associated with TD ().

Potential benefits of other therapeutic options:inhaled insulin and incretin-based therapiesFor several years, alternative routes of insulin ad-ministration have been investigated to overcome thelimitations related to its subcutaneous adminis-tration (). Thus, the approval of delivering in-sulin by inhalation has resulted in insights into therelationship between TD and the lung, especially


REVIEWD

ownloaded from



when insulin receptors have been revealed in type IIpneumocytes, favoring surfactant synthesis ().Using a dry-powder inhalation device, insulin iscombined with an excipient that dissolves instan-taneously in lung fluid and releases recombinantinsulin for absorption, acting as a prandial insulin.Small reductions in FEV have been shown withinhaled insulin, although this outcome is not pro-gressive over time and is reversed when the treatmentis discontinued (). The potential effect of inhaledinsulin on subjects with TD with some degree oflung involvement will need to be addressed in thenext future.

The underlying deficit of GLP- in TD could alsobe involved in the impairment of airway caliber. TheGLP- receptor is abundant in the lungs, and it playsa role in the stimulation of pulmonary surfactantproduction by AT-II cells in experimental studies (,). Therefore, the potential pulmonary benefit ofincretin-based therapies is particularly relevant, andefforts might be undertaken to answer the question ofwhether GLP- analogs could improve pulmonaryfunction. Moreover, the pulmonary effects of alter-native routes to subcutaneous administration, such asGLP- adsorbed onto Technosphere microparticlesfor oral inhalation, that produce plasma levels of GLP- like those of parenteral administration need to beevaluated (). Finally, deserving of further attentionis whether the pharmacological inhibition of thedipeptidyl peptidase-, preventing the inactivation ofthe endogenous GLP- and prolonging their enhancedsecretion after meal ingestion, exerts any effect onpulmonary function.

Clinical Gaps, Controversies, andFuture Directions

The general assumption that the lung is a newtarget of diabetic complications should strengthenthe relationship not only between endocrinologistsand pulmonologist, but also with primary carephysicians. However, there are still some researchcontroversies and clinical gaps that need furtherinvestigation.

Which patients with T2D are at risk forpulmonary disease?This review offers a better understanding of theconditions underlying pulmonary involvement inTD, a relatively common but generally under-recognized complication. Although this involvementappears to be mediated primarily by pathologicalmechanisms associated with a restrictive pulmonarypattern, obstructive mechanisms are also implicated.The aim is a comprehensive diagnosis that will lead tothe treatment that can be adapted for both lungdisease and TD. As in other chronic complications

associated with diabetes microangiopathy, lungdysfunction appears to be more frequent in patientswith longer duration of the disease and poorermetabolic control. Similarly, the relationship betweenTD and obesity is of such interdependence that theterm “diabesity” has been coined. Therefore, excessweight must be considered as a risk factor contrib-uting to restrictive lung disease in subjects with TD(, ). In this regard, cross-sectional studies havereported a reduction in FVC and FEV in obesity(–). The mechanical effects of truncal obesityand metabolic effects of adipose tissue partly explainthe impairment of pulmonary function in obesesubjects ().

Is it recommended to screen patients with T2D forpulmonary dysfunction and sleep disturbances? Ifso, how should it be done?The first question concerns the most cost-effectivemanner of screening pulmonary dysfunction andsleep-breathing abnormalities in patients with TD.This is an important issue in terms of communityhealth care and its economic burden. Appropriatestudies aimed at determining the cost-effectivenessof including in routine visits such simple diagnostictests as forced spirometry maneuvers and ques-tionnaires to evaluate daytime sleepiness and sleepquality are needed.

Will lung dysfunction modify therapy for T2D?Another subject that merits exploration is the impactof improving metabolic control on pulmonary func-tion and sleep breathing, as well as the differentialactions and effects of the various treatments. In thisregard, as previously mentioned, the potential bene-ficial effects of incretin-based therapies, mainly GLP-receptor agonists, on pulmonary function seem partic-ularly relevant because of their capacity to increase lungsurfactant production and to enhance airway stability anddiameter. In addition, the treatment of patients with TDand pulmonary involvement includes improvements inlifestyle habits (smoking cessation, healthy diet, anda good level of physical activity), nutritional measure-ments, and muscle training.

How do lung dysfunction and sleep-breathingdisorders affect the patient with T2D?Both TD and the degree of poor glycemic control arerelated to respiratory function impairment and alsoseem to adversely affect breathing during sleep, be-coming an independent risk factor for higher ratesof sleep apnea while leading to more severe noctur-nal hypoxemia and daytime sleepiness. The complexpathomechanisms underlying breathing disorders inTD need to be further investigated, but it seems thatthe metabolic pathways related to IR and low-gradechronic inflammation are crucial in initiating lungabnormalities.

“It seems that the metabolicpathways related to IRand low-grade chronicinflammation are crucial ininitiating lung abnormalities.”


REVIEWD

ownloaded from



http://dx.doi.org/10.1210/er.2017-00173


In summary, the current evidence strongly supportsthe link between TD and respiratory dysfunction andindicates that pulmonary function should be taken intoconsideration by health care providers. Specific pilotscreening programs would be very useful for obtaining

preliminary results, which could provide further generalguidance on this issue. The current evidence points to thelung as an end target for TD complications and supportsthe recommendation that patients with TD be con-sidered a vulnerable group for pulmonary dysfunction.

References1. Unwin N, Gan D, Whiting D. The IDF Diabetes Atlas:

providing evidence, raising awareness and pro-moting action. Diabetes Res Clin Pract. 2010;87(1):2–3.

2. Forbes JM, Cooper ME. Mechanisms of diabeticcomplications. Physiol Rev. 2013;93(1):137–188.

3. Soriguer F, Goday A, Bosch-Comas A, Bordiu E,Calle-Pascual A, Carmena R, Casamitjana R, Castaño L,Castell C, Catala M, Delgado E, Franch J, GaztambideS, Girbes J, Gomis R, Gutierrez G, Lopez-Alba A,Martınez-Larrad MT, Menendez E, Mora-Peces I,Ortega E, Pascual-Manich G, Rojo-Martınez G,Serrano-Rios M, Valdes S, Vazquez JA, Vendrell J.Prevalence of diabetes mellitus and impairedglucose regulation in Spain: the [email protected] Study.Diabetologia. 2012;55(1):88–93.

4. Harris MI, Eastman RC. Early detection of un-diagnosed diabetes mellitus: a US perspective. Di-abetes Metab Res Rev. 2000;16(4):230–236.

5. Sampol G, Lecube A. Type 2 diabetes and the lung:a bidirectional relationship. Endocrinol Nutr. 2012;59(2):95–97.

6. van den Borst B, Gosker HR, Zeegers MP, ScholsAMWJ. Pulmonary function in diabetes: a meta-analysis. Chest. 2010;138(2):393–406.

7. Lecube A, Sampol G, Muñoz X, Hernandez C, MesaJ, Simo R. Type 2 diabetes impairs pulmonaryfunction in morbidly obese women: a case-controlstudy. Diabetologia. 2010;53(6):1210–1216.

8. Davis WA, Knuiman M, Kendall P, Grange V, DavisTM; Fremantle Diabetes Study. Glycemic exposureis associated with reduced pulmonary function intype 2 diabetes: the Fremantle Diabetes Study.Diabetes Care. 2004;27(3):752–757.

9. Lecube A, Sampol G, Lloberes P, Romero O, Mesa J,Hernandez C, Simo R. Diabetes is an independent riskfactor for severe nocturnal hypoxemia in obese pa-tients. A case-control study. PLoS One. 2009;4(3):e4692.

10. Enright PL, Kronmal RA, Higgins M, Schenker M,Haponik EF. Spirometry reference values for womenand men 65 to 85 years of age. Cardiovascular healthstudy. Am Rev Respir Dis. 1993;147(1):125–133.

11. Klein OL, Krishnan JA, Glick S, Smith LJ. Systematicreview of the association between lung functionand type 2 diabetes mellitus. Diabet Med. 2010;27(9):977–987.

12. Barrett-Connor E, Frette C. NIDDM, impaired glu-cose tolerance, and pulmonary function in olderadults. The Rancho Bernardo Study. Diabetes Care.1996;19(12):1441–1444.

13. Davis TM, Knuiman M, Kendall P, Vu H, Davis WA.Reduced pulmonary function and its associations intype 2 diabetes: the Fremantle Diabetes Study.Diabetes Res Clin Pract. 2000;50(2):153–159.

14. Lange P, Parner J, Schnohr P, Jensen G. CopenhagenCity Heart Study: longitudinal analysis of ventilatorycapacity in diabetic and nondiabetic adults. EurRespir J. 2002;20(6):1406–1412.

15. Walter RE, Beiser A, Givelber RJ, O’Connor GT,Gottlieb DJ. Association between glycemic state

and lung function: the Framingham Heart Study.Am J Respir Crit Care Med. 2003;167(6):911–916.

16. Lawlor DA, Ebrahim S, Smith GD. Associations ofmeasures of lung function with insulin resistanceand type 2 diabetes: findings from the BritishWomen’s Heart and Health Study. Diabetologia.2004;47(2):195–203.

17. Litonjua AA, Lazarus R, Sparrow D, Demolles D,Weiss ST. Lung function in type 2 diabetes: theNormative Aging Study. Respir Med. 2005;99(12):1583–1590.

18. Chance WW, Rhee C, Yilmaz C, Dane DM, PrunedaML, Raskin P, Hsia CC. Diminished alveolar micro-vascular reserves in type 2 diabetes reflect sys-temic microangiopathy. Diabetes Care. 2008;31(8):1596–1601.

19. Yeh HC, Punjabi NM, Wang NY, Pankow JS, DuncanBB, Cox CE, Selvin E, Brancati FL. Cross-sectional andprospective study of lung function in adults withtype 2 diabetes: the Atherosclerosis Risk in Com-munities (ARIC) study. Diabetes Care. 2008;31(4):741–746.

20. Oda E, Kawai R. A cross-sectional relationship be-tween vital capacity and metabolic syndrome andbetween vital capacity and diabetes in a sampleJapanese population. Environ Health Prev Med. 2009;14(5):284–291.

21. Oda E, Kawai R. A cross-sectional relationship be-tween vital capacity and diabetes in Japanese men.Diabetes Res Clin Pract. 2009;85(1):111–116.

22. Yeh F, Dixon AE, Marion S, Schaefer C, Zhang Y, BestLG, Calhoun D, Rhoades ER, Lee ET. Obesity inadults is associated with reduced lung function inmetabolic syndrome and diabetes: the Strong HeartStudy. Diabetes Care. 2011;34(10):2306–2313.

23. Klein OL, Kalhan R, Williams MV, Tipping M, Lee J,Peng J, Smith LJ. Lung spirometry parameters anddiffusion capacity are decreased in patients withtype 2 diabetes. Diabet Med. 2012;29(2):212–219.

24. Huang H, Guo Q, Li L, Lin S, Lin Y, Gong X, Yao J,Liang J, Lin L, Wen J, Chen G. Effect of type 2 diabetesmellitus on pulmonary function. Exp Clin EndocrinolDiabetes. 2014;122(6):322–326.

25. Klein OL, Aviles-Santa L, Cai J, Collard HR, KanayaAM, Kaplan RC, Kinney GL, Mendes E, Smith L,Talavera G, Wu D, Daviglus M. Hispanics/Latinoswith type 2 diabetes have functional andsymptomatic pulmonary impairment mirroringkidney microangiopathy: Findings from theHispanic Community Health Study/Study ofLatinos (HCHS/SOL). Diabetes Care. 2016;39(11):2051–2057.

26. Lange P, Groth S, Kastrup J, Mortensen J, AppleyardM, Nyboe J, Jensen G, Schnohr P. Diabetes mellitus,plasma glucose and lung function in a cross-sectional population study. Eur Respir J. 1989;2(1):14–19.

27. Dharwadkar AR, Dharwadkar AA, Banu G, BagaliS. Reduction in lung functions in type-2 diabetesin Indian population: correlation with glycemic

status. Indian J Physiol Pharmacol. 2011;55(2):170–175.

28. Asanuma Y, Fujiya S, Ide H, Agishi Y. Characteristicsof pulmonary function in patients with diabetesmellitus. Diabetes Res Clin Pract. 1985;1(2):95–101.

29. Kim HK, Kim CH, Jung YJ, Bae SJ, Choe J, Park JY,Lee KU. Association of restrictive ventilatorydysfunction with insulin resistance and type 2diabetes in koreans. Exp Clin Endocrinol Diabetes.2011;119(1):47–52.

30. Weir DC, Jennings PE, Hendy MS, Barnett AH, BurgePS. Transfer factor for carbon monoxide in patientswith diabetes with and without microangiopathy.Thorax. 1988;43(9):725–726.

31. Marvisi M, Bartolini L, del Borrello P, Brianti M,Marani G, Guariglia A, Cuomo A. Pulmonaryfunction in non-insulin-dependent diabetes melli-tus. Respiration. 2001;68(3):268–272.

32. Lange P, Groth S, Mortensen J, Appleyard M, NyboeJ, Schnohr P, Jensen G. Diabetes mellitus and ven-tilatory capacity: a five year follow-up study. EurRespir J. 1990;3(3):288–292.

33. Engstrom G, Janzon L. Risk of developing diabetes isinversely related to lung function: a population-basedcohort study. Diabet Med. 2002;19(2):167–170.

34. Engstrom G, Hedblad B, Nilsson P, Wollmer P,Berglund G, Janzon L. Lung function, insulin re-sistance and incidence of cardiovascular disease:a longitudinal cohort study. J Intern Med. 2003;253(5):574–581.

35. Ford ES, Mannino DM; National Health and Nu-trition Examination Survey Epidemiologic Follow-upStudy. Prospective association between lung func-tion and the incidence of diabetes: findings fromthe National Health and Nutrition ExaminationSurvey Epidemiologic Follow-up Study. DiabetesCare. 2004;27(12):2966–2970.

36. Fernandez-Real JM, Ricart W. Insulin resistance andinflammation in an evolutionary perspective: thecontribution of cytokine genotype/phenotype tothriftiness. Diabetologia. 1999;42(11):1367–1374.

37. Lazarus R, Sparrow D, Weiss ST. Impaired ventilatoryfunction and elevated insulin levels in nondiabeticmales: the Normative Aging Study. Eur Respir J. 1998;12(3):635–640.

38. Lecube A, Sampol G, Muñoz X, Lloberes P,Hernandez C, Simo R. Insulin resistance is related toimpaired lung function in morbidly obese women:a case-control study. Diabetes Metab Res Rev. 2010;26(8):639–645.

39. Sagun G, Gedik C, Ekiz E, Karagoz E, Takir M, Oguz A.The relation between insulin resistance and lungfunction: a cross sectional study. BMC Pulm Med.2015;15:139.

40. Rastogi D, Bhalani K, Hall CB, Isasi CR. Association ofpulmonary function with adiposity and metabolicabnormalities in urban minority adolescents. AnnAm Thorac Soc. 2014;11(5):744–752.

41. Petersen KF, Dufour S, Befroy D, Garcia R, Shulman GI.Impairedmitochondrial activity in the insulin-resistant


REVIEWD

ownloaded from



offspring of patients with type 2 diabetes.N Engl J Med.2004;350(7):664–671.

42. Morino K, Petersen KF, Shulman GI. Molecularmechanisms of insulin resistance in humans andtheir potential links with mitochondrial dysfunc-tion. Diabetes. 2006;55(Suppl 2):S9–S15.

43. Lazarus R, Sparrow D, Weiss ST. Handgrip strengthand insulin levels: cross-sectional and prospectiveassociations in the Normative Aging Study. Meta-bolism. 1997;46(11):1266–1269.

44. Rastogi D, Fraser S, Oh J, Huber AM, Schulman Y,Bhagtani RH, Khan ZS, Tesfa L, Hall CB, Macian F.Inflammation, metabolic dysregulation, and pul-monary function among obese urban adolescentswith asthma. Am J Respir Crit Care Med. 2015;191(2):149–160.

45. Lim SY, Rhee EJ, Sung KC. Metabolic syndrome,insulin resistance and systemic inflammation as riskfactors for reduced lung function in Korean non-smoking males. J Korean Med Sci. 2010;25(10):1480–1486.

46. Thuesen BH, Husemoen LLN, Hersoug LG, PisingerC, Linneberg A. Insulin resistance as a predictor ofincident asthma-like symptoms in adults. Clin ExpAllergy. 2009;39(5):700–707.

47. Arshi M, Cardinal J, Hill RJ, Davies PSW, WainwrightC. Asthma and insulin resistance in children.Respirology. 2010;15(5):779–784.

48. Forno E, Han YY, Muzumdar RH, Celedon JC. Insulinresistance, metabolic syndrome, and lung functionin US adolescents with and without asthma.J Allergy Clin Immunol. 2015;136(2):304–311.e8.

49. Singh S, Bodas M, Bhatraju NK, Pattnaik B, GhewareA, Parameswaran PK, Thompson M, Freeman M,Mabalirajan U, Gosens R, Ghosh B, Pabelick C,Linneberg A, Prakash YS, Agrawal A. Hyper-insulinemia adversely affects lung structure andfunction. Am J Physiol Lung Cell Mol Physiol. 2016;310(9):L837–L845.

50. Singh S, Prakash YS, Linneberg A, Agrawal A. Insulinand the lung: connecting asthma and metabolicsyndrome. J Allergy (Cairo). 2013;2013: 627384.

51. Kim KM, Kim SS, Lee SH, Song WJ, Chang YS, MinKU, Cho SH. Association of insulin resistance withbronchial hyperreactivity. Asia Pac Allergy. 2014;4(2):99–105.

52. Shapiro DL, Livingston JN,ManiscalcoWM, FinkelsteinJN. Insulin receptors and insulin effects on type IIalveolar epithelial cells. Biochim Biophys Acta. 1986;885(2):216–220.

53. Frisch CM, Zimmermann K, Zilleßen P, Pfeifer A,Racke K, Mayer P. Non-small cell lung cancer cellsurvival crucially depends on functional insulinreceptors. Endocr Relat Cancer. 2015;22(4):609–621.

54. Ruttenstock E, Doi T, Dingemann J, Puri P. Insulinreceptor is downregulated in the nitrofen-inducedhypoplastic lung. J Pediatr Surg. 2010;45(5):948–952.

55. Fernandez-Real JM, Chico B, Shiratori M, Nara Y,Takahashi H, Ricart W. Circulating surfactant pro-tein A (SP-A), a marker of lung injury, is associatedwith insulin resistance. Diabetes Care. 2008;31(5):958–963.

56. Perez-Perez A, Vilariño-Garcıa T, Fernandez-Riejos P,Martın-Gonzalez J, Segura-Egea JJ, Sanchez-MargaletV. Role of leptin as a link between metabolism andthe immune system. Cytokine Growth Factor Rev.2017;35:71–84.

57. Chen H, Liang ZW, Wang ZH, Zhang JP, Hu B, XingXB, Cai WB. Akt activation and inhibition of cy-tochrome C release: Mechanistic insights intoleptin-promoted survival of type II alveolar epi-thelial cells. J Cell Biochem. 2015;116(10):2313–2324.

58. De Blasio MJ, Boije M, Kempster SL, Smith GC,Charnock-Jones DS, Denyer A, Hughes A, Wooding

FB, Blache D, Fowden AL, Forhead AJ. Leptin ma-tures aspects of lung structure and function in theovine fetus. Endocrinology. 2016;157(1):395–404.

59. Mendivil CO, Koziel H, Brain JD. Metabolic hor-mones, apolipoproteins, adipokines, and cytokinesin the alveolar lining fluid of healthy adults: com-partmentalization and physiological correlates. PLoSOne. 2015;10(4):e0123344.

60. Bergen HT, Cherlet TC, Manuel P, Scott JE. Iden-tification of leptin receptors in lung and isolatedfetal type II cells. Am J Respir Cell Mol Biol. 2002;27(1):71–77.

61. Nair P, Radford K, Fanat A, Janssen LJ, Peters-GoldenM, Cox PG. The effects of leptin on airway smoothmuscle responses. Am J Respir Cell Mol Biol. 2008;39(4):475–481.

62. Arteaga-Solis E, Zee T, Emala CW, Vinson C, Wess J,Karsenty G. Inhibition of leptin regulation ofparasympathetic signaling as a cause of extremebody weight-associated asthma [published cor-rection appears in Cell Metab. 2013;17(3):463-464].Cell Metab. 2013;17(1):35–48.

63. Park HK, Ahima RS. Physiology of leptin: energyhomeostasis, neuroendocrine function and meta-bolism. Metabolism. 2015;64(1):24–34.

64. Rehman Khan A, Awan FR. Leptin resistance: Apossible interface between obesity and pulmonary-related disorders. Int J Endocrinol Metab. 2016;14(1):e32586.

65. Sin DD, Man SF. Impaired lung function and serumleptin in men and women with normal bodyweight: a population based study. Thorax. 2003;58(8):695–698.

66. Hickson DA, Burchfiel CM, Petrini MF, Liu J,Campbell-Jenkins BW, Bhagat R, Marshall GD.Leptin is inversely associated with lung function inAfrican Americans, independent of adiposity: theJackson Heart Study. Obesity (Silver Spring). 2011;19(5):1054–1061.

67. Eising JB, Uiterwaal CS, Evelein AM, Visseren FL, vander Ent CK. Relationship between leptin and lungfunction in young healthy children. Eur Respir J.2014;43(4):1189–1192.

68. Naveed B, Weiden MD, Kwon S, Gracely EJ,Comfort AL, Ferrier N, Kasturiarachchi KJ, CohenHW, Aldrich TK, Rom WN, Kelly K, Prezant DJ,Nolan A. Metabolic syndrome biomarkers pre-dict lung function impairment: a nested case-control study. Am J Respir Crit Care Med. 2012;185(4):392–399.

69. Khrisanapant W, Sengmeuang P, KukongviriyapanU, Pasurivong O, Pakdeechotel P. Plasma leptinlevels and a restrictive lung in obese Thai childrenand adolescents. Southeast Asian J Trop Med PublicHealth. 2015;46(1):116–124.

70. Newson RB, Jones M, Forsberg B, Janson C, BossiosA, Dahlen SE, Toskala EM, Al-Kalemji A, KowalskiML, Rymarczyk B, Salagean EM, van Drunen CM,Bachert C, Wehrend T, Kramer U, Mota-Pinto A,Burney P, Leynaert B, Jarvis D. The association ofasthma, nasal allergies, and positive skin prick testswith obesity, leptin, and adiponectin. Clin Exp Al-lergy. 2014;44(2):250–260.

71. Leão da Silva P, de Mello MT, Cheik NC, SanchesPL, Munhoz da Silveira Campos R, Carnier J,Inoue D, do Nascimento CM, Oyama LM, Tock L,Tufik S, Damaso AR. Reduction in the leptinconcentration as a predictor of improvement inlung function in obese adolescents. Obes Facts.2012;5(6):806–820.

72. Duncan BB, Schmidt MI, Pankow JS, Ballantyne CM,Couper D, Vigo A, Hoogeveen R, Folsom AR, HeissG; Atherosclerosis Risk in Communities Study.Low-grade systemic inflammation and the

development of type 2 diabetes: the atheroscle-rosis risk in communities study. Diabetes. 2003;52(7):1799–1805.

73. McClean KM, Kee F, Young IS, Elborn JS. Obesityand the lung: 1. Epidemiology. Thorax. 2008;63(7):649–654.

74. Schmidt MI, Duncan BB, Sharrett AR, Lindberg G,Savage PJ, Offenbacher S, Azambuja MI, Tracy RP,Heiss G. Markers of inflammation and prediction ofdiabetes mellitus in adults (Atherosclerosis Risk inCommunities study): a cohort study. Lancet. 1999;353(9165):1649–1652.

75. Douni E, Kollias G. A critical role of the p75 tumornecrosis factor receptor (p75TNF-R) in organ in-flammation independent of TNF, lymphotoxin a, orthe p55TNF-R. J Exp Med. 1998;188(7):1343–1352.

76. Mazzon E, Cuzzocrea S. Role of TNF-alpha in lungtight junction alteration in mouse model of acutelung inflammation. Respir Res. 2007;8:75.

77. Berry MA, Hargadon B, Shelley M, Parker D, ShawDE, Green RH, Bradding P, Brightling CE, WardlawAJ, Pavord ID. Evidence of a role of tumor necrosisfactor alpha in refractory asthma. N Engl J Med. 2006;354(7):697–708.

78. Lecube A, Sampol G, Muñoz X, Ferrer R, HernandezC, Simo R. TNF-a system and lung function im-pairment in obesity. Cytokine. 2011;54(2):121–124.

79. Vernooy JH, Kuçukaycan M, Jacobs JA, ChavannesNH, Buurman WA, Dentener MA, Wouters EF.Local and systemic inflammation in patients withchronic obstructive pulmonary disease: solubletumor necrosis factor receptors are increased insputum. Am J Respir Crit Care Med. 2002;166(9):1218–1224.

80. Aronson D, Roterman I, Yigla M, Kerner A, AvizoharO, Sella R, Bartha P, Levy Y, Markiewicz W. Inverseassociation between pulmonary function andC-reactive protein in apparently healthy subjects.Am J Respir Crit Care Med. 2006;174(6):626–632.

81. Wannamethee SG, Shaper AG, Rumley A, Sattar N,Whincup PH, Thomas MC, Lowe GD. Lung func-tion and risk of type 2 diabetes and fatal andnonfatal major coronary heart disease events:possible associations with inflammation. DiabetesCare. 2010;33(9):1990–1996.

82. Dennis RJ, Maldonado D, Rojas MX, Aschner P,Rondon M, Charry L, Casas A. Inadequate glu-cose control in type 2 diabetes is associated withimpaired lung function and systemic in-flammation: a cross-sectional study. BMC PulmMed. 2010;10:38.

83. Hsia CC, Raskin P. The diabetic lung: relevance ofalveolar microangiopathy for the use of inhaledinsulin. Am J Med. 2005;118(3):205–211.

84. Isotani H, Nakamura Y, Kameoka K, Tanaka K,Furukawa K, Kitaoka H, Ohsawa N. Pulmonarydiffusing capacity, serum angiotensin-convertingenzyme activity and the angiotensin-convertingenzyme gene in Japanese non-insulin-dependentdiabetes mellitus patients. Diabetes Res Clin Pract.1999;43(3):173–177.

85. Agarwal AS, Fuladi AB, Mishra G, Tayade BO.Spirometry and diffusion studies in patients withtype-2 diabetes mellitus and their association withmicrovascular complications. Indian J Chest DisAllied Sci. 2010;52(4):213–216.

86. Lau AC, Lo MK, Leung GT, Choi FP, Yam LY,Wasserman K. Altered exercise gas exchange asrelated to microalbuminuria in type 2 diabeticpatients. Chest. 2004;125(4):1292–1298.

87. Guvener N, Tutuncu NB, Akcay S, Eyuboglu F,Gokcel A. Alveolar gas exchange in patients withtype 2 diabetes mellitus. Endocr J. 2003;50(6):663–667.


REVIEWD

ownloaded from



http://dx.doi.org/10.1210/er.2017-00173


88. Ljubic S, Metelko Z, Car N, Roglic G, Drazic Z.Reduction of diffusion capacity for carbon mon-oxide in diabetic patients. Chest. 1998;114(4):1033–1035.

89. Mousa K, Onadeko BO, Mustafa HT, MohamedM, Nabilla A, Omar A, Al-Bunni A, Elgazzar A.Technetium 99mTc-DTPA clearance in the eval-uation of pulmonary involvement in patientswith diabetes mellitus. Respir Med. 2000;94(11):1053–1056.

90. Kuziemski K, Specjalski K, Jassem E. Diabetic pul-monary microangiopathy - fact or fiction? Endok-rynol Pol. 2011;62(2):171–176.

91. Kumar A, Bade G, Trivedi A, Jyotsna VP, Talwar A.Postural variation of pulmonary diffusing capacityas a marker of lung microangiopathy in Indianpatients with type 2 diabetes mellitus. Indian JEndocrinol Metab. 2016;20(2):238–244.

92. Zochodne DW. Diabetes mellitus and the pe-ripheral nervous system: manifestations andmechanisms. Muscle Nerve. 2007;36(2):144–166.

93. Bottini P, Scionti L, Santeusanio F, Casucci G,Tantucci C. Impairment of the respiratory system indiabetic autonomic neuropathy. Diabetes NutrMetab. 2000;13(3):165–172.

94. Murashima S, Takeda K, Matsumura K, YamakadoK, Sakuma H, Kitano T, Nakagawa T, Ichihara T,Yamakado T, Murata K. Increased lung uptake ofiodine-123-MIBG in diabetics with sympatheticnervous dysfunction. J Nucl Med. 1998;39(2):334–338.

95. Heaton RW, Guy RJ, Gray BJ, Watkins PJ, Costello JF.Diminished bronchial reactivity to cold air in di-abetic patients with autonomic neuropathy. BrMed J (Clin Res Ed). 1984;289(6438):149–151.

96. Behera D, Das S, Dash RJ, Jindal SK. Cough reflexthreshold in diabetes mellitus with and withoutautonomic neuropathy. Respiration. 1995;62(5):263–268.

97. Rhind GB, Gould GA, Ewing DJ, Clarke BF, DouglasNJ. Increased bronchial reactivity to histamine indiabetic autonomic neuropathy. Clin Sci (Lond).1987;73(4):401–405.

98. Douglas NJ, Campbell IW, Ewing DJ, Clarke BF,Flenley DC. Reduced airway vagal tone in diabeticpatients with autonomic neuropathy. Clin Sci(Lond). 1981;61(5):581–584.

99. Tantucci C, Bottini P, Dottorini ML, Puxeddu E,Casucci G, Scionti L, Sorbini CA. Ventilatory re-sponse to exercise in diabetic subjects with auto-nomic neuropathy. J Appl Physiol (1985). 1996;81(5):1978–1986.

100. Kaminski DM, Schaan BD, da Silva AM, Soares PP,Plentz RD, Dall’Ago P. Inspiratory muscle weaknessis associated with autonomic cardiovascular dys-function in patients with type 2 diabetes mellitus.Clin Auton Res. 2011;21(1):29–35.

101. Tesfaye S, Boulton AJ, Dyck PJ, Freeman R, HorowitzM, Kempler P, Lauria G, Malik RA, Spallone V, VinikA, Bernardi L, Valensi P; Toronto Diabetic Neu-ropathy Expert Group. Diabetic neuropathies: up-date on definitions, diagnostic criteria, estimation ofseverity, and treatments [published correctionappears in Diabetes Care. 2010;33(12):2725]. Di-abetes Care. 2010;33(10):2285–2293.

102. Kabitz HJ, Sonntag F, Walker D, Schwoerer A,Walterspacher S, Kaufmann S, Beuschlein F, SeufertJ, Windisch W. Diabetic polyneuropathy is associ-ated with respiratory muscle impairment in type 2diabetes. Diabetologia. 2008;51(1):191–197.

103. Guerrero N, Bunout D, Hirsch S, Barrera G, Leiva L,Henrıquez S, De la Maza MP. Premature loss ofmuscle mass and function in type 2 diabetes. Di-abetes Res Clin Pract. 2016;117:32–38.

104. Correa AP, Ribeiro JP, Balzan FM, Mundstock L,Ferlin EL, Moraes RS. Inspiratory muscle training intype 2 diabetes with inspiratory muscle weakness.Med Sci Sports Exerc. 2011;43(7):1135–1141.

105. Kiliçli F, Dokmetas S, Candan F, Ozsahin S, KorkmazS, Amasyali E, Fakioglu K, Dal K, Acibucu F, Cakir I.Inspiratory muscle strength is correlated with car-nitine levels in type 2 diabetes. Endocr Res. 2010;35(2):51–58.

106. Wanke T, Formanek D, Auinger M, Popp W, ZwickH, Irsigler K. Inspiratory muscle performance andpulmonary function changes in insulin-dependentdiabetes mellitus. Am Rev Respir Dis. 1991;143(1):97–100.

107. Fuso L, Pitocco D, Longobardi A, Zaccardi F,Contu C, Pozzuto C, Basso S, Varone F, GhirlandaG, Antonelli Incalzi R. Reduced respiratorymuscle strength and endurance in type 2 di-abetes mellitus. Diabetes Metab Res Rev. 2012;28(4):370–375.

108. Buchmann N, Norman K, Steinhagen-Thiessen E,Demuth I, Eckardt R. [Pulmonary function in elderlysubjects with metabolic syndrome and type II di-abetes : Data from the Berlin Aging Study II]. ZGerontol Geriatr. 2016;49(5):405–415.

109. Kohn RR, Schnider SL. Glycosylation of humancollagen. Diabetes. 1982;31:47–51.

110. Laurent GJ. Lung collagen: more than scaffolding.Thorax. 1986;41(6):418–428.

111. Sternberg M, Cohen-Forterre L, Peyroux J. Con-nective tissue in diabetes mellitus: biochemical al-terations of the intercellular matrix with specialreference to proteoglycans, collagens and basementmembranes. Diabete Metab. 1985;11(1):27–50.

112. Schnapf BM, Banks RA, Silverstein JH, RosenbloomAL, Chesrown SE, Loughlin GM. Pulmonary functionin insulin-dependent diabetes mellitus with limitedjoint mobility. Am Rev Respir Dis. 1984;130(5):930–932.

113. Katsuoka F, Kawakami Y, Arai T, Imuta H, FujiwaraM, Kanma H, Yamashita K. Type II alveolar epithelialcells in lung express receptor for advanced glycationend products (RAGE) gene. Biochem Biophys ResCommun. 1997;238(2):512–516.

114. Ota C, Ishizawa K, Yamada M, Tando Y, He M,Takahashi T, Yamaya M, Yamamoto Y, YamamotoH, Kure S, Kubo H. Receptor for advanced glycationend products expressed on alveolar epithelial cells isthe main target for hyperoxia-induced lung injury.Respir Investig. 2016;54(2):98–108.

115. Mukherjee TK, Mukhopadhyay S, Hoidal JR. Im-plication of receptor for advanced glycation endproduct (RAGE) in pulmonary health and patho-physiology. Respir Physiol Neurobiol. 2008;162(3):210–215.

116. Milutinovic PS, Alcorn JF, Englert JM, Crum LT, OuryTD. The receptor for advanced glycation endproducts is a central mediator of asthma patho-genesis. Am J Pathol. 2012;181(4):1215–1225.

117. Sin DD, Pahlavan PS, Man SF. Surfactant protein D:a lung specific biomarker in COPD? Ther Adv RespirDis. 2008;2(2):65–74.

118. Ju CR, Liu W, Chen RC. Serum surfactant protein D:biomarker of chronic obstructive pulmonary dis-ease. Dis Markers. 2012;32(5):281–287.

119. Osaka A, Yanagihara K, Yamada Y, Hasegawa H,Inokuchi N, Hayashi T, Komoda M, Nakamura S,Aoyama M, Sawada T, Kamihira S. Elevation ofserum KL-6 glycoprotein or surfactant protein-D inadult T-cell leukemia with distinct pulmonarycomplications. Tohoku J Exp Med. 2009;218(2):99–105.

120. Bonella F, Volpe A, Caramaschi P, Nava C, Ferrari P,Schenk K, Ohshimo S, Costabel U, Ferrari M.

Surfactant protein D and KL-6 serum levels insystemic sclerosis: correlation with lung and sys-temic involvement. Sarcoidosis Vasc Diffuse LungDis. 2011;28(1):27–33.

121. Korner M, Stockli M, Waser B, Reubi JC. GLP-1receptor expression in human tumors and humannormal tissues: potential for in vivo targeting. J NuclMed. 2007;48(5):736–743.

122. Benito E, Blazquez E, Bosch MA. Glucagon-likepeptide-1-(7-36)amide increases pulmonary sur-factant secretion through a cyclic adenosine 39,59-monophosphate-dependent protein kinasemechanismin rat type II pneumocytes. Endocrinology. 1998;139(5):2363–2368.

123. Vara E, Arias-Dıaz J, Garcıa C, Balibrea JL, Blazquez E.Glucagon-like peptide-1(7-36) amide stimulatessurfactant secretion in human type II pneumocytes.Am J Respir Crit Care Med. 2001;163(4):840–846.

124. Romanı-Perez M, Outeiriño-Iglesias V, Gil-LozanoM, Gonzalez-Matıas LC, Mallo F, Vigo E. PulmonaryGLP-1 receptor increases at birth and exogenousGLP-1 receptor agonists augmented surfactant-protein levels in litters from normal and nitrofen-treated pregnant rats. Endocrinology. 2013;154(3):1144–1155.

125. Romanı-Perez M, Outeiriño-Iglesias V, Moya CM,Santisteban P, Gonzalez-Matıas LC, Vigo E, Mallo F.Activation of the GLP-1 receptor by liraglutideincreases ACE2 expression, reversing right ventriclehypertrophy, and improving the production of SP-Aand SP-B in the lungs of type 1 diabetes rats. En-docrinology. 2015;156(10):3559–3569.

126. Nicolaie T, Zavoianu C, Nuta P. Pulmonary in-volvement in diabetes mellitus. Rom J Intern Med.2003;41(4):365–374.

127. Weynand B, Jonckheere A, Frans A, Rahier J. Di-abetes mellitus induces a thickening of the pul-monary basal lamina. Respiration. 1999;66(1):14–19.

128. Popov D, Simionescu M. Structural and transportproperty alterations of the lung capillary endo-thelium in diabetes. Ital J Anat Embryol. 2001;106(2,Suppl 1)405–412.

129. Kida K, Utsuyama M, Takizawa T, Thurlbeck WM.Changes in lung morphologic features and elasticitycaused by streptozotocin-induced diabetes mellitusin growing rats. Am Rev Respir Dis. 1983;128(1):125–131.

130. Foster DJ, Ravikumar P, Bellotto DJ, Unger RH, HsiaCC. Fatty diabetic lung: altered alveolar structureand surfactant protein expression. Am J Physiol LungCell Mol Physiol. 2010;298(3):L392–L403.

131. Ofulue AF, Thurlbeck WM. Experimental diabetesand the lung. II. In vivo connective tissue meta-bolism. Am Rev Respir Dis. 1988;138(2):284–289.

132. Popov D, Simionescu M. Alterations of lungstructure in experimental diabetes, and diabetesassociated with hyperlipidaemia in hamsters. EurRespir J. 1997;10(8):1850–1858.

133. Schnider SL, Kohn RR. Effects of age and diabetesmellitus on the solubility and nonenzymatic glu-cosylation of human skin collagen. J Clin Invest. 1981;67(6):1630–1635.

134. Reddy GK. Cross-linking in collagen by non-enzymatic glycation increases the matrix stiffness inrabbit achilles tendon. Exp Diabesity Res. 2004;5(2):143–153.

135. Throckmorton DC, Brogden AP, Min B, RasmussenH, Kashgarian M. PDGF and TGF-beta mediatecollagen production by mesangial cells exposed toadvanced glycosylation end products. Kidney Int.1995;48(1):111–117.

136. Twigg SM, Cao Z, MCLennan SV, Burns WC,Brammar G, Forbes JM, Cooper ME. Renal connec-tive tissue growth factor induction in experimental


REVIEWD

ownloaded from



diabetes is prevented by aminoguanidine. Endocri-nology. 2002;143(12):4907–4915.

137. Mir SH, Darzi MM. Histopathological abnormalitiesof prolonged alloxan-induced diabetes mellitus inrabbits. Int J Exp Pathol. 2009;90(1):66–73.

138. Sun Z, Liu L, Liu N, Liu Y. Muscular response andadaptation to diabetes mellitus. Front Biosci. 2008;13:4765–4794.

139. Wagner U, Fehmann HC, Bredenboker D, Yu F,Barth PJ, von Wichert P, Goke B. The stimulatoryeffect of amylin on mucus secretion in isolated rattrachea. Res Exp Med (Berl). 1993;193(6):347–352.

140. Wagner U, Bredenbroker D, Barth PJ, Fehmann HC,von Wichert P. Amylin immunoreactivity in the rattrachea and characterization of the interaction ofamylin and somatostatin on airway mucus secre-tion. Res Exp Med (Berl). 1995;195(5):289–296.

141. Khemtemourian L, Gazit E, Miranker A. Recentinsight in islet amyloid polypeptide morphology,structure, membrane interaction, and toxicity intype 2 diabetes. J Diabetes Res. 2016;2016:2535878.

142. Kodolova IM, Lysenko LV, Saltykov BB. [Changes inthe lungs in diabetes mellitus]. Arkh Patol. 1982;44(7):35–40.

143. Kuziemski K, Pienkowska J, Słominski W, Jassem E,Studniarek M. Pulmonary capillary permeability andpulmonary microangiopathy in diabetes mellitus.Diabetes Res Clin Pract. 2015;108(3):e56–e59.

144. Vracko R, Thorning D, Huang TW. Basal lamina ofalveolar epithelium and capillaries: quantitativechanges with aging and in diabetes mellitus. Am RevRespir Dis. 1979;120(5):973–983.

145. Matsubara T, Hara F. [The pulmonary function andhistopathological studies of the lung in diabetesmellitus]. Nippon Ika Daigaku Zasshi. 1991;58(5):528–536.

146. Fariña J, Furio V, Fernandez-Aceñero MJ, Muzas MA.Nodular fibrosis of the lung in diabetes mellitus.Virchows Arch. 1995;427(1):61–63.

147. O’Sullivan BP, Freedman SD. Cystic fibrosis. Lancet.2009;373(9678):1891–1904.

148. Moran A, Dunitz J, Nathan B, Saeed A, Holme B,Thomas W. Cystic fibrosis-related diabetes: currenttrends in prevalence, incidence, and mortality. Di-abetes Care. 2009;32(9):1626–1631.

149. Moran A, Becker D, Casella SJ, Gottlieb PA, KirkmanMS, Marshall BC, Slovis B; CFRD Consensus Con-ference Committee. Epidemiology, pathophysiology,and prognostic implications of cystic fibrosis-relateddiabetes: a technical review. Diabetes Care. 2010;33(12):2677–2683.

150. Adler AI, Shine BS, Chamnan P, Haworth CS, BiltonD. Genetic determinants and epidemiology of cysticfibrosis-related diabetes: results from a British co-hort of children and adults. Diabetes Care. 2008;31(9):1789–1794.

151. Parkins MD, Rendall JC, Elborn JS. Incidence and riskfactors for pulmonary exacerbation treatmentfailures in patients with cystic fibrosis chronicallyinfected with Pseudomonas aeruginosa. Chest. 2012;141(2):485–493.

152. Bouvet GF, Maignan M, Arslanian E, Coriati A,Rabasa-Lhoret R, Berthiaume Y. Association be-tween serum YKL-40 level and dysglycemia in cysticfibrosis. Cytokine. 2015;71(2):296–301.

153. Barrio R. Management of endocrine disease: Cysticfibrosis-related diabetes: novel pathogenic insightsopening new therapeutic avenues. Eur J Endocrinol.2015;172(4):R131–R141.

154. Hillman M, Eriksson L, Mared L, Helgesson K,Landin-Olsson M. Reduced levels of active GLP-1 inpatients with cystic fibrosis with and without di-abetes mellitus. J Cyst Fibros. 2012;11(2):144–149.

155. Ntimbane T, Comte B, Mailhot G, Berthiaume Y,Poitout V, Prentki M, Rabasa-Lhoret R, Levy E. Cysticfibrosis-related diabetes: from CFTR dysfunction tooxidative stress. Clin Biochem Rev. 2009;30(4):153–177.

156. Carrington M, Krueger LJ, Holsclaw DS, Jr, IannuzziMC, Dean M, Mann D. Cystic fibrosis-related di-abetes is associated with HLA DQB1 allelesencoding Asp-57- molecules. J Clin Immunol. 1994;14(6):353–358.

157. Adler AI, Shine B, Haworth C, Leelarathna L,Bilton D. Hyperglycemia and death in cysticfibrosis-related diabetes. Diabetes Care. 2011;34(7):1577–1578.

158. Koch C, Rainisio M, Madessani U, Harms HK,Hodson ME, Mastella G, McKenzie SG, Navarro J,Strandvik B; Investigators of the European Epide-miologic Registry of Cystic Fibrosis. Presence ofcystic fibrosis-related diabetes mellitus is tightlylinked to poor lung function in patients with cysticfibrosis: data from the European EpidemiologicRegistry of Cystic Fibrosis. Pediatr Pulmonol. 2001;32(5):343–350.

159. Sims EJ, Green MW, Mehta A. Decreased lungfunction in female but not male subjects withestablished cystic fibrosis-related diabetes. DiabetesCare. 2005;28(7):1581–1587.

160. West NE, Beckett VV, Jain R, Sanders DB, Nick JA,Heltshe SL, Dasenbrook EC, VanDevanter DR,Solomon GM, Goss CH, Flume PA; STOP in-vestigators. Standardized Treatment of PulmonaryExacerbations (STOP) study: Physician treatmentpractices and outcomes for individuals with cysticfibrosis with pulmonary Exacerbations. J CystFibros. 2017;16(5):600–606.

161. Rosenecker J, Hofler R, Steinkamp G, Eichler I,Smaczny C, Ballmann M, Posselt HG, Bargon J, vonder Hardt H. Diabetes mellitus in patients withcystic fibrosis: the impact of diabetes mellitus onpulmonary function and clinical outcome. Eur J MedRes. 2001;6(8):345–350.

162. Lavie M, Fisher D, Vilozni D, Forschmidt R, Sarouk I,Kanety H, Hemi R, Efrati O, Modan-Moses D.Glucose intolerance in cystic fibrosis as a de-terminant of pulmonary function and clinical sta-tus. Diabetes Res Clin Pract. 2015;110(3):276–284.

163. Milla CE, Warwick WJ, Moran A. Trends in pul-monary function in patients with cystic fibrosiscorrelate with the degree of glucose intolerance atbaseline. Am J Respir Crit Care Med. 2000;162(3 Pt 1):891–895.

164. Lanng S, Thorsteinsson B, Nerup J, Koch C. Influenceof the development of diabetes mellitus on clinicalstatus in patients with cystic fibrosis. Eur J Pediatr.1992;151(9):684–687.

165. Terliesner N, Vogel M, Steighardt A, Gausche R,Henn C, Hentschel J, Kapellen T, Klamt S, GebhardtJ, Kiess W, Prenzel F. Cystic-fibrosis related-diabetes(CFRD) is preceded by and associated with growthfailure and deteriorating lung function. J PediatrEndocrinol Metab. 2017;30(8):815–821.

166. Leclercq A, Gauthier B, Rosner V, Weiss L, Moreau F,Constantinescu AA, Kessler R, Kessler L. Early as-sessment of glucose abnormalities during contin-uous glucose monitoring associated with lungfunction impairment in cystic fibrosis patients.J Cyst Fibros. 2014;13(4):478–484.

167. Schwarzenberg SJ, Thomas W, Olsen TW, Grover T,Walk D, Milla C, Moran A. Microvascular compli-cations in cystic fibrosis-related diabetes. DiabetesCare. 2007;30(5):1056–1061.

168. Brennan AL, Geddes DM, Gyi KM, Baker EH. Clinicalimportance of cystic fibrosis-related diabetes. J CystFibros. 2004;3(4):209–222.

169. Lecube A, Pachon G, Petriz J, Hernandez C, Simo R.Phagocytic activity is impaired in type 2 diabetesmellitus and increases after metabolic improve-ment. PLoS One. 2011;6(8):e23366.

170. Ip BC, Hogan AE, Nikolajczyk BS. Lymphocyte rolesin metabolic dysfunction: of men and mice. TrendsEndocrinol Metab. 2015;26(2):91–100.

171. Brennan AL, Gyi KM, Wood DM, Johnson J, HollimanR, Baines DL, Philips BJ, Geddes DM, Hodson ME,Baker EH. Airway glucose concentrations and effecton growth of respiratory pathogens in cystic fibrosis.J Cyst Fibros. 2007;6(2):101–109.

172. Ntimbane T, Krishnamoorthy P, Huot C, Legault L,Jacob SV, Brunet S, Levy E, Gueraud F, Lands LC,Comte B. Oxidative stress and cystic fibrosis-relateddiabetes: a pilot study in children. J Cyst Fibros. 2008;7(5):373–384.

173. Moran A, Milla C, Ducret R, Nair KS. Proteinmetabolism in clinically stable adult cystic fibrosispatients with abnormal glucose tolerance. Diabetes.2001;50(6):1336–1343.

174. Yaochite JN, Caliari-Oliveira C, Davanso MR, CarlosD, Malmegrim KC, Cardoso CR, Ramalho LN, PalmaPV, da Silva JS, Cunha FQ, Covas DT, Voltarelli JC.Dynamic changes of the Th17/Tc17 and regulatoryT cell populations interfere in the experimentalautoimmune diabetes pathogenesis. Immunobiol-ogy. 2013;218(3):338–352.

175. Chan YR, Chen K, Duncan SR, Lathrop KL, LatocheJD, Logar AJ, Pociask DA, Wahlberg BJ, Ray P, Ray A,Pilewski JM, Kolls JK. Patients with cystic fibrosishave inducible IL-17+IL-22+ memory cells in lungdraining lymph nodes. J Allergy Clin Immunol. 2013;131(4):1117–1129, 1129.e1–1129.e5.

176. Hunt WR, Helfman BR, McCarty NA, Hansen JM.Advanced glycation end products are elevated incystic fibrosis-related diabetes and correlate withworse lung function. J Cyst Fibros. 2016;15(5):681–688.

177. Widger J, Ranganathan S, Robinson PJ. Progressionof structural lung disease on CT scans in childrenwith cystic fibrosis related diabetes. J Cyst Fibros.2013;12(3):216–221.

178. Onady GM, Stolfi A. Insulin and oral agents formanaging cystic fibrosis-related diabetes. CochraneDatabase Syst Rev. 2016;4:CD004730.

179. Laguna TA, Nathan BM, Moran A. Managing di-abetes in cystic fibrosis. Diabetes Obes Metab. 2010;12(10):858–864.

180. Kolouskova S, Zemkova D, Bartosova J, Skalicka V,Sumnık Z, Vavrova V, Lebl J. Low-dose insulintherapy in patients with cystic fibrosis and early-stage insulinopenia prevents deterioration of lungfunction: a 3-year prospective study. J PediatrEndocrinol Metab. 2011;24(7-8):449–454.

181. van den Berg JM, Kouwenberg JM, Heijerman HG.Demographics of glucose metabolism in cystic fi-brosis. J Cyst Fibros. 2009;8(4):276–279.

182. Hameed S, Jaffe A, Verge CF. Cystic fibrosis re-lated diabetes (CFRD)–the end stage of progres-sive insulin deficiency. Pediatr Pulmonol. 2011;46(8):747–760.

183. Tasali E, Mokhlesi B, Van Cauter E. Obstructive sleepapnea and type 2 diabetes: interacting epidemics.Chest. 2008;133(2):496–506.

184. Punjabi NM, Polotsky VY. Disorders of glucosemetabolism in sleep apnea. J Appl Physiol (1985).2005;99(5):1998–2007.

185. Lamond N, Tiggemann M, Dawson D. Factorspredicting sleep disruption in type II diabetes. Sleep.2000;23(3):415–416.

186. Taub LF, Redeker NS. Sleep disorders, glucose reg-ulation, and type 2 diabetes. Biol Res Nurs. 2008;9(3):231–243.


REVIEWD

ownloaded from



http://dx.doi.org/10.1210/er.2017-00173


187. Cuellar NG, Ratcliffe SJ. A comparison of glycemiccontrol, sleep, fatigue, and depression in type 2diabetes with and without restless legs syndrome.J Clin Sleep Med. 2008;4(1):50–56.

188. Luyster FS, Dunbar-Jacob J. Sleep quality and qualityof life in adults with type 2 diabetes. Diabetes Educ.2011;37(3):347–355.

189. Knutson KL, Ryden AM, Mander BA, Van Cauter E.Role of sleep duration and quality in the risk andseverity of type 2 diabetes mellitus. Arch Intern Med.2006;166(16):1768–1774.

190. Lecube A, Sanchez E, Gomez-Peralta F, Abreu C,Valls J, Mestre O, Romero O, Martınez MD, SampolG, Ciudin A, Hernandez C, Simo R. Global assess-ment of the impact of type 2 diabetes on sleepthrough specific questionnaires. A case-controlstudy. PLoS One. 2016;11(6):e0157579.

191. Dixon JB, Dixon ME, Anderson ML, Schachter L,O’brien PE. Daytime sleepiness in the obese: not assimple as obstructive sleep apnea. Obesity (SilverSpring). 2007;15(10):2504–2511.

192. Brown RE, Basheer R, McKenna JT, Strecker RE,McCarley RW. Control of sleep and wakefulness.Physiol Rev. 2012;92(3):1087–1187.

193. Jordan AS, McSharry DG, Malhotra A. Adult ob-structive sleep apnoea. Lancet. 2014;383(9918):736–747.

194. Weisbrod CJ, Eastwood PR, O’Driscoll G, Green DJ.Abnormal ventilatory responses to hypoxia in type2 diabetes. Diabet Med. 2005;22(5):563–568.

195. Nishimura M, Miyamoto K, Suzuki A, Yamamoto H,Tsuji M, Kishi F, Kawakami Y. Ventilatory and heartrate responses to hypoxia and hypercapnia in pa-tients with diabetes mellitus. Thorax. 1989;44(4):251–257.

196. Duffin J. The chemoreflex control of breathingand its measurement. Can J Anaesth. 1990;37(8):933–942.

197. Duffin J. Measuring the respiratory chemoreflexesin humans. Respir Physiol Neurobiol. 2011;177(2):71–79.

198. Cheng L, Khoo MC. Modeling the autonomic andmetabolic effects of obstructive sleep apnea:a simulation study. Front Physiol. 2012;2:111.

199. Sobotka PA, Liss HP, Vinik AI. Impaired hypoxicventilatory drive in diabetic patients with auto-nomic neuropathy. J Clin Endocrinol Metab. 1986;62(4):658–663.

200. Malli F, Papaioannou AI, Gourgoulianis KI, Daniil Z.The role of leptin in the respiratory system: anoverview. Respir Res. 2010;11:152.

201. Campo A, Fruhbeck G, Zulueta JJ, Iriarte J, Seijo LM,Alcaide AB, Galdiz JB, Salvador J. Hyperleptinaemia,respiratory drive and hypercapnic response in obesepatients. Eur Respir J. 2007;30(2):223–231.

202. Bassi M, Furuya WI, Zoccal DB, Menani JV,Colombari E, Hall JE, da Silva AA, do Carmo JM,Colombari DS. Control of respiratory and car-diovascular functions by leptin. Life Sci. 2015;125:25–31.

203. Bassi M, Furuya WI, Zoccal DB, Menani JV,Colombari DS, Mulkey DK, Colombari E. Facilitationof breathing by leptin effects in the central nervoussystem. J Physiol. 2016;594(6):1617–1625.

204. Ramar K, Surani S. The relationship between sleepdisorders and stroke. Postgrad Med. 2010;122(6):145–153.

205. Deak MC, Kirsch DB. Sleep-disordered breathing inneurologic conditions. Clin Chest Med. 2014;35(3):547–556.

206. Mims KN, Kirsch D. Sleep and Stroke. Sleep Med Clin.2016;11(1):39–51.

207. Peltier AC, Consens FB, Sheikh K, Wang L, Song Y,Russell JW. Autonomic dysfunction in obstructive

sleep apnea is associated with impaired glucoseregulation. Sleep Med. 2007;8(2):149–155.

208. Young T, Palta M, Dempsey J, Skatrud J, Weber S,Badr S. The occurrence of sleep-disordered breath-ing among middle-aged adults. N Engl J Med. 1993;328(17):1230–1235.

209. Foster GD, Sanders MH, Millman R, Zammit G,Borradaile KE, Newman AB, Wadden TA, Kelley D,Wing RR, Sunyer FX, Darcey V, Kuna ST; SleepAHEAD Research Group. Obstructive sleep apneaamong obese patients with type 2 diabetes. Di-abetes Care. 2009;32(6):1017–1019.

210. Trakada GP, Steiropoulos P, Nena E, ConstandinidisTC, Bouros D. Prevalence and clinical characteristicsof obesity hypoventilation syndrome among in-dividuals reporting sleep-related breathing symp-toms in northern Greece. Sleep Breath. 2010;14(4):381–386.

211. Vgontzas AN, Legro RS, Bixler EO, Grayev A, Kales A,Chrousos GP. Polycystic ovary syndrome is asso-ciated with obstructive sleep apnea and daytimesleepiness: role of insulin resistance. J Clin EndocrinolMetab. 2001;86(2):517–520.

212. Resnick HE, Redline S, Shahar E, Gilpin A, NewmanA, Walter R, Ewy GA, Howard BV, Punjabi NM; SleepHeart Health Study. Diabetes and sleep distur-bances: findings from the Sleep Heart Health Study.Diabetes Care. 2003;26(3):702–709.

213. Lecube A, Romero O, Sampol G, Mestre O, CiudinA, Sanchez E, Hernandez C, Caixas A, Vigil L, Simo R.Sleep biosignature of type 2 diabetes: a case-controlstudy. Diabet Med. 2017;34(1):79–85.

214. Dement W, Kleitman N. The relation of eyemovements during sleep to dream activity: anobjective method for the study of dreaming. J ExpPsychol. 1957;53(5):339–346.

215. Pallayova M, Donic V, Gresova S, Peregrim I, TomoriZ. Do differences in sleep architecture exist betweenpersons with type 2 diabetes and nondiabeticcontrols? J Diabetes Sci Technol. 2010;4(2):344–352.

216. Knutson KL, Van Cauter E, Zee P, Liu K, Lau-derdale DS. Cross-sectional associations betweenmeasures of sleep andmarkers of glucose metabolismamong subjects with and without diabetes: the Cor-onary Artery Risk Development in Young Adults(CARDIA) Sleep Study. Diabetes Care. 2011;34(5):1171–1176.

217. Ekstedt M, Akerstedt T, Soderstrom M. Micro-arousals during sleep are associated with increasedlevels of lipids, cortisol, and blood pressure. Psy-chosom Med. 2004;66(6):925–931.

218. Chouchou F, Pichot V, Pepin JL, Tamisier R, Celle S,Maudoux D, Garcin A, Levy P, Barthelemy JC, RocheF; PROOF Study Group. Sympathetic overactivitydue to sleep fragmentation is associated with el-evated diurnal systolic blood pressure in healthyelderly subjects: the PROOF-SYNAPSE study. EurHeart J. 2013;34(28):2122–2131, 2131a.

219. Lavie L. Obstructive sleep apnoea syndrome–anoxidative stress disorder. Sleep Med Rev. 2003;7(1):35–51.

220. Lecube A, Sampol G, Hernandez C, Romero O,Ciudin A, Simo R. Characterization of sleepbreathing pattern in patients with type 2 diabetes:sweet sleep study. PLoS One. 2015;10(3):e0119073.

221. Ficker JH, Dertinger SH, Siegfried W, Konig HJ, PentzM, Sailer D, Katalinic A, Hahn EG. Obstructive sleepapnoea and diabetes mellitus: the role of cardio-vascular autonomic neuropathy. Eur Respir J. 1998;11(1):14–19.

222. Chen X, Walther FJ, Sengers RM, Laghmani H, SalamA, Folkerts G, Pera T, Wagenaar GT. Metforminattenuates hyperoxia-induced lung injury in neo-natal rats by reducing the inflammatory response.

Am J Physiol Lung Cell Mol Physiol. 2015;309(3):L262–L270.

223. RamadanW, Dewasmes G, PetitjeanM,WiernspergerN, Delanaud S, Geloen A, Libert JP. Sleep apnea isinduced by a high-fat diet and reversed andprevented by metformin in non-obese rats. Obesity(Silver Spring). 2007;15(6):1409–1418.

224. Hein MS, Schlenker EH, Patel KP. Altered control ofventilation in streptozotocin-induced diabetic rats.Proc Soc Exp Biol Med. 1994;207(2):213–219.

225. Kim HJ, Lee JY, Jung HS, Kim DK, Lee SM, Yim JJ,Yang SC, Yoo CG, Chung HS, Kim YW, Han SK, ShimYS, Lee CH. The impact of insulin sensitisers on lungfunction in patients with chronic obstructive pul-monary disease and diabetes. Int J Tuberc Lung Dis.2010;14(3):362–367.

226. Sexton P, Metcalf P, Kolbe J. Respiratory effects ofinsulin sensitisation with metformin: a prospectiveobservational study. COPD. 2014;11(2):133–142.

227. Vargas HA, Rondon M, Dennis R. Pharmacologicaltreatment and impairment of pulmonary functionin patients with type 2 diabetes: a cross-sectionalstudy. Biomedica. 2016;36(2):276–284.

228. Guazzi M, Oreglia I, Guazzi MD. Insulin improvesalveolar-capillary membrane gas conductance in type 2diabetes. Diabetes Care. 2002;25(10):1802–1806.

229. Lecube A, Ciudin A, Sampol G, Valladares S,Hernandez C, Simo R. Effect of glycemic control onnocturnal arterial oxygen saturation: a case-controlstudy in type 2 diabetic patients. J Diabetes. 2015;7(1):133–138.

230. Vgontzas AN, Zoumakis E, Lin HM, Bixler EO,Trakada G, Chrousos GP. Marked decrease insleepiness in patients with sleep apnea by eta-nercept, a tumor necrosis factor-alpha antagonist.J Clin Endocrinol Metab. 2004;89(9):4409–4413.

231. McGill JB, Ahn D, Edelman SV, Kilpatrick CR, SantosCavaiola T. Making insulin accessible: does inhaledinsulin fill an unmet need? Adv Ther. 2016;33(8):1267–1278.

232. Santos Cavaiola T, Edelman S. Inhaled insulin:a breath of fresh air? A review of inhaled insulin. ClinTher. 2014;36(8):1275–1289.

233. Marino MT, Costello D, Baughman R, Boss A,Cassidy J, Damico C, van Marle S, van Vliet A,Richardson PC. Pharmacokinetics and pharmaco-dynamics of inhaled GLP-1 (MKC253): proof-of-concept studies in healthy normal volunteersand in patients with type 2 diabetes. Clin PharmacolTher. 2010;88(2):243–250.

234. Steele RM, Finucane FM, Griffin SJ, Wareham NJ,Ekelund U. Obesity is associated with altered lungfunction independently of physical activity andfitness. Obesity (Silver Spring). 2009;17(3):578–584.

235. Biring MS, Lewis MI, Liu JT, Mohsenifar Z. Pulmonaryphysiologic changes of morbid obesity. Am J MedSci. 1999;318(5):293–297.

236. Chen Y, Rennie D, Cormier YF, Dosman J. Waistcircumference is associated with pulmonary func-tion in normal-weight, overweight, and obesesubjects. Am J Clin Nutr. 2007;85(1):35–39.

237. Carpio C, Santiago A, Garcıa de Lorenzo A,Alvarez-Sala R. [Changes in lung function testingassociated with obesity]. Nutr Hosp. 2014;30(5):1054–1062.

238. Brazzale DJ, Pretto JJ, Schachter LM. Optimizingrespiratory function assessments to elucidate theimpact of obesity on respiratory health. Respirology.2015;20(5):715–721.

AcknowledgmentsWe thank Jose Marıa Martı from G Comunicacion forsupport in the elaboration of figures.


REVIEWD

ownloaded from



Financial Support: This work was supported by grantsfrom Instituto de Salud Carlos III (Fondo de InvestigacionSanitaria PI06/0476, PI12/00803, and PI15/00260), the Euro-pean Union (European Regional Development Fund, FondoEuropeo de Desarrollo Regional, “Una manera de hacerEuropa”), and Fundacion Sociedad Española Endocrinologıa yNutricion. Centro de Investigacion Biomedica en Red deDiabetes y Enfermedades Metabolicas Asociadas and Centrode Investigacion Biomedica en Red de Enfermedades Res-piratorias are an initiative of the Instituto de Salud Carlos III.Correspondence and Reprint Requests: Rafael Simo,

MD, PhD, Endocrinology and Nutrition Department, Hos-pital Universitari Vall d’Hebron, Diabetes and MetabolismResearch Unit, VHIR, Passeig de la Vall d’Hebron 119-129,08035 Barcelona, Spain. E-mail: [email protected] Summary: A.L. received a grant from Novo

Nordisk that is unrelated to the submitted work. N.M.P.

received grants from ResMed and Philips Respironics that areunrelated to the submitted work. The remaining authorshave nothing to disclose.Search strategy and selection criteria: references for this

review were identified through searches of PubMed for ar-ticles published from January 1979 to May 2017 by use of theterms type 2 diabetes, pulmonary function, pulmonarydysfunction, spirometry, sleep apnea syndrome, sleepbreathing disorders, fasting plasma glucose, and glycatedhemoglobin. Selection for inclusion was based on our ex-pertise and our perception of the relevance and impact onthe field of lung diseases and T2D.

AbbreviationsAGE, advanced glycation end product; AHI, Apnea-HypopneaIndex; ARIC, Atherosclerosis Risk in Communities; AT-II, alveolar

type II cells; BL, basal lamina; BMI, body mass index; CF, cysticfibrosis; CFRD, cystic fibrosis–related diabetes; CRP, C-reactive protein; DAN, diabetic autonomic neuropathy;DLCO, diffusing capacity for carbon monoxide; FEV1,forced expiratory volume in 1 second; FPG, fastingplasma glucose; FVC, forced vital capacity; GLP-1,glucagon-like peptide 1; HbA1c, glycated hemoglobin;HOMA-IR, homeostatic model assessment of insulinresistance; IGT, impaired glucose tolerance; IR, insulin re-sistance; MVV, maximum voluntary ventilation; NGT, normalglucose tolerance; OGTT, oral glucose tolerance test; PEF, peakexpiratory flow; RAGE, receptor for advanced glycation endproduct; REM, rapid eye movement; SAS, sleep apnea syn-drome; STZ, streptozotocin; T1D, type 1 diabetes; T2D, type 2diabetes; Th, T helper; TLC, total lung capacity; TNF, tumornecrosis factor; TNF-R, tumor necrosis factor receptor; VC, vitalcapacity.


REVIEWD

ownloaded from



mailto:[email protected]

http://dx.doi.org/10.1210/er.2017-00173


Two New Targets for Type 2 Diabetes Care

Documents