Importance of Metabolic Acidosis as a Health Risk in Chronic Kidney Disease

Importance of Metabolic Acidosis as a Health Risk in Chronic Kidney DiseaseAnita Vincent-Johnson and Julia J. Scialla
Adv C
Human kidneys are well adapted to excrete the daily acid load from diet and metabolism in order to maintain homeostasis. In
approximately 30% of patientswithmore advanced stages of CKD, these homeostatic processes are no longer adequate, result-
ing in metabolic acidosis. Potential deleterious effects of chronic metabolic acidosis in CKD, including muscle wasting, bone
demineralization, hyperkalemia, andmore rapid progression of CKD, have been well cataloged. Based primarily upon concerns
related to nutrition and bone disease, early Kidney DiseaseOutcomesQuality Initiative guidelines recommended treatingmeta-
bolic acidosis with alkali therapy targeting a serum bicarbonate$22mEq/L. More recent guidelines have suggested similar tar-
gets based upon potential slowing of CKD progression. However, appropriately powered, long-term, randomized controlled
trials to study efficacy and safety of alkali therapy for these outcomes are largely lacking. As a result, practice among physicians
varies, underscoring the complexity of treatment of chronic metabolic acidosis in real-world CKD practice. Novel treatment ap-
proaches and rigorous phase 3 trials may resolve some of this controversy in the coming years. Metabolic acidosis is an impor-
tant complication of CKD, and where it “falls” in the priority schema of CKD care will depend upon the generation of strong
clinical evidence.
Q 2022 The Authors. Published by Elsevier Inc. on behalf of the National Kidney Foundation, Inc. This is an open access article
under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Epidemiology, Clinical trial, Bicarbonate, Kidney failure
n essential function of the kidneys is to respond to
From the Department of Medicine and University of Virginia School of Med- icine, Charlottesville, VA (A.V.-J.); and the Department of Public Health Sci- ences, University of Virginia School of Medicine, Charlottesville, VA (J.J.S.).
Financial Disclosure: J.J.S. received consulting fees for a Tricida advisory board meeting in 2019 and modest research support for clinical event activities related to trials sponsored by Sanofi and GlaxoSmithKline. A.V.-J. has nothing to disclose.
Address correspondence to Julia J. Scialla, MD, MHS, Division of Nephrology, PO Box 800133, Charlottesville, VA 22908. E-mail: JS7RK@ hscmail.mcc.virginia.edu
2022 The Authors. Published by Elsevier Inc. on behalf of the National Kidney Foundation, Inc. This is an open access article under the CC BY-NC- ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1548-5595 https://doi.org/10.1053/j.ackd.2022.05.002
Adifferences in daily acid load and maintain a rela- tively constant systemic pH. Under normal physiologic conditions, the human kidneys are typically faced with a load of nonvolatile acids that are generated by diet and daily metabolism.1 For instance, metabolism of di- etary protein typically results in net acids as sulfur- containing amino acids are oxidized and excreted as sulfate. Other foods may produce acid or base loads based on the relative content of noncombustible anions and cations. The most common modern diets typically generate a daily acid load.2 For this reason, the human kidney is well adapted to excrete acid. Ammonium (NH4
1) is produced from glutamine and glutamate pri- marily in the proximal tubule, with additional urine acidification (H1 secretion) occurring in the alpha- intercalated cells of the distal tubule.3,4 When either of these processes fail to excrete the daily acid load, meta- bolic acidosis will occur. Failure of these homeostatic mechanisms is common in CKD.5-8
PREVALENCE AND IMPACT OF METABOLIC ACIDOSIS IN CKD In the United States, over 26 million Americans have CKD.9 Of these, nearly 20%, or close to 5 million Ameri- cans, have CKD stages G3b-5.10 These advanced stages of CKD are frequently complicated by metabolic acidosis. For instance, data from theCKDPrognosis Consortium, an international collaboration conducting large meta- analyses of established cohorts, estimate that 12-20% of pa- tients with CKD stage G3b and 27-38% of patients with stage G4-5 CKD have metabolic acidosis.8 These rates are similar to those reported by other cohorts in the United States and internationally.5,11,12 However, these types of cohorts are typically focused on selected populations with qualifying health conditions or that are recruited from health care settings. The National Health and Nutri- tion Examination Survey is a representative survey of the noninstitutionalized US population. Estimates from
hronic Kidney Dis. 2022;29(4):329-336
National Health and Nutrition Examination Survey are similar, suggesting that approximately 18% of Americans with CKD stage G3b and 32% of Americans with CKD stage G4 or G5 have metabolic acidosis defined as serum bicarbonate ,22 mEq/L.7 As patients with CKD progress and approach dialysis initiation, over 50% of patients may experience metabolic acidosis.13 Certain conditions can predispose patients to metabolic acidosis earlier in CKD, a phenotype often described as a renal tubular acidosis.14,15 For instance, patients with interstitial kidney diseases, including patients with connective tissue disor- ders, such as Sjogren’s syndrome, or with sickle cell hemo- globinopathy, often exhibit metabolic acidosis at a higher glomerular filtration rate (GFR).16-19 Patients with diabetic nephropathy also typically have a higher risk of metabolic acidosis for a given GFR due to hyporeninemic hypoaldosteronism.14 These differences in serum bicarbonate level in patients with diabetes have been widely observed in CKD cohorts and large meta-analyses.8,20
Vincent-Johnson and Scialla330
POTENTIAL CONSEQUENCES OF METABOLIC ACIDOSIS IN CKD The potential effects of chronic metabolic acidosis are numerous and include impacts on muscle, bone, and the kidney, among others. Metabolic acidosis directly induces protein catabolism via effects on the ubiquitin-proteolysis pathway.21 Some small studies have shown that alkali sup- plements can improve nitrogen balance and muscle vol- ume, measured as an increase in midarm muscle circumference and lean body weight.22-24 A study of 20 patients with CKD found that measures of physical functioning, such as sit-to-stand time, improved with al- kali supplements.22 However, the impact on physical func- tioning has been less consistent with several larger studies showing no significant difference between alkali therapy and placebo including studies using clinical assessments of functioning, such as patient-reported outcomes.25,26
Bone mineralization and growth are other important tar- gets of therapy. Patients with chronic metabolic acidosis, particularly with the distal renal tubular acidosis pheno-
CLINICAL SUMMARY
Close to 5 million Americans have CKD G3b-5, and studies
suggest 12-20%of patientswith CKD stageG3b and 27-38%
with G4-5 CKD have chronic metabolic acidosis.
Chronic metabolic acidosis in CKD is associated with
potential deleterious health effects including bone
mineralization and impaired growth in children, muscle
loss, hyperkalemia, andprogression of CKD, amongothers.
Although Kidney Disease Outcomes Quality Initiative and
Kidney Disease: Improving Global Outcomes guidelines
suggest use of alkali to treat chronic metabolic acidosis,
published commentaries recognize low-quality evidence
and potential risks including worsening hypertension and
edema due to sodium loading.
Novel agents to treat metabolic acidosis and the
accompanying clinical trials may help resolve controversy
about the role of metabolic acidosis treatment in CKD care.
type, demonstrate marked calciuria resulting in nephro- lithiasis and nephrocalcino- sis.14 In children, distal renal tubular acidosis and chronic metabolic acidosis more generally associate with impaired linear growth.27,28 Metabolic acidosis promotes a shift in potassium from the intracel- lular to extracellular space and results in higher serum potassium and increased risk of clinically important hyperkalemia.29 Several small studies demonstrate small, but statistically signif- icant, reduction in serum po- tassium with alkali supplements, even when
metabolic acidosis is not present
or is only mild.22,23,26,30 These impacts may be larger for patients with more significant acidosis.29 Meta-analyses of clinical trials studying alkali interventions have concluded that these benefits, such as improving potas- sium balance, body composition or nitrogen balance, and measures of bone disease, are possible, but the overall data are limited.31
Metabolic acidosis may also impact the kidney and car- diovascular system, potentially promoting CKD progres- sion. Direct mechanisms, such as excessive activation of pathways that promote acid excretion, have been pro- posed.32 For instance, excess ammonium generation and concentration in the renal medullary interstitium has been hypothesized to activate complement and cellular injury.33 Aldosterone and endothelin, which rise in meta- bolic acidosis to increase urine acidification in the distal nephron, may promote profibrotic pathways.34 Effects
on endothelial function have been proposed based on experimental impacts of alkali on flow-mediated dilation.35
How these biological changes translate to clinical out- comes is less well understood. A large body of observa- tional data generally support associations between metabolic acidosis and CKD progression.11,20,36,37 Howev- er, there are a number of major challenges in studying the consequences of metabolic acidosis using observational data in CKD. The first is that most of these studies rely exclusively upon the serum bicarbonate concentration that can also be affected by respiratory acid-base disorders and medications often used in CKD including renin- angiotensin-aldosterone inhibitors and diuretics. A second challenge is the tight coupling of serum bicarbonate with kidney function, and thus the potential for residual con- founding. In fact, acidosis itself may be amarker of kidney tubular function that predicts progression independently from GFR.38
Modest amounts of clinical trial data are now available
Adv Chronic K
to support or refute some of the experimental and obser- vational data described above. One of the first ran- domized controlled trials was a small (n ¼ 134) open- label study published in 2009.23 The authors demon- strated a higher creatinine clearance, an indirect mea- sure of GFR, at 24 months for patients with advanced CKD and metabolic acidosis (mean baseline serum bicarbonate 20 mEq/L) than for those treated with sodium bicarbonate to a target serum bicarbonate of 23 mEq/L vs nontreated pa- tients. Fewer patients in the treatment arm required dial- ysis over follow-up. Unlike
later studies, the control group in this study did not have a prespecified serum bicarbonate level at which a “rescue” therapy was advised,22,26 and thus, clinically meaningful acidosis may have influenced initiation of dialysis. Another trial in which patients had metabolic acidosis at baseline (mean serum bicarbonate approxi- mately 18 mEq/L) showed some improvement in esti- mated GFR after 6 months of treatment with sodium bicarbonate.24 These results must be interpreted cautiously because the study was of short duration and could reflect sodium/volume loading, the study was not designed for this outcome, and GFR was estimated from serum creatinine which could be biased by the impact of acidosis on muscle. Additional trials have demonstrated potential benefits of treating metabolic acidosis on CKD progression or dialysis initiation but have been limited by small, single-center and open-label
idney Dis. 2022;29(4):329-336
Table 1. Evolution of US Clinical Practice Guidelines Related to Metabolic Acidosis in CKD
Clinical Practice Guideline Recommendations Predominant Rationale Rating of Evidence
KDOQI 2003 Guideline for Bone
Metabolism and Disease in CKD (1) “Serum levels of total CO2 [serum bi-
carbonate] should be maintained at
.22 mEq/L.”
salts should be given to achieve this
goal.”
with CKD
levels of total CO2 [serum bicarbon-
ate] should be maintained at .22
mEq/L; in neonates and young infants
below age 2, serum levels of total CO2
should be maintained at $20 mEq/L.”
(2) “If necessary, supplemental alkali
salts should be given to achieve this
goal.”
Children with CKD
corrected to at least the lower limit of
normal (22 mmol/L) in children with
CKD stages 2 to 5 and 5D.”
Preserve linear growth B (moderate)
KDIGO 2012 Guidelines for Evaluation
and Management of CKD
serum bicarbonate concentrations less
bicarbonate supplementation be given
the normal range, unless
Update 2020 (1) “In adults with CKD 3-5D, we recom-
mend reducing net acid production
(NEAP) through increased bicarbon-
lution supplementation (1C) in order
to reduce the rate of decline of resid-
ual kidney function.”
able to maintain serum bicarbonate
levels at 24-26 mmol/L (OPINION).”
Prevent GFR decline
quality evidence)
(2) OPINION
Abbreviations: GFR, glomerular filtration rate; KDIGO, Kidney Disease: Improving Global Outcomes; KDOQI, Kidney Disease Outcome Quality Initiative. Note: Total CO2 approximates serumbicarbonate. For thesemeasurements, 1mmol/L¼ 1mEq/L. CKD stages refer toG stages,with 5D reflecting stage 5 treatedwith dialysis and 5 reflecting stage 5 not treated with dialysis.
M e ta b o lic
A c id o sis
in C K D
K id n e y D is.
2 0 2 2 ;2 9 (4 ):3
2 9 -3 3 6
Vincent-Johnson and Scialla332
designs, limited reporting of outcome ascertainment, poor reporting of randomized sequence generation or lack of randomization, and lack of sodium or volume expansion control.39,40 Dietary interventions that reduce dietary acid load through augmented intake of fruits and vegetables may also improve metabolic acidosis. However, results of these interventions are challenging to interpret purely as treatments of metabolic acidosis because these complex interventions deliver many phyto- nutrients and alter a broad matrix of nutrients and foods. Interpreting delays in dialysis initiation with alkali
therapy is challenging because acidosis itself is often an indication for initiation. For instance, in the Sodium Bicarbonate Therapy for Patients with Severe Metabolic Acidaemia in the Intensive Care Unit (BICAR-ICU) study of approximately 400 patients with critical illness and acute metabolic acidosis, initiation of sodium bicar- bonate compared to control did not reduce mortality but did strongly prevent or delay the need for a renal replacement therapy.41 This impact on renal replace- ment therapy was not likely related to altering the nat- ural history of disease but to removing an acute indication for dialysis, namely severe acidosis. In the chronic setting, the use of alkali may also improve sys- temic pH and serum potassium, contributing to a delay in dialysis initiation that is due to its role as a support- ive, but not disease-modifying, treatment. This is an important distinction and will require trials to be con- ducted in the setting of more mild metabolic acidosis in moderate stages of CKD for the purpose of slowing disease progression. Meta-analyses including these studies and others
have largely concluded that there may be a modest benefit of alkali interventions on kidney function decline but recognize substantial limitations in the meta-analyzed studies.42-45 Limitations of some of the key studies include small size, single-center experience, lack of blinding, extensive dropout, evaluation of low- risk populations, and short follow-up intervals. With a dearth of large, high-quality studies, the evidence base is inadequate to clearly recommend treatment of metabolic acidosis to slow CKD progression at this time.44,45
CLINICAL PRACTICE GUIDELINES ON TREATMENT OF METABOLIC ACIDOSIS IN CKD US CKD clinical practice guidelines have advocated treating metabolic acidosis in CKD since their incep- tion. The first clinical practice guidelines in the United States focused primarily on patients with kidney failure treated with dialysis. These guidelines recommended maintaining serum bicarbonate concentration at or above 22 mEq/L in order to prevent protein catabolism and bone demineralization.46 No specific guidelines were provided for patients with CKD not requiring dialysis. In 2002-2003, the Kidney Disease Outcomes Quality Initiative (KDOQI) published guidelines for earlier stages of kidney disease.47,48 Guidelines for
Bone Metabolism and Disease also suggested maintain- ing serum bicarbonate levels at or above 22 mEq/L (Table 1).47 The rationale in these guidelines was based upon impacts on bone demineralization and induction of secondary hyperparathyroidism. Guidelines for chil- dren with CKD in 2005 and 2008 emphasized similar targets with the additional concern related to linear growth.49,50 Subsequent guideline updates on bone dis- eases did not specifically comment on metabolic acidosis as a summary recommendation.51,52 The next major update to US clinical practice guidelines com- menting on metabolic acidosis came in 2012 with the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines for the evaluation and management of CKD.53 KDIGO guidelines again suggested maintaining serum bicarbonate at or above 22 mEq/L but now stressed prevention of CKD progression and mortality, with the evidence rated as 2C (ie, suggestion based on low-quality evidence). The 2012 KDOQI commentary on the guideline generally agreed but supported focusing first on higher priority treatment targets such as hypertension and diabetes management.54 The most recent major national guideline to comment on treatment of metabolic acidosis in CKD is the new joint guideline of the KDOQI and the Academy of Nutrition and Dietetics on nutrition in CKD. This joint group rated the evidence differently, providing a stronger recommendation. The KDOQI group recommends treatment with alkali supplementation to slow CKD progression without reference to whether metabolic acidosis is present, rating this as based on low-quality evidence (1C). They also suggest higher target serum bicarbonate levels of 24-26 mEq/L based on observa- tional data and opinion (Table 1).55
A careful review of the evolution of these guidelines demonstrates a shifting focus from primarily bone and mineral metabolism and nutritional concerns to CKD progression. Caution is advised based on the generally low-quality evidence base. Most of these treatments have not been tested broadly in multicenter studies for the total balance of risks and benefits in patients with CKD. We highlight some practical challenges and need for additional evidence in the remainder of this perspective.
REAL-WORLD GAPS IN EVIDENCE AND FUTURE RESEARCH NEEDS
Case 1 A 68-year-old male presents with CKD stage G3b due to obstructive nephropathy. He has a history of paraplegia complicated by development of a chronic stage IV sacral ulcer requiring a diverting colostomy and chronic indwelling urinary catheter for wound healing. He was recently discharged following hospitalization for acute kidney injury in the setting of catheter-associated urinary tract infection and dehydration. He presents to clinic with the following pertinent laboratory history:
Adv Chronic Kidney Dis. 2022;29(4):329-336
2 wk
Prior to
eGFR (mL/min/
1.73 m2)
2
years
1
year
Metabolic Acidosis in CKD 333
There are several factors to consider in the decision about treatment with alkali supplementation for presumed metabolic acidosis. Current guidelines largely assume low serum bicarbonate is synonymous with metabolic acidosis in the setting of CKD. In patients with an acute illness, confirmation of metabolic acidosis typically re- quires an arterial blood gas although this may not always be performed. Here, both gastrointestinal and renal com- ponents may also contribute to a complex acid-base disor- der, scenarios not typically encountered in small trials conducted to date.44 A second dilemma concerns when a provider should consider alkali therapy after and during recovery from acute kidney injury. Treatment of metabolic acidosis during acute kidney injury is currently controver- sial and the subject of recent,41 as well as ongoing, multi- center trials, such as the Multicentre Evaluation of Sodium Bicarbonate in Acute Kidney Injury in Critical Care (https://www.isrctn.com/ISRCTN14027629).
Case 2 Your patient is a 34-year-old male being followed up in clinic for CKD stage G3a with subnephrotic range of pro- teinuria related to longstanding, now well-controlled, hy- pertension. Since his last visit 6 months ago, he is doing well with no major complaints. Blood pressure is 128/78. He has no edema, and he is currently treated with an angiotensin receptor blocker and diuretic. He is physically active and works full time, preferring to limit his medica- tions to once daily. His pertinent labs are shown below:
2 years ago 1 year ago Current
Blood urea nitrogen
eGFR (mL/min/
1.73 m2)
Adv Chronic Kidney Dis. 2022;29(4):329-336
At face value, this appears to be a straightforward example. Blood pressure is well controlled, meeting cur- rent targets. Serum bicarbonate has been modestly below guideline-suggested targets on several repeated assess- ments. At his young age, the opportunity to slow CKD progressionwill be particularly important to lower his life- time risk of kidney failure. During real-life decision-mak- ing, considerations may include the obligate sodium load of current alkali supplements and their potential to impact blood pressure control. Although many trials have not demonstrated clear evidence of worsened blood pressure with alkali supplementation, some studies have raised caution. The Bicarbonate Administration to Stabi- lize Estimated Glomerular Filtration Rate (BASE) trial evaluated sodium bicarbonate supplementation at a dose of 0.5-0.8 mEq/kg/d in individuals with CKD stage G3 andG4. Over 28weeks, they noted a small, but statistically significant, increase in albuminuria measured as urine albumin-to-creatinine ratio in the sodium bicarbonate compared to the placebo group.30 This finding may raise a concern about volume expansion, subtle but important increases in blood pressure, and changes in glomerular permeability induced by the use of alkali supplements. Additionally, alkali supplements are…