Risk of Heart Failure Following Community Acquired Pneumonia… · heart failure from 6 hospitals and 7 Emergency Departments in Edmonton (Alberta, Canada) were prospectively recruited

Confidential: For Review O

nly

Risk of Heart Failure Following Community Acquired

Pneumonia: A Prospective Controlled Study with 10-Years of Follow-up

Journal: BMJ

Manuscript ID BMJ.2016.035000

Article Type: Research

BMJ Journal: BMJ

Date Submitted by the Author: 11-Aug-2016

Complete List of Authors: Eurich, Dean; University of Alberta, School of Public Health Marrie, Thomas; Dalhousie University, Minhas-Sandhu, Jashu; University of Alberta, Majumdar, Sumit; University of Alberta, Medicine

Keywords: Pneumonia, heart failure, prospective cohort, mortality

https://mc.manuscriptcentral.com/bmj

BMJ


nly

1

Risk of Heart Failure Following Community Acquired Pneumonia: A Prospective

Controlled Study with 10-Years of Follow-up

Dean T. Eurich, Professor1,2 [email protected]; Thomas J. Marrie, Professor3 [email protected]; Jasjeet K Minhas-Sandhu, Research Associate2 [email protected]; Sumit R. Majumdar, Professor1,2,4 [email protected]

1 School of Public Health, University of Alberta, Edmonton, Alberta, Canada, T6G 2G3; 2ACHORD, 2-040 Li Ka Shing Center, University of Alberta, Edmonton, Alberta, Canada, T6G 2E1;3Department of Medicine, Faculty of Medicine, Dalhousie University, Halifax, Nova Scotia, Canada B3H 4H7; 4Department of Medicine, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada T6G 2B7;

Corresponding Author and address to which reprints should be addressed:

Dean T. Eurich, 2-040 Li Ka Shing Center for Health Research Innovation, University of Alberta, Edmonton, Alberta, Canada, T6G 2E1; Phone 780-492-6333; fax 780-492-7455; email: [email protected];

Abstract: 256; Text: 2724

Key Words: Pneumonia, Heart Failure,

Running Title: Heart Failure Following Pneumonia

What this paper adds:

Section 1: What is already known on this subject

CAP has been associated with new-onset heart failure; however, the true attributable risk of CAP on heart failure incidence relative to a control population is unknown, particularly over the long-term.

Section 2: What this study adds

Our data suggests that the ten-year risk of developing incident HF following a CAP event is very high. Indeed, for every twenty patients presenting with CAP, we would expect to see one additional case of heart failure compared to patients accessing the health system without pneumonia.

Page 1 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

2

Abstract

Objective: CAP has been associated with new-onset heart failure; however, the true

attributable risk of CAP on heart failure incidence relative to a control population is

unknown, particularly over the long-term.

Design: Cohort study

Setting and Participants: Between 2000-2002, 4988 adults with CAP and no history of

heart failure from 6 hospitals and 7 Emergency Departments in Edmonton (Alberta,

Canada) were prospectively recruited and matched on age, sex, and site of treatment with

up to 5 non-pneumonia controls without prevalent heart failure (n=29,402).

Main Outcome Measure: Risk of incident heart failure hospitalization or a combined

endpoint of heart failure or death through to 2012 were evaluated using multivariable

Cox proportional hazards analyses.

Results: Average age was 55 years [35% ≥ 65 years], 2649 (53%) were male, and the

majority were managed as outpatients (63%). Over a median of 9.9 years, (592) 12% of

CAP patients had incident heart failure compared with (1712) 7% of controls (adjusted

hazard ratio (aHR) 1.61, 95%CI 1.44 to 1.81, p<0.001). CAP patients were also more

likely to achieve the combined endpoint of heart-failure or all-cause mortality than the

controls (41% vs 26%; aHR=1.53, 95% CI=1.44 to 1.63, p<0.001). The increased risk for

CAP patients was observed in both the short-term (i.e., 90 days, p=0.015) and

intermediate term (i.e., 1 year, p<0.001) following acute illness. Results were consistent

based on inpatient or outpatient treatment and robust to several sensitivity analyses.

Conclusion. Our results demonstrate that CAP patients are at high risk for the future

development of heart failure and this should be considered when formulating post-

Page 2 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

3

discharge care plans and preventive strategies.

Page 3 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

4

Introduction

Community acquired pneumonia (CAP) is a common condition associated with

substantial morbidity and mortality irrespective of age.1 CAP remains the seventh

leading cause of death in the United States, accounts for an estimated $9 billion in direct

health care costs and 600 000 admissions to hospital each year.2-4 Although often

considered an acute event, the long-term risks of CAP are substantial (>50% mortality

over 5 years and near twofold higher risk of subsequent CAP than non-CAP controls).1

Moreover, despite major advances in risk stratification and management, downstream

sequelae in survivors of pneumonia have not substantively improved over the last

decade.5

There is increasing attention between the link of CAP and subsequent

cardiovascular diseases (CVD). Several studies have reported a significant short-term

risk in major CVD following acute respiratory infections, although the longer-term

impact of CAP on CVD is less certain.6 One cardiovascular complication that seems to

be highly prevalent in patients with CAP is heart failure (HF).7-22 The mechanism by

which CAP may influence heart failure is multidimensional. Acute infections can result

in reduced myocardial function (e.g. septic shock), increased oxygen consumption,

tachycardia, and circulatory problems which may all lead to increased cardiac workload

and risk of heart failure.23-25

Overall, studies have reported incident HF rates of 1.4% in outpatient CAP

populations to as high as 24% among CAP inpatients.6-20 Key study limitations have been

Page 4 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

5

highly selected samples (cohorts derived from trial participants or from hospitalized

patients from a single medical center),6 7 9-15 17-20 inclusion of existing HF exacerbation

with new onset incident HF,6 7 9 11-20, small sample sizes (<500),7 12 13 15 16 short follow-

ups (e.g. in-hospital events only),6 7 9 11-16 20 and relatively small number of events.7 12 13 15

16 18 Moreover, the majority of studies have only focused on hospitalized patients 7 9 11-20

and none have been able to distinguish the true attributable risk associated with CAP on

HF events relative to a control population without pneumonia or prevalent HF.

Thus, it remains unclear if CAP truly increases the risk of incident HF relative to

non-CAP patients in the short or longer-term. We therefore assembled a representative

population-based cohort of patients with pneumonia and age-sex matched population

controls without pneumonia, to better understand the natural history of an episode of

pneumonia and its relative impact with respect to short and long-term incidence of HF.

Methods

Cases – Pneumonia patients.

This prospective population-based clinical registry has been extensively described in

previous publications.1 All patients >17 years of age with CAP admitted from 2000-2002

to any of the six hospitals or 7 Emergency Departments in Edmonton, Alberta, Canada,

which serves a population of approximately 1 million people, were enrolled in a clinical

registry. Subjects who had tuberculosis or cystic fibrosis, were immunocompromised

(e.g., >10mg/day prednisone equivalents), or were pregnant were excluded. All cases had

Page 5 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

6

to have radiographic confirmation of pneumonia by treating physicians and at least two

signs or symptoms of CAP (cough; pleuritic chest pain; shortness of breath; temperature

>38°C; and crackles or bronchial breath sounds). All patients were treated according to a

validated clinical CAP pathway which outlined preferred processes of care. 26 All data

were prospectively collected by trained research nurses from presentation to discharge

using standardized abstract forms. These data included demographic information,

physician-confirmed clinical and laboratory data, and prescription medication use within

the week prior to admission. Ethics approval for the study was granted by the Health

Research Ethics Board at the University of Alberta.

Matched Controls – Non-pneumonia patients.

Each CAP subject was frequency matched with up to five age (5 year age bands) and sex

matched controls. Controls were required to be alive at the time of the CAP case,

presenting at same site of care (same hospital for inpatients or same Emergency

Department for outpatients) as the case within the same month and year for a non-

pneumonia related diagnosis, and have no history of CAP in the 1 year prior based on

International Classification of Disease 9th Revision Clinical Modification codes 480-487

or ICD 10 codes J10-J18.

Outcomes

The primary outcome of interest was incident HF, which we defined as any heart-failure

related hospitalization following the initial CAP event (or matched control index event)

through to March 31st, 2012. Hospitalizations with any diagnostic code of HF were

Page 6 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

7

identified using the International Classification of Diseases, Ninth and Tenth Revision

Clinical Modification codes (ICD-9-CM 428.x; ICD-10-CM I50) and ascertained by

linking patients to comprehensive provincial healthcare administrative databases and vital

statistics file.27 The quality and validity of the provincial administrative databases are

routinely checked both provincially and federally with processes to resolve data issues

where identified. Secondarily, we evaluated the risk of incident HF in the short-term (90

days post index discharge) and intermediate term (1 year post index discharge

composite). We also evaluated a composite endpoint of “incident HF or all-cause

mortality” over the follow-up period.

Statistical Analysis

As we were most interested in new-onset heart failure post discharge, we excluded all

prevalent heart failure patients defined as any ICD-9-DM or ICD-10-CM codes for heart

failure occurring prior to or during their initial CAP or matched control visit (minimum

of 3 years prior). Furthermore, we restricted our cohort to only those cases and controls

that survived their initial index hospitalization or ED visit. Time to heart-failure related

hospitalization from the index CAP event (or matched control event) were described

using Kaplan-Meier curves.

Multivariate Cox proportional hazard regression accounting for the matched nature of

this data were used to evaluate the outcomes for CAP patients relative to controls and

presented as hazard ratios (HR) and 95% confidence intervals (CI). No violations of

proportional hazards assumptions were found via log-log plots and Schoenfeld residuals.

Page 7 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

8

Censoring occurred at departure from Alberta Health or March 31, 2012, providing a

maximum follow-up of about 11 years. Repeat events were not considered; only time to

first event was evaluated.

To ensure differences in outcomes were not driven by differences in comorbidity, all

analyses were adjusted using the 31 co-existing conditions as defined by the Elixhauser

index. The Elixhauser index conditions were calculated based on diagnostic codes within

the three years prior to the index event for cases and matched controls. Adjustments were

also made for socioeconomic status (based on government level of support provided) and

Aboriginal status, as well as the number of physician visits in the one year prior to the

index event as an additional marker of disease burden. In addition, age was included in

the models to control for any residual confounding not accounted for in the matching

process. In addition, all analyses were stratified by site of care (inpatient and outpatient)

for initial visit to ensure results were consistent among inpatients and outpatients. All

analyses were conducted using SAS V9.4 (City, State, SAS Institute Inc.).

Sensitivity Analysis

First, as the primary analysis excluded patients with new-onset heart failure during their

initial index event visit, we expanded the endpoint to include heart failure events that

occurred within the initial CAP or matched control visit. Thus, the endpoint considered

any incident heart-failure diagnosis during the index hospitalization or any subsequent

heart-failure related hospitalization as having incident heart-failure. Second, we expanded

Page 8 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

9

our outcome definition to include not only heart-failure related hospitalizations post

discharge but also included heart failure-related ED visits. For this analysis, only time to

first ED or hospitalization visit for heart failure was considered. Third, given the

relatively high rates of mortality in this older population, we re-ran our primary analyses

using a competing risk framework.28 Fourth, we specifically evaluated new-onset HF in

patients with a history of myocardial infarction or in those who experienced a myocardial

infarction during their initial index event. Finally, we specifically evaluated the risk of

new-onset HF in patients with Streptococcus pneumoniae bacteremia.15

Page 9 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

10

Results

Of the 6874 CAP patients in the cohort, 314 (5%) died in hospital, 857 (12%) had

prevalent heart failure, and 309 (5%) had a heart failure code occurring within their initial

CAP visit and were excluded from our primary analyses. Of the remaining 5394 CAP

patients, 4988 were successfully matched with at least one control (n=23060; 3676 [74%]

with 5 controls). Of the 406 unmatched CAP patients, 22 (0.5%) could not be linked to

administrative databases and 384 (7%) did not have suitable controls (Supplemental

Figure). The median follow-up was 9.9 years (IQR=4.8), with maximum follow-up being

11.4 years. CAP patients were 55 years old (SD=20) with 1762 (35%) over 65 years old,

and were mostly treated on an outpatient basis (63%). While cases and controls were

evenly distributed by sex (p=0.60), controls were slightly younger (53 vs 55 years,

p<0.001) and had fewer comorbidities (Table 1).

The primary endpoint of incident heart-failure occurred in 592 (12%) of CAP cases

compared to 1712 (7%) of controls (p<0.001); this translates into HF rates of 1.7 per 100

person years vs 0.9 per 100 person years respectively (p<0.001) (Figure 1). After

adjustment, CAP patients had a higher rate of incident heart-failure relative to the

matched controls (adjusted hazard ratio (aHR)=1.61, 95% confidence interval (CI)=1.44

to 1.81, p<0.001)(Table 2). Moreover, compared to controls, CAP patients were also

more likely to have an incident heart-failure hospitalization within 90 days post index

discharge (1.4% vs 0.6%; aHR=1.52, 95% CI=1.08 to 2.13, p=0.015) and also a higher

risk at 1 year (3.3% vs. 1.43%; aHR=1.86, 95% CI=1.50 to 2.32, p<0.001. Results were

also consistent when stratified by site of care (inpatient vs outpatient) (Table 2).

Page 10 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

11

Over the entire follow-up, 1917 (38%) CAP cases and 5509 (24%) controls died from

any-cause (p<0.001) ); this translates into mortality rates of 5.2 per 100 person years vs

2.9 per 100 person years respectively (p<0.001). A substantial number of both cases and

controls experienced the composite endpoint of incident heart-failure or all-cause

mortality with 2035 (41%) of cases and 6041 (26%) of controls having the composite

endpoint. After adjustment, CAP patients were also more likely to achieve the combined

endpoint of incident heart-failure or all-cause mortality than the control patients

(aHR=1.53, 95% CI=1.44 to 1.63, p<0.001) (Table 2). Again, increased risk of the

composite endpoint was observed for both inpatients and outpatients relative to controls

(Table 2).

Sensitivity Analyses

Sensitivity analyses were robust and showed similar results to the primary analyses

(Figure 2). After expanding the endpoint to include incident heart failure occurring within

the initial index visit, compared to their matched controls, CAP patients were still at an

increased risk of incident heart-failure related events compared to their matched controls

(777 (15%) vs 2220 (8.9%); vs aHR=1.82, 95%CI=1.65 to 2.01, p<0.001). Second, after

including incident heart failure-related emergency department visits into the outcome,

patients with CAP were still more likely to have heart failure hospitalizations (706 (14%)

vs 2 022 (8.8%) vs aHR=1.62, 95%CI=1.45 to 1.80, p<0.001). Third, our results were

similar when re-evaluated using a competing risk framework (aHR 1.37, 95%CI=1.25 to

1.51, p<0.001). Finally, compared to their matched controls (n=460), patients with

Page 11 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

12

pneumococcal bacteremia (n=98) were at significant risk of incident heart failure (13

(13.3%) vs 34 (7.4%)13%; aHR 2.4, 95%CI=1.08 to 5.3).

Discussion

In this large, population-based cohort study of CAP patients, we have highlighted the

significant increased risk of incident HF after an episode of CAP. Our data suggests that

the ten-year risk of developing incident HF following a CAP event is approximately 12%.

Further, compared to age-sex matched controls, there was over a 50% relative increase in

the risk of incident HF. This increased risk of HF following a CAP event is conferred

relatively early following discharge (within 90 days) and in many patients the occurrence

of “new” HF and CAP occur simultaneously during the same acute episode of illness.

To our knowledge, this is the longest outcomes study of CAP patients and HF reported to

date. Only one other previous study evaluated the longer-term risk of HF following an

episode of CAP. In a highly select study evaluating the effect of clarithromycin on

cardiovascular events, Schembri et al observed 3% of patients with CAP developed new

or worsening HF within one year. This is consistent with our results at one year where we

also observed 3% of CAP patients having a HF hospitalization, although among

inpatients our results were slightly higher whereby 6% were hospitalized for HF within 1

year. Our results are less consistent with a previous meta-analysis that suggested an

overall short-term risk of 14.1% among CAP inpatients.21 Although 18% of inpatients

will develop HF over the longer term, substantially fewer patients develop HF in the

short term. Indeed, by 90 days of discharge only 1.4% of patients developed HF (2%

Page 12 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

13

among inpatients) and only approximately 5% had HF when in-hospital heart failure

events were included. The discrepancy in results is likely fully explained by the fact

previous studies were conducted in highly select populations with small sample sizes and

events, and often included patients with prevalent HF.6 7 9-20 Moreover, no previous study

has been conducted relative to a control population. Indeed, although age-sex matched

controls also have significant long-term risk of HF, CAP confers a much higher risk of

HF, resulting in an absolute risk of HF of associated with CAP of 5% relative to controls.

Our study also builds on the evidence for patients with CAP managed as outpatients.

Only a single previous study has evaluated the impact of CAP on HF in patients managed

on an outpatient basis. In 907 outpatients, the PORT investigators observed an incident

rate of HF of 1.4% within 30 days.10 Conversely, within 90 days only 1% of outpatients

had incident HF in our study (with 8% developing HF over the entire follow-up). Again,

the PORT study was fairly restrictive in the spectrum of patients included in their study

and unlike our study was not population based.

The putative role of CAP on HF, and cardiac disease in general, is still evolving. CAP

raises systematic oxidative stress and inflammatory markers (e.g. circulatory cytokines),

in both the short and long term, leading increased risk of thrombogenesis, destabilization

of atherosclerotic plaques and endothelial dysfunction potentially leading to increased

rates of ischemic heart disease, atrial fibrillation, and reduced ventricular function.6 15 29

Furthermore, this increased host response to infection has been shown to occur even in

those with less severe infections and persistent long after a patient clinically recovers

Page 13 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

14

from the acute infection itself.25 Whether CAP truly causes HF per se or whether HF is

simply the endgame in the cardiac cascade triggered by acute CAP is not fully

elucidated. Other factors, which are common to both CAP and HF, are also likely at play

including advanced age, reduced renal function, and the presence of other major

comorbidities (e.g., atherosclerosis, diabetes, hypertension), and collectively all likely

contribute to the elevated risk of HF in CAP patients.6 A major limitation of our study is

we could not examine these potential mechanisms and etiologies of HF related to CAP.

Additional limitations include the fact that our study did not include certain high risk

patient groups, most notably those who are immunocompromised. Although this would

not be expected to lead directly to heart failure, immunocompromise is associated with

frailty and potentially sicker patients who may be at increased risk of complications and

chronic diseases including heart failure. Second, our diagnosis of incident heart failure

was based on administrative data as opposed to clinical data. However, heart failure

codes in administrative data have previously been shown to be highly valid with high

positive and negative predictive values as compared to medical records.30 Importantly,

both our CAP cases and controls used the same method to identify incident heart failure

and no differential bias would be expected. Last, due to the administrative coding, we are

unable to assess the degree of severity of heart failure in either of our cohorts. Moreover,

we were unable to account for some clinical differences (e.g., blood pressure) that may

have influenced heart failure events between the groups. We did adjust our models,

however, for a well-known and validated comorbidity index that would have captured

physician diagnosed hypertension. Despite these limitations, our study has several major

Page 14 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

15

strengths including the large population based sample, the use of controls to fully

evaluate the potential incremental risk of CAP on HF, and long-term follow-up data.

In conclusion, our results demonstrate that CAP patients are at very high risk for the

future development of heart failure. Indeed, for every twenty patients presenting with

CAP for either inpatient or outpatient management, we would expect to see one

additional case of heart failure compared to patients accessing the health system without

pneumonia. Whether CAP is simply a marker of a high risk population or contributes to

the underlying mechanism for the development of cardiovascular diseases and heart

failure is strongly debated.6 21 Regardless, our results clearly show the high risk nature of

these patients which will assist front-line clinicians in formulating post-discharge care

plans and preventive strategies.

Page 15 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

16

Competing Interests:

All authors have completed the ICMJE uniform disclosure form at

www.icmje.org/coi_disclosure.pdf and declare: no support from any organisation for the

submitted work; no financial relationships with any organisations that might have an

interest in the submitted work in the previous three years; no other relationships or

activities that could appear to have influenced the submitted work.

Details of contributions

DTE & JMS were responsible for the study design, statistical analysis and interpretation

of data, drafting the manuscript. All authors contributed to the study design,

interpretation of data and critical revision of manuscript. All authors, external and

internal, had full access to all of the data (including statistical reports and tables) in the

study and can take responsibility for the integrity of the data and the accuracy of the data

analysis. DTE affirms that the manuscript is an honest, accurate, and transparent account

of the study being reported; that no important aspects of the study have been omitted; and

that any discrepancies from the study as planned (and, if relevant, registered) have been

explained.

Funding

DTE receives salary support through a Canada Research Chair Award from the

Government of Canada. SRM holds the Endowed Chair in Patient Health Management

from the Faculties of Medicine and Dentistry and Pharmacy and Pharmaceutical

Sciences, all at the University of Alberta. Grants-in-aid from Capital Health, and

Page 16 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

17

unrestricted grants from Abbott Canada, Pfizer Canada, and Janssen-Ortho Canada have

been issued to TJM. All authors declare that DTE, TJM, JMS, and SRM have no non-

financial interests that may be relevant to the submitted work. The study sponsors played

no role in study design or conduct; collection, analysis, interpretation of data; writing of

the report; or in the decision to submit the paper for publication.

Data Sharing: No additional data available.

Page 17 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

18

Table 1 – Baseline Characteristics of Cases and Controls

N=28 048 Control

(n=23 060)

Case

(n=4 988)

p-value

N (%) or Mean (SD)

Age (SD) 53 (20) 55 (20) <0.001

Age Categories Under 25 26-45 46-65 66-80 Over 80

1919 (8)

7403 (32) 6684 (30) 4901 (21) 2153 (9)

373 (8)

1482 (30) 1371 (27) 1118 (22) 644 (13)

<0.001

Sex - Male 12152 (53) 2649 (53)

Welfare 1524 (6) 542 (11) <0.001 Aboriginal 807 (4) 320 (6) <0.001

Location of Treatment Outpatient Inpatient

15092 (65) 7968 (35)

3163 (63) 1825 (37)

1.00

MD visits in 1 year prior to index (SD) 11 (12) 12 (14) <0.001

Average length of follow-up in years 8.3 (4) 7.4 (4) <0.001

Median length of follow-up in years 10 (IQR=4.1) 9.6 (IQR=6.7)

Charlson Index 0.46 (1) 0.73 (2) <0.001

Elixhauser index Cardiac arrhythmia Valvular disease Pulmonary circulation disorders Peripheral vascular disorders Hypertension uncomplicated Other neurological disorders COPD Diabetes uncomplicated Diabetes complicated Hypothyroidism Renal failure Liver disease Solid tumor without metastasis Rheumatoid arthritis/collagen Obesity Fluid and electrolyte disorders Alcohol abuse Drug abuse Psychoses Depression

921 (4) 289 (1) 108 (1) 428 (2)

2303 (10) 556 (2)

1186 (5) 892 (4) 269 (1) 559 (2) 222 (1) 347 (2) 908 (4) 234 (1) 495 (2)

2096 (9) 816 (4) 551 (2) 393 (2)

1179 (5)

309 (6) 82 (2) 73 (2)

131 (3) 601 (12) 263 (5)

796 (16) 275 (6) 80 (2)

194 (4) 83 (2)

150 (3) 193 (4) 96 (2)

156 (3) 629 (13) 324 (7) 225 (5) 98 (2)

365 (7)

<0.001 0.029

<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.012

<0.001 <0.001 <0.001

0.82 <0.001 <0.001 <0.001 <0.001 <0.001 0.203

<0.001

Page 18 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

19

Table 2 – Risk of Heart Failure and /or Death Among Cases and Controls

N=28 048 Control

(n=23 060)

Case

(n=4 988)

Adjusted Hazard

Ratio with 95%

Confidence

interval

p-value

Primary Analysis (n, %)

Incident HF

Combined

Inpatients

Outpatients

1712 (7)

875 (11)

837 (5.6)

592 (12)

334 (18)

258 (8.2)

1.61 (1.44 to 1.81)

1.94 (1.64 to 2.29)

1.33 (1.12 to 1.57)

<0.001

<0.001

<0.001

Secondary Analyses (n, %)

Incident HF

within 90 days

Combined

Inpatients

Outpatients

145 (0.6)

80 (1)

65 (0.4)

68 (1.4)

41 (2)

27 (0.9)

1.52 (1.08 to 2.13)

1.45 (0.90 to 2.34)

1.38 (0.79 to 2.39)

<0.001

0.1

0.3

Incident HF

within 1 year

Combined

Inpatients

Outpatients

321 (1.4)

181 (2)

140 (1)

164 (3.3)

106 (6)

58 (2)

1.86 (1.5 to 2.32)

1.96 (1.46 to 2.64)

1.54 (1.08 to 2.20)

<0.001

<0.001

0.019

Incident HF or

All-cause Mortality

Combined

Inpatients

6041 (26%)

3172 (40)

2035 (41%)

1178 (65)

1.53 (1.44 to 1.63)

1.77 (1.62 to 1.93)

<0.001

<0.001

Page 19 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

20

Outpatients 2869 (19) 857 (27) 1.34 (1.23 to 1.47) <0.001

Page 20 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

21

Figure 1. Cumulative Incidence of Heart Failure

Legend

CAP Case

Control

Page 21 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

22

Figure 2. Adjusted Hazard Ratios for CAP Patients Relative to Controls Incident

HF

1.61

1.82

1.62

1.37

2.40

0.5 1.5 2.5 3.5 4.5 5.5HR:

Reduced Risk Increased Risk

Incident HF

Main Analyses

Competing Risk Analysis with Death

Inclusion of HF During Index Event

Sensitivty Analyses

Inclusion of Emergency Department HF

Events

Pneumococcal Bacteremia

Page 22 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

23

Roles: All authors contributed to the conception and design, DTE, SRM, JMS, TJM contributed to the analysis and interpretation of data. DTE, JMS drafted the article, all authors revised it critically for important intellectual content, and all authors provided final approval of the version to be published. DTE and JMS had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Role of Funding

DTE receives salary support through a Canada Research Chair Award. SRM holds an Endowed Chair in Patient Health Management supported by the Faculties of Medicine and Dentistry and Pharmacy and Pharmaceutical Sciences at the University of Alberta. Grants-in-aid from Capital Health, and unrestricted grants from Abbott Canada, Pfizer Canada, and Janssen-Ortho Canada have been issued to TJM. All authors declare that DTE, TJM, JMS, and SRM have no non-financial interests that may be relevant to the submitted work. The study sponsors played no role in study design or conduct; collection, analysis, interpretation of data; writing of the report; or in the decision to submit the paper for publication.

Exclusive License

The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of all authors, a worldwide licence (http://www.bmj.com/sites/default/files/BMJ%20Author%20Licence%20March%202013.doc) to the Publishers and its licensees in perpetuity, in all forms, formats and media (whether known now or created in the future), to i) publish, reproduce, distribute, display and store the Contribution, ii) translate the Contribution into other languages, create adaptations, reprints, include within collections and create summaries, extracts and/or, abstracts of the Contribution and convert or allow conversion into any format including without limitation audio, iii) create any other derivative work(s) based in whole or part on the on the Contribution, iv) to exploit all subsidiary rights to exploit all subsidiary rights that currently exist or as may exist in the future in the Contribution, v) the inclusion of electronic links from the Contribution to third party material where-ever it may be located; and, vi) licence any third party to do any or all of the above. All research articles will be made available on an Open Access basis (with authors being asked to pay an open access fee—see http://www.bmj.com/about-bmj/resources-authors/forms-policies-and-checklists/copyright-open-access-and-permission-reuse). The terms of such Open Access shall be governed by a Creative Commons licence—details as to which Creative Commons licence will apply to the research article are set out in our worldwide licence referred to above.

Page 23 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

24

1. Eurich DT, Marrie TJ, Minhas-Sandhu JK, et al. Ten-Year Mortality after Community-acquired Pneumonia. A Prospective Cohort. Am J Respir Crit Care Med 2015;192(5):597-604. doi: 10.1164/rccm.201501-0140OC

2. Huang SS, Johnson KM, Ray GT, et al. Healthcare utilization and cost of pneumococcal disease in the United States. Vaccine 2011;29(18):3398-412. doi: 10.1016/j.vaccine.2011.02.088

3. Marrie TJ, Wu L. Factors influencing in-hospital mortality in community-acquired pneumonia: a prospective study of patients not initially admitted to the ICU. Chest 2005;127(4):1260-70. doi: 10.1378/chest.127.4.1260

4. McAlister FA, Majumdar SR, Blitz S, et al. The relation between hyperglycemia and outcomes in 2,471 patients admitted to the hospital with community-acquired pneumonia. Diabetes Care 2005;28(4):810-5.

5. Wang Y, Eldridge N, Metersky ML, et al. National trends in patient safety for four common conditions, 2005-2011. N Engl J Med 2014;370(4):341-51. doi: 10.1056/NEJMsa1300991

6. Corrales-Medina VF, Musher DM, Wells GA, et al. Cardiac complications in patients with community-acquired pneumonia: incidence, timing, risk factors, and association with short-term mortality. Circulation 2012;125(6):773-81. doi: 10.1161/CIRCULATIONAHA.111.040766

7. Cabre M, Serra-Prat M, Force L, et al. Functional status as a risk factor for mortality in very elderly patients with pneumonia. Med Clin (Barc) 2008;131(5):167-70.

8. Aliberti S, Ramirez JA. Cardiac diseases complicating community-acquired pneumonia. Current opinion in infectious diseases 2014;27(3):295-301. doi: 10.1097/QCO.0000000000000055

9. Viasus D, Garcia-Vidal C, Manresa F, et al. Risk stratification and prognosis of acute cardiac events in hospitalized adults with community-acquired pneumonia. J Infect 2013;66(1):27-33. doi: 10.1016/j.jinf.2012.09.003

10. Fine MJ, Stone RA, Singer DE, et al. Processes and outcomes of care for patients with community-acquired pneumonia: results from the Pneumonia Patient Outcomes Research Team (PORT) cohort study. Arch Intern Med 1999;159(9):970-80.

11. Marrie TJ, Durant H, Yates L. Community-acquired pneumonia requiring hospitalization: 5-year prospective study. Rev Infect Dis 1989;11(4):586-99.

12. Esposito AL. Community-acquired bacteremic pneumococcal pneumonia. Effect of age on manifestations and outcome. Arch Intern Med 1984;144(5):945-8.

13. Janssens JP, Gauthey L, Herrmann F, et al. Community-acquired pneumonia in older patients. J Am Geriatr Soc 1996;44(5):539-44.

14. Fernandez-Sabe N, Carratala J, Roson B, et al. Community-acquired pneumonia in very elderly patients: causative organisms, clinical characteristics, and outcomes. Medicine (Baltimore) 2003;82(3):159-69. doi: 10.1097/01.md.0000076005.64510.87

Page 24 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

25

15. Musher DM, Rueda AM, Kaka AS, et al. The association between pneumococcal pneumonia and acute cardiac events. CID 2007;45(2):158-65. doi: 10.1086/518849

16. Becker T, Moldoveanu A, Cukierman T, et al. Clinical outcomes associated with the use of subcutaneous insulin-by-glucose sliding scales to manage hyperglycemia in hospitalized patients with pneumonia. Diabetes Res Clin Pract 2007;78(3):392-7. doi: 10.1016/j.diabres.2007.05.003

17. Adamuz J, Viasus D, Jimenez-Martinez E, et al. Incidence, timing and risk factors associated with 1-year mortality after hospitalization for community-acquired pneumonia. J Infect 2014;68(6):534-41. doi: 10.1016/j.jinf.2014.02.006

18. Yende S, Angus DC, Ali IS, et al. Influence of comorbid conditions on long-term mortality after pneumonia in older people. J Am Geriatr Soc 2007;55(4):518-25. doi: 10.1111/j.1532-5415.2007.01100.x

19. Schembri S, Williamson PA, Short PM, et al. Cardiovascular events after clarithromycin use in lower respiratory tract infections: analysis of two prospective cohort studies. BMJ 2013;346:f1235. doi: 10.1136/bmj.f1235

20. Mortensen EM, Halm EA, Pugh MJ, et al. Association of azithromycin with mortality and cardiovascular events among older patients hospitalized with pneumonia. JAMA 2014;311(21):2199-208. doi: 10.1001/jama.2014.4304

21. Corrales-Medina VF, Suh KN, Rose G, et al. Cardiac complications in patients with community-acquired pneumonia: a systematic review and meta-analysis of observational studies. PLoS Med 2011;8(6):e1001048. doi: 10.1371/journal.pmed.1001048

22. Corrales-Medina VF, Taljaard M, Fine MJ, et al. Risk stratification for cardiac complications in patients hospitalized for community-acquired pneumonia. Mayo

Clin Proc 2014;89(1):60-8. doi: 10.1016/j.mayocp.2013.09.015 23. Corrales-Medina VF, Madjid M, Musher DM. Role of acute infection in triggering

acute coronary syndromes. Lancet Infect Dis 2010;10(2):83-92. doi: 10.1016/S1473-3099(09)70331-7

24. Bazaz R, Marriott HM, Francis SE, et al. Mechanistic links between acute respiratory tract infections and acute coronary syndromes. J Infect 2013;66(1):1-17. doi: 10.1016/j.jinf.2012.09.009

25. Yende S, D'Angelo G, Mayr F, et al. Elevated hemostasis markers after pneumonia increases one-year risk of all-cause and cardiovascular deaths. PLoS One 2011;6(8):e22847. doi: 10.1371/journal.pone.0022847

26. Marrie TJ, Lau CY, Wheeler SL, et al. A controlled trial of a critical pathway for treatment of community-acquired pneumonia. CAPITAL Study Investigators. Community-Acquired Pneumonia Intervention Trial Assessing Levofloxacin. JAMA 2000;283(6):749-55. [published Online First: 2000/02/22]

27. Corrales-Medina VF, Alvarez KN, Weissfeld LA, et al. Association between hospitalization for pneumonia and subsequent risk of cardiovascular disease. JAMA 2015;313(3):264-74. doi: 10.1001/jama.2014.18229

28. Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. Journal of the American Statistical Association 1999;94(446):496-509.

Page 25 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

26

29. Van Eeden S, Leipsic J, Paul Man SF, et al. The relationship between lung inflammation and cardiovascular disease. Am J Respir Crit Care Med 2012;186(1):11-6. doi: 10.1164/rccm.201203-0455PP

30. Goff DC, Jr., Pandey DK, Chan FA, et al. Congestive heart failure in the United States: is there more than meets the I(CD code)? The Corpus Christi Heart Project. Archives of internal medicine 2000;160(2):197-202.

Page 26 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nlySupplemental Figure of Patient Flow Diagram

Page 27 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

Page 28 of 28


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Risk of Heart Failure Following Community Acquired Pneumonia… · heart failure from 6 hospitals and 7 Emergency Departments in Edmonton (Alberta, Canada) were prospectively recruited

Documents