The Impact of Chronic Obstructive Pulmonary Disease

Original Report: Patient-Oriented, Translational Research American

Journal of

Nephrology

Am J Nephrol 2012;36:287–295 DOI: 10.1159/000342207

Received: April 2, 2012 Accepted: July 23, 2012 Published online: September 7, 2012

The Impact of Chronic Obstructive Pulmonary Disease and Smoking on Mortality and Kidney Transplantation in End-Stage Kidney Disease Brian D. Kent a, b Elhadi E. Eltayeb a Alastair Woodman a Arif Mutwali a Hoang T. Nguyen a Austin G. Stack c, d a

Regional Kidney Centre, Letterkenny General Hospital, Health Services Executive-West, Donegal, Pulmonary and Sleep Disorders Unit, St. Vincent’s University Hospital, Dublin, c Department of Medicine, University Hospital Limerick, and d Graduate Entry Medical School, University of Limerick, Limerick , Ireland

b

Key Words End-stage kidney disease ⴢ Smoking ⴢ Transplantation ⴢ Mortality

Abstract Background: Chronic obstructive pulmonary disease (COPD) and tobacco use are leading causes of morbidity and mortality. The prevalence and clinical impact of COPD on mortality and kidney transplantation among patients who begin dialysis therapy is unclear. Methods: We explored the clinical impact of COPD and continued tobacco use on overall mortality and kidney transplantation in a national cohort study of US dialysis patients. National data on all dialysis patients (n = 769,984), incident between May 1995 and December 2004 and followed until October 31, 2006, were analyzed from the United States Renal Data System. Prevalence and period trends were determined while multivariable Cox regression evaluated relative hazard ratios (RR) for death and kidney transplantation. Results: The prevalence of COPD was 7.5% overall and increased from 6.7 to 8.1% from 1995– 2004. COPD correlated significantly with older age, cardiovascular conditions, cancer, malnutrition, poor functional status, and tobacco use. Adjusted mortality risks were significantly higher for patients with COPD (RR = 1.20, 95% CI

© 2012 S. Karger AG, Basel 0250–8095/12/0363–0287$38.00/0 Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/ajn

1.18–1.21), especially among current smokers (RR = 1.28, 95% CI 1.25–1.32), and varied inversely with advancing age. In contrast, the adjusted risks of kidney transplantation were significantly lower for patients with COPD (RR = 0.47, 95% CI 0.41–0.54, for smokers and RR = 0.54, 95% CI 0.50–0.58, for non-smokers) than without COPD [RR = 0.72, 95% CI 0.70– 0.75, for smokers and RR = 1.00 for non-smokers (referent category)]. Conclusions: Patients with COPD who begin dialysis therapy in the US experience higher mortality and lower rates of kidney transplantation, outcomes that are far worse among current smokers. Copyright © 2012 S. Karger AG, Basel

Introduction

Chronic obstructive pulmonary disease (COPD) is a leading global cause of morbidity and disability. By 2020, COPD is estimated to be the third leading cause of death worldwide [1]. Furthermore, it represents a significant The study was performed at the Regional Kidney Centre, Department of Medicine, Letterkenny General Hospital, Donegal, and Graduate Entry Medical School, Faculty of Education and Health Sciences, University of Limerick, Limerick, Ireland.

Austin G. Stack, MD, MSc, FRCPI Department of Medicine, Clinical Academic Liaison Building St. Nessans Road, Graduate Entry Medical School (GEMS), University of Limerick Limerick (Ireland) Tel. +353 61 234 959, E-Mail austin.stack @ ul.ie

economic burden to society – direct and indirect costs were estimated to amount to approximately 32 billion USD in the US alone in 2002 [2]. Emerging evidence suggests that chronic kidney disease is highly prevalent in populations with COPD and when present contributes independently to increased mortality [3]. However, there is little data available regarding the burden of COPD among patients with advanced chronic kidney disease. Rarer still are studies that have evaluated the contribution of COPD to overall mortality among those with endstage kidney disease (ESKD). Mortality rates in patients who develop ESKD are between 10- to 1,000-fold higher than in the general population [4]. The cause of this excess mortality is unclear, but evidence suggests that it is in part attributable to advancing age, diabetes, underlying cardiovascular conditions, and poor nutritional status as well as several dialysis-related factors [5–7]. Although it is speculated that COPD is also a major mortality determinant in this population, the relative extent of its contribution to overall patient survival is unclear. Similarly, little data is available on the adverse impact of tobacco use on patients with pre-existing COPD and whether continued smoking contributes synergistically to overall mortality in this highrisk population [8]. The purpose of this study was (1) to quantify the prevalence of COPD in a national dialysis cohort and evaluate period trends in successive cohorts from 1995 to 2004; (2) to evaluate the impact of COPD on mortality and kidney transplantation, and (3) to determine whether current tobacco use modifies the impact of COPD on these clinical outcomes.

Methods Data We tested our hypotheses in a historical prospective cohort of patients who commenced dialysis in the US between May 1995 and December 2004 using data from the standard analysis files of the United States Renal Data System. These data are derived from the Centers for Medicare and Medicaid Services (CMS) Medical Evidence Form, a government document that is completed for all new dialysis patients in the US [9]. The CMS Medical Evidence Form captured data on demographic details, comorbid conditions, baseline laboratory values, and dates of dialysis initiation. The presence or absence of clinician-diagnosed COPD was recorded at the onset of dialysis therapy. Additional covariates representing the following variables were included for our analyses: age, gender, race, the presence or absence of diabetes, hypertension, coronary artery disease, congestive heart failure, history of cardiac arrhythmia or cardiac arrest, cerebrovascular disease, current tobacco use, malignant neoplasm or cancer, and acquired

288

Am J Nephrol 2012;36:287–295

immunodeficiency syndrome (AIDS). Body mass index was estimated from recorded weight and height measurements and residual renal function at dialysis initiation was estimated from the Modification of Diet in Renal Disease formula [10]. The study had approval from the Ethics Committee of Letterkenny General Hospital. Statistical Analysis The prevalence of COPD overall and by gender was estimated for each annual incident cohort and compared across years using ␹2 analysis. Multivariate logistic regression was performed to determine the independent relationships of demographic, clinical, laboratory, lifestyle variables and elements of pre-dialysis care with COPD. These associations were represented by adjusted odds ratios (AOR) and 95% confidence intervals (CI). Multivariate Cox regression explored the relationship of COPD with mortality and kidney transplantation in separate analyses while adjusting for potential confounders. Patients were followed up until death, kidney transplantation, loss to follow-up, or the end of the study period (October 2006). Both unadjusted and adjusted hazard ratios were calculated and models of increased complexity were constructed to explore associations of COPD with mortality and kidney transplantation. In the final model, we adjusted for a comprehensive set of clinical, laboratory, lifestyle, functional, and pre-dialysis factors. Interactions between COPD and other exposure variables were tested for in the multivariable Cox model with a p value !0.01 determined as significant. Where a significant interaction was found, the stratumspecific relative hazard ratios (RR) for each subgroup were determined. Statistical analysis was performed using SAS statistical software (version 9.1.2, SAS institute, Cary, N.C., USA).

Results

Baseline Characteristics The study cohort included 769,984 individuals who commenced dialysis between May 1995 and December 2004. The mean age of subjects was 62.7 years, 54% were male, and the majority were white. Subjects with COPD were older, were more frequently males and whites, and had a greater prevalence of cardiovascular conditions than those without (table 1). Similarly, current tobacco use, a history of alcohol dependence, and poor mobility were more common among those with COPD. Prevalence and Correlates of COPD In total, 57,436 (7.5%) individuals commenced dialysis with a diagnosis of COPD. There was a significant increase in prevalence from 6.7% in 1995 to 8.1% in 2004 (fig. 1). Overall, COPD was more prevalent among men than women and increased significantly over time but disproportionately more in females (fig. 2). COPD was strongly related to advancing age (table 2). Several cardiovascular conditions were significantly associated with COPD including heart Kent /Eltayeb /Woodman /Mutwali / Nguyen /Stack

Table 1. Baseline characteristics of patients with and without COPD at dialysis onset in the US

Demographics Age, years Male gender, % Race, % White Black Asian Comorbid conditions, % Diabetes (cause of ESKD) Diabetes (cause and comorbid) Coronary disease Myocardial infarction Cerebrovascular disease Heart Failure Peripheral vascular disease Cardiac dysrhythmia Hypertension AIDS Malignancy BMI Lifestyle factors, % Alcohol dependence Tobacco use Functional status, % Inability to walk independently Inability to transfer independently Laboratory variables Serum Creatinine, mg/dl eGFR, ml/min/1.73 m2 Albumin, mg/l Hematocrit, % Pre-ESRD care, % Erythropoietin use pre-dialysis Dialysis modality, % Peritoneal dialysis Hemodialysis

All (n = 769,984)

With COPD (n = 57,436)

Without COPD (n = 712,548)

62.9815.2 53.8

69.5810.9 60.5

62.4815.4* 53.2*

65.7 29.6 3.6

80.6 17.4 1.3

64.5* 30.6* 3.8*

45.4 52.9 27.1 8.8 9.4 32.1 14.5 6.0 77.1 1.5 5.8 26.486.2

38.8 49.1 50.0 18.8 16.0 57.5 30.8 14.8 82.3 1.6 9.3 25.786.5

45.9* 53.2* 25.2* 8.0* 8.8* 30.0* 13.2* 5.3* 76.7* 1.5 5.5* 26.486.2*

1.4 5.3

2.9 17.8

1.3* 4.3*

4.1 1.5

8.0 3.0

3.8* 1.3*

7.482.9 8.183.2 3.280.7 29.585.4

6.582.5 8.883.3 3.180.6 30.085.2

7.482.9* 8.083.2* 3.280.7* 29.485.4*

29.5

27.6

29.6*

7.8 92.2

4.2 95.8

8.1* 91.9*

Values are means 8 SD unless otherwise indicated. eGFR = Estimated glomerular filtration rate, based on the abbreviated Modification of Diet in Renal Disease Study equation. * p < 0.0001 for all bivariate comparisons.

failure (AOR = 2.15), coronary disease (AOR = 1.49), cardiac arrhythmia/arrest (AOR = 1.47), peripheral vascular disease (AOR = 1.56), and stroke (AOR = 1.21). Patients with cancer (AOR = 1.34), AIDS (AOR = 1.17), and poor functional status [inability to walk (AOR = 1.32) and transfer (AOR = 1.20) independently] were more likely to have COPD, but the strongest independent correlate was current tobacco use (AOR = 5.38). The AOR of COPD increased progressively from 1995 to 2004 (p ! 0.001).

Association of COPD and Smoking with Mortality The overall mortality at the end of the study period was significantly higher in the COPD group (83.1 vs. 63.5%; p ! 0.001) with most deaths ascribed to cardiovascular causes (36.5 vs. 27.2%; p ! 0.001; table 3). In the unadjusted model, mortality risks were significantly higher among patients with COPD than among those without (RR = 1.77, 95% CI 1.75–1.78; table 4). With adjustment for demographic factors, the magnitude of this relation-

COPD and Outcomes in Kidney Failure

Am J Nephrol 2012;36:287–295

289

12

Prevalence (%)

10 8

7.4

6.7

6.9

7.2

7.2

1995

1996

1997

1998

1999 2000 Incident year

8.3

8.1

8.1

7.1

7.7

7.7

8.0

8.1

2001

2002

2003

2004

8.6

8.5

8.6

8.7

6 4 2 0

Fig. 1. The Prevalence of COPD among in-

cident dialysis patients in the United States by calendar year 1995–2004. ␹2 p ! 0.0001.

12

Male Female

Prevalence (%)

10 8 6

8.0

7.8

5.5

5.7

6.0

6.3

8.1

6.6

6.6

7.3

6.8

7.4

6.0

4 2 0

Fig. 2. The Prevalence of COPD among

1995

men and women at dialysis initiation from 1995–2004. ␹2 p ! 0.0001.

ship was reduced (RR = 1.42, 95% CI 1.41–1.44). With further adjustment for other known mortality predictors, the risk was further attenuated but remained significant (RR = 1.20, 95% CI 1.18–1.21). The magnitude of mortality risk varied inversely with age, with the greatest magnitude of risk in younger patients (age 18–39 years, RR = 1.35, 95% CI 1.17–1.56) and the least risk in older patients (age 70–79 years, RR = 1.19, and age 80–89 years, RR = 1.15). The adjusted mortality risks of COPD varied significantly by current smoking status as illustrated in table 4. Compared to patients without COPD and who did not smoke (RR = 1.00, referent group), patients with COPD who continued to smoke experienced the greatest mortality risks (RR = 1.28, 95% CI 1.25–1.32). The risks were 290

Am J Nephrol 2012;36:287–295

1996

1997

1998

1999 2000 Incident year

2001

2002

2003

2004

modestly lower but still significant for COPD patients who did not currently smoke (RR = 1.21, 95% CI 1.20– 1.23). Similarly, patients classified as not having COPD at dialysis initiation who continued to smoke experienced a significantly elevated mortality risk (RR = 1.16, 95% CI 1.14–1.18) compared to non-smokers without COPD. Association of COPD and Smoking with Kidney Transplantation Overall transplantation rates were substantially lower among those with COPD compared to those without (2.4 vs. 12.8%; p ! 0.001; table 3). In the unadjusted model, the likelihood of a kidney transplantation was significantly lower for patients with COPD compared to those without (RR = 0.26, 95% CI 0.25–0.28). With adjustment for conKent /Eltayeb /Woodman /Mutwali / Nguyen /Stack

founders, the risk of kidney transplantation improved modestly but remained significantly and substantially lower than among those without COPD (RR = 0.56, 95% CI 0.52–0.60). Across age categories, the risk of kidney transplantation decreased with advancing age (table 5) with patients in the oldest age groups having the least opportunity to receive a transplant. Transplantation risks also varied significantly by current smoking status (table 5). Compared with patients without COPD and who did not smoke (RR = 1.00, referent category), the adjusted risk of a kidney transplantation was significantly lower for non-COPD smokers (RR = 0.72, 95% CI 0.70–0.75) and lowest for COPD patients (RR = 0.47, 95% CI 0.41– 0.54, for smokers and RR = 0.54, 95% CI 0.50–0.58, for non-smokers).

Discussion

Table 2. Factors associated with the presence of COPD at dialysis

onset among US dialysis patientsa Demographics

AOR (95% CI)

Age 18–39 years (referent) 40–49 years 50–59 years 60–69 years 70–79 years ≥80 years Gender Male vs. female Race White (referent) Black Asian Native American Comorbid conditions Heart failure, yes vs. no Coronary disease, yes vs. no Arrhythmia/cardiac arrest, yes vs. no Peripheral vascular disease, yes vs. no Stroke, yes vs. no Hypertension, yes vs. no Cancer, yes vs. no AIDS, yes vs. no Lifestyle factors Current tobacco use, yes vs. no Alcohol dependence, yes vs. no Functional Status Inability to transfer, yes vs. no Inability to walk, yes vs. no Erythropoietin use prior to dialysis yes vs. no Calendar period 1995–1998 (referent) 1999–2001 2002–2004

1.00 1.67 (1.56–1.80) 3.05 (2.86–3.24) 4.65 (4.38–4.94) 5.27 (4.96–5.60) 4.35 (4.08–4.63) 1.18 (1.16–1.20) 1.00 0.84 (0.84–0.85) 0.63 (0.61–0.65) 0.63 (0.57–0.70) 2.15 (2.11–2.19) 1.49 (1.45–1.51) 1.47 (1.43–1.51) 1.56 (1.53–1.59) 1.21 (1.18–1.24) 1.19 (1.16–1.21) 1.34 (1.30–1.38) 1.17 (1.09–1.26)

In this nationally representative study, we highlight the substantial burden of COPD at dialysis initiation and demonstrate a worrying trend of increasing prevalence in a contemporary cohort. The prevalence increased in men and women, although disproportionately more so in women. We found that several cardiovascular conditions and lifestyle characteristics were significantly correlated with COPD and that continued tobacco use had by far the strongest association. Importantly, we demonstrated that both COPD and continued smoking were independent and significant predictors of overall mortality among new dialysis patients. Patients with COPD who continued to smoke experienced the highest mortality risks. Equally important, a diagnosis of COPD at dialysis initiation predicted lower risks of kidney transplantation. These findings demonstrate that COPD is an important contributor to overall poor outcomes for patients who begin dialysis therapy in the US. It confirms a synergistic and detrimental impact of continued tobacco use on patient survival and kidney transplantation among all patients who reach ESKD. We found that the prevalence of COPD increased significantly over the study period among men and women. The rise in prevalence rates among both sexes may well be explained by improved survival of COPD in the general population over time [11], but also suggests greater acceptance of overall sicker patients for dialysis therapy by nephrologists. It is also tempting to speculate that increased use of tobacco over this period was a major contributory factor, although adjustment for trends in tobacco use did not explain the observed rise in COPD prevalence. Consistent with previous studies, we found that

COPD was more common in men than women but, quite surprisingly, we found that the rate of increase was greater in women than men over the study period. Whether this can be explained by differential rates of tobacco use between the sexes, improvements in disease recognition among women or improved survival of women over men with COPD prior to dialysis initiation is unclear [12, 13].


Am J Nephrol 2012;36:287–295

5.38 (5.24–5.53) 1.60 (1.51–1.69) 1.20 (1.12–1.27) 1.32 (1.26–1.37) 0.89 (0.87–0.91) 1.00 1.06 (1.04–1.09) 1.16 (1.14–1.19)

a Multivariable analysis of factors associated with COPD at dialysis initiation. Model included demographic, clinical, biochemical, lifestyle and functional status indicators measured at dialysis onset as well as grouped calendar periods. n = 769,984 for adjusted model.

291

Table 3. Overall mortality, cause-specific mortality, and transplantation outcomes among patients with and without COPD who began

dialysis treatment in the US

Deaths Cause of death Cardiovascular Cerebrovascular Infection Malignancy Other Unknown Transplantation Lost to follow-up

Total (n = 769,984)

COPD (n = 57,436)

No COPD (n = 712,548)

499,926 (64.9)

47,720 (83.1)

452,206 (63.5)

214,452 (27.9) 20,672 (2.7) 64,168 (8.3) 20,144 (2.6) 78,909 (10.3) 20,476 (2.7) 92,770 (12.1) 6,251 (0.81)

20,984 (36.5) 1,481 (2.6) 5,645 (9.8) 1,945 (3.4) 8,559 (14.9) 1,804 (3.1) 1,375 (2.4) 363 (0.63)

193,468 (27.2)* 19,191 (2.7) 58,523 (8.2)* 18,199 (2.6)* 70,350 (9.9)* 18,672 (2.6)* 91,395 (12.8)* 5,888 (0.83)*

Figures in parentheses are percentages. * p < 0.0001 for all bivariate comparisons.

Table 4. Mortality RR for COPD among new dialysis patients in the US

COPD RR (95% CI)

Without COPD (referent)

All patients Unadjusted Demographic-adjusteda Adjusted for all covariatesb

1.77 (1.75–1.78) 1.42 (1.41–1.44) 1.20 (1.18–1.21)

1.00 1.00 1.00

Age groupc 18–39 years 40–49 years 50–59 years 60–69 years 70–79 years 80–89 years

1.35 (1.17–1.56) 1.35 (1.26–1.44) 1.25 (1.21–1.30) 1.26 (1.23–1.28) 1.19 (1.17–1.21) 1.15 (1.12–1.19)

1.00 1.00 1.00 1.00 1.00 1.00

Group

COPD

Unadjusted Demographic-adjusteda Adjusted for all covariatesb

smokers RR (95% CI)

non-smokers RR (95% CI)

smokers RR (95% CI)

non-smokers (referent)

1.59 (1.55–1.62) 1.51 (1.48–1.55) 1.28 (1.25–1.32)

1.80 (1.78–1.82) 1.42 (1.40–1.43) 1.21 (1.20–1.23)

0.90 (0.89–0.91) 1.21 (1.20–1.23) 1.16 (1.14–1.18)

1.00 1.00 1.00

a Demographic variables of age, gender, and race. b Final multivariable model adjusted for demographic (age, sex, and race), clinical (diabetes, hypertension, coronary disease, heart failure, cardiac arrhythmia/arrest, peripheral vascular disease, cerebrovascular disease, pulmonary disease, cancer, AIDS, tobacco use,

292

Without COPD

Am J Nephrol 2012;36:287–295

BMI, difficulty in walking and transfers, erythropoietin use before dialysis, and alcohol use) and laboratory factors (serum albumin, hematocrit, and eGFR at dialysis initiation). c Each age group modeled in a separate Cox model.

Kent /Eltayeb /Woodman /Mutwali / Nguyen /Stack

Table 5. Kidney transplantation RR by COPD status among new dialysis patients in the US

COPD RR (95% CI)

Without COPD (referent)

All patients Unadjusted Demographic-adjusteda Adjusted for all covariatesb

0.26 (0.25–0.28) 0.41 (0.39–0.43) 0.56 (0.52–0.60)

1.00 1.00 1.00

Age groupc 18–39 years 40–49 years 50–59 years 60–69 years 70–79 years 80–89 years

0.69 (0.56–0.86) 0.57(0.49–0.67) 0.57 (0.51–0.64) 0.52 (0.46–0.59) 0.45 (0.34–0.61) 0.35 (0.46–2.70)

1.00 1.00 1.00 1.00 1.00 1.00

Subgroup

COPD

Unadjusted Demographic-adjusteda Adjusted for all covariatesb

Without COPD

smokers RR (95% CI)

non-smokers RR (95% CI)

smokers RR (95% CI)

non-smokers (referent)

0.31 (0.28–0.35) 0.36 (0.32–0.40) 0.47 (0.41–0.54)

0.25 (0.24–0.28) 0.41 (0.35–0.44) 0.54 (0.50–0.58)

0.95 (0.92–0.98) 0.67 (0.65–0.70) 0.72 (0.70–0.75)

1.00 1.00 1.00

a Demographic variables of age, gender, and race. b Final multivariable model adjusted for demographic (age, sex, and race), clinical (diabetes, hypertension, coronary disease, heart failure, cardiac arrhythmia/arrest, peripheral vascular disease, cerebrovascular disease, pulmonary disease, cancer, AIDS, tobacco use,

BMI, difficulty in walking and transfers, erythropoietin use before dialysis, and alcohol use), and laboratory factors (serum albumin, hematocrit, and eGFR at dialysis initiation). c Each age group modeled in a separate Cox model.

What is striking from this study is the negative impact of both COPD and smoking on risks of kidney transplantation and overall mortality. The mortality risks associated with COPD were equal in magnitude to that of coronary disease or heart failure. Similarly, COPD conferred a detrimental effect on the opportunities for kidney transplantation. We highlighted two important interactions. First, the detrimental impact of COPD on mortality varied across age groups with younger patients experiencing the greatest risk. Second, continued use of tobacco among patients with established COPD at dialysis initiation was associated with significantly poorer outcomes compared to those who had ceased. There are several potential mechanisms through which COPD may contribute to the excess mortality of dialysis patients. First, studies in the general population have demonstrated strong independent associations between COPD and risk of myocardial infarction and stroke [14, 15]. The finding of increased cardiovascular deaths among COPD patients in our analysis strongly suggests

that accelerated cardiovascular disease may be a potential mechanism in dialysis patients. Second, it is also possible that the excess mortality in COPD patients commencing dialysis may in part be mediated by a greater underlying burden of cardiovascular conditions [16]. In our analysis, a diagnosis of COPD was strongly associated with the presence of known cardiovascular disease. Third, tobacco use is an obvious potential link between COPD, dialysis, and mortality. A history of smoking is associated with an increased risk of progression of kidney disease, while ongoing tobacco use is an independent predictor of morbidity and mortality in dialysis cohorts [8, 17, 18]. Although the overall mechanism of increased mortality risk associated with COPD is unclear, the primary pathologic process may well be systemic inflammation. Inflammatory markers are elevated in subjects with COPD and predict future hospitalization and increased mortality [19, 20]. Similarly, among dialysis cohorts, elevations in C-reactive protein are common and also predict future cardiovascular death [21, 22]. It is conceivable


Am J Nephrol 2012;36:287–295

293

that chronic activation of the inflammatory response in patients with COPD contributes to the development of conditions such as malnutrition and anemia, both of which are common in dialysis patients and predict future mortality [23]. The findings reported here should be considered in the context of an observational study and its inherent limitations. A potential concern is the accurate identification of COPD patients among this cohort by data recorded in the CMS Medical Evidence Form. Previous comparisons of prevalence between this form and the Choices for Health Outcomes in caring for ESRD (CHOICE) cohort study found a significantly higher prevalence of COPD (6.0 vs. 17%, respectively) and suggested an underreporting based on medical chart abstraction [24]. On the other hand, it is possible that authors of the CHOICE study may have inadvertently overinflated prevalence estimates due to inclusion of other lung disease categories such as asthma and interstitial lung disease. We admit, however, that the present study and the CHOICE study are no substitute for well-designed prospective studies that could accurately confirm COPD based on clinical findings as well as functional studies of respiratory function. We also recognize the limitations of observational studies in identifying cause and effect associations. Although we adjusted for several well-known predictors of mortality, other factors omitted from the model may completely or partly explain the effect of COPD on subsequent mortality. Despite these, there are several strengths to this study, namely its large sample size, the capture of successive incident dialysis cohorts over a 10year period, and the presence of several demographic, clinical and laboratory and lifestyle characteristics for adjustment. Finally, the outcome measurements are based

on well-validated instruments that have been used in previous studies of this nature [4–9]. In conclusion, the present study highlights the dual importance of COPD and tobacco use among new dialysis patients. Both contributed independently and synergistically to higher mortality and lower success of kidney transplantation. The detrimental impact of COPD was similar for men and women, varied inversely with age, and was significantly greater for patients who continued to smoke compared to those who did not. This study suggests the need for greater recognition of COPD as a common disease at dialysis initiation and predictor of poor clinical outcomes. It also reaffirms a key message for all healthcare providers that smoking cessation strategies should be relentlessly pursued in all patients who continue to smoke on dialysis. Such efforts could yield tangible long-term improvements in outcomes for both COPD and non-COPD subjects commencing renal replacement therapy.

Acknowledgments A.G.S. and H.T.N. had full access to all of the data in the study and take responsibility for the integrity of the data and data analysis. All authors collaborated on the study design, analysis, editing and final manuscript. The data reported here have been supplied by the United States Renal Data System. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as the official policy or interpretation of the US Government.

Disclosure Statement All authors declare that they have no conflicts of interest.

References 1 Murray CJ, Lopez AD: Alternative projections of mortality and disability by cause 1990–2020: Global Burden of Disease Study. Lancet 1997;349:1498–1504. 2 Rabe KF, Hurd S, Anzueto A, Barnes PJ, Buist SA, Calverley P, Fukuchi Y, Jenkins C, Rodriguez-Roisin R, van Weel C, Zielinski J: Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med 2007; 176: 532–555. 3 Incalzi RA, Corsonello A, Pedone C, Battaglia S, Paglino G, Bellia V: Chronic renal fail-

294

ure: a neglected comorbidity of COPD. Chest 2010;137:831–837. 4 USRDS System: USRDS 2009 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States. National Institutes of Health, Bethesda, MD, 2009. 5 Stack AG, Bloembergen WE: Prevalence and clinical correlates of coronary artery disease among new dialysis patients in the United States: a cross-sectional study. J Am Soc Nephrol 2001; 12:1516–1523. 6 Leavey SF, Strawderman RL, Jones CA, Port FK, Held PJ: Simple nutritional indicators as

Am J Nephrol 2012;36:287–295

independent predictors of mortality in hemodialysis patients. Am J Kidney Dis 1998; 31:997–1006. 7 Ganesh SK, Hulbert-Shearon T, Port FK, Eagle K, Stack AG: Mortality differences by dialysis modality among incident ESRD patients with and without coronary artery disease. J Am Soc Nephrol 2003; 14: 415– 424. 8 Wagner M, Ansell D, Kent DM, Griffith JL, Naimark D, Wanner C, Tangri N: Predicting mortality in incident dialysis patients: an analysis of the United Kingdom Renal Registry. Am J Kidney Dis 2011;57:894–902.

Kent /Eltayeb /Woodman /Mutwali / Nguyen /Stack

9 Chapter IV. Patient characteristics at the start of ESRD: data from the HCFA medical evidence form. Am J Kidney Dis 1999; 34(2 suppl 1):S63–S73. 10 Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D: A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461–470. 11 Almagro P, Salvado M, Garcia-Vidal C, Rodriguez-Carballeira M, Delgado M, Barreiro B, Heredia JL, Soriano JB: Recent improvement in long-term survival after a COPD hospitalisation. Thorax 2010;65:298–302. 12 Han MK, Postma D, Mannino DM, Giardino ND, Buist S, Curtis JL, Martinez FJ: Gender and chronic obstructive pulmonary disease: why it matters. Am J Respir Crit Care Med 2007;176:1179–1184. 13 Soriano JB, Maier WC, Egger P, Visick G, Thakrar B, Sykes J, Pride NB: Recent trends in physician diagnosed COPD in women and men in the UK. Thorax 2000;55:789–794. 14 Sin DD, Man SF: Chronic obstructive pulmonary disease as a risk factor for cardiovascular morbidity and mortality. Proc Am Thorac Soc 2005;2:8–11.


15 Donaldson GC, Hurst JR, Smith CJ, Hubbard RB, Wedzicha JA: Increased risk of myocardial infarction and stroke following exacerbation of COPD. Chest 2010;137:1091– 1097. 16 Cheung AK, Sarnak MJ, Yan G, Berkoben M, Heyka R, Kaufman A, Lewis J, Rocco M, Toto R, Windus D, Ornt D, Levey AS: Cardiac diseases in maintenance hemodialysis patients: results of the HEMO Study. Kidney Int 2004; 65:2380–2389. 17 Stack AG, Murthy BV: Cigarette use and cardiovascular risk in chronic kidney disease: an unappreciated modifiable lifestyle risk factor. Semin Dial 2010;23:298–305. 18 Liebman SE, Lamontagne SP, Huang LS, Messing S, Bushinsky DA: Smoking in dialysis patients: a systematic review and metaanalysis of mortality and cardiovascular morbidity. Am J Kidney Dis 2011; 58: 257– 265. 19 Pinto-Plata VM, Mullerova H, Toso JF, Feudjo-Tepie M, Soriano JB, Vessey RS, Celli BR: C-reactive protein in patients with COPD, control smokers and non-smokers. Thorax 2006;61:23–28. 20 Dahl M, Vestbo J, Lange P, Bojesen SE, Tybjaerg-Hansen A, Nordestgaard BG: C-reactive protein as a predictor of prognosis in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2007;175:250–255.

21 Owen WF, Lowrie EG: C-reactive protein as an outcome predictor for maintenance hemodialysis patients. Kidney Int 1998; 54: 627–636. 22 deFilippi C, Wasserman S, Rosanio S, Tiblier E, Sperger H, Tocchi M, Christenson R, Uretsky B, Smiley M, Gold J, Muniz H, Badalamenti J, Herzog C, Henrich W: Cardiac troponin T and C-reactive protein for predicting prognosis, coronary atherosclerosis, and cardiomyopathy in patients undergoing long-term hemodialysis. JAMA 2003;290:353–359. 23 Aparicio M, Cano N, Chauveau P, Azar R, Canaud B, Flory A, Laville M, Leverve X: Nutritional status of haemodialysis patients: a French national cooperative study. French Study Group for Nutrition in Dialysis. Nephrol Dial Transplant 1999;14:1679–1686. 24 Longenecker JC, Coresh J, Klag MJ, Levey AS, Martin AA, Fink NE, Powe NR: Validation of comorbid conditions on the end-stage renal disease medical evidence report: the CHOICE study. Choices for Healthy Outcomes in Caring for ESRD. J Am Soc Nephrol 2000;11:520–529.

Am J Nephrol 2012;36:287–295

295