Abstract
Chronic obstructive pulmonary disease (COPD) may cause significant symptoms and have an impact on survival. Smoking is an important risk factor for COPD and is common in candidates for liver transplantation; however, the risk factors for and outcomes of COPD in this population are unknown. We performed a prospective cohort study of 373 patients being evaluated for liver transplantation at 7 academic centers in the United States. COPD was characterized by expiratory airflow obstruction and defined as follows: prebronchodilator forced expiratory volume in 1 second/forced vital capacity < 0.70. Patients completed the Liver Disease Quality of Life Questionnaire 1.0, which included the Short Form-36. The mean age of the study sample was 53 ± 9 years, and 234 (63%) were male. Sixty-seven patients (18%, 95% confidence interval 14%–22%) had COPD, and 224 (60%) had a history of smoking. Eighty percent of patients with airflow obstruction did not previously carry a diagnosis of COPD, and 27% were still actively smoking. Older age and any smoking (odds ratio = 3.74, 95% confidence interval 1.94–7.23, P < 0.001) were independent risk factors for COPD. Patients with COPD had worse New York Heart Association functional class and lower physical component summary scores on the 36-Item Short Form but had short-term survival similar to that of patients without COPD. In conclusion, COPD is common and often undiagnosed in candidates for liver transplantation. Older age and smoking are significant risk factors of COPD, which has adverse consequences on functional status and quality of life in these patients.
Chronic obstructive pulmonary disease (COPD) is characterized by progressive airflow limitation that is not fully reversible. The physiologic changes associated with COPD lead to dyspnea, exercise limitation, susceptibility to infections and exacerbations, and significant morbidity and mortality. Patients with advanced liver disease have well-known abnormalities in pulmonary function. Lung volumes are commonly decreased by hepatomegaly, ascites, basal atelectasis, or pleural effusions.1–5 Screening studies of patients with chronic liver disease have demonstrated arterial blood gas abnormalities in as many as 45%.2 However, little is known about the prevalence, risk factors, and impact of COPD in patients with advanced liver disease considered for liver transplantation (LT).
Smoking is the most important risk factor for COPD in the general population and is commonly reported by patients with advanced liver disease. Smoking may also contribute to the severity of various liver diseases and hepatocellular carcinoma.6–8 Smoking and COPD may be particularly important in patients with advanced liver disease being evaluated for LT, for whom perioperative outcomes and postoperative recovery could be compromised and long-term interactions with immunosuppression are unknown.
Therefore, we aimed to determine the prevalence and predictors of COPD in patients with advanced liver disease evaluated for LT. We hypothesized that smoking would be not only common but also a strong predictor of COPD. We also aimed to determine the impact of COPD on functional status, health-related quality of life, and overall survival.
PATIENTS AND METHODS
Study Design and Study Sample
The Pulmonary Vascular Complications of Liver Disease study enrolled a cohort of 536 patients evaluated for LT at 7 centers in the United States between 2003 and 2006.9 The cohort included patients with clinical portal hypertension and excluded patients who had evidence of active infection or recent (<2 weeks) gastrointestinal bleeding or who had previously undergone liver or lung transplantation. The sample for this study included newly evaluated patients in the cohort with available prebronchodilator spirometry and data regarding smoking status.
Data Collection
Demographic information was collected from each patient; age was categorized by quartiles for the analysis. Data regarding the etiology of underlying liver disease, past medical history, and social history were recorded. Patients underwent a physical examination that included anthropometry and blood pressure measurements and underwent a laboratory assessment. The Model for End-Stage Liver Disease (MELD) score was calculated.10
Chest radiography was interpreted locally at each center. Spirometry, lung volumes, and diffusing capacity for carbon monoxide were measured; results are expressed with standard gender- and race-specific prediction equations where appropriate.11–13 Arterial blood gas sampling was performed on room air in the seated position. Contrast transthoracic echocardiography was performed and interpreted at each center. Agitated saline was injected via a peripheral vein during imaging. The appearance of microbubbles in the left heart after 3 or more cardiac cycles (late) after venous injection of agitated saline was considered to indicate intrapulmonary shunting. The appearance of microbubbles in the left heart after fewer than 3 cardiac cycles (early) after venous injection was considered to indicate intracardiac shunting. The presence of hepatopulmonary syndrome was defined by (1) contrast echocardiography with late appearance of microbubbles after venous injection of agitated saline and (2) an alveolar-arterial oxygen gradient ≥ 15 mm Hg (or ≥ 20 mm Hg if age > 64 years).14 Significant pulmonary hypertension was defined as right ventricular systolic pressure (RVSP) > 50 mm Hg by echocardiography (if estimable).
We administered the Liver Disease Quality of Life Questionnaire 1.0.15 This is a validated questionnaire that not only includes questions specific to advanced liver disease but also includes the Medical Outcomes Study Short Form-36 (SF-36; version 2.0).16 Patients either completed the questionnaire during their visit to the clinical center or returned the questionnaire by mail.
The study was approved by the institutional review board of each center, and patients provided informed consent before undergoing study procedures.
COPD Definition
COPD was defined with modified Global Initiative on Obstructive Lung Disease criteria as follows: prebronchodilator forced expiratory volume in 1 second (FEV1)/forced vital capacity (FVC) < 0.70.17 Mild COPD was defined as FEV1 % predicted ≥ 80%, moderate as FEV1 % predicted ≥ 50% and < 80%, severe as FEV1 % predicted ≥ 30% and < 50%, and very severe as FEV1 % predicted < 30%.
Statistical Analysis
Continuous data were summarized as mean ± standard deviation or median (interquartile range). Categorical variables were summarized as n (%). Unpaired Student t tests, Wilcoxon rank sum tests, chi-square tests, and Fisher’s exact tests were used as appropriate. Bivariate logistic regression was performed with COPD status as the dependent variable and potential predictors as independent variables, with results expressed as odds ratios (ORs). Multivariate logistic regression was performed, including all variables that had P values < 0.20 on bivariate analysis or were hypothesized to predict case status. With the number of cases, we determined that the final multivariate model should include 4 or fewer predictors to prevent overfitting. The final multivariate model was assessed with the Hosmer-Lemeshow goodness-of-fit test; dbetas were calculated to assess influential subjects. Proportional hazards regression models were used for survival analyses.
All analyses were performed with Stata/IC version 10.0 (StataCorp, College Park, TX). A 2-sided P value of <0.05 was considered statistically significant.
RESULTS
Of 536 patients in the entire cohort, 473 (88%) patients were undergoing a new evaluation for LT. Of these, 373 (79%) had pulmonary function testing and available smoking history recorded, and they composed the study sample. The study sample was similar to the excluded patients [new LT evaluations without pulmonary function or smoking data (n = 100)] in terms of age, gender, and MELD score (data not shown). Excluded patients were more likely to be non-Hispanic white (92%) than included patients (79%).
The mean age of the study sample was 53 ± 9 years, and 234 (63%) were male. Two hundred ninety-three (79%) were non-Hispanic white, 39 (10%) were Hispanic, and 22 (6%) were non-Hispanic black. Sixty-seven patients [18%, 95% confidence interval (CI) 14%–22%] had COPD. Eighteen (29%) had mild COPD, 39 (58%) had moderate COPD, 8 (12%) had severe COPD, and 2 (3%) had very severe COPD. Patients with COPD were somewhat older than patients without COPD and tended to be male (Table 1). COPD patients were somewhat less likely to be Hispanic white.
TABLE 1.
Demographics and Liver Disease Variables
| Variable | n | COPD | No COPD | OR | 95% CI | P Value |
|---|---|---|---|---|---|---|
| Age, years | 373 | |||||
| 18–47 | 11 (16%) | 70 (23%) | 1.0 | — | — | |
| 48–53 | 13 (19%) | 87 (28%) | 0.95 | 0.40–2.25 | 0.91 | |
| 54–59 | 18 (27%) | 71 (23%) | 1.61 | 0.71–3.66 | 0.25 | |
| 60–79 | 25 (37%) | 78 (26%) | 2.03 | 0.94–4.45 | 0.07 | |
| Gender, male | 373 | 48 (72%) | 186 (61%) | 1.62 | 0.91–2.91 | 0.10 |
| Race/ethnicity | 373 | 0.47 | ||||
| Non-Hispanic white | 60 (90%) | 233 (76%) | 1.0 | — | — | |
| Hispanic white | 3 (5%) | 36 (12%) | 0.32 | 0.10–1.09 | 0.07 | |
| Non-Hispanic black | 3 (5%) | 19 (6%) | 0.61 | 0.18–2.14 | 0.44 | |
| Other | 1 (2%) | 18 (6%) | 0.22 | 0.03–1.65 | 0.14 | |
| Etiology of liver disease | 373 | |||||
| Alcohol | 38 (57%) | 116 (38%) | 2.15 | 1.26–3.67 | 0.005 | |
| Hepatitis C infection | 32 (48%) | 138 (45%) | 1.11 | 0.66–1.89 | 0.69 | |
| Autoimmune hepatitis | 1 (2%) | 12 (4%) | 0.37 | 0.05–2.90 | 0.35 | |
| Nonalcoholic fatty liver disease | 3 (5%) | 36 (12%) | 0.35 | 0.10–1.18 | 0.09 | |
| Hepatitis B infection | 2 (3%) | 15 (5%) | 0.60 | 0.13–2.67 | 0.50 | |
| Alpha-1 antitrypsin deficiency | 1 (2%) | 9 (3%) | 0.50 | 0.06–4.01 | 0.51 | |
| Primary sclerosing cholangitis | 1 (2%) | 16 (5%) | 0.27 | 0.04–2.11 | 0.21 | |
| Primary biliary cirrhosis | 1 (2%) | 12 (4%) | 0.37 | 0.05–2.90 | 0.35 | |
| Cryptogenic cirrhosis | 6 (9%) | 29 (9%) | 0.94 | 0.37–2.36 | 0.89 | |
| MELD score | 371 | 14 ± 5 | 13 ± 5 | 1.02 | 0.98–1.07 | 0.34 |
NOTE: Data are shown as mean ± standard deviation or n (%).
Abbreviations: CI, confidence interval; COPD, chronic obstructive pulmonary disease; MELD, Model for End-Stage Liver Disease; OR, odds ratio.
There were significant differences in the etiology of liver disease between patients with and without COPD (Table 1); patients may have had more than 1 etiology of liver disease. Patients with COPD were significantly more likely than those without it to have alcohol-related liver disease. Those without COPD tended to have more nonalcoholic fatty liver disease, primary sclerosing cholangitis (PSC), or primary biliary cirrhosis (PBC) in comparison with those with COPD. The groups were similar in terms of hepatitis C infection and alpha-1-antitrypsin deficiency–related liver disease. We had available genotyping on 5 patients with reported alpha-1-antitrypsin deficiency (all non-COPD), which showed the ZZ genotype in only 1 patient. The severity of liver dysfunction appeared similar between the groups, as reflected by the MELD score, and the median time from diagnosis of liver disease to referral for LT evaluation was 3 years in both groups.
Patients with COPD were more likely to have a history of ascites than those without COPD (P = 0.018; Table 2). Other liver disease complications and the prevalence of diabetes, hypertension, coronary artery disease, and other medical comorbidities were not different between the groups (data not shown). Notably, 80% of patients with COPD by spirometry had this diagnosis first made at the time of LT evaluation.
TABLE 2.
Past Medical History
| Variable | n | COPD | No COPD | OR | 95% CI | P Value |
|---|---|---|---|---|---|---|
| Ascites | 371 | 48 (73%) | 173 (57%) | 2.03 | 1.13–3.67 | 0.018 |
| Variceal bleeding | 372 | 14 (21%) | 69 (22%) | 0.90 | 0.47–1.72 | 0.76 |
| Encephalopathy | 371 | 29 (44%) | 138 (45%) | 0.95 | 0.55–1.62 | 0.85 |
| Spontaneous bacterial peritonitis | 372 | 3 (5%) | 22 (7%) | 0.60 | 0.18–2.08 | 0.40 |
| Hepatocellular carcinoma | 372 | 5 (8%) | 28 (9%) | 0.80 | 0.30–2.14 | 0.70 |
| Hepatic hydrothorax | 373 | 3 (5%) | 14 (5%) | 0.98 | 0.27–3.50 | 0.97 |
| Transjugular intrahepatic portosystemic shunt | 372 | 5 (8%) | 22 (7%) | 1.04 | 0.38–2.84 | 0.94 |
| Smoking | 373 | |||||
| Never | 13 (19%) | 136 (44%) | 1.0 | — | — | |
| Past or current | 54 (81%) | 170 (56%) | 3.32 | 1.74–6.34 | 0.001 | |
| Chronic alcohol use | 370 | |||||
| Never | 12 (18%) | 103 (34%) | 1.0 | — | — | |
| Past or current | 55 (82%) | 200 (66%) | 2.36 | 1.21–4.60 | 0.01 | |
| Intravenous drug use | 364 | 12 (19%) | 52 (17%) | 1.08 | 0.54–2.15 | 0.84 |
| Blood transfusion | 318 | 22 (37%) | 99 (38%) | 0.93 | 0.52–1.67 | 0.81 |
NOTE: Data are shown as n (%).
Abbreviations: CI, confidence interval; COPD, chronic obstructive pulmonary disease; OR, odds ratio.
Eighty-one percent of patients with COPD had ever smoked versus 56% of those without COPD (P = 0.001; Table 2). Of the COPD patients who were ever-smokers, more than one-quarter (27%) reported active smoking at the time of their LT evaluation. For those who were ever smokers with available pack-year data (n = 131), 20 pack years or more of smoking was associated with a dramatically increased risk of COPD in comparison with those with a lesser smoking history (OR = 4.17, 95% CI 1.12–15.47, P = 0.033). For all patients with smoking status and pack-year data (n = 280), 20 pack years or more of smoking continued to show a very significant increased risk of COPD versus fewer pack years or no smoking history (OR = 6.09, 95% CI 2.97–12.48, P = 0.001). Alcohol use was also associated with an increased risk of COPD (P = 0.008).
Patients with COPD had a significantly lower body mass index in comparison with patients without COPD (27 ± 6 kg/m2 versus 29 ± 6 kg/m2, respectively, P = 0.004; Table 3). There were no other differences in physical examination or laboratory values.
TABLE 3.
Physical Examination and Laboratory Results
| Variable | n | COPD | No COPD | P Value |
|---|---|---|---|---|
| Physical examination | ||||
| Body mass index, kg/m2 | 373 | 27 ± 6 | 29 ± 6 | 0.007 |
| Pulse, beats per minute | 372 | 75 ± 14 | 77 ± 14 | 0.41 |
| Systolic blood pressure, mm Hg | 373 | 123 ± 19 | 120 ± 18 | 0.26 |
| Diastolic blood pressure, mm Hg | 373 | 70 ± 11 | 70 ± 11 | 0.82 |
| Room air oxygen saturation, % | 218 | 97 ± 4 | 97 ± 3 | 0.84 |
| Ascites | 368 | 31 (48%) | 113 (37%) | 0.12 |
| Lower extremity edema | 368 | 32 (49%) | 165 (54%) | 0.44 |
| Clubbing | 363 | 6 (9%) | 26 (9%) | 0.90 |
| Asterixis | 366 | 4 (6%) | 19 (6%) | 0.96 |
| Spider angiomata | 356 | 25 (40%) | 102 (35%) | 0.46 |
| Laboratory results | ||||
| Blood urea nitrogen, mg/dL | 365 | 16 ± 9.0 | 16 ± 12 | 0.60 |
| Creatinine, mg/dL | 373 | 1.2 ± 1.0 | 1.0 ± 1.0 | 0.37 |
| Hemoglobin, g/dL | 370 | 12.4 ± 2.2 | 12.6 ± 2.0 | 0.46 |
| Platelet count, 109/L | 368 | 117 ± 63 | 124 ± 89 | 0.55 |
| International normalized ratio | 371 | 1.4 ± 0.3 | 1.3 ± 0.3 | 0.24 |
| Alanine aminotransferase, U/L | 372 | 48 ± 35 | 55 ± 42 | 0.18 |
| Aspartate aminotransferase, U/L | 372 | 71 ± 50 | 76 ± 48 | 0.41 |
| Total bilirubin, mg/dL | 373 | 2.5 ± 2.6 | 2.8 ± 3.0 | 0.63 |
| Alkaline phosphatase, U/L | 369 | 146 ± 70 | 172 ± 149 | 0.72 |
| Total protein, g/dL | 358 | 7.1 ± 0.8 | 7 ± 0.8 | 0.52 |
| Albumin, g/dL | 365 | 3.1 ± 0.6 | 3.1 ± 0.7 | 0.60 |
NOTE: Data are shown as mean ± standard deviation or n (%).
Abbreviation: COPD, chronic obstructive pulmonary disease.
By definition, pulmonary function varied significantly between the 2 groups (Table 4). Patients with COPD had significantly lower FEV1 % predicted, FEV1/FVC, and (in whom it was measured) diffusing capacity for carbon monoxide and higher alveolar-arterial oxygen gradients (all P < 0.05). Chest radiography and most echocardiographic measurements were similar between the 2 groups (Table 5). However, patients with COPD had significantly higher estimated RVSP than patients without COPD and more commonly had significant pulmonary hypertension [RVSP > 50 mm Hg; COPD = 9 (24%) versus no COPD = 20 (10%), P = 0.02, n = 230]. Imputing low RVSP estimates for those patients with transthoracic echocardiography with missing RVSP estimates did not change these findings (data not shown). Approximately 60% of both COPD and non-COPD patients demonstrated intrapulmonary shunting; 3 (27%) patients with COPD and 69 (33%) patients without COPD (with complete data) had hepatopulmonary syndrome (P = 1.0, n = 218).
TABLE 4.
Pulmonary Function Testing and Arterial Blood Gas Results
| Variable | n | COPD | No COPD | P Value |
|---|---|---|---|---|
| Pulmonary function testing | ||||
| Forced vital capacity, % predicted | 373 | 85 ± 19 | 85 ± 16 | 0.71 |
| Forced expiratory volume in 1 second, % predicted | 373 | 69 ± 19 | 86 ± 16 | <0.0001 |
| Forced expiratory volume in 1 second/forced vital capacity | 373 | 62 ± 8 | 79 ± 5 | <0.0001 |
| DLCOcorr, % predicted | 355 | 55 ± 13 | 60 ± 17 | 0.02 |
| Arterial blood gas | ||||
| pH | 333 | 7.44 ± 0.03 | 7.44 ± 0.04 | 0.79 |
| pCO2, mm Hg | 333 | 34 ± 5 | 35 ± 5 | 0.69 |
| pO2, mm Hg | 333 | 81 ± 12 | 85 ± 16 | 0.11 |
| Oxygen saturation, % | 326 | 94 ± 3 | 95 ± 4 | 0.72 |
| Alveolar-arterial oxygen gradient, mm Hg | 333 | 24 (13–31) | 15 (7–26) | 0.003 |
NOTE: Data are shown as mean ± standard deviation, median (interquartile range), or n (%).
Abbreviations: COPD, chronic obstructive pulmonary disease; DLCOcorr, diffusing capacity for carbon monoxide corrected for hemoglobin; pCO2, partial pressure of carbon dioxide; pO2, partial pressure of oxygen.
TABLE 5.
Echocardiography and Chest Radiography Results
| Variable | n | COPD | No COPD | P Value |
|---|---|---|---|---|
| Chest radiography | ||||
| Cardiomegaly | 351 | 8 (13%) | 30 (10%) | 0.60 |
| Large pulmonary arteries | 351 | 2 (3%) | 7 (2%) | 0.67 |
| Interstitial lung disease | 351 | 2 (3%) | 9 (3%) | 1.0 |
| Hyperinflation | 351 | 2 (3%) | 3 (1%) | 0.22 |
| Pleural effusion | 351 | 7 (11%) | 46 (16%) | 0.33 |
| Echocardiography | ||||
| Right atrial dilation | 332 | 10 (17%) | 52 (19%) | 0.66 |
| Right ventricular dilation | 344 | 20 (32%) | 88 (31%) | 0. 95 |
| Right ventricular hypertrophy | 343 | 12 (19%) | 57 (20%) | 0.82 |
| Right ventricular dysfunction | 346 | 6 (9%) | 15 (5%) | 0.24 |
| Paradoxical septal motion | 324 | 1 (2%) | 2 (1%) | 0.47 |
| Right ventricular systolic pressure, mm Hg | 230 | 44 ± 19 | 37 ± 13 | 0.01 |
| Left atrial size, cm | 309 | 4 ± 1 | 4 ± 1 | 0.63 |
| Left ventricular hypertrophy | 349 | 15 (24%) | 91 (32%) | 0.24 |
| Pericardial effusion | 341 | 8 (13%) | 33 (12%) | 0.81 |
| Shunting | 307 | 0.46 | ||
| None | 17 (32%) | 91 (36%) | ||
| Intrapulmonary | 31 (58%) | 150 (59%) | ||
| Intracardiac | 5 (9%) | 13 (5%) | ||
NOTE: Data are shown as mean ± standard deviation or n (%).
Abbreviation: COPD, chronic obstructive pulmonary disease.
Multivariate Analysis
We assessed variables from Tables 1 and 2 in multivariate logistic regression. Older age and any smoking were independently associated with an increased risk of COPD (Table 6). Alcohol use and etiology of liver disease were no longer associated with the risk of COPD in our adjusted model and were likely confounded by smoking. Race/ethnicity and gender were not found to be predictors of COPD in our study sample. There were no significant interactions between the final factors in the model. The model fit was adequate (P = 0.51), and there were no particularly influential subjects.
TABLE 6.
Multivariate Logistic Regression
| Variable | OR | 95% CI | P Value |
|---|---|---|---|
| Age | |||
| 18–47 | 1.0 | — | — |
| 48–53 | 0.84 | 0.35–2.03 | 0.71 |
| 54–59 | 1.69 | 0.73–3.90 | 0.22 |
| 60–79 | 2.27 | 1.02–5.06 | 0.044 |
| Smoking (past or current versus never) | 3.74 | 1.94–7.23 | <0.001 |
Abbreviations: CI, confidence interval; OR, odds ratio.
Functional Status, Quality of Life, and Socioeconomic Status
Overall, patients with COPD had worse New York Heart Association functional class than patients without COPD (Table 7). Of the 373 patients in the study sample, 214 completed the Liver Disease Quality of Life Questionnaire 1.0. Patients who returned the questionnaire were similar to those from the study sample who did not in terms of age, gender, race/ethnicity, MELD score, smoking, and probability of COPD (data not shown). There were no differences in the liver disease–specific scales between patients with and without COPD (data not shown). However, patients with COPD scored lower than non-COPD patients on the physical component summary score of the SF-36 questionnaire, indicating worse quality of life (31 ± 10 versus 35 ± 12, respectively, P = 0.047, n = 209) Adjustment for age did not affect these results. There were no differences in marital status, educational status, or income between the 2 groups (data not shown).
TABLE 7.
New York Heart Association Functional Class: COPD Versus No COPD (n = 372)
| Class | COPD | No COPD |
|---|---|---|
| I | 12 (18%) | 88 (29%) |
| II | 29 (43%) | 138 (45%) |
| III | 25 (37%) | 66 (22%) |
| IV | 1 (2%) | 13 (4%) |
NOTE: Wilcoxon rank sum test, P = 0.03.
Abbreviation: COPD, chronic obstructive pulmonary disease.
LT and Survival
Two hundred four patients in the study sample were listed for LT. Patients with COPD were significantly less likely to be listed for LT than patients without COPD [28 (42%) versus 176 (58%), respectively, P = 0.019]. Of the patients with COPD who were listed for LT, 39% had mild COPD, 50% had moderate COPD, 11% had severe COPD, and none had very severe COPD. Patients with COPD who were denied listing for LT had significantly more severe disease than patients with COPD who were listed (P = 0.01). One hundred thirteen (30%) patients in the study sample underwent LT; there were no differences in performance of LT between patients with and without COPD (data not shown).
All patients were followed until December 31, 2006; there were no patients lost-to follow-up. The 1-year survival of the entire cohort was 87% (95% CI 83%–90%), and there were 81 deaths overall. The median follow-up time was 601 days (interquartile range, 390–1319 days). Patients with COPD had a survival similar to that of patients without COPD overall (hazard ratio = 1.0, 95% CI 0.50–1.70, P = 0.89); adjustment for the performance of LT did not change this result. The severity of COPD was not associated with the risk of death in the study sample (data not shown). We found no difference between groups in post-LT survival (hazard ratio = 0.50, 95% CI 0.06–3.70, P = 0.48). Similarly, there was no association between smoking and overall or posttransplant mortality (data not shown).
DISCUSSION
To the best of our knowledge, this is the largest prospective, multicenter cohort study of lung function ever performed in candidates for LT. We have shown that 18% of new patients undergoing LT evaluation in the United States have COPD. Older age and any history of smoking increase the risk of COPD in this population. Notably, 27% of patients with COPD reported active smoking, and 80% had the diagnosis of COPD made for the first time during their LT evaluation. In addition, COPD was associated with a worse New York Heart Association functional class and with a worse quality of life in terms of physical limitations. Despite the impact on functional status and quality of life, COPD did not significantly affect survival.
The prevalence of COPD in the LT candidates in our study was greater than the prevalence in the general population found in some studies and similar to that found in others.18–20 The COPD prevalence in our study was greater than that found in older, smaller studies performed in LT candidates; however, differences between our study population and the others may explain these findings. In 1991, Hourani et al.2 published a study of 116 consecutive patients admitted for evaluation for LT and found that 7% had obstruction by spirometry (defined as FEV1/FVC less than 2 standard deviations below the predicted value). Although 55% of these patients were current or former smokers, the mean age was 45 years, and most patients were female, in contrast to our cohort, which was some 10 years older on average and mostly male. In addition, only 8% of the study population of Hourani et al. had cirrhosis due to alcohol; this was a much lower prevalence than that in our study. Another study found a prevalence of obstruction of 12% by spirometry in 58 patients listed for LT.4 Of note, this study excluded symptomatic patients and patients who were evaluated but not listed for LT. This cohort had a smoking prevalence of 29%, included mostly females, and included mostly patients with PBC and PSC. Krowka et al.3 studied 95 patients who underwent LT who had preoperative and postoperative pulmonary function testing. Although 75% of these patients had either PBC or PSC, these investigators found that 17% had an FEV1/FVC ratio <75%, which did not change after LT.
It may be difficult to generalize the prevalence and risk factors for COPD from these studies. The most common modern indications for LT evaluation are alcohol-related liver disease and hepatitis C infection, which affect very different populations in terms of age and gender in comparison with those included in these older studies. Importantly, gender and alcohol use are tightly related to smoking, so our results are likely much more applicable to the LT population evaluated in the United States in the modern era.
Patients with advanced liver disease commonly demonstrate coexistent restriction due to hepatomegaly and ascites, which could increase the FEV1/FVC ratio.1,2,5 Therefore, it is possible that we have underestimated the actual prevalence of COPD in our cohort. On the other hand, some studies have shown that diuresis or large-volume paracentesis in patients with tense ascites increases FEV1 and FVC in parallel with an unchanged FEV1/FVC ratio.21,22 Less than 50% of patients in our study had detectable ascites by physical examination, and this makes it less likely that this had a significant impact on our conclusions.
The most important contributor to the increased prevalence of COPD in this population was smoking, which was reported by 81% of the patients with COPD and 56% of the study sample overall. Others have shown virtually identical frequencies of smoking in patients with advanced liver disease.2,23 These smoking rates are significantly higher than those seen in the general population, and this is likely attributable to the association between smoking and other habits predisposing to advanced liver disease, such as chronic alcohol use and intravenous drug use. The increased risk for older patients in our study may be due to either age-related changes in lung function or residual confounding by pack years (that is, older age potentially represents more years of smoking). Although we recorded collected information on pack years, these data were incomplete and may suffer from misclassification.
In addition to the impact of smoking on COPD, studies have also shown that tobacco use is associated with increased severity of liver fibrosis,6 hepatocellular carcinoma,7 and hepatic lesions in patients with hepatitis C infection.8 There is also evidence to indicate that patients with a history of smoking have an increased risk for posttransplant complications.23 We did not find differences in overall survival or post-transplant survival by smoking status; however, we neither assessed perioperative outcomes (such as length of stay) nor had sufficient numbers of transplanted patients and length of follow-up to assess post-LT complications, such as neoplasm. Studies have found increased risks of various cancers in patients with alcohol-related liver disease undergoing LT in comparison with others.24,25 This association may very well be confounded by smoking, which is more likely responsible for many of these cancers.
Patients with COPD had higher RVSP by transthoracic echocardiography and more commonly had significant pulmonary hypertension in comparison with patients without COPD. This may be explained by ventilation-perfusion mismatch and local hypoxic vasoconstriction, hypoxia-related pulmonary vascular remodeling, parenchymal destruction, or elevated cardiac output or pulmonary venous pressure. Although there were no differences between COPD and non-COPD patients in terms of the partial pressure of oxygen or oxygen saturation in arterial blood, there was a significant difference in the alveolar-arterial oxygen gradient. Because right heart catheterization was not routinely performed in these patients, it is difficult to know whether the pulmonary hypertension in these patients was attributable to the lung disease itself or to a cardiac etiology. It is certainly possible, however, that the pulmonary hypertension in patients with COPD contributed to the functional limitation, although there were no differences in right-sided cardiac morphology attributable to COPD. It should be noted that it is unknown whether these patients had portopulmonary hypertension, which requires invasive hemodynamic measurement and the absence of other significant cardiac or pulmonary disease for diagnosis.
Chronic liver disease is associated with reduced quality of life, notwithstanding other comorbid conditions.26 Even so, COPD had a significant additional negative impact not only on functional class but also on quality of life in our cohort. LT candidates with COPD had more impairment in the performance of usual physical activities, as reflected by the SF-36 physical component score. Considering this already severely compromised patient population, we believe that the treatment of COPD and smoking cessation are likely important for a positive impact on patients’ physical functioning and quality of life both before and after LT.
Although 18% of our study patients were found to have COPD as defined by prebronchodilator FEV1/FVC < 0.70, 80% of these patients did not have a previous clinical diagnosis of COPD. It may be reasonable, therefore, for all patients evaluated for LT to undergo screening spirometry. The diagnosis of unsuspected COPD in the LT candidate presents an important opportunity for a smoking cessation intervention by transplant health professionals. The risks of respiratory tract cancers, particularly in the setting of posttransplant immunosuppression, make this an important part of the pretransplant evaluation. Lastly, the impact on functional status and physical activity–related quality of life makes diagnosis and treatment of COPD (including smoking cessation) potentially clinically important interventions in a candidate for LT.
There are some limitations to our study. Our study sample included only patients who were evaluated for LT and only those with pulmonary function testing and a provided smoking history. This may have resulted in a selection bias. Although postbronchodilator spirometry is the gold standard for diagnosing COPD, only prebronchodilator spirometry data were available. Thus, we may have included patients with other types of obstructive disease (such as asthma) in our diseased group. We recorded a variety of potential risk factors for COPD; nonetheless, our findings could be accounted for by another confounding variable or by residual confounding. Given the relatively small size of our study sample, there was limited power to detect weaker risk factors for COPD; however, this is the largest study of lung function ever performed in LT candidates.
In summary, older age and smoking were independently associated with an increased risk of COPD in patients with advanced liver disease undergoing LT evaluation. Screening and intervention may be warranted to minimize the short- and long-term impact of COPD in this patient population.
Acknowledgments
Funded by National Institutes of Health grants DK064103, DK065958, RR00645, RR00585, RR00046, RR00032, and HL67771.
Abbreviations
- CI
confidence interval
- COPD
chronic obstructive pulmonary disease
- DLCOcorr
diffusing capacity for carbon monoxide corrected for hemoglobin
- FEV1
forced expiratory volume in 1 second
- FVC
forced vital capacity
- LT
liver transplantation
- MELD
Model for End-Stage Liver Disease
- OR
odds ratio
- PBC
primary biliary cirrhosis
- pCO2
partial pressure of carbon dioxide
- pO2
partial pressure of oxygen
- PSC
primary sclerosing cholangitis
- RVSP
right ventricular systolic pressure
- SF-36
Medical Outcomes Study Short Form-36
Footnotes
The Pulmonary Vascular Complications of Liver Disease Study Group also includes the following: Jeffrey Okun, B.A., Daniel Rabinowitz, Ph.D., Evelyn M. Horn, M.D., Lori Rosenthal, N.P., and Sonja Olsen, M.D., from Columbia University; Vijay Shah, M.D., and Russell Wiesner, M.D., from the Mayo Clinic; J. Stevenson Bynon, M.D., Devin Eckhoff, M.D., Harpreet Singh, Rajasekhar Tanikella, Raymond L. Benza, M.D., and Keith Wille, M.D., from the University of Alabama; Lisa Forman, M.D., and David Badesch, M.D., from the University of Colorado; Roshan Shrestha, M.D., from The University of North Carolina at Chapel Hill; Darren B. Taichman, M.D., Ph.D., Vivek Ahya, M.D., Harold Palevsky, M.D., and Rajender Reddy, M.D., from the University of Pennsylvania; and Neil Kaplowitz, M.D., and James Knowles, M.D., Ph.D., from the University of Southern California.
References
- 1.Angueira CE, Kadakia SC. Effects of large-volume paracentesis on pulmonary function in patients with tense cirrhotic ascites. Hepatology. 1994;20:825–828. doi: 10.1002/hep.1840200409. [DOI] [PubMed] [Google Scholar]
- 2.Hourani JM, Bellamy PE, Tashkin DP, Batra P, Simmons MS. Pulmonary dysfunction in advanced liver disease: frequent occurrence of an abnormal diffusing capacity. Am J Med. 1991;90:693–700. [PubMed] [Google Scholar]
- 3.Krowka MJ, Dickson ER, Wiesner RH, Krom RA, Atkinson B, Cortese DA. A prospective study of pulmonary function and gas exchange following liver transplantation. Chest. 1992;102:1161–1166. doi: 10.1378/chest.102.4.1161. [DOI] [PubMed] [Google Scholar]
- 4.Mohamed R, Freeman JW, Guest PJ, Davies MK, Neuberger JM. Pulmonary gas exchange abnormalities in liver transplant candidates. Liver Transpl. 2002;8:802–808. doi: 10.1053/jlts.2002.33746. [DOI] [PubMed] [Google Scholar]
- 5.Yao EH, Kong BC, Hsue GL, Zhou AC, Wang H. Pulmonary function changes in cirrhosis of the liver. Am J Gastroenterol. 1987;82:352–354. [PubMed] [Google Scholar]
- 6.Zein CO, Beatty K, Post AB, Logan L, Debanne S, McCullough AJ. Smoking and increased severity of hepatic fibrosis in primary biliary cirrhosis: a cross validated retrospective assessment. Hepatology. 2006;44:1564–1571. doi: 10.1002/hep.21423. [DOI] [PubMed] [Google Scholar]
- 7.Trichopoulos D, MacMahon B, Sparros L, Merikas G. Smoking and hepatitis B-negative primary hepatocellular carcinoma. J Natl Cancer Inst. 1980;65:111–114. [PubMed] [Google Scholar]
- 8.Pessione F, Ramond MJ, Njapoum C, Duchatelle V, Degott C, Erlinger S, et al. Cigarette smoking and hepatic lesions in patients with chronic hepatitis C. Hepatology. 2001;34:121–125. doi: 10.1053/jhep.2001.25385. [DOI] [PubMed] [Google Scholar]
- 9.Kawut SM, Krowka MJ, Trotter JF, Roberts KE, Benza RL, Badesch DB, et al. Clinical risk factors for portopulmonary hypertension Hepatology 2008// 10.1002/hep.22275Available at: http://www3.interscience.wiley.com [DOI] [PMC free article] [PubMed]
- 10.Kamath PS, Wiesner RH, Malinchoc M, Kremers W, Therneau TM, Kosberg CL, et al. A model to predict survival in patients with end-stage liver disease. Hepatology. 2001;33:464–470. doi: 10.1053/jhep.2001.22172. [DOI] [PubMed] [Google Scholar]
- 11.Crapo RO, Morris AH. Standardized single breath normal values for carbon monoxide diffusing capacity. Am Rev Respir Dis. 1981;123:185–189. doi: 10.1164/arrd.1981.123.2.185. [DOI] [PubMed] [Google Scholar]
- 12.Crapo RO, Morris AH, Clayton PD, Nixon CR. Lung volumes in healthy nonsmoking adults. Bull Eur Physiopathol Respir. 1982;18:419–425. [PubMed] [Google Scholar]
- 13.Hankinson JL, Odencrantz JR, Fedan KB. Spirometric reference values from a sample of the general U.S. population. Am J Respir Crit Care Med. 1999;159:179–187. doi: 10.1164/ajrccm.159.1.9712108. [DOI] [PubMed] [Google Scholar]
- 14.Rodriguez-Roisin R, Krowka MJ, Herve P, Fallon MB. Pulmonary-hepatic vascular disorders (PHD) Eur Respir J. 2004;24:861–880. doi: 10.1183/09031936.04.00010904. [DOI] [PubMed] [Google Scholar]
- 15.Gralnek IM, Hays RD, Kilbourne A, Rosen HR, Keeffe EB, Artinian L, et al. Development and evaluation of the Liver Disease Quality of Life instrument in persons with advanced, chronic liver disease—the LDQOL 1.0. Am J Gastroenterol. 2000;95:3552–3565. doi: 10.1111/j.1572-0241.2000.03375.x. [DOI] [PubMed] [Google Scholar]
- 16.Ware JE, Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care. 1992;30:473–483. [PubMed] [Google Scholar]
- 17.Rabe KF, Hurd S, Anzueto A, Barnes PJ, Buist SA, Calverley P, et al. 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. doi: 10.1164/rccm.200703-456SO. [DOI] [PubMed] [Google Scholar]
- 18.Mannino DM, Braman S. The epidemiology and economics of chronic obstructive pulmonary disease. Proc Am Thorac Soc. 2007;4:502–506. doi: 10.1513/pats.200701-001FM. [DOI] [PubMed] [Google Scholar]
- 19.Mannino DM, Doherty DE, Buist SA. Global Initiative on Obstructive Lung Disease (GOLD) classification of lung disease and mortality: findings from the Atherosclerosis Risk in Communities (ARIC) study. Respir Med. 2006;100:115–122. doi: 10.1016/j.rmed.2005.03.035. [DOI] [PubMed] [Google Scholar]
- 20.Mannino DM, Homa DM, Akinbami LJ, Ford ES, Redd SC. Chronic obstructive pulmonary disease surveillance—United States, 1971–2000. MMWR Surveill Summ. 2002;51:1–16. [PubMed] [Google Scholar]
- 21.Chang SC, Chang HI, Chen FJ, Shiao GM, Wang SS, Lee SD. Therapeutic effects of diuretics and paracentesis onlung function in patients with non-alcoholic cirrhosis and tense ascites. J Hepatol. 1997;26:833–838. doi: 10.1016/s0168-8278(97)80249-8. [DOI] [PubMed] [Google Scholar]
- 22.Chao Y, Wang SS, Lee SD, Shiao GM, Chang HI, Chang SC. Effect of large-volume paracentesis on pulmonary function in patients with cirrhosis and tense ascites. J Hepatol. 1994;20:101–105. doi: 10.1016/s0168-8278(05)80474-x. [DOI] [PubMed] [Google Scholar]
- 23.Ehlers SL, Rodrigue JR, Widows MR, Reed AI, Nelson DR. Tobacco use before and after liver transplantation: a single center survey and implications for clinical practice and research. Liver Transpl. 2004;10:412–417. doi: 10.1002/lt.20087. [DOI] [PubMed] [Google Scholar]
- 24.Fung JJ, Jain A, Kwak EJ, Kusne S, Dvorchik I, Eghtesad B. De novo malignancies after liver transplantation: a major cause of late death. Liver Transpl. 2001;7:S109–S118. doi: 10.1053/jlts.2001.28645. [DOI] [PubMed] [Google Scholar]
- 25.Oo YH, Gunson BK, Lancashire RJ, Cheng KK, Neuberger JM. Incidence of cancers following orthotopic liver transplantation in a single center: comparison with national cancer incidence rates for England and Wales. Transplantation. 2005;80:759–764. doi: 10.1097/01.tp.0000173775.16579.18. [DOI] [PubMed] [Google Scholar]
- 26.Younossi ZM, Boparai N, Price LL, Kiwi ML, McCormick M, Guyatt G. Health-related quality of life in chronic liver disease: the impact of type and severity of disease. Am J Gastroenterol. 2001;96:2199–2205. doi: 10.1111/j.1572-0241.2001.03956.x. [DOI] [PubMed] [Google Scholar]