Increased mortality of acute upper gastrointestinal bleeding in patients with chronic obstructive pulmonary disease. A case controlled, multiyear study of 53 consecutive patients.

The etiology, clinical presentation, and mortality of acute upper gastrointestinal bleeding in patients with chronic obstructive pulmonary disease (COPD) were analyzed in a case-controlled study of 53 consecutive patients admitted from 1985 through 1990 to a university teaching hospital. The primary controls were 40 consecutive patients with acute upper gastrointestinal bleeding and without COPD admitted from June through November 1990 to the same hospital. COPD patients had a significantly increased mortality from gastrointestinal bleeding as compared to controls with gastrointestinal bleeding and without COPD (mortality in COPD = 32%, controls = 10%, odds ratio = 4.3, confidence interval of odds ratio = 1.22-14.8, P < 0.01, Fisher's exact test) and as compared to a second control group of 53 consecutive COPD patients without gastrointestinal bleeding (mortality in second controls = 11%, odds ratio = 3.7, confidence interval of odds ratio = 1.25-11.0, P < 0.02, chi square). The study COPD patients had a significantly greater likelihood of being older, smokers, alcoholics, and taking corticosteroids than the primary controls. However, an increased mortality was still present when controlling for these differences by population stratification (eg, mortality in patients > or = 60 years old: COPD = 36%, controls = 13%, odds ratio = 4.6, P < 0.05). The two groups had similar mean values of parameters of bleeding severity, such as lowest hematocrit and units of packed erythrocytes transfused. The increased mortality was correlated with COPD severity (eg, four of five patients with prior endotracheal intubation for COPD died, 13 of 48 COPD patients without prior intubation died, odds ratio = 10, P < 0.04, Fisher's exact test).(ABSTRACT TRUNCATED AT 250 WORDS)