Bacterial infections in terminally ill hospice patients

Terminally ill patients are very susceptible to infections, which are the result of disease-related processes and/or therapy-induced mechanisms. These patients are already subject to multiple severe symptoms and associated comorbid conditions, with much resultant distress. Infection increases this symptom burden and further reduces quality of life. We have retrospectively investigated the prevalence of infection and clinical course in 102 consecutive patients who died after admission to a tertiary palliative care unit and assessed the site-specific frequency of infection, pathogenic organisms involved, and the pattern of antibiotic agents used. The prevalence of symptoms and comorbid conditions on admission and during the progress phase of care were noted. Median overall survival of the total cohort was 12 days. The median survival of patients with infections was 22 days. Thirty-seven patients (36.3%) were diagnosed with 42 separate infections. The sites of infections were the urinary tract (42.5%), the respiratory tract (22.9%), blood (12.5%), skin and subcutaneous tissues (12.5%), and the eyes (10.0%). There were 20 separate positive cultures isolated from specimens obtained from 13 individual patients. Three isolates were obtained from 1 patient, 2 isolates obtained from 5 patients, and 1 isolate was obtained from each of the 7 remaining patients. Escherichia coli was the most common pathogen isolated. Eleven patients with infections (31.4%) were diagnosed on admission, and antibiotic treatment was commenced within 48 hours of admission in 21 patients (60%). Overall antibiotic response and symptom control of infections was observed to be a minimum of 40%. Psychological distress was common in this group of patients (P = 0.001) as were disabling symptoms on admission, such as pain, immobility, and weakness. Symptoms indicating poor survival, such as severe pain and dyspnea, were not significantly associated with infection. Decreased patient survival in this cohort was not significantly associated with the presence of bacterial infection (P = 0.07), irrespective of whether or not a positive culture isolate was obtained. We conclude that appropriate management of infection resulted in enhanced palliative symptom control.