Revisiting Combination Antibiotic Therapy for Community-Acquired Invasive Streptococcus pneumoniae Pneumonia

Sir—In a previously published article [1], we showed that adults ⩾50 years of age with invasive Streptococcus pneumoniae pneumonia given combination antibiotic therapy consisting of a macrolide and a β-lactam had a significantly lower case-fatality rate, compared with those given monotherapy with a β-lactam (7% vs. 22%, respectively). Because we did not analyze the clinical comparability of our 2 treatment groups, other investigators who subsequently reported similar findings in invasive pneumococcal pneumonia [2–7] cited our omission as a shortcoming. In this report, we provide that analysis. For 2 groups of adults with invasive pneumococcal pneumonia, who were aged ⩾50 years, were admitted to hospitals from 1978 to 1997, and were treated with combination therapy or monotherapy, data were analyzed for comparability regarding age, severity of illness, and underlying comorbid conditions [1]. Treatment with antibiotics in both groups was started on the day of admission. S. pneumoniae was isolated from blood or pleural fluid samples. Clinical and demographic data were abstracted from the hospital chart. Only deaths that occurred during the first 7 days of hospitalization were counted, unlike in our previous article [1], when all deaths were counted. The institutional review board of Marshall University and of each of the affiliated hospitals approved this research study. The clinical features of the combination antibiotic therapy group and the monotherapy group were comparable (table 1). They did not differ significantly on the basis of all parameters examined, including age, vital signs at admission, total leukocyte count at admission, number of lobes involved, and preexisting underlying diseases. Both groups had, on average, slight hypothermia, tachycardia and tachypnea, normal blood pressure, and leukocytosis. Mainly, only 1 lobe was involved. Approximately four-fifths of each group had ⩾1 underlying comorbid condition(s). The case-fatality rate in the first 7 days of hospitalization was significantly lower in the combination therapy group, compared with the monotherapy group (1.8% vs. 14.5%, respectively; P = .009, 2-tailed Fisher's exact test). Two persons in the combination therapy group also received a fluoroquinolone (ciprofloxacin) and neither died. Penicillin resistance was not a factor, because ∼94% of persons were infected with penicillin-susceptible strains. Infection with penicillin-resistant strains was equally common in the 2 treatment groups, but no such case was fatal. No patients had HIV or AIDS. Comparison of attributes of patients aged ⩾50 years who had invasive community-acquired pneumonia due to Streptococcus pneumoniae and were given either combination therapy or monotherapy. Thus, compared with monotherapy, combination antibiotic therapy effectively lowered the case-fatality rate of invasive pneumococcal pneumonia among ill adults ⩾50 years of age. Reason dictates that a β-lactam is the antibiotic of choice for the treatment of invasive pneumococcal pneumonia. Why does use of 2 antibiotics, 1 of which is a macrolide, more effectively lower case-fatality rates in invasive pneumococcal pneumonia than does use of a β-lactam alone? Apparently, it is not because of antibiotic synergy, because treatment with erythromycin and a penicillin or a cephalosporin failed to show synergy in vitro [8]. It is unclear whether a proportion of patients with invasive pneumococcal pneumonia also have a second infection with an atypical pathogen, such as Mycoplasma pneumoniae or Chlamydia pneumoniae, and these pathogens respond to treatment with the macrolide. It is important to know that the best evidence supports use of combination antibiotic therapy, and prospective randomized clinical trials might settle this point [2, 4, 6, 7, 10]. However, until such trials can be performed, it seems reasonable to provide severely ill adults who have community-acquired pneumococcal pneumonia—and especially those who have bacteremia—with combination antibiotic therapy [11]. We thank Sally Wells and the staffs of the microbiology departments of the affiliated hospitals, for their cooperation. Potential conflicts of interest. M.A.M. and R.J.S.: no conflicts.