The Yield of Sputum Culture in Bacteremic Pneumococcal Pneumonia After Initiation of Antibiotics

To the Editor—The sputum Gram stain and culture are crucial components of the diagnostic workup in patients with community-acquired pneumonia (CAP). Prior studies have demonstrated a reduced yield of sputum studies when antibiotics are initiated prior to obtaining a sputum specimen [1–4], but we are aware of only a single study that relates this yield of sputum studies to the time after initiation of antibiotic therapy [3]. In that study, however, the number of patients was insufficient to determine the time at which the yield of sputum studies began to decline after antibiotics had been started. For at least 2 reasons, it is important to determine the maximal time interval during which antibiotics can be administered without a significant reduction in the yield of sputum cultures. First, these data can help guide recommendations on the chronology of diagnostic and therapeutic interventions. In the absence of such data, current Infectious Diseases Society of America (IDSA) guidelines recommend obtaining a sputum sample prior to administering antibiotics [5]. Second, such results would provide useful information to clinicians who are faced with the common dilemma of whether it is still worthwhile to submit sputum for culture from patients who have already begun antibiotics. We identified all patients with bacteremic pneumococcal pneumonia (BPP) hospitalized at the Michael E. DeBakey Veterans Affairs Medical Center during 1998–2013. BPP was defined by ≥1 blood culture positive for Streptococcus pneumoniae and the presence of a pulmonary infiltrate on chest radiography in a patient with clinical findings consistent with pneumonia. We selected BPP as an inclusion criterion to ensure that all patients had proven CAP and the same pathogen was responsible. For the present study, we included only those patients who were able to produce an adequate sputum sample, defined as one that contains areas with ≥10 white blood cells for each epithelial cell. A Gram-stained specimen was considered positive if a predominance of gram-positive pairs (diplococci) and chains were identified at a ×1000 magnification. In our electronic medical records, a barcode system enters the time that a sputum sample is obtained and the time at which a medication is administered. Eighty-five patients with BPP who provided an adequate sputum specimen were identified. No patient received antibiotics during the week prior to presentation. Sixty (71%) patients received the first dose of an antibiotic active against S. pneumoniae prior to sputum culture collection. The median duration of therapy prior to obtaining sputum cultures was 5.6 hours (patients with sputum studies prior to antibiotics considered as 0 hours for analysis). Overall, S. pneumoniae grew from the sputum culture of 36 patients. The yield of sputum Gram stain and culture declined significantly in specimens that were obtained ≥10 hours after antibiotic therapy had been started (Table 1). Yield of Sputum Studies at Various Time Intervals After Initiation of Antibiotics Abbreviation: NS, not significant. a Interval refers to the time from initiation of intravenous antibiotics until a sputum sample was obtained. Yield of Sputum Studies at Various Time Intervals After Initiation of Antibiotics Abbreviation: NS, not significant. a Interval refers to the time from initiation of intravenous antibiotics until a sputum sample was obtained. Although IDSA guidelines recommend obtaining a sputum specimen prior to initiating antibiotics, our data suggest that the yield of sputum studies is not significantly reduced during the first 10 hours of treatment. To avoid delays in treatment, a practical approach might involve initiating antibiotics prior to sputum collection, provided that efforts are made to obtain a sputum sample within 10 hours. Potential conflicts of interest. Both authors: No reported conflicts. Both authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.