By Dean L. Winslow, MD, FACP, FIDSA, FPIDS

Professor of Medicine, Division of General Medical Disciplines, Division of Infectious Diseases and Geographic Medicine, Stanford University School of Medicine

Dr. Winslow reports no financial relationships relevant to this field of study.

SYNOPSIS: Researchers analyzed the diagnostic accuracy and yield of sputum Gram stain (SGS) in community-acquired pneumonia across 24 studies of 4,533 adult patients in a meta-analysis. SGS was specific for the diagnosis of Streptococcus pneumoniae and Haemophilus influenzae infection.

SOURCE: Ogawa H, et al. Sputum Gram stain for bacterial pathogen diagnosis in community-acquired pneumonia: A systematic review and Bayesian meta-analysis of diagnostic accuracy and yield. Clin Infect Dis 2019 Sep 5. pii: ciz876. doi: 10.1093/cid/ciz876. [Epub ahead of print].

The researchers examined MEDLINE, Embase, and Cochrane databases through August 2018 to identify studies on the diagnostic accuracy, yield (percentage of patients with any pathogen[s] correctly identified by sputum Gram stain [SGS]), and clinical outcomes of SGS in adult patients with community-acquired pneumonia (CAP). Investigators included 24 studies reporting on 4,533 patients in the meta-analysis. When good-quality sputum specimens (defined as having 25 white blood cells [WBCs] and < 10 epithelial cells per low power field) were selected, SGS exhibited a sensitivity of 0.69 (95% credible intervals [CrI], 0.56-0.80) and specificity of 0.91 (CrI, 0.83-0.96) for detecting Streptococcus pneumoniae, and it displayed a sensitivity of 0.76 (CrI, 0.60-0.87) and specificity of 0.97 (CrI, 0.91-0.99) for Haemophilus influenzae. Bacterial pathogens were identified in 73% (CrI, 26-96%) of good-quality specimens, and in 36% (CrI, 22-53%) of all specimens, regardless of quality. There were too few bacterial pathogens other than S. pneumoniae and H. influenzae to enable analysis of sensitivity and specificity for these other bacterial pathogens.


For a variety of reasons, the routine examination of Gram-stained sputum specimens prior to prescription of antibiotics for patients admitted to the hospital for CAP has fallen by the wayside in the past 30 years. During my own training in internal medicine and infectious diseases in the late 1970s, we took great pride in standing by the bedside, even clapping our patients on their backs, and exhorting them (“Really deep! You can do it, sir!”) to produce a good-quality sputum specimen for us to walk down the hall to the “stat lab.” We often would prepare the smear ourselves, stain it (a mark of distinction as an ID fellow in the old days was having purple fingertips from the stain), and examine it carefully under the microscope, usually before we gave antibiotics. Now in the United States, one must be an ASCP-trained and certified laboratory technologist to perform almost any diagnostic test, the micro lab may be located off-site several miles from the hospital, and inflexible quality measures mandate (with weak supporting evidence) administration of empiric antibiotics within one hour of triage, even in patients who are relatively stable and not in shock.1

This paper gave me hope that the SGS remains an incredibly useful rapid diagnostic test that can help us safely “streamline” our antibiotics up-front in many cases of CAP. The advantages of doing this should be apparent to all readers of ID Alert in terms of reducing the selection of antibiotic-resistant organisms, reducing the incidence of side effects from overly broad-spectrum antibiotics, and likely reducing Clostridioides difficile infection rates. The challenges to having the Gram stain come back into practice as an important clinical decision-making tool in the treatment of CAP include systems issues (hospital labs need to make the facilities and personnel available to offer this test in a rapid and efficient fashion) and the need to re-educate practicing doctors and trainees on the importance of this simple test. Although the authors of this meta-analysis did not look at SGS cost-effectiveness, it has to be more cost-effective than all of the thousands of serum troponin levels and CT angiograms that are ordered every day in U.S emergency departments for less-than-certain indications.


  1. Gilbert D, Kalil A, Klompas M, et al. IDSA Position Statement: Why IDSA Did Not Support the Surviving Sepsis Campaign Guidelines. Clin Infect Dis 2018;66:1631-1635.