Abstract
Background: The decision to treat a suspected case of pertussis with antibiotics is usually based on a clinical diagnosis rather than waiting for laboratory confirmation. The current guideline focuses on making the clinical diagnosis of pertussis-associated cough in adults and children. Methods: The American College of Chest Physicians (CHEST) methodologic guidelines and the Grading of Recommendations, Assessment, Development, and Evaluation framework were used. The Expert Cough Panel based their recommendations on findings from a systematic review that was recently published on the topic; final grading was reached by consensus according to Delphi methodology. The systematic review was carried out to answer the Key Clinical Question: In patients presenting with cough, how can we most accurately diagnose from clinical features alone those who have pertussis-associated cough as opposed to other causes of cough? Results: In adults, after pre-specified meta-analysis exclusions, pooled estimates of sensitivity and specificity were generated for only 4 clinical features: paroxysmal cough, post-tussive vomiting, inspiratory whooping, and absence of fever. Both paroxysmal cough and absence of fever had high sensitivity (93.2% [95% CI, 83.2-97.4] and 81.8% [95% CI, 72.2-88.7], respectively) and low specificity (20.6% [95% CI, 14.7-28.1] and 18.8% [95% CI, 8.1-37.9]). Inspiratory whooping and posttussive vomiting had a low sensitivity (32.5% [95% CI, 24.5-41.6] and 29.8% [95% CI, 18.0-45.2]) but high specificity (77.7% [95% CI, 73.1-81.7] and 79.5% [95% CI, 69.4-86.9]). In children, after pre-specified meta-analysis exclusions, pooled estimates of sensitivity and specificity were generated for only 1 clinical feature in children (0-18 years): posttussive vomiting. Posttussive vomiting in children was only moderately sensitive (60.0% [95% CI, 40.3-77.0]) and specific (66.0% [95% CI, 52.5-77.3]). Conclusions: In adults with acute (< 3 weeks) or subacute (3-8 weeks) cough, the presence of whooping or posttussive vomiting should rule in a possible diagnosis of pertussis, whereas the lack of a paroxysmal cough or the presence of fever should rule it out. In children with acute (< 4 weeks) cough, posttussive vomiting is suggestive of pertussis but is much less helpful as a clinical diagnostic test. Guideline suggestions are made based upon these findings and conclusions.