Persistent cough following an acute respiratory tract infection is common in children, but clinicians may find it difficult to give accurate prognostic information on likely duration of cough without a microbiologic diagnosis. This study estimates the prevalence of Mycoplasma pneumoniae (Mp) and assesses the prognostic value of detecting Mp and respiratory viruses in children with persistent cough.
We retrospectively analyzed blood samples, nasopharyngeal aspirates (NPAs), and cough duration data from 179 children with persistent cough lasting 14 days or longer. Of these children, 37% had serologically confirmed Bordetella pertussis (pertussis). We detected Mp by polymerase chain reaction of NPAs and IgM serology, and respiratory viruses (human rhinoviruses, influenza viruses, respiratory syncytial viruses, and human metapneumovirus) by polymerase chain reaction of NPAs. We used Kaplan–Meier analyses to calculate median cough durations with 95% confidence intervals (CIs).
We detected Mp in 22 of 170 children with sufficient blood and/or NPAs (12.9%, 95% CI: 8.7–18.8). Cough duration in children with positive Mp serology (median: 39 days, 95% CI: 24–54) was significantly shorter than in children with positive pertussis serology (median: 118 days, 95% CI: 82–154, P < 0.001). The presence of respiratory viruses did not significantly lengthen cough duration in children with pertussis (median: 154 days, 95% CI: 74–234, P = 0.810). Only 3 children had both Mp and respiratory virus infections.
Mp is an important infection in children with persistent cough and is associated with a significantly shorter duration of cough than pertussis. However, detecting respiratory viruses does not add prognostic value in children with pertussis.
From the *Department of Primary Health Care, University of Oxford, Oxford, United Kingdom; †Respiratory and Systemic Infection Laboratory, London Health Protection Agency, London, United Kingdom; and ‡Respiratory Virus Unit, Virus Reference Department, London Health Protection Agency, London, United Kingdom.
Accepted for publication July 14, 2011.
Supported by the National Institute for Health Research (to K.W.).
The authors have no other funding or conflicts of interest to disclose.
Address for correspondence: Kay Wang, MRCGP, Department of Primary Health Care, University of Oxford, 23-38 Hythe Bridge St, Oxford, United Kingdom OX1 2ET. E-mail: email@example.com.