Background Pneumonia is a major cause of morbidity and mortality in under-five children with about 5million deaths annually in developing countries1. CMV is responsible for serious morbidity and mortality in immunodeficient children with pneumonia2.
Objective To determine the incidence of CMV associated pneumonia in children with severe lower respiratory tract infection (LRTI).
Methods Under-5year children with severe LRTI were enrolled over a 12months period. Severity criteria:accessory muscle use, supplemental oxygen, or assisted ventilation. Anthropometry and HIV status were recorded. Throat swabs were taken for CMV PCR and CMV serology was done. Consent and ethical approval obtained.
Results 107children, aged 2weeks to 46 months (mean 5.96 mths) enrolled. Incidence of laboratory confirmed CMV was 40%(35/87); 67% among HIV-infected and 28% among HIV-uninfected (p<0.05)).Of 100 children tested for HIV infection, 30% were positive(30/100).The mean ages of HIV-infected and uninfected children were similar (5.83±5.77 vs 5.99±9.43 respectively). There was a slight difference in height-for-age Z-scores between HIV-infected (–2.51±3.22) and uninfected (–1.17±3.41) (p=0.07). Incidence of CMV was not associated with age or nutritional status. There were 18 deaths,17% mortality; this was significantly higher (p<0.01) among HIV-infected children (40%) than in HIV-uninfected (9%). Mortality was higher amongst those with positive CMV throat swabs (20%) compared to negative CMV throat swabs (12%), (not statistically significant). Children with a positive throat CMV were likely to receive assisted ventilation (17%) compared to those with negative throat CMV (11.5%); not significant (p=0.058).
Conclusion Many under-5 children with severe LRTI had laboratory confirmed CMV infection. Incidence and mortality rate of CMV is higher in HIV-infected children and these patients are likely to require assisted ventilation.