Validity of complete blood count in detecting influenza among clinically suspected children in Sri Lanka: a retrospective study from low resource setting country.

introductionInfluenza in children ranges from subclinical to severe illness with complications. However, confirmation of its diagnosis based on a positive PCR test for nasopharyngeal swab has become a challenge due to its cost and the discomfort of the procedure. The study aimed to assess the validity of Complete Blood Count (CBC) parameters in detecting influenza among clinically suspected children.

methodsA retrospective study was conducted in a sample of children clinically suspected of influenza at three private hospitals in Galle. All children with clinical suspicion of influenza by a consultant pediatrician who had both CBC and PCR tests for nasopharyngeal swab, as laboratory investigations extracted from available databases at relevant laboratories. The PCR test for nasopharyngeal swab was considered the gold standard test. Known group comparison and criterion validity were used in the validation process at a 0.05 significance level.

resultsThe study included 166 children (50% males). Median (IQR) age of the sample was 4 (2 to 6) years. Fifty-nine children (35.5%) had positive PCR results for influenza. There was no statistically significant difference in CBC parameters according to gender or age group. The median (IQR) WBC(p < 0.001), neutrophil(p = 0.005), lymphocytes(p = 0.002), and platelet count(p < 0.001) were statistically significantly lower among children with Influenza, while for PCV(p = 0.021), vice versa. They all had statistically significant diagnostic accuracy (p < 0.05) based on ROC curve analysis. WBC count had the highest diagnostic validity (AUC = 0.71; 95% CI 0.63-0.79). The cut-off points in detecting influenza were in the normal range for all parameters. All the CBC parameters had satisfactory diagnostic accuracy in detecting influenza among clinically suspected children. CONCLUSIONS & RECOMMENDATIONS: There is a statistically significant difference in full blood count parameters between the influenza-positive and influenza-negative groups. The cut-off points in detecting influenza were in the normal range for all parameters. CBC shows some predictive value but cannot replace PCR due to modest diagnostic accuracy. CBC findings combined with clinical presentation can be recommended as a cost-effective method instead of PCR to use in clinical practice, especially in low-resource settings. CLINICAL TRIAL REGISTRATION NUMBER: Not applicable.