Using clinical criteria on seasonal and pandemic influenza: what to look for?

The reliability of clinical criteria on the diagnosis of influenza has been a question which has generated much debate in the scientific community. In Portugal, Influenza-like illness (ILI) cases have been notified by two sentinel networks, one of general practitioners (“Médicos-Sentinela”) and another of emergency rooms of hospitals and health centres, to the National Influenza Reference Laboratory in the context of the National Influenza Surveillance Programme, using clinical criteria adapted from the International Classification of Health Problems in Primary Care. With the emergence of a novel influenza A(H1N1) virus circulating in humans, the clinical definition of an influenza case was reformulated to accommodate the clinical features observed at that time. But how clinically different was the new variant A(H1N1) infection when compared to the seasonal influenza? In this study we propose to evaluate the signs and symptoms present in influenza cases diagnosed in Portugal during a seasonal influenza winter with those reported during the recent A(H1N1) pandemic. Also, we intend to evaluate the clinical criteria used for the notification of influenza cases during these two periods. 2. Materials and methods ILI cases were diagnosed for influenza at the National Influenza Reference Laboratory in the context of the National Influenza Surveillance Programme during the 2008/2009 influenza winter season and 2009/2010 pandemic period, from week 38/2008 through week 20/2010. Clinical information on those cases included the presence of signs and symptoms related with ILI as defined by the International Classification on Health Problems in Primary Care and on the case definition introduced by the European Commission Decision 202/253/EC (signs/symptoms analysed: sudden onset of symptoms, presence of fever, weakness, headache, myalgia, cough, sore throat, respiratory difficulty, chills and contact with an influenza patient). The odds ratio (OR) of being positive for influenza for each sign/symptom was calculated, independently and using multivariable logistic regression with all the signs/symptoms. Results were compared with the clinical definition of influenza used during both periods. 3. Results During the 2008/2009 winter, when seasonal influenza A(H3) viruses were dominant, all signs/symptoms analysed revealed an OR associated with a risk of being positive for influenza. Those with a higher risk were cough (OR 9.5, CI95% 4.7-19.3), fever (OR 4.5, CI95% 2.8-7.1), chills (OR 2.3, CI95% 1.5-3.5), myalgia (OR 1.7, CI95% 1.1-2-7) and contact with another ILI patient (OR 1.5, CI95% 1.1-2.0). Adjusting for all the signs/symptoms, the multivariable logistic regression reveals cough (OR 10.7, CI95% 5.2-22.0), fever (OR 5.1, CI95% 3.2-8.2) and contact with another patient (OR 1.4, CI95% 1.0-1.9) to be statistically significant. For the pandemic season 2009/2010, the signs/symptoms associated with a higher risk of being positive for influenza were fever (OR 4.5, CI95% 2.5-8.3), cough (OR 3.2, CI95% 2.1-4.8) and contact with another patient (OR 2.0, CI95% 1.4-2.8). Multivariable logistic regression also indicates these signs/symptoms to be statistically significant, after adjustment. 4. Conclusions Although initial evidences that the new influenza pandemic variant A(H1N1) could cause a clinically different infection, and fears that the disease would present a more severe profile than seasonal influenza, data collected in our country through the National Influenza surveillance Programme indicates that both situations were clinically similar in their presentation. This fact has been supported by studies from other countries