Ali S. Raja1 and Christopher R. Carpenter2 1 Department of Emergency Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA 2 Department of Emergency Medicine, Washington University School of Medicine, St. Louis, MO, USA In typical years, up to 20% of residents in the United States develop influenza, resulting in between 140,000 and 710,000 annual hospitalizations between 2010 and 2020.1 While the incidence of influenza has been significantly less during the coronavirus disease 2019 (COVID‐19) pandemic,2 once mask restrictions and physical distancing recommendations are withdrawn, influenza rates will likely return to their baseline levels. Influenza increases morbidity and mortality for a number of specific patient populations, including the very young and older adults and those with chronic lung disease.3,4 Geriatric patients (>65 years) account for 71–85% of annual influenza deaths in the United States.5 In general, healthy children and young adults do not experience life‐threatening complications from influenza, but antigenic shifts in 1918, 1957, and 1968 were associated with viral pneumonia, acute respiratory distress syndrome, and multiorgan failure in these populations. Interpandemic influenza is an extremely unpleasant experience even for healthy individuals, with significant costs accrued due to diminished productivity, absenteeism, and related healthcare expenditures. In the United States, influenza results in between $2.0 and $5.8 billion in direct healthcare costs and $16.3 billion in lost earnings each year.6,7 Historically, emergency department (ED) resources have been occasionally strained by influenza‐like illness (ILI) pandemics such as severe acute respiratory syndrome (SARS) in 2003 and H1N1 swine flu in 2009. The sustained multiyear COVID‐19 pandemic has, of course, fundamentally changed the this into a long‐term and devastating stress rather than an occasional periodic strain. Since infectious patients need to be sequestered during such pandemics in order to limit the exposure of the general population, diagnostic testing may be unavailable, emphasizing the importance of emergency clinicians’ understanding of the diagnostic accuracy of bedside testing to distinguish influenza from ILI.8 ILI is defined by the Centers for Disease Control (CDC) as a temperature higher than 37.8 °C with a cough or sore throat, but patients with influenza may present to the ED with a variety of atypical manifestations including delirium, falls, vomiting, incontinence, or diarrhea. Clinicians must integrate the chief complaint with a thorough review of systems within the context of their own regional influenza prevalence.9 Two strains of influenza exist (Type A and Type B) and are clinically indistinguishable. However, the distinction remains relevant because older antiviral agents (amantadine and rimantadine) are effective for Type A only – and even then, only somewhat. Many strains of influenza A, including the 2009 H1N1 variant, have been found to be resistant to these medications.10 More modern antiviral medications like oseltamivir may be effective against both types – especially in high‐risk populations – but only if used within 48 hours of symptom onset, emphasizing the importance of early diagnosis.11 Unfortunately, the “viral syndrome” is not unique to influenza and may be the presentation for a variety of viral (coronavirus, rhinovirus, adenovirus, and parainfluenza virus) and bacterial (Legionella, Mycoplasma, and Streptococcal) respiratory infections that do not respond to anti‐influenza therapy. The prevalence of influenza is in a constant state of flux, influenced by the geographical region, time of year, and patient population. In the United States, the CDC maintains a weekly report of influenza activity stratified by region and influenza subtype (http://www.cdc.gov/flu/weekly/). Although these estimates are derived from office‐based practices rather than the ED, they can be used to estimate regional influenza pretest probabilities while interpreting physical exam findings and contemplating viral testing and therapy. For example, in the week of 26 February 2022, the proportion of tested specimens that demonstrated influenza was 4.1% and ILI represented 1.5% of office visits. Therefore, the prevalence of influenza was 4.1% × 1.5% = 0.062% (the pretest probability). Similar influenza epidemiological data outside the United States are also available at the World Health Organization’s Global Influenza Surveillance and Response System (https://www.who.int/initiatives/global‐influenza‐surveillance‐and‐response‐system). What is the diagnostic accuracy of the history and physical exam for influenza? One diagnostic meta‐analysis, which intentionally neglected manuscripts in the immediate post‐SARS period, focusing instead on influenza. Based upon six studies of 7105 patients (five prospective, none ED based), they reported the summary estimates as shown in Table 35.1.12 Another systematic review that included two additional studies reported several more elements of the history that may be useful to rule in (rigors LR+ 7.2, fever and presenting within 3 days LR+ 4.0, and sweating LR+ 3.0) or rule out (any systemic symptoms LR− 0.36, coughing LR− 0.38, and inability to cope with daily activities LR− 0.39) influenza.13 Another systematic review evaluated the accuracy of clinical decision rules or combinations of signs and symptoms.14 They were able to find 10 heterogeneous studies that presented 14 clinical decision rules. Only one of the rules (the Flu‐Score) had been validated prospectively and externally,15 with an area under the curve (AUC) of only 0.66 (confidence interval [CI] 0.63–0.70) in the validation study, limiting its clinical utility. Stein et al
Chapter 35
Influenza
Background
Clinical question
Full access? Get Clinical Tree
