Influenza season is in full swing. Most areas of the US are seeing widespread influenza activity.Though some patients can become very ill, and even die most do well. The main purpose of this post is to encourage you to use clinical history and physical examination to allow you to make the diagnosis of influenza, especially when disease prevalence is high. The bottom line is that you are smart, and if you think it is the flu you are probably right.
What are the classic symptoms of influenza?
The classic symptoms of uncomplicated influenza are varied and depend on the age of the child. The most common story is abrupt onset of fever, headache, myalgia, malaise and upper respitratory tract symptoms. Younger children may also have febrile seizures, more prominent respiratory symptoms (more croup-like or bronchiolitis-like) and more GI symptoms. One retrospective study of children with flu managed as an outpatients noted that the most common symptoms, in descending order are;
- Fever in 95+%
- Cough (77%) and rhinitis (78%) in more than three fourths
- Headache in 26-39%
- Myalgias 7% in children 3-13 years, 13% 7-13 years
Most kids improve over approximately one week (with or without oseltamivir); cough can last longer though. Some children will even have several weeks of mild exertion fatigue. Clearly symptoms are variable. Even the individual virus plays a role – influenza B strains for instance seem to cause more musculoskeletal symptoms than A strains. So with these variable symptoms it would make sense to offer testing right?
What are the available tests for the flu?
Test #1 Your brain – Make the diagnosis based on clinical suspicion and using your H&P
Test #2 Point of care/rapid antigen, RNA
- These tests take less than 15 minutes and can be obtained from swab, aspirate or wash of the nasal passages or throat
- The overall sensitivity, even when prevalence is high is 50-70%
- The specificity in times of high prevalence is 90-95%
- False negative rates are especially high when community prevalence is high as well
- False positive rates are much more common when disease prevalence rate is low
- Cost as low as $20-25; more at retail clinics ~$35-40
Test #3 Molecular/PCR testing
- These tests take at least 1 to 8 hours and can be obtained from swab, aspirate or wash of the nasal passages or throat
- They are often a part of a respiratory viral panel
- Per a metaanalysis from Merckx et al the Influenza A sensitivity was 92% (95% CI 85-96) and 95% (95% CI 87-98) for influenza B
- The aforementioned metaanalysis showed a specificity of >99% for influenza A & B
- Cost is $80-110 dollars for the influenza PCR alone; considerably higher for the multi-virus panels
Who and when to test
There is no perfect test for the flu. If you really must rule it in or out then you are looking at molecular?PCR testing which is not fast, nor cheap. Antigenic testing has a low sensitivity making it not the best at ruling out the flu. I think that this flow chart from the CDC website is awesome as it does a great job laying out the thought process for testing versus not.
OK, so you should test when the results will influence your clinical management. Obviously this is patient/situation dependent; but here are some examples of situations when testing makes sense:
- Suspected influenza in patients with high risk of complications
- Chronic diseases like CF, congenital heart disease, sickle cell, diabetes, CP, muscular dystrophy etc,.
- The immunosuppressed – chronic steroids, cancer treatment, HIV
- Pregnant women
- Children on long term aspirin treatment
- Children under the age of 2
- Children admitted to the hospital for respiratory or neurologic complications of the flu
- A situation where the diagnosis will influence the ongoing workup – the febrile neonate is a perfect example
The antigen +/- PCR question should consider the above conditions plus the overall cost. Only test if it will make a difference in whether or not you treat (see below) or in complex/admitted patients. Otherwise you are not necessarily any more accurate than your clinical suspicion. And yes, there may be the assumption that parents expect or prefer a test. Even in a busy urgent care setting with rapid turnover I think that this mindset is a fallacy. Parents want to know that their child is safe. They may have read about pediatric deaths from the flu and want to be reassured. You can do this with words and empathy alone in most cases. If you go into the room expecting that they want a test then you are already biased. Resist the temptation to test before seeing the patient – as this, in my experience, just begets more testing. Then, we increase costs and run low on resources for the patients that truly need them. Spend time developing a good communication script for diagnostic discussions and make sure that your educational materials are broadly appropriate and readily available. A patient who gets a test doesn’t receive better care necessarily – contrary to what you may believe the test is not what parents want, rather they want an accurate diagnosis and appropriate education and reassurance delivered in a professional manner.
The CDC recommendations on the use of antiviral agents
General dosing of oseltamivir (Tamiflu) 5 day course
- 0 to 3 months – 12 mg bid
- 4 to 5 months – 17 mg bid
- 6 to 11 months – 24 mg bid
- 1 to 12 years 4 mg/kg/day divided bid
- <15 kg 60 mg/kg divided bid
- 15-23 kg 90 mg/day divided bid
- 23-40 kg 120 mg/day divided bid
- >13 years or >40 kg 150 mg/day divided bidReferences
Merckx J, Wali R, Schiller I, Caya C, Gore GC, Chartrand C, Dendukuri N, Papenburg J. Diagnostic Accuracy of Novel and Traditional Rapid Tests for Influenza Infection Compared With Reverse Transcriptase Polymerase Chain Reaction: A Systematic Review and Meta-analysis. Ann Intern Med. 2017 Sep 19;167(6):394-409. doi: 10.7326/M17-0848. Epub 2017 Sep 5.
Mansour et al. Comparative Cost Analysis Between PCR Testing and DFA Testing for Diagnosing Respiratory Virus Infections. American Journal of Clinical Pathology, Volume 144, Issue suppl_2, 1 October 2015, Pages A209.