Have you read my post on the exam based approach to the patient with a sore throat? Cool, you should also be using the Centor Criteria to help decide who needs to be tested for strep.
These criteria can be used to assess the likelihood of bacterial infection in patients with a sore throat. It was studied in adult patients, and assigns 1 point to each of 4 criteria:
- History of fever
- Tonsillar exudate
- Tender anterior cervical adenopathy
- Absence of cough
The modified Centor Criteria added age into the mix
- <15 years add 1 point
- >44 years subtract 1 point
The risk of Group A Streptococcal pharyngitis differs based on the number of “points”
- <2 points: Risk of GAS infection <10% – no antibiotics or testing necessary
- 2 points: Risk of strep 15%
- 3 points: Risk of GAS 32%
- >4 points: Risk of GAS 56% with a subsequent throat culture positive predictive value of 40-60%
If you have zero of the 4 original variables the negative predictive value is 80%. Though not implicitly stated in the criteria they can be extrapolated to help rule out strep throat and enable you to avoid swabbing for a rapid strep. Now, with any kid under 15 years of age you’ll always assign at least 1 point – thus, the above NPV doesn’t apply, but you still see the risk of strep at <10% if the child is afebrile, has a cough, and does not have tonsillar exudate or tender anterior cervical lymph nodes.
This is important if you don’t have rapid strep antigen testing available. If you do, then use it – the specificity is in the upper 90%s. So, it is great at “ruling in” strep. The sensitivity is 90-95%. So you could get 1/10 or 1/20 patients with a false negative result. You’ll probably send a backup culture – but do this with the knowledge that in the USA we’ll likely see 1-2 cases of acute rheumatic fever per state per year.
Overall the Centor Criteria are great, and important to teach, and especially to practice in a location where rapid strep testing is not available/feasible.
