Distinct response patterns of the 22-item Sino-Nasal Outcome Test (SNOT-22) may be used to distinguish nonsinogenic headaches from chronic rhinosinusitis, according to a study published in the International Forum of Allergy & Rhinology.
The investigators of this retrospective cohort study sought to examine the usefulness of SNOT-22 and identify score patterns to develop negative predictors of chronic rhinosinusitis in patients diagnosed with nonsinogenic headache.
A total of 724 adults were included in the study, with 560 diagnosed with chronic rhinosinusitis and 164 with nonsinogenic headache. An additional 412 independent patients (324 with chronic rhinosinusitis and 88 with nonsinogenic headache) were recruited to validate the negative predictive values calculated in the discovery population. Nonsinogenic headache was defined as headache accompanied by facial pain and facial pressure in the absence of sinonasal inflammation (using endoscopy and sinus-dedicated computed tomography).
From the SNOT-22 results in the discovery population, researchers selected potential predictors of chronic rhinosinusitis based on differences between the 2 groups. A receiver operating characteristic curve was used to assess the negative predictive value of SNOT-22 score patterns by generating optimal cutoff points for each potential predictor.
Within the discovery population, some SNOT-22 patterns were significantly higher in patients with nonsinogenic headache vs chronic rhinosinusitis, including domain 3 (ear/facial) and domain 4 (psychological dysfunction); others were significantly lower, including domain 1 (rhinologic) and domain 2 (extranasal rhinologic). Among individual item scores, 2, 4, 5, 6, and 8 were lower in the nonsinogenic headache group than in patients with chronic rhinosinusitis, but items 9, 10, 11, 12, 13, 17, 18, 19, and 20 were higher in the patients with nonsinogenic headache.
Based on areas under the receiver operating characteristic curve, the best predictors of nonsinogenic headache were the domain 3/ domain 1 ratio, item 12, domain 3, and the (item 12 + item 10)/(item 4 + item 5) ratio. The best negative predictor of chronic rhinosinusitis was a domain 3/domain 1 ratio >0.66, which had high negative predictive values in both the discovery (0.880) and validation (0.889) populations. Differences in total SNOT-22 scores between nonsinogenic headache and CRS were not significant.
Limitations to the study include the retrospective design and the lack of a formal diagnosis of nonsinogenic headache by a neurologist. Furthermore, the researchers did not analyze long-term patient data, and it is possible that a proportion of nonsinogenic headache patients may have developed chronic rhinosinusitis during follow-up.
”Our study showed that patients with nonsinogenic headache have distinct SNOT-22 score patterns as compared [with] patients with [chronic rhinosinusitis],” concluded the study authors. “A domain 3 (Ear/facial)/domain 1 (Rhinologic) score ratio greater than 0.66 is capable of ruling out the presence of [chronic rhinosinusitis] in nearly 9 of 10 patients in an aggregate population of over 1600 patients. These findings suggest this ratio may be used as a rapid and noninvasive method to counsel patients and help guide rational diagnostic and therapeutic algorithms.”
Wu D, Gray ST, Holbrook EH, BuSaba NY, Bleier BS. SNOT-22 score patterns strongly negatively predict chronic rhinosinusitis in patients with headache [published online October 3, 2018]. Int Forum Allergy Rhinol. doi:10.1002/alr.22216