It has been clear for quite some years, at least for anyone dealing daily with chronic rhinosinusitis (CRS) patients, that CRS is an ‘umbrella’ diagnosis. There are significant differences between patients, including different demographic data, different endoscopic and radiographic images, different disease burden and different responses to surgery and medications. The holy grail of modern rhinology has been to phenotype CRS patients, searching for its own “unifying theory of everything” that can explain these differences. The authors of this clever study have started the other way round. Instead of creating another difficult to prove theory, they looked closely at their data. They performed cluster analysis, a statistical method of defining groups within patients, using endoscopic, radiographic, age, gender and patient outcome data, without using any a priori classification model. The result was five clusters, each characterised by different objective signs of disease (endoscopy and CT scans) and different PROM measures. Cluster 5 was the smaller group but the one worst affected in terms of loss of productivity and quality of life, with half of its patients reporting depression, and the second worst objective measures of disease. On the other hand, cluster 2 had the oldest patients (mean age of 61 years), the biggest number of male patients, with the worst objective signs of disease but only intermediately depressed QOL. Overall an interesting read, food for thought and open to many interpretations. Unfortunately the authors were unable to collect any biomarker data – hopefully someone will follow up on this study and attempt to explore any underlying connections. Watch this space.

Identification of chronic rhinosinusitis phenotypes using cluster analysis.
Soler ZM, Madison Hyer J, Ramakrishnan V, Smith TL, Mace J, Rudmik L, Schlosser RJ.
Share This
Christos Georgalas

Academic Medical Center, The Netherlands.

View Full Profile