Identification of chronic rhinosinusitis phenotypes using cluster analysis

Zachary M. Soler, J. Madison Hyer, Viswanathan Ramakrishnan, Timothy Smith, Jess Mace, Luke Rudmik, Rodney J. Schlosser

Research output: Contribution to journalArticle

43 Citations (Scopus)

Abstract

Background: Current clinical classifications of chronic rhinosinusitis (CRS) have been largely defined based upon preconceived notions of factors thought to be important, such as polyp or eosinophil status. Unfortunately, these classification systems have little correlation with symptom severity or treatment outcomes. Unsupervised clustering can be used to identify phenotypic subgroups of CRS patients, describe clinical differences in these clusters and define simple algorithms for classification. Methods: A multi-institutional, prospective study of 382 patients with CRS who had failed initial medical therapy completed the Sino-Nasal Outcome Test (SNOT-22), Rhinosinusitis Disability Index (RSDI), Medical Outcomes Study Short Form-12 (SF-12), Pittsburgh Sleep Quality Index (PSQI), and Patient Health Questionnaire (PHQ-2). Objective measures of CRS severity included Brief Smell Identification Test (B-SIT), CT, and endoscopy scoring. All variables were reduced and unsupervised hierarchical clustering was performed. After clusters were defined, variations in medication usage were analyzed. Discriminant analysis was performed to develop a simplified, clinically useful algorithm for clustering. Results: Clustering was largely determined by age, severity of patient reported outcome measures, depression, and fibromyalgia. CT and endoscopy varied somewhat among clusters. Traditional clinical measures, including polyp/atopic status, prior surgery, B-SIT and asthma, did not vary among clusters. A simplified algorithm based upon productivity loss, SNOT-22 score, and age predicted clustering with 89% accuracy. Medication usage among clusters did vary significantly. Conclusion: A simplified algorithm based upon hierarchical clustering is able to classify CRS patients and predict medication usage. Further studies are warranted to determine if such clustering predicts treatment outcomes.

Original languageEnglish (US)
Pages (from-to)399-407
Number of pages9
JournalInternational Forum of Allergy and Rhinology
Volume5
Issue number5
DOIs
StatePublished - May 1 2015

Fingerprint

Cluster Analysis
Phenotype
Smell
Polyps
Endoscopy
Fibromyalgia
Discriminant Analysis
Nose
Eosinophils
Sleep
Asthma
Outcome Assessment (Health Care)
Prospective Studies
Depression
Health

Keywords

  • Cluster analysis
  • Phenotype
  • Quality of life
  • Sinusitis
  • Staging

ASJC Scopus subject areas

  • Immunology and Allergy
  • Otorhinolaryngology

Cite this

Soler, Z. M., Hyer, J. M., Ramakrishnan, V., Smith, T., Mace, J., Rudmik, L., & Schlosser, R. J. (2015). Identification of chronic rhinosinusitis phenotypes using cluster analysis. International Forum of Allergy and Rhinology, 5(5), 399-407. https://doi.org/10.1002/alr.21496

Identification of chronic rhinosinusitis phenotypes using cluster analysis. / Soler, Zachary M.; Hyer, J. Madison; Ramakrishnan, Viswanathan; Smith, Timothy; Mace, Jess; Rudmik, Luke; Schlosser, Rodney J.

In: International Forum of Allergy and Rhinology, Vol. 5, No. 5, 01.05.2015, p. 399-407.

Research output: Contribution to journalArticle

Soler, ZM, Hyer, JM, Ramakrishnan, V, Smith, T, Mace, J, Rudmik, L & Schlosser, RJ 2015, 'Identification of chronic rhinosinusitis phenotypes using cluster analysis', International Forum of Allergy and Rhinology, vol. 5, no. 5, pp. 399-407. https://doi.org/10.1002/alr.21496
Soler, Zachary M. ; Hyer, J. Madison ; Ramakrishnan, Viswanathan ; Smith, Timothy ; Mace, Jess ; Rudmik, Luke ; Schlosser, Rodney J. / Identification of chronic rhinosinusitis phenotypes using cluster analysis. In: International Forum of Allergy and Rhinology. 2015 ; Vol. 5, No. 5. pp. 399-407.
@article{c0bb943aca82469b9f03d271bdda8d12,
title = "Identification of chronic rhinosinusitis phenotypes using cluster analysis",
abstract = "Background: Current clinical classifications of chronic rhinosinusitis (CRS) have been largely defined based upon preconceived notions of factors thought to be important, such as polyp or eosinophil status. Unfortunately, these classification systems have little correlation with symptom severity or treatment outcomes. Unsupervised clustering can be used to identify phenotypic subgroups of CRS patients, describe clinical differences in these clusters and define simple algorithms for classification. Methods: A multi-institutional, prospective study of 382 patients with CRS who had failed initial medical therapy completed the Sino-Nasal Outcome Test (SNOT-22), Rhinosinusitis Disability Index (RSDI), Medical Outcomes Study Short Form-12 (SF-12), Pittsburgh Sleep Quality Index (PSQI), and Patient Health Questionnaire (PHQ-2). Objective measures of CRS severity included Brief Smell Identification Test (B-SIT), CT, and endoscopy scoring. All variables were reduced and unsupervised hierarchical clustering was performed. After clusters were defined, variations in medication usage were analyzed. Discriminant analysis was performed to develop a simplified, clinically useful algorithm for clustering. Results: Clustering was largely determined by age, severity of patient reported outcome measures, depression, and fibromyalgia. CT and endoscopy varied somewhat among clusters. Traditional clinical measures, including polyp/atopic status, prior surgery, B-SIT and asthma, did not vary among clusters. A simplified algorithm based upon productivity loss, SNOT-22 score, and age predicted clustering with 89{\%} accuracy. Medication usage among clusters did vary significantly. Conclusion: A simplified algorithm based upon hierarchical clustering is able to classify CRS patients and predict medication usage. Further studies are warranted to determine if such clustering predicts treatment outcomes.",
keywords = "Cluster analysis, Phenotype, Quality of life, Sinusitis, Staging",
author = "Soler, {Zachary M.} and Hyer, {J. Madison} and Viswanathan Ramakrishnan and Timothy Smith and Jess Mace and Luke Rudmik and Schlosser, {Rodney J.}",
year = "2015",
month = "5",
day = "1",
doi = "10.1002/alr.21496",
language = "English (US)",
volume = "5",
pages = "399--407",
journal = "International Forum of Allergy and Rhinology",
issn = "2042-6976",
publisher = "Wiley-Blackwell",
number = "5",

}

TY - JOUR

T1 - Identification of chronic rhinosinusitis phenotypes using cluster analysis

AU - Soler, Zachary M.

AU - Hyer, J. Madison

AU - Ramakrishnan, Viswanathan

AU - Smith, Timothy

AU - Mace, Jess

AU - Rudmik, Luke

AU - Schlosser, Rodney J.

PY - 2015/5/1

Y1 - 2015/5/1

N2 - Background: Current clinical classifications of chronic rhinosinusitis (CRS) have been largely defined based upon preconceived notions of factors thought to be important, such as polyp or eosinophil status. Unfortunately, these classification systems have little correlation with symptom severity or treatment outcomes. Unsupervised clustering can be used to identify phenotypic subgroups of CRS patients, describe clinical differences in these clusters and define simple algorithms for classification. Methods: A multi-institutional, prospective study of 382 patients with CRS who had failed initial medical therapy completed the Sino-Nasal Outcome Test (SNOT-22), Rhinosinusitis Disability Index (RSDI), Medical Outcomes Study Short Form-12 (SF-12), Pittsburgh Sleep Quality Index (PSQI), and Patient Health Questionnaire (PHQ-2). Objective measures of CRS severity included Brief Smell Identification Test (B-SIT), CT, and endoscopy scoring. All variables were reduced and unsupervised hierarchical clustering was performed. After clusters were defined, variations in medication usage were analyzed. Discriminant analysis was performed to develop a simplified, clinically useful algorithm for clustering. Results: Clustering was largely determined by age, severity of patient reported outcome measures, depression, and fibromyalgia. CT and endoscopy varied somewhat among clusters. Traditional clinical measures, including polyp/atopic status, prior surgery, B-SIT and asthma, did not vary among clusters. A simplified algorithm based upon productivity loss, SNOT-22 score, and age predicted clustering with 89% accuracy. Medication usage among clusters did vary significantly. Conclusion: A simplified algorithm based upon hierarchical clustering is able to classify CRS patients and predict medication usage. Further studies are warranted to determine if such clustering predicts treatment outcomes.

AB - Background: Current clinical classifications of chronic rhinosinusitis (CRS) have been largely defined based upon preconceived notions of factors thought to be important, such as polyp or eosinophil status. Unfortunately, these classification systems have little correlation with symptom severity or treatment outcomes. Unsupervised clustering can be used to identify phenotypic subgroups of CRS patients, describe clinical differences in these clusters and define simple algorithms for classification. Methods: A multi-institutional, prospective study of 382 patients with CRS who had failed initial medical therapy completed the Sino-Nasal Outcome Test (SNOT-22), Rhinosinusitis Disability Index (RSDI), Medical Outcomes Study Short Form-12 (SF-12), Pittsburgh Sleep Quality Index (PSQI), and Patient Health Questionnaire (PHQ-2). Objective measures of CRS severity included Brief Smell Identification Test (B-SIT), CT, and endoscopy scoring. All variables were reduced and unsupervised hierarchical clustering was performed. After clusters were defined, variations in medication usage were analyzed. Discriminant analysis was performed to develop a simplified, clinically useful algorithm for clustering. Results: Clustering was largely determined by age, severity of patient reported outcome measures, depression, and fibromyalgia. CT and endoscopy varied somewhat among clusters. Traditional clinical measures, including polyp/atopic status, prior surgery, B-SIT and asthma, did not vary among clusters. A simplified algorithm based upon productivity loss, SNOT-22 score, and age predicted clustering with 89% accuracy. Medication usage among clusters did vary significantly. Conclusion: A simplified algorithm based upon hierarchical clustering is able to classify CRS patients and predict medication usage. Further studies are warranted to determine if such clustering predicts treatment outcomes.

KW - Cluster analysis

KW - Phenotype

KW - Quality of life

KW - Sinusitis

KW - Staging

UR - http://www.scopus.com/inward/record.url?scp=84929048991&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84929048991&partnerID=8YFLogxK

U2 - 10.1002/alr.21496

DO - 10.1002/alr.21496

M3 - Article

C2 - 25694390

AN - SCOPUS:84929048991

VL - 5

SP - 399

EP - 407

JO - International Forum of Allergy and Rhinology

JF - International Forum of Allergy and Rhinology

SN - 2042-6976

IS - 5

ER -