2016 Section 5 Green Book

Familial risk of CRSwNP and CRSsNP

(CRSsNP), a Mendelian inheritance pattern is not sup- ported, but rather a multifactorial etiology with both ge- netic and environmental influences. The complex nature of this pathogenesis necessitates a large-scale study to better elucidate familial patterns that may indicate potential for an underlying genetic susceptibility. Our objective in this study was to determine whether relatives of patients with CRSwNP and CRSsNP are at an increased risk based on observed familial clustering in a unique population research database. Subjects and methods The Utah Population Database (UPDB) is a dynamic, shared resource located at the University of Utah and con- sists of computerized data records for over 7 million in- dividuals. It is the only database of its kind in the United States and one of a few in the world; most families living in Utah are represented in the UPDB. 8 For example, of all individuals born in Utah in 1950, 79% have grandparent information and 67% have 5 or more previous generations documented in this resource. The UPDB includes statewide vital records and hospital inpatient and ambulatory facil- ity data that are linked to individuals in multigenerational pedigrees. Case population We identified patients ages 18 years or older at index di- agnosis with an International Classification of Diseases, 9th revision, clinical modification (ICD-9-CM) diagnosis of CRS, defined as 1 or more of ICD-9-CM codes 473.x appearing in the medical record (any diagnostic position) from 1996 to 2011. Patients without a diagnosis of nasal polyps, defined as an absence of ICD-9-CM 471.x, were considered CRSsNP and those with this ICD-9-CM code were considered CRSwNP. Individuals with a history of CF (ICD-9-CM277.0), inverted papilloma (ICD-9-CM212.0), or head/facial trauma (ICD-9-CM 801.0–804.9) were ex- cluded. We elected to also use as inclusion criteria Cur- rent Procedural Terminology (CPT) codes for diagnostic nasal endoscopy (CPT 31231) and sinus surgery CPT codes 31233, 31237 (nasal polypectomy), 31254, 31255, 31256, 31276, and 31287 to increase the probability of accurately identifying a clinical diagnosis of CRSwNP or CRSsNP based on medical claims data. The resulting case popu- lation included 1638 probands with CRSwNP and 24,200 probands with CRSsNP with genealogy in UPDB to exam- ine familial risk. Control population Controls with no history of CRSwNP or CRSsNP were randomly selected from the Utah population and matched 5:1 to corresponding CRSwNP or CRSsNP cases on sex and birth year. To appropriately match exposure periods, a control had to have follow-up (known to reside in Utah)

at least as long as the date of diagnosis for their respective case as described. 9

Statistical analysis Using a software suite developed specifically for the UPDB and in conjunction with the software package R (Rx64 2.14.2 Windows Server 2008), 10 the magnitude of familial risk was estimated by Cox regression in order to assess the relative risk of CRSwNP and CRSsNP for relatives and spouses of CRSwNP and CRSsNP cases diagnosed in Utah between 1996 and 2011 compared to matched controls. 11 Estimates of familial risk were based on a haz- ard rate ratio (HR) of familial recurrence. This represents the ratio of the hazard rate for the occurrence of CRSwNP and CRSsNP among relatives of the index cases with the comparable hazard rate among the relatives of the con- trols. Cox models included cases and controls with ad- equate follow-up in Utah, who linked to a UPDB pedi- gree comprising at least 2 generations. Model covariates included sex and birth year. As observations within fam- ilies are non-independent, a Huber-White sandwich esti- mator of variance for clustered data was used. 10,12 This approach corrects for any families that were analyzed mul- tiple times because of multiple CRSwNP or CRSsNP cases within the family. Analyses were performed separately in which specific groups of relatives of the cases were com- pared to the comparable relatives of the matched controls as follows: first-degree relatives (1stDRs), including parents, children, and siblings; second-degree relatives (2ndDRs), including grandparents, grandchildren, aunts/uncles, and nieces/nephews; third-degree relatives (3rdDRs), primarily first cousins; fourth-degree relatives (first cousins once re- moved); and fifth-degree relatives (second cousins). We also examined risk in spouses of probands to assess evidence of disease susceptibility from a shared environment. Spouses were defined as the married or unmarried co-parent of the index cases or controls who had children determined from UPDB records (77% and 71% of all CRSwNP cases or con- trols, respectively; 78% and 72% of all CRSsNP cases or controls, respectively). We randomly selected 81 patients from a total of 682 indi- viduals treated in the University of Utah Healthcare system during 2008 to 2011 for whom we had access to electronic charts. This patient sample met our inclusion criteria based on the ICD-9 and CPT codes outlined in the Case popu- lation section and a chart review was performed to com- pare diagnoses based on ICD-9/CPT codes to the clinical record. We used criteria published in 2007 by Rosenfeld et al., 13 which give cardinal symptoms for CRS and then require verification of the symptomatic diagnosis by ei- ther computed tomography (CT) or nasal endoscopy. Two electronic charts were unavailable, leaving 79 available for Results Verification of diagnosis

International Forum of Allergy & Rhinology, Vol. 5, No. 4, April 2015

72