Premium
Risk‐adjusted relationship between voriconazole utilization and non‐melanoma skin cancer among lung and heart/lung transplant patients
Author(s) -
McLaughlin J.M.,
Equils O.,
Somerville K.T.,
Aram J.A.,
Schlamm H.T.,
Welch V.L.,
Mardekian J.,
Barbers R.G.
Publication year - 2013
Publication title -
transplant infectious disease
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.69
H-Index - 67
eISSN - 1399-3062
pISSN - 1398-2273
DOI - 10.1111/tid.12063
Subject(s) - medicine , voriconazole , hazard ratio , lung cancer , proportional hazards model , confidence interval , lung , confounding , skin cancer , cancer , dermatology , antifungal
Background We examined the relationship between voriconazole utilization and non‐melanoma skin cancer ( NMSC ) development among adult lung and heart/lung transplant patients who were continuously enrolled in a large U.S. commercial health plan. Methods Cox proportional hazards regression models were constructed to assess both the crude and adjusted effect of voriconazole usage on NMSC development. Overall, 467 adult lung (98%) and heart/lung (2%) transplant patients (60% male) with median age of 58 years were analyzed. Results Fifty‐seven (12%) patients developed NMSC over a median follow‐up time of 610 days. At the crude level, patients with any (vs. none) claim for voriconazole were more likely to develop NMSC (19% vs. 12%, hazard ratio [ HR ]: 1.74, 95% confidence interval [ CI ]: 1.02, 2.96, P = 0.04). However, after statistical adjustment for demographic and clinical factors, the effect was largely diminished and no longer statistically significant ( HR : 1.23, 95% CI : 0.71, 2.14, P = 0.45). Results were similar when modeling average and total dose of voriconazole. Risk factors significantly related to NMSC development were being male, older age, sun exposure, history of chronic obstructive pulmonary disorder, and history of immune disorder. Conclusion Results suggest that the relationship between voriconazole utilization and NMSC among lung transplant patients may be a result of confounding by indication, and that controlling for underlying patient characteristics is paramount.