Background/Purpose: Psoriatic arthritis (PsA) is an inflammatory arthritis affecting approximately 520,000 patients in the US and up to one third of patients with psoriasis. Studies in the past have found an increased risk of psoriasis and worsening disease with smoking. However, the association between smoking and psoriatic arthritis in psoriatic patients and the general population remains much debated with some studies showing an increased risk while other showing an inverse relationship. We aimed to study the risk of psoriatic arthritis with smoking among the general population and patients with psoriasis using a large inpatient US database.

Methods: Using the Nationwide Inpatient Sample (NIS) data from 2009-2011, we identified current smokers based on International Classification of Diseases, Ninth Revision (ICD-9) code 305.1 and identified previous smokers who had ICD-9 code V15.82 but without 305.1. NIS is the largest publicly available all-payer inpatient care database in the United States and is sponsored by the Agency for Healthcare Research and Quality as a part of Healthcare Cost and Utilization Project.  Patients with Psoriasis were selected based on ICD-9 codes 696.1 and 696.8 while PsA were identified based on 696.0. Univariate and multivariate logistic regressions were used to derive odds ratio for measures of association. Statistical analysis was done using STATA version 13.0 (College Station, TX).

Results: NIS database from 2009-2011 contained 23,634,793 (weighted counts in the whole US population: N=117,033,987) records of discharges. Out of those, 13,850 (weighted N=68,392) had psoriatic arthropathy. Among the whole population, 81.66 % were non-smokers, 11.28 were active smokers while 7.06 % were previous smokers. Multivariate regression analysis after controlling for confounders (table 1) showed significantly higher incidence of PsA in active smokers as compared to non-smokers (OR 1.19, CI 01.12- 1.27, p ≤0.0001) in the general population. The risk was further higher in active smokers with psoriasis (OR 3.02, CI 2.04- 4.47, p ≤0.0001). Similarly, previous smokers were also at increased risk for PsA in the general population (OR 1.40, CI 1.31- 1.50, p ≤0.0001) but the risk in patients with psoriasis was not statistically significant (OR 1.59, CI 0.93- 2.70, p=0.091). Hypertension, hyperlipidemia, obesity, diabetes mellitus and rheumatoid arthritis were all found to be independent risk factors in our multivariate regression model.

Conclusion: Active tobacco use was associated with a higher risk of PsA in the general population and a much higher risk was seen in patients with psoriasis. This is in sharp contrast to previous studies showing an inverse relationship.  Moreover, the risk was significantly lower in previous smokers as compared to active smokers, suggesting that quitting smoking may lower the risk of PsA in patients with psoriasis. This lower risk was however not seen among the general population. As PsA usually develops 8-10 years later in patients with psoriasis, it may be considered a more severe phenotype that may develop in genetic susceptible patients (eg. HLA-C*06) with certain environmental exposures such as tobacco use. Hence, quitting smoking in psoriatic patients may prevent development of PsA. More studies are needed to get a better understanding of the mechanisms that explain this finding and identify gene-environment interactions.

Table 1.  Multivariate logistic regression showing risk of psoriatic arthritis with smoking (active and previous) in patients with psoriasis and general population.