2018 Section 5 - Rhinology and Allergic Disorders

represented among those undergoing surgery, making up 40.3% of the California population, but 63.4% of our patient population. At least some of this discrepancy may be due to a lower prevalence of CRS in Hispanics. A study that analyzed data from the National Health Interview Survey of 27,731 US residents, which included 610 Hispanics, showed a sinusitis prevalence of 8.8% in Hispanics compared to 13% in Caucasians. 29 An addi- tional explanation for the discrepancy in both initial sur- gery and likelihood of revision in Hispanics may be access to healthcare. In the study by Soler et al., 28 His- panics were more likely to be uninsured, delay medical care due to cost-related concerns, and were less likely to have seen a medical specialist or undergone a surgical procedure in the previous 12 months. Our multivariate model controlled for known variables associated with diminished access to healthcare including income, insur- ance status, and urban setting, and yet the diminished revision rates of Hispanics remained significant. It is still quite plausible, however, that reduced access to care resulted in lower revision rates in Hispanics, as our data only included insurance status and other variables at a single time point, and these variables can easily change over time. It should be noted that the revision rates presented in this study are overall on the lower end of reported revision rates 5–7,9–11,30–32 and may be in part a reflection of the limitations of the database, as there was no guar- anteed follow-up for specific patients, and we therefore may not be capturing a portion of revision patients. Alternatively, the lower revision rate may be due to an overall lower disease burden in our population, which unlike many of the prior cohort studies included patients from both university tertiary practices as well as com- munity practices. Our period of observation (mean 42 months) may also not be long enough to capture patients who will ultimately undergo repeat surgery. However, we did find that 43% of patients who get revision sur- gery will have it done within the first year. Further- more, these data did not capture the number of patients that had already failed prior surgery, which is a known risk factor for failing a second surgery. 30,33 Further limi- tations of the present study are inherent in administra- tive data in a lack of granularity of likely clinically important factors such as patient-reported and physician-reported measures of disease severity as well as concurrent therapies and variable coding practices among surgeons. The findings of the present study war- rant further investigation in a prospective manner to clarify why patients of female gender and Hispanic eth- nicity experience variable revision rates after ESS. CONCLUSION Over 61,000 patients were identified who under- went outpatient ESS in CA between 2005 and 2011. In this large dataset we found that overall revision-free survival at 5 years post-ESS was 91.4%. Factors that increased the rate of revision surgery included the pres- ence of nasal polyps and female gender. Patients of His- panic ethnicity were less likely to undergo revision

TABLE III. Cox Proportional Hazards Model Showing Which Variables Effected Survival/Need for Revision Surgery

b (Standard Error)

Predictor

HR (95% CI)

P Value

Nasal polyps

0.17 (0.03)

1.19 (1.11-1.27)

< .001* < .001*

Female sex

0.22 (0.03)

1.24 (1.16-1.32)

Income

2 0.18 (0.13) 2 0.18 (0.13) 2 0.13 (0.14) 2 0.14 (0.14) 0.09 (0.04) 0.13 (0.07) 0.02 (0.04) 0.14 (0.11) 2 0.06 (0.07) 2 0.06 (0.08)

1st quartile

0.83 (0.64-1.08)

.173

2nd quartile

0.84 (0.64-1.09) 0.89 (0.69-1.17)

.179 .406

3rd quartile

4th quartile

0.87 (0.66-1.13 .289

Public insurance

1.1 (1.03-1.19)

.008*

Urban setting

1.14 (0.99-1.31)

.063

Race

White

1.02 (0.94-1.11) 1.15 (0.92-1.42)

.649 .218

Black

Hispanic

0.94(0.83-1.07)

.366

Asian/Pacific

0.94 (0.80-1.12)

.505 .967

Native American

NA

NA

2 0.02 (0.12)

Other ethnicity

0.98 (0.7801.22)

.831

*Significant differences ( P < .05). CI 5 confidence interval; HR 5 hazard ratio; NA 5 not applicable.

activating eosinophils and allowing autoreactive antibody-producing B cells to escape tolerance. In our cohort, women accounted for a similar minority of 38.4% of patients with CRSwNP, and women in the CRSwNP cohort were more likely to have revision surgery. How- ever, we found that female gender was a risk factor for revision surgery in the CRSsNP cohort as well, sugges- ting there may be additional factors beyond disease severity that is driving increased revision rates in women. It has been shown that high levels of anxiety and depression are common in patients with CRS, and that psychiatric comorbidity is associated with increased symptoms in CRS. 22 Increased rates of psychiatric comorbidities such as depression in women 23 may be one factor influencing the higher rates of revision surgery. Previous research has supported the notion that women have a higher prevalence of musculoskeletal pain, likely due to a combination of biological and psychosocial fac- tors, 24 and are more likely to report pain. 25 It is there- fore possible that women are more likely to experience and report sinus symptoms and therefore require reoper- ation when compared to their male counterparts. Histor- ically, there have also been gender disparities in patients being offered orthopedic and vascular sur- gery, 26,27 which may exist in sinus surgery and lead to women not being offered intervention until their disease has progressed and is less likely to respond optimally to intervention. The present study also found that patients of His- panic ethnicity were less likely to undergo repeat sur- gery. Data from the 2010 Census 28 indicate that 37.6% of the California population is Hispanic, though this group represented only 10% of our study population. Non-Hispanic whites were disproportionately

Stein et al.: Revision Rates Following ESS

131