Abstract
Purpose: :
To assess complication rates among Medicare beneficiaries undergoing pars plana vitrectomy (PPV) over the past decade and to determine whether adverse outcomes and the need for additional surgery are changing with the increasing use of smaller-gauge PPV in recent years.
Methods: :
Claims data on adults aged ≥68 years in the 5% Medicare sample were reviewed to identify all beneficiaries undergoing a first PPV during 1994-1995, 1999-2000, and 2004_2005. 1 year rates of severe complications (endophthalmitis, suprachoroidal hemorrhage, retinal detachment), less-severe complications, the need for additional surgery, and blindness were calculated for the 2004-2005 cohort and compared with 1-year rates of each of the other two periods. A multivariable analysis with adjustment for potential confounding variables was performed to estimate the hazard of experiencing the four types of adverse outcomes.
Results: :
In this 5% Medicare sample, the total numbers of PPVs performed were 3263 in 1994-1995, 5064 in 1999-2000, and 5263 in 2004-2005. The incidence of severe adverse events following PPV ranged from 4.8% to 5.5% for all 3 cohorts. No significant difference was observed in the 1-year rate of severe complications among the 3 groups. The likelihood of experiencing a less-severe complication within 1 year was lower in the earlier surgery cohorts (adjusted HR for 1994-1995, 0.78 [CI, 0.66-0.91]; for 1999-2000, 0.75 ([CI, 0.69- 0.85] than in the 2004-2005 group, as was the need for additional surgery (adjusted HR for 1994-1995, 0.6 [CI, 0.55-0.66]; for 1999-2000, 0.89 [CI, 0.83-0.96]). The likelihood of endophthalmitis was higher among black patients [adjusted HR = 2.32 (95% CI 0.94 - 5.68) (p=0.066) and other nonwhite patients [adjusted HR = 3.06 (95% CI 1.18 - 7.95)] (p=0.0218), compared with whites.
Conclusions: :
In this 5% Medicare sample, the likelihood of severe, sight-threatening complications, including endophthalmitis, was no different for beneficiaries in all 3 cohorts and similar to those reported in recent observational studies. However, there was a 22-25% increased hazard of experiencing a less severe complication and an 11-40% increased hazard of requiring additional surgery for those undergoing PPV in 2004-2005. Future studies should assess the factors contributing to the differences in less-severe adverse outcomes and need for additional surgery among the groups and explore why the hazard of post-PPV endophthalmitis is elevated among nonwhites and identify ways to reduce these racial disparities in surgical outcomes.
Keywords: vitreoretinal surgery • clinical (human) or epidemiologic studies: outcomes/complications • endophthalmitis