Trends in parathyroidectomy rates in US hemodialysis patients from 1992 to 2007

Background Parathyroidectomy rates in hemodialysis patients increased from 1992 to 2002, when medication choices to manage secondary hyperparathyroidism expanded. Study Design Retrospective follow-up registry study. Setting & Participants We evaluated annual cohorts of point-prevalent US hemodialysis patients with Medicare as primary payer for 1992-2007 (n = 1,063,258 for 1992-1999; 757,207 for 2000-2003; 902,119 for 2004-2007). Predictor Comorbid conditions, vitamin D use, previous kidney transplant, and parathyroid hormone testing were assessed in the previous year. Available bone and mineral disorder treatment patterns were evaluated. Outcomes We examined incidence rate trends and patient characteristics through 2007 to estimate the association between parathyroidectomy and patient factors. Follow-up was from January 1 of each study year to the earliest in the same year of parathyroidectomy, death, or December 31. Measurements We used χ² analysis to compare patient characteristics in 3 time frames. Unadjusted and adjusted parathyroidectomy rates were calculated. Cox regression was used to test the association of parathyroidectomy and covariates. Results Adjusted parathyroidectomy rates increased from 1998 (7.0/1,000 patient-years; 1,045 events), peaked in 2002 (12.8/1,000 patient-years; 2,229 events), decreased through 2005 (5.4/1,000 patient-years; 1,078 events), and increased in 2006 (8.6/1,000 patient-years; 1,743 events) and 2007 (8.8/1,000 patient-years; 1,832 events). Vitamin D use, virtually undetectable in 1991, subsequently steadily increased; >80% of patients received vitamin D in 2006. Limitations The study was not designed to provide causal explanations for observed changes; oral medication use trend data were limited to one large dialysis provider and may not reflect use patterns in all dialysis facilities; because Medicare is not the primary payer for all US hemodialysis patients, results do not describe the entire US hemodialysis population; parathyroid hormone values are lacking in the database. Conclusions Adjusted parathyroidectomy rates varied substantially from 1992 through 2007. Rates were highest in 1994 and 2002 and lowest in 1998 and 2005, likely influenced by changing medication use patterns and guideline publication.