Portal de Periódicos da CAPES

Regional Patterns of Olmesartan Prescription and the Prevalence of Duodenal Villous Atrophy Throughout the United States

2017; Elsevier BV; Volume: 16; Issue: 4 Linguagem: Inglês

10.1016/j.cgh.2017.09.058

1542-7714

Eugenia Uche-Anya, Peter H.R. Green, Robert M. Genta, Benjamin Lebwohl,

Helicobacter pylori-related gastroenterology studies

A severe, sprue-like enteropathy has been described in patients exposed to olmesartan.1Rubio-Tapia A. Herman M.L. Ludvigsson J.F. et al.Severe spruelike enteropathy associated with olmesartan.Mayo Clin Proc. 2012; 87: 732-738Google Scholar Although histologically similar to celiac disease, patients typically are seronegative and only respond to drug cessation. Studies suggest that olmesartan-associated enteropathy is a rare adverse event among olmesartan prescription users.2Marthey L. Cadiot G. Seksik P. et al.Olmesartan-associated enteropathy: results of a national survey.Aliment Pharmacol Ther. 2014; 40: 1103-1109Google Scholar, 3Greywoode R. Braunstein E.D. Arguelles-Grande C. et al.Olmesartan, other antihypertensives, and chronic diarrhea among patients undergoing endoscopic procedures: a case-control study.Mayo Clin Proc. 2014; 89: 1239-1243Google Scholar, 4Menne J. Haller H. Olmesartan and intestinal adverse effects in the ROADMAP study.Mayo Clin Proc. 2012; 87: 1230-1231Google Scholar However, among patients with villous atrophy, it is uncertain whether olmesartan is a rare or common but under-recognized cause.5DeGaetani M. Tennyson C.A. Lebwohl B. et al.Villous atrophy and negative celiac serology: a diagnostic and therapeutic dilemma.Am J Gastroenterol. 2013; 108: 647-653Google Scholar We therefore aimed to determine whether regional variations in olmesartan prescription in the United States exert an ecologic effect on the prevalence of villous atrophy. We obtained histologic data containing all duodenal biopsy specimens submitted from patients ≥ 18 years to Miraca Life Sciences (Irving, TX), a large national pathology laboratory, from January 2, 2008, to April 30, 2015. All specimens were analyzed and reported by a single group of fellowship-trained gastrointestinal histopathologists. The database does not contain serologic data; therefore, we could not determine which patients had celiac disease, although celiac disease is the most common cause of seronegative villous atrophy in the United States.5DeGaetani M. Tennyson C.A. Lebwohl B. et al.Villous atrophy and negative celiac serology: a diagnostic and therapeutic dilemma.Am J Gastroenterol. 2013; 108: 647-653Google Scholar We cross-referenced patients’ residential ZIP codes from publicly available olmesartan prescription data from the Centers for Medicare and Medicaid Services beneficiaries in 2013.6Centers for Medicaid and Medicare Services. Provider utilization and payment data: part D prescriber data CY 2013. Available at: https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Medicare-Provider-Charge-Data/Part-D-Prescriber.html. Accessed July 25, 2016.Google Scholar We identified all olmesartan prescriptions by Medicare providers in each ZIP code. These ZIP codes were ranked by quartiles based on the number of olmesartan prescriptions per physician, and again by olmesartan prescriptions per capita. Data containing the number of registered physicians and the residential population in each ZIP code were obtained from the Centers for Medicare and Medicaid Services6Centers for Medicaid and Medicare Services. Provider utilization and payment data: part D prescriber data CY 2013. Available at: https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Medicare-Provider-Charge-Data/Part-D-Prescriber.html. Accessed July 25, 2016.Google Scholar and US 2010 Census Data,7U.S. Census Bureau. Census 2010; social explorer. Available at: http://old.socialexplorer.com/pub/reportdata/HtmlResults.aspx?reportid=R11205065. Accessed July 28, 2016.Google Scholar respectively. We compared the prevalence of villous atrophy in each olmesartan prescription quartile using the Cochran–Armitage test for trend. The analysis was repeated, restricting the population to patients older than age 65. The study had 80% power to detect a 0.11% difference in the prevalence of villous atrophy between the lowest and highest quartiles. There were 443,479 patients ≥ 18 years from 729 ZIP codes in 48 states. We excluded 4016 patients from 7 ZIP codes without corresponding olmesartan prescription information available, yielding a remaining sample size of 439,463 individuals from 722 ZIP codes. The median age was 54 years and 67% were female. Of these 439,463 patients, 7556 (1.72%) had villous atrophy. The prevalence of villous atrophy from the lowest to highest quartile of olmesartan prescriptions per capita was 1.69%, 1.83%, 1.66%, and 1.70% (P = .45). With olmesartan prescriptions per physician, there was an inverse relationship between increasing olmesartan prescription rates and the prevalence of villous atrophy: 1.77%, 1.75%, 1.76%, and 1.61% (P = .0056) (Figure 1). Similarly, for patients ≥ 65 years of age, higher quartiles of olmesartan prescriptions per capita/physician were not associated with a higher prevalence of villous atrophy. The prevalence of villous atrophy from the lowest to highest quartile of olmesartan prescriptions per capita was 1.50%, 1.38%, 1.48%, and 1.34% (P = .22), and the prevalence of villous atrophy from the lowest to highest quartile of olmesartan prescriptions per physician was 1.45%, 1.40%, 1.45%, and 1.40% (P = .74). Results were unchanged when restricted to biopsy specimens submitted in 2011 to 2012. In contrast to a French nationwide cohort study that used individual patient data, this study examined enteropathy on a national level so as to determine if olmesartan is a rare or common cause of villous atrophy in the United States.8Basson M. Mezzarobba M. Weill A. et al.Severe intestinal malabsorption associated with olmesartan: a French nationwide observational cohort study.Gut. 2016; 65: 1664-1669Google Scholar In this analysis, higher regional olmesartan prescription rates were not associated with a higher prevalence of villous atrophy. Because olmesartan enteropathy is seen predominantly in the older population, we repeated the analysis for patients ≥ 65 years and found similarly null results. Therefore, despite increasing case reports/series characterizing olmesartan-induced sprue-like enteropathy, olmesartan does not appear to be a major cause of villous atrophy on a national level. The strengths of this study included its large, nationally representative sample size and uniform methodology of biopsy specimen interpretation. Limitations included its retrospective study design, although a prospective study is impractical to detect what appears to be a rare outcome.4Menne J. Haller H. Olmesartan and intestinal adverse effects in the ROADMAP study.Mayo Clin Proc. 2012; 87: 1230-1231Google Scholar In addition, this was an ecologic study and therefore we are unable to make inferences about individual risk. Although designed to be nationally representative, the study likely undersampled patients in ZIP codes with high population densities where endoscopists may be unlikely to send samples to a commercial pathology service. Finally, olmesartan prescription data are limited to Medicare beneficiaries and may not account for other factors affecting physician prescribing behavior such as economics, physician supply, and marketing forces. In conclusion, there was no relationship between regional rates of olmesartan prescription and the prevalence of villous atrophy. Although clinically significant in the individual, olmesartan enteropathy appears to be an uncommon cause of villous atrophy in the United States.