Rationale: Care of patients with pulmonary hypertension is complex. Although pulmonary vasodilators are effective for Group 1 pulmonary hypertension, clinical guidelines and the Choosing Wisely Campaign recommend against routine use for Groups 2 and 3 pulmonary hypertension (the most common types of pulmonary hypertension) because of a lack of benefit, potential for harm, and high cost ($10,000-$13,000 per patient per year treated). Little is known about how these medications are used in practice.
Objectives: To determine national patterns of phosphodiesterase-5 inhibitor prescribing for pulmonary hypertension in the Veterans Health Administration.
Methods: Retrospective analysis of Veterans prescribed phosphodiesterase-5 inhibitor for pulmonary hypertension between 2005 and 2012 at any Veterans Health Administration site. Patients were identified by presence of an International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis code for pulmonary hypertension and one or more outpatient prescriptions for daily phosphodiesterase-5 inhibitor therapy. We developed and validated, using gold-standard chart abstraction, an International Classification of Diseases, Ninth Revision, Clinical Modification-based algorithm to assign pulmonary hypertension group. Our primary outcome was the proportion of patients who received potentially inappropriate phosphodiesterase-5 inhibitor, as determined by guideline recommendations (Group 1 pulmonary hypertension: appropriate; Groups 2/3: potentially inappropriate; Groups 4/5: uncertain value), among all patients prescribed phosphodiesterase-5 inhibitor for pulmonary hypertension. Secondary outcomes included proportion of treated patients who received guideline-recommended right heart catheterization.
Results: Among 108,777 Veterans with pulmonary hypertension, 2,790 (2.6% [95% confidence interval, 2.5-2.7%]) received daily phosphodiesterase-5 inhibitor therapy. Among treated patients, 541 (19.4% [95% confidence interval, 18.0-20.9%]) received appropriate treatment, 1,711 (61.3% [95% confidence interval, 59.5-63.1%]) potentially inappropriate treatment, and 358 (12.8% [95% confidence interval, 11.6-14.1%]) treatment of uncertain value. The number of potentially inappropriately treated patients per year increased substantially over the study period (53 in 2005, 748 in 2012). On the basis of chart abstraction in a randomly selected subset of patients treated with phosphodiesterase-5 inhibitor, half (110 of 230, 47.8% [95% confidence interval, 41.3-54.5%]) had documented right heart catheterization to confirm presence or type of pulmonary hypertension. After factoring presence of and data from right heart catheterization into our treatment appropriateness algorithm, only 11.7% (95% confidence interval, 8.0-16.8%) received clearly appropriate treatment.
Conclusions: Most Veterans with pulmonary hypertension do not receive phosphodiesterase-5 inhibitor therapy. However, among treated Veterans, almost two-thirds of phosphodiesterase-5 inhibitor prescriptions are inconsistent with pulmonary hypertension guidelines, exposing patients to potential harm and creating a financial burden on the healthcare system. Further study is warranted to clarify the effects of these prescription patterns on pulmonary hypertension outcomes.
Keywords: prescriptions; pulmonary hypertension; utilization.