α-Blocker Treatment Linked to Worse Cardiovascular Outcomes in Patients With Benign Prostatic Hyperplasia

For patients with benign prostatic hyperplasia (BPH), treatment with α-blockers (ABs) was linked to worse cardiovascular outcomes compared with treatment 5-α reductase inhibitors (5-ARIs).

These findings were published in JAMA Network Open and are important as ABs are the most prescribed class of medications for patients with BPH, but their cardiovascular safety profile is not well understood for these patients.

Doctor taking patient's blood pressure | Image credit: Pixel-Shot – stock.adobe.com

To compare the safety of ABs and 5-ARIs, researchers conducted a new-user cohort study that utilized insurance claims data from a 20% random sample of Medicare beneficiaries spanning from 2007 to 2019. The aim was to assess the 1-year risk of adverse cardiovascular outcomes in men aged between 66 and 90 years who initiated either treatment. Patients from the random sample were included if they had 12 months of continuous enrollment and at least 1 diagnosis code for BPH within 12 months prior to initiating the treatment. Follow-up began after patients refilled their prescription of the drug, with tamsulosin being the most frequently filled AB.

The main outcomes of interest included hospitalization due to heart failure (HF), a composite of major adverse cardiovascular events (MACE)—encompassing hospitalization for stroke, myocardial infarction, or death—, a combination of MACE or hospitalization for HF, and overall mortality.

In total, the study included 189,868 older men with BPH. The 163,829 who initiated ABs had a mean (SD) age of 74.6 (6.2) years, and the 26,039 who initiated 5-ARIs had a similar mean age of 75.3 (6.4) years. Both groups were comprised of at least 80% White participants.

When compared with 5-ARIs, ABs were associated with an elevated 1-year risk of MACE, composite MACE and HF hospitalization, and death, though there was no increased risk for HF hospitalization alone for ABs.

Specifically, the incidence of MACE alone was 8.95% (95% CI, 8.81%-9.09%) for ABs and 8.32% (95% CI, 7.92%-8.72%) for 5-ARIs (RR, 1.08; 95% CI, 1.02-1.13). The risk difference (RD) per 1000 individuals was 6.26 (95% CI, 2.15-10.37), suggesting an additional risk of MACE per 1000 individuals when treated with ABs.

For composite MACE and HF, The RR was 1.07 (95% CI, 1.03-1.12) with an RD per 1000 individuals of 7.40 (95% CI, 2.88-11.93). For death, the RR was also 1.07 (95% CI, 1.01-1.14), with an RD per 1000 individuals of 3.85 (95% CI, 0.40-7.29).

The authors noted that potential residual confounding may have influenced these findings. However, even a slight contrast in cardiovascular outcomes between ABs and 5-ARIs could have significant public health implications, considering the widespread prescription of ABs for BPH.

To gauge the extent of the exposure-outcome association for ABs, the researchers computed the number needed to harm (NNH)—the average number of patients who are exposed to an intervention before 1 patient is harmed—by taking the inverse of the RD estimate for outcomes where the 95% CIs did not intersect with 0. This led to a NNH of 160 individuals for MACE alone, 136 individuals for composite MACE and HF, and 260 individuals for death.

In current medical practice, the most commonly prescribed ABs for BPH belong to the α-1A subtype-selective antagonists, known for their minimal impact on blood pressure. The current study observed that even when the analysis was limited to subtype-selective ABs in sensitivity assessments, there was a sustained increase in the risks of adverse cardiovascular outcomes. This result was expected, given that 86.3% (141,398) of the prescribed ABs in the study fell into the subtype-selective category. Collectively, the authors said these findings support the notion that the activation of endogenous cardiac α1-A adrenergic receptors is cardioprotective, as evidenced in preclinical studies.

The study has several limitations worth noting, including susceptibility to variable misclassification inherent in the use of insurance claims data. While efforts were made to control for confounding, residual confounding, especially regarding smoking, may influence the results. Additionally, the study's generalizability is limited as it excluded Medicare Advantage plan beneficiaries, required multiple medication fills, and only considered claims through 2019. The 1-year follow-up also may not capture reasons for treatment changes, and differences in variables not accounted for could impact the transportability of results to other populations.

“Although we present the most extensive analysis, to our knowledge, of the cardiovascular safety of ABs heretofore, further investigation with more detailed clinical data is warranted to guide ongoing clinical practice,” the authors noted.

Reference

Zhang J, Latour CD, Olawore O, et al. Cardiovascular outcomes of α-blockers vs 5-α reductase inhibitors for benign prostatic hyperplasia. JAMA Netw Open. Published online November 14, 2023. doi:10.1001/jamanetworkopen.2023.43299