Perhexiline, originally used as a first-line prophylactic antianginal agent, is now regarded primarily as a treatment for otherwise refractory myocardial ischemia. Recent studies have also demonstrated its short-term utility in heart failure, hypertrophic cardiomyopathy, and inoperable aortic stenosis. Its benefits on myocardial energetics state are potentially counter-balanced by risk of hepatotoxicity and peripheral neuropathy during long-term treatment if drug accumulation occurs. Since perhexiline exhibits complex pharmacokinetics with wide inter-individual variability, its long-term use requires regular plasma concentration monitoring. In this study, the risk of neuro- and hepato-toxicity during long-term perhexiline therapy in relation to the intensity of therapeutic drug monitoring was investigated. Furthermore, determinants of mortality during perhexiline treatment were evaluated.Methods:
In 170 patients treated with perhexiline for a median of 50 months (interquartile range: 31–94 months), outcomes and relationship to plasma drug concentrations were documented.Results:
Rationale for treatment with perhexiline included myocardial ischemia in 88% and severe systolic heart failure in 38%. Plasma concentrations were within the therapeutic range of 150–600 ng/mL on 65% of assay occasions and toxic levels accounted for 8.8% of measurements. No patient developed hepatotoxicity attributable to perhexiline while 3 developed peripheral neuropathy possibly induced by treatment. Actuarial 5-year survival rate was 83% overall, and 76.3% in patients with associated systolic heart failure.Conclusions:
This first audit of a large population treated long-term perhexiline demonstrates the following: (1) Although the frequency of monitoring is less than ideal, therapeutic drug monitoring effectively limits occurrence of toxic drug concentrations and virtually eliminates long-term hepato- and neuro-toxicity and (2) Mortality rates during long-term therapy, notably for patients with concomitant heart failure, are surprisingly low.