Study objectives: IV pentamidine therapy in HIV-infected patients has been associated in case reports and one uncontrolled prospective series with frequent prolongation of the rate-corrected QT interval (QTc) and a high risk for potentially lethal ventricular arrhythmias, especially torsade de pointes. The aim of this study was to prospectively examine in a controlled manner the effect of IV pentamidine therapy on the QT interval and the incidence of ventricular arrhythmias.
Design: Open, nonrandomized, prospective evaluation of ventricular arrhythmia incidence in HIV-infected patients receiving pentamidine or trimethoprim-sulfamethoxazole (TMP-SMX) utilizing Holter monitoring prior to and during therapy with these agents.
Setting: Staten Island University Hospital, Staten Island, NY.
Patients: Twenty-seven HIV-infected patients, of whom 16 received IV pentamidine and 11 received IV TMP-SMX.
Measurements and results: Study patients underwent Holter monitoring prior to therapy and during the first 3 days and last 2 days of therapy with pentamidine or TMP-SMX, 12-lead ECG prior to and every 24 to 48 h, serum electrolytes prior to and on days 3, 6, 9, and 12 of therapy, and baseline transthoracic two-dimensional and Doppler echocardiography. In the pentamidine group, the results for each monitoring period were as follows (means are presented ± SEM): pretherapy, 1.66±1.03 (median=0) premature ventricular complexes (PVCs) per hour, zero nonsustained ventricular tachycardia (NSVT), zero sustained ventricular tachycardia (VT); early therapy, 1.55±0.91 (median=0.04) PVCs per hour, two NSVT (both ≤5 complexes), zero sustained VT; late therapy, 1.69±1.17 (median=0.08) PVCs per hour, zero NSVT, zero sustained VT (p value not significant for early or late therapy as compared to pretherapy for PVCs per hour, NSVT, or sustained VT). In the TMP-SMX group, the Holter monitoring results were as follows: pretherapy, 1.36±1.27 (median=0) PVCs per hour, zero NSVT, zero sustained VT; early therapy, 0.71±0.53 (median=0.03) PVCs per hour, two NSVT, zero sustained VT; late therapy, 0.56±0.51 (median=0) PVCs per hour, zero NSVT, zero sustained VT (p value not significant for pretherapy, early therapy, or late therapy with TMP-SMX as compared to pentamidine for PVCs per hour, NSVT, or VT). The QTc also did not significantly differ during therapy with pentamidine as compared to TMP-SMX. The mean QTc in the pentamidine group decreased during therapy as compared to pretherapy with the difference approaching significance for days 2, 4, and 6 with pentamidine (p<0.06).
Conclusions: QTc prolongation during therapy with pentamidine in HIV-infected patients is not as frequent an occurrence as has been reported previously. In the absence of QTc prolongation, pentamidine therapy was not associated with a significant increase in PVCs, NSVT, or sustained VT as compared to pretherapy recordings or as compared to therapy with TMP-SMX.