QTc Prolongation and Torsades de Pointes in an Elderly Woman Taking Fluoxetine
To the Editor:
We describe an 83-year-old woman who experienced multiple episodes of torsades de pointes, probably because she was taking fluoxetine.
Transient unexplained loss of conscience in Ms. A led to her admission. Routine physical and laboratory examinations revealed no abnormalities (e.g., ischemia, electrolyte disturbances) besides a left bundle branch block and a prolonged QTc interval (478 msec, corrected for QRS interval widening because of the left bundle branch block; reference <450 msec). Her medications at admission were acetylsalicylic acid, 30 mg/day, and fluoxetine, 20 mg/day (serum levels at admission: fluoxetine, 204 μg/liter; norfluoxetine, 138 μg/liter); fluoxetine had been started 6 months earlier. During the night, continuous ECG recordings revealed recurrent short episodes of torsades de pointes. Multiple syncopal episodes occurred before admission, each time resolving spontaneously, which had never been observed before the initiation of fluoxetine.
Ms. A started having symptoms after the initiation of fluoxetine treatment and the documented pause-dependent polymorphic ventricular tachycardia characteristic of drug-induced QTc prolongation (1), which render fluoxetine use the most likely cause. Therefore, fluoxetine was discontinued. ECG recordings 2 and 8 months after fluoxetine discontinuation were normal apart from the left bundle branch block (QTc interval, 421 msec and 408 msec, respectively, corrected for the widening of the QRS interval). No further episodes of syncope or tachyarrhythmias were seen.
ECG recordings in fluoxetine preregistration trials (N=312) showed no heart-block-inducing ECG conductance disturbances. However, four case reports implicating fluoxetine in QTc prolongation (2, 3) or torsades de pointes (4, 5) have been published. One involved intentional overdose (5), and one involved concomitant verapamil treatment (3). Because of fluoxetine’s widespread use, the absence of cardiac conductance disturbances in the limited relatively healthy population in the preregistration phase is not very informative. However, at postregistration, thousands of ECG recordings in patients taking fluoxetine have been performed; in two studies focused on those with heart disease, no conduction disturbances were recorded. In our patient, there was a suggestive temporal relationship between the use of fluoxetine and the occurrence of QT prolongation and repeated episodes of torsades de pointes. After withdrawal of fluoxetine, the QTc interval returned to normal. In the absence of other plausible explanations, we conclude that the use of fluoxetine was the probable cause of the development of recurrent torsades de pointes. Older age, preexistent left bundle branch block, and female gender were likely additional risk factors (1).
Prescribers should be aware of the potential capacity of fluoxetine to prolong the QTc interval. Although it is probably extremely rare, it can have serious consequences for the patients involved.
1. Roden DM: Drug-induced prolongation of the QT interval. N Engl J Med 2004; 350:1013–1022Crossref, Medline, Google Scholar
2. Ravina T, Suarez ML, Mendez-Castrillon J: Fluoxetine-induced QTU interval prolongation, T wave alternans and syncope. Int J Cardiol 1998; 65:311–313Crossref, Medline, Google Scholar
3. Varriale P: Fluoxetine (Prozac) as a cause of QT prolongation. Arch Intern Med 2001; 161:612Crossref, Medline, Google Scholar
4. Appleby M, Mbewu A, Clarke B: Fluoxetine and ventricular torsades—is there a link? Int J Cardiol 1995; 49:178–180Crossref, Medline, Google Scholar
5. Lherm T, Lottin F, Larbi D, Bray M, Legall C, Caen D: [Torsades de pointes after poisoning with fluoxetine alone] (letter). Presse Med 2000; 29:306–307 (French)Medline, Google Scholar
