Pharmacogenomics, Antibiotics, and the Heart
abstract & commentary
Synopsis: A mutation in the gene encoding a cardiac potassium ion channel that predisposes to the development of QT interval prolongation and torsade de pointes after the administration of a variety of antibiotics is present in approximately 1.6% of the population.
Source: Sesti F, et al. Proc Natl Acad Sci U S A 2000;97: 10613-10618.
A prolonged qt interval predisposes to the development of malignant ventricular arrhythmias, and may, in some cases, be caused by exposure to medications that block cardiac potassium channels. Among the drugs that have been associated with this effect are some antibiotics, particularly macrolides and fluoroquinolones. The infrequency of such occurrences could be explained by the presence of a genetic predisposition in a small proportion of the population. Single nucleotide polymorphisms (SNPs) occur in humans at a frequency of approximately 1 in 500 to 1000 nucleotides. The study reviewed here examined the role of SNPs in predisposing to drug-induced QT prolongation.
Sesti and colleagues, in Nashville, New Haven, and Pavia, Italy, evaluated 98 patients who had either prolonged QT interval (> 600 msec) or who had developed torsades de pointes while receiving medication. They were examined for the presence of genetic polymorphisms in KCNE2, a gene encoding a subunit of the cardiac rapidly activating delayed rectifier channel, IKr. Previously discovered mutations in IKr have been shown to encode for altered channel proteins known to be associated with inheritied LQTS have been described.
Three individuals were found to have sporadic mutations, while one had a polymorphism in this gene that had previously been found to be relatively common. This fourth individual, with a history of Marfan syndrome and a normal QT interval at baseline, developed marked QT prolongation after three orally-administered doses of trimethoprim-sulfamethoxazole. He had a missense mutation causing a substitution of alanine for threonine at MiRP1 position 8 (T8A). A prior study had demonstrated a prevalence of this mutation of approximately 1.6% in the general population.
The normal gene product of KCNE2, T8Q-MiRP1, encodes a subunit of the cardiac potassium channel IKr. The effects of the identified mutations in vitro were examined by their transfection into Chinese hamster ovary cells and measurement of the resultant changes in current density, a reflection of potassium flux. Each of the three sporadic mutations was associated with diminished potassium flux at baseline, with no change in the presence of various drugs. In the case of the T8A mutation, however, flux was normal at baseline but was significantly inhibited in the presence of either trimethoprim or sulfamethoxazole at concentrations similar to those achieved clinically.
Comment by Stan Deresinski, MD, FACP
This study demonstrates the presence in a significant proportion of the population of a "form fruste" of prolonged QT interval that is only evident in the presence of a stressor, such as a drug affecting cardiac potassium efflux channels. Among the drugs capable of having this effect are a number of antimicrobials.1
Prolongation of the QT interval predisposes to cardiac arrhythmias, in particular, the highly characteristic ventricular tachycardia known as torsade de pointes ("twisting of the points"), so named because of the "twisting" of the QRS complexes of varying amplitude about the isoelectric line.
The QT interval is a measure of the duration of cardiac repolarization, an electrical activity resulting from the outflow from myocardial cells of potassium exceeding the declining inflow of sodium and calcium. These flows occur via specialized ion channels, the malfunction of which causes an excess of intracellular positive charge, which results in prolongation of repolarization and, hence, of the QT interval. Ensuing early after-depolarisations in turn allow the onset of torsade de pointes.
KCNE2 encodes MinK-related peptide 1 (MiRP1), which combines with HERG (human-ether-a-go-go; so named because the initial discovery of this gene was by observation of fruit flies with mutations who "go-go danced") to form a cardiac IKr channel. Mutations in KCNE2, such as T8A, diminish potassium egress and thus prolong cardiac repolarization.
Most cases of inherited LQTS are the result of mutations in one of five ion-channel genes. Mutations in SCN5A affect the depolarizing sodium channel, while mutations affecting the other four channels each decrease the egress of potassium from the myocardial cell that is associated with cardiac repolarization.
The T8A mutation, encoding T8A-MiRP1, appears to be clinically silent unless exposed by administration of a drug, such as trimethoprim-sulfamethoxazole in the case described here. In this case, T8A-MiRP1 predisposed to arrhythmia by increasing the inhibitory effect of SMX on the channel. This mutation is, unfortunately, relatively common; T8A had previously been identified in 16 of 1010 controls, one of 230 individuals with sporadic long QTS, and one of 20 with drug-induced arrhythmia.
A wide variety of drugs are associated with QT prolongation. Among antimicrobials, these include macrolides, fluoroquinolones, trimethoprim-sulfamethoxazole, pentamidine, quinine, and halofantrine. Antimicrobials such as amphotericin B and foscarnet, may also cause QT prolongation indirectly, by causing hypokalemia or hypomagnesemia. Pharmacokinetic drug-drug interaction may have a similar effect—e.g., azole antifungals that inhibit the metabolism of drugs such as cisapride, or the inhibition of pimozide metabolism by clarithromycin.2
Cardiac arrhythmias, including torsade de pointes, have long been associated with the use of erythromycin. Forty-nine episodes of life-threatening ventricular arrythmias and deaths associated with IV administration of erythromycin lactobionate were reported to the FDA between 1970 and 1996.3 Two-thirds of these cases involved women; in vitro studies suggest that erythromycin causes significantly greater QT prolongation in perfused female rabbit hearts than in their male counterparts.3 Erythromycin has previously been shown to inhibit the rapidly activating component (IKr), of the delayed rectifier potassium current (IK).4 All drugs that have been associated with torsade are believed to inhibit potassium ion flux through this channel.
It has been shown that intravenous administration of erythromycin to patients with community-acquired pneumonia results in significant prolongation of QTc interval, which resolves five minutes after the end of the antibiotic infusion.5 Other macrolides, including azithromycin and clarithromycin, also prolong QT interval. For instance, QT prolongation and cardiac arrest in newborns receiving the macrolide antibiotic, spiramycin, have been reported.6 QT prolongation with torsade has been associated with clarithromycin administration in the absence of any interacting drug.7 The reported incidence of possibly associated serious cardiac dysrhythmias is three per million treatments with clarithromycin.8
There has been recent interest in the potential for QT interval prolongation in association with fluoroquinolone administration. In October 1999, grepafloxacin was removed from the U.S. market because of cardiac side effects believed to be related to prolongation of the QT interval. Sparfloxacin is reported to be associated with a mean increase in rate-corrected QT (QTc) interval of 10 msec, compared to 2 msec in patients receiving comparator antibiotics in clinical trials, a finding that occurred with greater frequency in the elderly.9 Sparfloxacin use has been associated with the development of torsade de pointes and the reported incidence of possibly associated serious cardiac dystrhytmias is 14.5 per million treatments.8,10 The U.S. FDA has reported that it was aware of 11 cases of torsade in approximately 3 million treatment episodes during post-marketing surveillance of levofloxacin.11 In contrast, the incidence of possibly associated serious cardiac dysrhythmias is reported to be one case per million ciprofloxacin treatment courses (among approximately 250 million administered)!8
Gemifloxacin, not yet approved for use in the U.S., is reported to cause a mean increase in QTc interval of 5.0 msec; no torsade was seen in 5249 clinical trial patients who received this drug.12 Little information is readily available concerning gatifloxacin, although recently presented studies reported no or "no clinically important" changes in QTc interval effect on QTc interval.13-15 However, the concentration of gatifloxacin required to inhibit IKr function is comparable to that of grepafloxacin: IC50s of 26.5 uM and 27.2 uM, respectively.16 These may be compared to the IC50 for IKr of sparfloxacin of 0.23 uM.
The most intensively studied antibiotic has been moxifloxacin. In Phase III trials, the mean change in QTc interval in 787 patients receiving moxifloxacin was an increase of 6 ± 26 msec. Prolongation was noted in 9.5% of moxifloxacin recipients and in 9.2% of recipients of comparator agents.11,17 Patients with known congenital QT prolongation and those taking class Ia or III antiarrhythmic agents were excluded after QT prolongation was recognized—a criterion applied to approximately two-thirds of those enrolled. No cardiovascular morbidity or mortality attributable to QTc prolongation was observed in more than 4000 patients receiving moxifloxacin in these trials. A separate analysis of patients in these trials who received a QTc prolonging drug together with moxifloxacin found a mean increase of QTc interval of 1 ± 35 msec and 4 ± 35 msec in those who took such drugs together with a comparator antibiotic.18 The frequency of clinical cardiovascular events was low and identical in patients receiving concomitant drugs together with either moxifloxacin or a comparator antibiotic, and none were related to QTc prolongation. Post-marketing surveillance has described only one highly questionable case of torsade among a population of moxifloxacin recipients that is approaching 2 million.17
Mean QTc changes in subjects in these trials who received comparator antibiotics were: amoxicillin, -4 ± 30 msec; doxycyline, -2 ± 23 msec; cefuroxime axetil, 2 ± 20 msec; clarithromycin 2 ± 23 msec; and cephalexin, 3 ± 16 msec. More important than these values, however, is the number who have significant prolongation of their QTc interval; this was seen in only one (0.13%) each of moxifloxacin and comparator recipients. No associated arrhythmias were noted.
In a separate study involving healthy volunteers receiving various doses reaching a maximum of single 800 mg or multiple 600 mg daily doses (normal dose, 400 mg), a mean QTc interval increase of 6.9 msec was seen at the time of peak moxifloxacin concentration while placebo recipients had a mean increase of 3.5 msec.19 The frequency with which the QTc was greater than 450 msec after administration was 2.7% in moxifloxacin recipients and 4.3% in those given placebo. The increase in QTc interval was greater than 60 msec in 2.7% and 2.1%. Finally, 2.1% each of moxifloxacin and placebo recipients exhibited a QTc greater than 450 msec together with an increase in the interval of greater than 30 msec.
These alterations need to be put into perspective; thus, the upper limit of normal QTc in men is 450 msec and in women is 470 msec. Furthermore, the normal intra-individual variation is reported to be as much as 70 msec. This information, as well as the study data, indicate that moxyfloxacin is a safe antibiotic.
Thus, while the effect varies among the individual agents, QT prolongation appears to be a class effect of the fluoroquinolones that occurs in a small proportion of recipients. The overall incidence of torsade with the fluoroquinolones in wide use in the United States appears to be somewhat less than that seen with erythromycin. The infrequency of torsade suggests that there is not only a genetic predispositon, such as the T8A-MiRP1 mutant ion channel, but that additional factors are necessary to trigger this malignant rhythm in most instances. Thus, drugs known to affect QT interval should be avoided in patients with known QT prolongation, those with uncorrected hypokalemia or hypomagnesemia, and in patients receiving class IA (e.g., quinidine, procainamide) or class III (e.g., amiodarone, sotalol) antiarrhytmic agents. It is likely that, in the not too distant future, individuals will be screened for known gene mutations before drug administration—welcome to the world of pharmacogenomics! (Editor’s Note: See the "International Registry for Drug-Induced Arrhythmias" at the following internet address: http://www.dml.georgetown.edu/depts/pharmacology/torsades.html, and also at: http://georgetowncert.org/qtdrugs_welcome.html.)
References
1. Viskin S. Lancet 1999;354:1625-1633.
2. Flockhart DA, et al. J Clin Psychopharmacol 2000;20: 317-324.
3. Drici MD, et al. JAMA 1998;280:1774-1776.
4. Antzelevitch C, et al. J Am Coll Cardiol 1996;28: 1836-1848.
5. Mishra A, et al. Chest 1999;115:983-986.
6. Stramba-Badiale M, et al. Am Heart J 1997;133: 108-111.
7. Kamochi H, et al. Jpn Circ J 1999;63:421-422.
8. FDC Report. FDA/PhRMA Task Force to assess QT risk by pre-clinical markers. The Pink Sheet—Prescription Pharmaceutical and Biotechnology 1999;61:15-16.
9. Lipsky BA, et al. Clin Ther 1999;21:148-159.
10. Dupont H, et al. European J Clin Microbiol Infect Dis 1996;15:350-351.
11. Ball P, et al. Drug Safety 1999;21:407-421.
12. Gemifloxacin. (2nd European Congress on Chemotherapy, #M130.
13. Grasela DM. Clin Infect Dis 2000;31Suppl 2:S51-S58.
14. Grasela DF, et al. Lack of effect of multiple-dose gatifloxacin (GAT) on oral glucose tolerance (OGTT), glucose and insulin homeostasis, and glyburide pharmacokinetics (PK) in patients with type II non-insulin dependent diabetes mellitus (NIDDM). 39th ICAAC, 1999.
15. Breen J, Skuba K, Grasela D. Safety and tolerability of oral gatifloxacin, a new 8-methoxy fluoroquinolone: Overview of clinical data. 39th ICAAC, 1999.
16. Anderson ME, et al. A comparison of K current antagonistic properties and proarrhythmnic consequences of gatifloxacin, grepafloxacin and sparfloxacin. 3rd European Congress Chemother, Madrid, 2000, #M161.
17. Ball P. J Antimicrob Chemother 2000;45:557-559.
18. Hollister AS, et al. Moxifloxacin has a favorable cardiovascular safety profile in patients taking concomitant QTc prolonging drugs. ICAAC 2000 #818, Toronto.
19. Kubitza D, Delesen H. Influence of oral moxifloxacin on the QTc interval of healthy volunteers. ICAAC 2000, Toronto, #811.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content