Long Q-T Syndrome Associated with Oral Erythromycin Used in Preoperative Bowel Preparation

We present the first reported case of a patient developing long Q-T syndrome (LQTS) after taking oral erythromycin for preoperative bowel preparation, discuss the perioperative management of LQTS, and review the associated rhythm of torsade de pointes. Case Report A 70-yr-old, 102-kg female patient was admitted to the hospital for bowel preparation 1 day before gastric partitioning and takedown of a jejuno-ileal bypass. Her medical history was significant for morbid obesity, esophageal reflux, frequent diarrhea, renal insufficiency, and kidney stones. Her medications included meclizine (for chronic nausea) and potassium supplements. Her surgical history included jejuno-ileal bypass (7 yr previously), appendectomy, cholecystectomy, inguinal herniorrhaphy, and multiple kidney stone operations. She had no known drug allergies. Her physical examination was remarkable only for morbid obesity. Complete blood count was normal, and serum electrolytes included sodium 146 mEq/L, potassium 4.0 mEq/L, chloride 113 mEq/L, carbon dioxide 15 mEq/L, blood urea nitrogen 20 mEq/L, glucose 104 mg/dL, calcium 9.3 mg/dL, and an elevated creatinine of 2.1 mg/dL. Admission electrocardiogram (ECG) showed a normal Q-T interval corrected for rate (QTc) of 0.426 s and was significant only for sinus bradycardia (heart rate 50 bpm). On arrival to the hospital, the patient received intravenous (IV) hydration with lactated Ringer's solution at a rate of 75 mL/h. Bowel preparation included erythromycin base 1 g per os (PO) and metronidazole 0.5 g PO every 2 h for four doses and 1 gallon PO GoLYTELYregistered sign (Braintree Laboratories, Inc., Braintree, MA), a mechanical bowel cleanser. She had no complaints overnight except for her usual frequent retching. On the morning of surgery, she received premedication of cefotetan 2 g IV and ranitidine 50 mg IV. At that time, her nurse noted that the patient was complaining of dizziness. On arrival to the preoperative holding area, the patient was sedated with midazolam 0.75 mg IV and fentanyl 50 micro g IV. She was monitored with a noninvasive blood pressure cuff and pulse oximeter during placement of a 16-gauge IV line, radial arterial catheter, and thoracic epidural catheter. An epidural test dose of 3 mL 1.5% lidocaine with 1:200,000 epinephrine was negative. Her vital signs remained stable as 5 mL of 2% lidocaine with 1:200,000 epinephrine were administered through the epidural catheter. A T6-12 level was established. While in transport from the holding area to the operating room, the patient again complained of dizziness. Ephedrine 15 mg IV was empirically given in divided doses and the dizziness resolved. In the operating room, five-lead ECG monitoring revealed that the patient was in ventricular bigeminy with runs of narrow-complex tachycardia. Blood pressure and oxygen saturation were normal, and the patient was without complaint. The surgeon and anesthesiologist decided to delay the operation to further evaluate her new dysrhythmia, and she was transferred to the postanesthesia care unit for monitoring and formal cardiology consultation. On 12-lead ECG evaluation, the cardiologist noted the patient's Q-T interval to be 0.5 s, increasing to 0.6 s after ectopic beats, and he attributed the abnormality to the erythromycin she had taken. Because of the known association between LQTS and malignant dysrhythmias, the patient's surgery was canceled. She was monitored overnight with continuous ECG (Figure 1) and was discharged home the next morning with a heart rate of 58 bpm and an improved QTc of 0.45 s. She has not rescheduled her surgery.Figure 1: Electrocardiogram recording from the patient 8.5 h after her rhythm was first noted in the operating room, showing a Q-T interval of 0.6 s and a Q-T interval corrected for a rate of 0.64 s.Discussion This is the first reported case of a patient without other risk factors developing LQTS (QTc > 0.44 s) after oral erythromycin. Additionally, this is the first reported case of LQTS occurring after oral erythromycin given as part of preoperative bowel preparation. The association between IV erythromycin and LQTS is well known 1-6. However, despite being substantially absorbed after ingestion 7, oral erythromycin has been previously associated with LQTS only twice: in one patient with a history of repeated light-headedness 8 and in another who was taking disopyramide 9. Oral erythromycin is implicated in this case because other known causes of LQTS (e.g., hypokalemia, hypocalcemia, hypomagnesemia, use of other cardioactive drugs) were absent. LQTS is of concern because of its association with torsade de pointes, a polymorphic ventricular tachycardia. One hypothesis regarding the etiology of LQTS suggests that it may be associated with an imbalance in sympathetic innervation of the heart 10. A potential relationship between this patient's thoracic epidural placement and her LQTS might therefore be postulated. However, a literature review reveals that there has never been a report of epidural catheter placement associated with development of LQTS, and, in fact, published reports suggest that regional anesthesia is safe in LQTS patients 11,12. Further, this patient complained of dizziness before her epidural was inserted, which suggests that she was already experiencing arrhythmias, although no vital signs were checked by the nursing staff at the time of her complaint. It is currently believed that sympathetic imbalance plays more of a role in the development of dysrhythmias in patients with LQTS than it does in the etiology of the syndrome itself 10. A second hypothesis, now widely accepted, is that LQTS is the result of an intrinsic defect in cardiac repolarization 10. Prolongation of the Q-T interval can be caused by either persistent inward current or reduced outward current during the plateau phase of the cardiac action potential 10. The latter can be affected by blocking the rapidly activating cardiac delayed rectifier potassium current, as is accomplished by several genetic mutations leading to congenital LQTS 10, and by drugs such as erythromycin that lead to acquired LQTS 13,14. In electrophysiological experiments, erythromycin is also implicated in the development of torsade de pointes, delaying repolarization and causing early after-depolarizations, which are thought to be important initiating events of the dysrhythmia 4. In this patient, the unusually large dose of oral erythromycin that she received probably also contributed to the development of LQTS. The usual preoperative dose of erythromycin base is 1 g PO at 1:00 PM, 2:00 PM, and 11:00 PM (3 g over 10 hours) 7. This patient, however, received 4 g over six hours. A 500-mg oral dose of erythromycin base has been shown to produce blood concentrations of 2-4 mg/L 15, and because this patient received 8 times that dose, her blood concentrations were probably in the range of 15-30 mg/L, although actual blood concentrations were not measured. There are no data on the blood levels of erythromycin likely to cause LQTS, but concentrations of 30 mg/L have been shown to produce significant cardiac repolarization changes 4, which suggests that this patient's LQTS could be explained by the amount of erythromycin she received. Patients with congenital LQTS are initially treated with beta-adrenergic blockers. beta-adrenergic blockade is not used to change the Q-T prolongation, but to control the associated dysrhythmias. If patients continue to have syncope despite beta-blockade, left cardiac sympathetic denervation has been shown to offer further protection 16. Subsets of congenital LQTS patients can also benefit from the addition of cardiac pacing to the beta-blocker regimen 17. Implantable cardioverter defibrillators have been tried in patients unresponsive to the previously mentioned therapies but are still considered experimental 10. Patients with acquired LQTS are treated by addressing the cause of the prolonged Q-T. Torsade de pointes in patients with congenital or acquired LQTS should be treated initially with IV magnesium sulfate 18. If magnesium is unsuccessful, lidocaine or phenytoin should be tried 18. Subsets of patients may benefit from cardiac pacing or from isoproterenol infusion 18. Patients with congenital LQTS might require cardioversion or defibrillation, but torsade de pointes in patients with acquired LQTS is often self-limited; therefore, such intervention is usually not required for them 18. Because data are limited, the intraoperative risk due to LQTS is difficult to fully assess. Case reports have previously described asymptomatic patients with prolonged Q-T intervals who develop ventricular dysrhythmias during anesthesia 19-22. Both IV and volatile anesthetics have been shown to prolong the Q-T interval 23,24, and although such effects may be small 18, any prolongation of the Q-T interval in a patient with preexisting LQTS increases the risk of developing a dysrhythmia 25. Nonetheless, general anesthesia has been successfully delivered in patients with LQTS 20,22,26, although patients with untreated congenital or acquired LQTS should have elective surgery postponed because of that risk. Congenital LQTS patients should receive beta-blockade before elective surgery, and patients with acquired LQTS should have the cause of their QT prolongation elucidated and corrected. Patients with congenital LQTS tend to be more prone to dysrhythmias after adrenergic stimulation, whereas patients with acquired LQTS tend to be more at risk of dysrhythmias triggered from bradycardias 10,18. Therefore, if a patient with congenital LQTS requires emergency surgery, they should be protected from sympathetic stimulation as much as possible (e.g., they should be kept calm, and the medications used should attenuate, not stimulate, the sympathetic system). As previously mentioned, both regional anesthesia (without the use of epinephrine in solutions) and general anesthesia have been safely given to congenital LQTS patients. Intraoperative dysrhythmias should be treated as outlined previously, and postoperative pain should be aggressively treated. Patients with acquired LQTS, on the other hand, should be protected from bradycardia. Because it is impossible to be certain of which stimulus increases risk in certain individuals, all LQTS patients should be treated with careful monitoring and avoidance of both extreme sympathetic stimulation and extreme bradycardia.