Received date: May 19, 2012; Accepted date: October 22, 2012; Published date: October 30, 2012
Citation: Waxler B, Margeta B, Tureanu L, Fogg L (2012) Duration of the Surgery and Age are Risk Factors for QTc Interval Prolongation under General Anesthesia with Volatile Anesthetics. J Anesth Clin Res 3:254. doi: 10.4172/2155-6148.1000254
Copyright: © 2012 Waxler B, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Anesthesia & Clinical Research
Background: Prolongation of the QT interval signals disordered cardiac repolarization which poses a significant risk to patients undergoing surgical procedures under anesthesia. Methods: This study analyzed the demographic, clinical, and pharmacological factors in relation to QT interval prolongation under different types of anesthesia. This prospective observational study compared demographic characteristics, clinical and pharmacological factors from patients who demonstrated a prolonged QTc interval under GA (general anesthesia) and in those who had QTc interval prolongation while receiving RA (regional) or MAC (monitored anesthesia care). Results: Duration of surgery correlated strongly with QTc interval prolongation in patients who were exposed to volatile anesthetics (r=.228, p=.010), but not in patients who received RA/MAC (r=.121, p=.444).
Anaesthetic agents; Diabetes; Heart; Age; Duration of surgery; Monitoring; Electrocardiography
QT interval prolongation indicates disordered cardiac repolarization and is associated with an increased risk for arrhythmia and sudden cardiac death [1,2]. Various risk factors for QT interval prolongation have been noted in a general population of patients, including electrolyte imbalances, age , hypertension (HTN) [4,5], diabetes mellitus (DM) [6,7], and previous cardiac events, such as myocardial ischemia, recent conversion from atrial fibrillation and congestive heart failure [8,9]. Although volatile anesthetics are well known to cause QT interval prolongation in in vitro and in vivo studies [10-12], little is known about associated risk factors that predispose to QT interval prolongation and resultant potential for significant morbidity in the perioperative setting. We therefore designed this observational study to analyze the demographic characteristics, clinical and pharmacological factors in relation to QT interval prolongation during different types of anesthesia.
Ethical approval for this study (No IRB # 04-083) was provided by Institutional Review Board of the Cook County Bureau of Health Services, Chicago, IL (Chairperson Lynda Brodsky, Director, Research Affairs, Cook County Bureau of Health Services) on June 6, 2006 under the protocol title: “Characterization of QT prolongation in perioperative period.”
We collected the data from 168 patients. Those individuals selected for the study met the following inclusion criteria: 1) patients undergoing non-cardiac surgery; 2) patients who had both a preoperative and postoperative 12-lead electrocardiogram (ECG); 3) age ≥18 years of either gender. Individuals with any of the following were excluded: pacemakers/defibrillators, atrial fibrillation, and a QRS interval greater than 120 msec (complete left or right branch bundle blocks and intraventricular conduction delay), acute myocardial infarction, pregnancy, incarceration, or a poor quality ECG.
Individuals who met the inclusion criteria for our study were first screened by the telemetry (MUSE Cardiology Information System, with version 5E, windows Server 2003, from GE Healthcare) in the post anesthetic care unit (PACU) for a prolonged QT interval. We obtained a postoperative 12-lead ECG to confirm either normal or abnormal QTc. Standard 12-lead ECGs were recorded in all subjects at a speed of 25 mm/s with a 10 mm/mV gain.
We compared QT intervals from the most recent preoperative ECG and the first ECG in PACU. A cardiologist, who was blinded to patient information, type of anesthesia and timing of ECG measured all QT intervals manually. Each QT interval was measured from the onset of the QRS to the end of T wave, which was defined as the point of return to the isoelectric line. QT interval duration was assessed from lead II (or other leads if it could not be measured in lead II). The QTc interval was calculated by the method of Bazett . R-R interval was determined by computer measurement if the rhythm was regular; if the rhythm was irregular, R-R was determined manually.
For this study, we defined prolongation of the QTc as an interval ≥440 msec in males and ≥450 msec in females. Based upon the results and the anesthetic procedures, participants were divided into two groups as shown in Table 1. Group 1 (n=61) included patients who received general anesthesia with volatile anesthetic agent and in whom preoperative QTc was normal and postoperative prolonged QTc. Group 2 (n=65) included patients who received GA with volatile anesthetic agent and had normal preoperative QTc and normal postoperative QTc. Group 3 (n=13) included patients who had either regional anesthesia (RA) or monitored anesthesia care (MAC) in whom preoperative QTc was normal and postoperative prolonged QTc. Group 4 (n=29) included patients who had RA or MAC and who had normal QTc had normal preoperative QTc and normal postoperative QTc. Of note, individuals in groups 3 and 4 were not exposed to any volatile anesthetic agents.
|Type of Anaesthesia||Preoperative normal QTc &postoperative normal QTc||Preoperative normal QTc &postoperative prolonged QTc||Totals|
|General Anaesthesia||65 (Group 2)||61 (Group 1)||126|
|Regional Anaesthesia or MAC||29 (Group 4)||13 (Group 3)||42|
Table 1: Distribution of patients according to type of anaesthesia and perioperative QTc interval changes.
We collected the following information from each patient: type of anesthesia received, preoperative and postoperative 12-lead ECG, demographical data (Table 2: age, gender, weight and race), clinical factors (duration of surgery, initial temperature in PACU, presence or absence of hypertension, diabetes mellitus, and status as a smoker. In addition, we assessed various pharmacological factors, such as the volatile anesthetic agent used, perioperative administration or use of 5-HT3 antagonists (dolasetron, dol, or ondansetron, ondan), and the use of metoclopramide, dexamethasone (or other steroids), insulin, and beta-blockers. These anesthetists have not administrated to droperidol and haloperidol to these patients.
|General Anaesthesia||Regional Anaesthesia and Monitored Anaesthesia Care|
|Group 1 (Prolonged)||Group 2 (Normal)||Group 3 (Prolonged)||Group 4 (Normal)|
|Preoperative QTc (msec)||426 ± 15||416 ± 18||428 ± 13||408 ± 24|
|Postoperative QTc (msec)||472 ± 22||425 ± 18||460 ± 17||415 ± 22|
|Change in QTc (msec)||46 ± 22||8 ± 19||29 ± 22||8 ± 19|
Table 2: Changes in QTc interval.
Statistical analysis of the data was performed with Statistical Product and Service Solutions (SPSS). Descriptive data for all groups and variables were expressed as mean ± SD for continuous measures or percent of a group for discrete measures. Continuous clinical factors were analyzed with Pearson correlations with change in QTc used as a dependent variable. Discrete clinical factors were analyzed with repeated measures analysis of variance analysis (RM-ANOVA). The RM-ANOVA was a 2×2 design with the within-subject factor of time (preoperative and postoperative) crossed with the diagnostic group (e.g., hypertensive versus normotensive). A significant interaction term would indicate a larger increase in QTc in one of the two groups. We then examined graphs of the means to interpret the direction of the effect. Statistical significance was assumed if the p< 0.05 throughout these analyses. Effect sizes analysis used a standard method of Cohen .
In order to analyze the influence of the type of anesthesia on QTc interval prolongation in the perioperative period, we collected demographic, clinical and pharmacological factors from the patients who demonstrated a prolonged QTc interval under GA (Group 1) and in those who had QTc interval prolongation while receiving RA or MAC (Group 3). Controls for each of these groups (Groups 2 and 4) consisted of patients who did not experience prolongation of the QTc interval. Distribution of patients was shown in Table 1, and the preoperative and postoperative QTc intervals for each group were summarized in Table 2.
Demographic characteristics of the patients who received either GA or RA/MAC are shown in Table 3. There were no differences in gender distribution among groups. Similarly, race distribution was not different among groups and was consistent with the patient population in our hospital (Table 3). There were no differences in age and weight between the group receiving GA and the group receiving RA/MAC (p=0.763 and 0.523, respectively). However, the patients who had prolonged QTc after GA were older as compared to the group who had normal QTc after receiving GA (Table 3).
|GENERAL ANAESTHESIA||NORMAL QTc||PROLONGED QTc||p-value|
|Gender (%)||Females 66.2
|Age (years) ± SD||53.8 ± 13.3||57.9 ± 12.3||.033|
|Weight (kg) ± SD||81.3 ± 20.4||81.7 ± 20.6||.958|
|RA & MAC||NORMAL QTc||PROLONGED QTc||p-value|
|Gender (%)||Females 48.3
|Age (years) ± SD||55.3 ± 12.7||56.0 ± 13.1||.906|
|Weight (kg) ± SD||81.5 ± 12.8||89.4 ± 24.3||.216|
Table 3: Demographic characteristics of patients who received general Anaesthesia and regional Anaesthesia /monitored Anaesthesia care.
Various clinical and pharmacological factors have been associated with the QTc interval prolongation in general population. We examined the association between these factors and the perioperative QTc interval prolongation in patients receiving different types of anesthesia (Tables 4 and 5). Our data show that the QTc interval prolongation was more likely in patients who had longer surgery under GA and in older patients (Table 4). Interestingly, these factors were not significantly different in patients who demonstrated prolonged QTc interval under RA/MAC, compared to the control (Table 4).
|Risk factors||Pearson correlation (r)||p value|
|Duration of Surgery||0.228||0.010|
|Regional Anaesthesia/Monitored Anaesthesia Care|
|Risk factors||Pearson correlation (r)||p value|
|Duration of Surgery||0.121||0.444|
Table 4: Clinical risk factors for perioperative QTc interval prolongation in different types of Anesthesia.
|Risk Factor||p-value||95% CI for mean (total)|
|Regional Anaesthesia/Monitored Anaesthesia Care|
|Risk Factor||p-value||95% CI for mean (total)|
|Steroids (IV)||0.064||0 .03-0.25|
Table 5: Pharmacologic risk factors for QTc prolongation.
A number of different classes of antiemetic have been associated with significant QTc interval prolongation in the general population. When we compared QTc interval change in relation to the most commonly used antiemetic intraoperatively (serotonin inhibitors, steroids and metoclopramide) we did not observe significant differences between the groups when these medications were administered during perioperative period (Table 5). In addition, perioperative use of betablockers did not have any significant effect on the QTc change regardless of the type of anesthesia (Table 5). Likewise, history of smoking was not associated with QTc interval prolongation perioperatively (Table 5). However; insulin administration to control intraoperative glucose level in diabetic patients was associated with QTc interval prolongation under GA (Table 5)
Because there were fewer patients who were not exposed to volatile agents (RA/MAC group) than those who had GA, we considered that a lack of effect might be attributable to the small sample size, rather than the lack of a relationship. Accordingly, we constructed a table of effect sizes for both anesthesia groups. It can be argued that effect sizes are better estimates of clinical effectiveness than probability values. Effect sizes do not change with different sample sizes, while probability values do . If the risk factor was continuous, the effect size is expressed in terms of the Pearson correlation coefficient ‘r’ (Table 6). Age demonstrated a strong positive correlation with QTc interval prolongation in patients who received GA, whereas it correlated negatively in patients who received RA/MAC (Table 6). Duration of surgery correlated strongly with QTc interval prolongation in patients who were exposed to volatile anesthetics, but not in patients who received RA/MAC (Table 6). These results confirm our earlier observations presented in Table 4 that duration of surgery and age are important risk factors in perioperative QTc interval prolongation only in patients exposed to volatile anesthetics.
|GENERAL ANESTHESIA||REGIONAL ANESTHESIA AND MONITORED ANESTHESIA CARE|
|Risk factors||Pearson correlation (r)||p value||Pearson correlation (r)||p value|
|Duration of Surgery||0.278||<0.001||0.115||0.430|
Table 6: Correlation of QTc interval change and clinical risk factors in different types of Anesthesia.
For discrete clinical risk factors (HTN and DM), we used the standardized mean difference (Cohen’s d). Our analysis revealed a small to medium effect size in age-matched diabetic patients in both GA and RA/MAC groups (Table 7). This implies that DM may be an important clinical risk factor for perioperative QTc interval prolongation regardless of the type of anesthesia. These data also suggest that a larger sample size is needed to obtain adequate statistical power to draw a definite conclusion. The sample size of 168 produced a power of .82 with a one-tailed alpha of .05, and a Cohen’s d of .40. Cohen would classify this effect size as just below a ‘medium’ effect (d=.50), but it also indicates that we would be unable to detect what he calls a ‘small’ effect (d=.20). In order to detect a small effect, we would need a sample of 600 participants to achieve a power of .79, keeping our other assumptions unchanged.
|GENERAL ANAESTHESIA||REGIONAL ANAESTHESIA AND MONITORED ANAESTHESIA CARE|
|Variable||QTc Interval Change (msec) (mean ± SD)||Cohen's d||QTc Interval Change (msec) (mean ± SD)||Cohen's d|
|HTN*||6.68 ± 40.18||0.17||8.93 ± 26.11||0.34|
|DM**||12.10 ± 32.51||0.37||11.69 ± 29.88||0.39|
Table 7: Effect sizes for QTc interval change in the presence of Hypertension and Diabetes mellitus in different types of Anaesthesia.
In age-matched hypertensive patients Cohen’s d showed less pronounced effect (Table 7). Thus, the effect of HTN on perioperative QTc interval prolongation cannot be excluded. Our analysis showed that HTN strongly correlated with age of the patients (older patients were more likely to have HTN), so QTc interval prolongation in older patients could be partially due to the presence of HTN.
In this prospective observational study we compared demographic, pharmacological and clinical risk factors for QTc interval prolongation in patients receiving general anesthesia, monitoring anesthesia care, and regional anesthesia for noncardiac surgery.
In an overall study population of 168 individuals, 94 patients exhibited QTc prolongation during the perioperative period. In patients who received general anesthesia, duration of surgery, age, and perioperative insulin administration emerged to be important risk factors associated with QTc prolongation. This effect was not observed in RA/MAC group. Since, volatile anesthetic are well known to cause QTc interval prolongation in both in vitro and in vivo studies [10-12,16], we concluded that the duration of exposure to volatile anesthetic poses a significant risk for QTc interval prolongation in susceptible patients.
Antiemetic medications that we included for analysis in our study, as well as beta-blockers and tobacco use, did not have significant effect on QTc prolongation during perioperative period. These results confirm our clinical observation that the antiemetic are safe to use for postoperative nausea and vomiting prophylaxis in the perioperative period.
Certain underlying clinical conditions are well known to have significant risk for cardiac complications during surgery. DM is now recognized as an important risk factor for the development of postoperative cardiac events due to the presence of subclinical cardiac disease, autonomic neuropathy affecting cardiac conduction system and cardiac electrical instability [17,18]. In general population, DM is associated with QTc interval prolongation [19,20]. Moreover, in diabetic patients the presence of QTc interval prolongation has been shown to predict sudden cardiac death [20,21]. Both genetic factors and acquired subclinical cardiac disease influence the QTc interval in this patient population [7,18,20,22,23]. Our study showed that DM could be an important clinical predictor for QTc interval prolongation during perioperative period, regardless of type of anesthesia. Larger prospective clinical studies are needed to define the association between the presence of DM and the perioperative QTc interval changes, since these changes could signify underlying cardiac electrical instability and may predispose these patients to the development of malignant arrhythmia and even sudden cardiac death in the postoperative period.
Hypertension has also been identified as a risk factor for QTc interval prolongation . In our study, HTN correlated strongly with age, and together they had significant influence on the QTc interval prolongation. It seems that age related changes in cardiac muscle, with underlying diastolic dysfunction due to increased stiffness [25,26] are more important that the presence of HTN itself.
In this study, we examined demographic, pharmacological and clinical factors for QTc interval prolongation in patients receiving different types of anesthesia. Our results show that duration of exposure to volatile anesthetic agents poses a significant risk for the QTc interval prolongation, which is especially pronounced in older patients. Diabetes mellitus seems to be an important clinical predictor for perioperative QTc interval prolongation regardless of the type of anesthesia, especially in patients who needed insulin for perioperative glucose control. Further, larger outcome clinical studies are needed to confirm whether these perioperative changes translate to increased risk for postoperative cardiac complications.
The authors thank Dorothy Gore (Supervisor) and Mamie James (Technician) from the ECG/Non-invasive Lab at Stroger Hospital for their assistance. The authors thank Dr. Sara Rabito (Chief Section of Research, Department of Anesthesiology and Pain Management at Stroger Hospital) and Ljuba Stojiljkovic, PhD, MD (Department of Anesthesiology, Northwestern University's Feinberg School of Medicine, Chicago, IL) for these contributions to the project. We also acknowledge the excellent work of Drs. Oyebisi Aremu, Raed Rahman, and Sumit Singh, anesthesia residents who reviewed patient charts and tabulated clinical data. None of the authors has any conflict of interest.
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals