bs_bs_banner
R E V I E W
Drug and Alcohol Review (March 2014), 33, 194–201 DOI: 10.1111/dar.12099
Effect of amphetamine on corrected-QT interval change during methadone maintenance treatment in Taiwan: A prospective cohort study NI-CHI LIN1, CHIEH-LIANG HUANG2,3, CHUNG-YING CHEN1, TSANG-YAW LIN1, HSIU-YI WANG1, YE-HSU LU4*, LIN-MEI CHEN5, VINCENT CHIN-HUNG CHEN6,7 & MICHAEL GOSSOP8 1
Tsaotun Psychiatric Centre, Ministry of Health and Welfare, Nan-Tou, Taiwan, 2College of Medicine, China Medical University, Taichung, Taiwan, 3Centre for Drug Abuse & Addiction, China Medical University Hospital, Taichung, Taiwan, 4Division of Cardiology, Department of Internal Medicine, Kaohsiung Medical University Chung-Ho Memorial Hospital, Kaohsiung City, Taiwan, 5Department of Psychiatry, Taichung Veterans General Hospital, Taichung, Taiwan, 6 Department of Psychiatry, Chung Shan Medical University, Taichung, Taiwan, 7Department of Psychiatry, Chung Shan Medical University Hospital, Taichung, Taiwan, and 8National Addiction Centre, Institute of Psychiatry, King’s College London, London, UK
Abstract Introduction and Aims. Previous studies have suggested that methadone is associated with prolonged corrected-QT (QTc) interval, but published prospective research studies in this area are relatively scarce.This study investigates QTc interval change among methadone maintenance patients and possible associated risk factors. One of the aims is to explore the effect of amphetamines. Design and Methods. This prospective cohort study with six-month follow up assesses the effect of methadone on QTc interval among a sample (n = 170) of heroin users in a methadone maintenance treatment program in Taiwan. Demographic data, substance use history, medical history and laboratory studies were collected at study enrollment.Twelve-lead electrocardiograms were performed for all participants both at study enrollment and six months later. Results. The median daily methadone dose was 41 mg. A mean increase of QTc interval (17.1 ms, SD = 50.0, P < 0.001) was found at six-month follow up. QTc interval prolongation in the sample at baseline was 2.9%, and at six months was 12.4%. A positive correlation was found between comorbid amphetamine use frequency in the past month and QTc interval change. Methadone dose was not associated with QTc change. Discussion and Conclusions. An increase of mean QTc interval was found among methadone maintenance patients at six-month follow up. Electrocardiogram monitoring should be performed among patients who are at risk of frequently using amphetamines during methadone maintenance treatment. [Lin N-C, Huang C-L, Chen C-Y, Lin T-Y, Wang H-Y, Lu Y-H, Chen L-M, Chen V C-H, Gossop M. Effect of amphetamine on QTc interval change during methadone maintenance treatment in Taiwan: A prospective cohort study. Drug Alcohol Rev 2014;33:194–201] Key words: methadone, amphetamine, electrocardiogram. Introduction Methadone has long been regarded as an effective treatment for opioid dependence. However, concerns about the safety of methadone have been raised in recent years. Some case studies reported lethal arrhythmias when patients were treated with methadone [1–3]. Several cross-sectional studies also showed that metha-
done was associated with corrected-QT (QTc) interval prolongation [4–7]. Few prospective research studies in this area are available in the published literature [8,9], and only one Asian short-term prospective study is currently available [10]. QTc prolongation is associated with torsade de pointes (TdP), which is often self-terminating and may cause a syncopal episode. If TdP is more persistent,
Ni-Chi Lin BMed,Visiting Staff, Chieh-Liang Huang BMed,Visiting Staff, Chung-Ying Chen BMed,Visiting Staff,Tsang-Yaw Lin BMed,Visiting Staff, Hsiu-Yi Wang BN, Nurse,Ye-Hsu Lu BMed, Visiting Staff, Lin-Mei Chen Mpsych, Clinical Psychologist, Vincent Chin-Hung Chen PhD, Associate Professor, Michael Gossop PhD, Professor. Correspondence to Dr Vincent Chin-Hung Chen, Chung San Medical University Hospital, Department of Psychiatry, No.110, Section 1, Jianguo N. Road,Taichung City 40201,Taiwan.Tel: +886 (4) 2473 9595; Fax: +886 (4) 2471 5842; E-mail: [email protected] *Dr. Ye-Hsu Lu contributes equally to Dr. Vincent Chin-Hung Chen. Received 30 June 2013; accepted for publication 15 November 2013. © 2013 Australasian Professional Society on Alcohol and other Drugs
Amphetamine, methadone and QTc change
ventricular fibrillation leading to cardiac arrest and sudden death can result [11]. Lengthening QTc interval especially at levels greater than 500 ms increases the risk for TdP [12]. The mechanism of proarrhythmic toxicity of methadone is related to blockade of the rapidly activating delayed rectifier potassium current (IKr) to result in prolongation of ventricular repolarisation [13,14]. Levacetylmethadol (Orlaam), a derivative of methadone, was withdrawn from the European market in 2001 after the European Medicines Agency noticed 10 patients treated with Orlaam developed lifethreatening cardiac disorders such as TdP [15]. In 2006, the US Food and Drug Administration issued drug safety information for patients and health-care professionals concerning the toxic effects of methadone on the heart (QT prolongation and TdP) [16]. However, prolonged QTc interval is not only associated with methadone usage and is commonly because of multiple factors [9,17,18]. Factors predicting TdP include female sex, organic heart disease, congenital long QT syndrome, electrolyte disturbances, bradycardia, atrioventricular and sinoatrial blocks, drug-related factors (e.g. concomitant use of a medication that interferes the metabolism of a QT-prolonging drug), and hepatic impairment [19,20]. Polysubstance use is often to be observed in opioid addicts [21], and co-use of amphetamines and heroin is common [22,23]. Amphetamine-type stimulants are the second most widely used drugs globally [24]. QT interval prolongation after the use of amphetamine-type stimulants has been reported [25,26], and few studies have reported the effects of concurrent use of amphetamines and methadone on QTc change [27]. Since 2003, the number of people infected with human immunodeficiency virus (HIV) has been increasing sharply in Taiwan [28]. In response to the rapid increase in the number of HIV cases among injecting drug users (especially heroin users), the Taiwan government introduced methadone maintenance treatment (MMT) since 2005. Methamphetamine and heroin are two of the most commonly consumed illegal drugs in Taiwan [29], and methamphetamine use has been increasing since the end of the 1980s [30]. In Taiwan, methamphetamine was the most commonly used concomitant illicit drug, and approximately 85% of heroin abusers had tried methamphetamine at least once [31]. The present study used a prospective follow-up design to investigate the effects of QTc interval change among opiate-dependent patients in a methadone treatment program and to identify possible associated risk factors. One of the aims was to explore the effect of amphetamines.
195
Methods Participants and procedures The sample comprised 170 heroin-dependent patients recruited at an MMT program (outpatient clinic) at Tsaotun Psychiatric Centre in Nantou County in Middle Taiwan. All attendees were interviewed by a senior psychiatrist to determine the diagnosis of heroin dependence. A trained psychiatric nurse interviewed all attendees to make diagnosis of comorbid-psychiatric disorder by Mini-International Neuropsychiatric Interview [32] which had been validated in Taiwan [31]. On the first day of the MMT program, all participants completed a set of self-reported questionnaires (demographic data, heroin and other substancesuse history, and psychosocial information before treatment) administered by a trained psychiatric nurse. Self-reported drug use (tobacco, amphetamines, benzodiazepines, club drugs, alcohol and other substances) was assessed at baseline, one-month, three-month and six-month follow up. Participants were asked to report the amount of drug use within a day and number of days of drug use in the past 30 days and six months. Exclusion criteria included illiteracy, age less than 18 years, mental retardation and significant past heart disease or other medical history [except HIV, hepatitis B virus (HBV) and hepatitis C virus (HCV) infection].
Measurements Twelve-lead electrocardiograms (ECG) were performed for all participants both at the time of study enrollment and six months later. For each participant, baseline and six-month follow-up ECGs were tested at the same period in a day. Resting supine 12-lead ECGs were performed using the same machine (Kenz Cardico 1210, Suzuken Company, Nagoya City, Japan). The ECGs were recorded at a paper speed of 25 mm s−1 and an amplification of 10 mm mV−1. The QT interval was defined as the time from the beginning of the QRS complex to the end of the T wave. Heart rate affects QT length, so QT interval is corrected for heart rate using Bazett’s formula (QTc = QT interval divided by the square root of the R to R interval). Automated QTc measurements have a clear utility [33]. A senior cardiologist, blinded to the study, interpreted the ECGs. We randomly selected 24 cases from baseline ECGs. In these 24 random cases, the QTc interval in lead II was manually measured by a senior cardiologist blinded to study. We used intraclass correlation coefficient to calculate the inter-rater reliability between automated QTc interval and manually measured QTc interval. The result showed excellent agreement with the Intraclass correlation coeffi© 2013 Australasian Professional Society on Alcohol and other Drugs
196
N.-C. Lin et al.
cient = 0.90 (0.79–0.95). Because of the powerful agreement, we applied automated QTc interval for all of the analyses. QTc interval prolongation was defined as a QTc value greater than 450 ms in men and greater than 470 ms in women [34]. The threshold of a significant risk of drug-induced arrhythmia is defined as a QTc value of more than 500 ms during drug treatment [12]. Concomitant findings of ECG were also recorded. Data were also collected on demographic characteristics, drug use history, medical history and psychosocial information before treatment. Clinical laboratory test (complete blood count, electrolyte levels, glucose, cholesterol, triglyceride, liver function tests, blood urea nitrogen, creatinine, thyroxin level, anti-HCV antibody, hepatitis B surface antigen, anti-HIV antibody, treponema pallidum haemagglutination and venereal disease research laboratory) results were collected at study enrollment. Methadone was administered daily under supervision. Dosages of methadone prescription were based on self-reported amounts per day of heroin use and clinical assessment of opioid withdrawal symptoms. Statistical analysis A descriptive analysis was carried out to describe the demographic characteristics, heroin-related issues, other substance use history, laboratory data and the prevalence of QTc interval prolongation at baseline and at six-month follow up. Simple descriptive statistics such as means, standard deviations and proportions were used when appropriate. The difference was analysed through χ2-test (categorical) and t-test (or Mann–Whitney U-test) for continuous data. Change of proportion of QTc interval prolongation was calculated by McNemar Test. All analyses were conducted using spss 15.0 (SPSS Inc., Chicago, IL, USA) for Windows. Differences were considered statistically significant if the P value was less than 0.05.
Table 1. Baseline demographic and clinical data of participants (n = 170) Characteristics Age, years Onset age of substance use, years First time of heroin use, years
Mean 37.9 24.9 26.8 n
Gender Male Female Route of heroin use Smoke Injection Self-reported substance abuse historya Benzodiazepines Amphetamines Alcohol Concomitant medications (by self-report) Yes No HCV Yes No HBV Yes No HIV Yes No Hypokalaemia Yes No Hypocalcaemia Yes No Hypomagnesaemia Yes No
SD 7.7 7.4 7.1 Percentage %
155 15
91.2 8.8
32 135
19.2 80.8
59 26 57
36.0 15.8 34.5
13 157
7.6 92.3
153 16
90.5 9.5
23 147
13.5 86.5
38 132
22.4 77.6
3 147
2.0 98.0
2 109
1.8 98.2
0 150
0 100
a
Any use of the substance in the past month. HBV, hepatitis B virus; HCV, hepatitis C virus; HIV, human immunodeficiency virus.
Ethics The study was approved by the institutional review board of the Tsaotun Psychiatric Centre. All subjects provided written informed consent before participation. Results The study sample comprised 170 heroin-dependent patients receiving MMT. Baseline demographic data and clinical characteristics of the participants are listed in Table 1. The median daily methadone dose was 41 mg (full dose range 5–86 mg). The mean age of the participants at the time of study enrollment was 37.9 © 2013 Australasian Professional Society on Alcohol and other Drugs
(SD = 7.7) years. Most patients were men (91%, n = 155). The route of heroin administration was by injection (81%) and smoking/chasing the dragon (19%). Concurrent use of amphetamines was reported by 16% (n = 26) of the sample, use of benzodiazepines by 36% (n = 59) and alcohol by 34.5% (n = 57). Concomitant use of prescribed medications from another treatment facility was not common (reported by 8%, n = 13).Thirty-eight patients (22%) had HIV infection, 23 patients (14%) had HBV infection, and most of the patients had HCV infection (91%). Electrolyte imbalance was rare in participants at baseline.
Amphetamine, methadone and QTc change
Table 2. Baseline electrocardiographic characteristics (n = 170) Electrocardiographic classification
n
Percentage %
Sinus tachycardia (>100 bpm) Sinus bradycardia (450 ms for men, >470 ms for women) at baseline was 2.9% and at six-month follow up was 12.4%. The finding of a QTc interval longer than 500 ms increased from 1.2% at baseline to 2.9% at follow up. Heart rate was not significantly changed from the baseline. No TdP or sudden deaths were reported during this study. Further analyses were conducted to investigate possible associations between other factors and QTc interval change. The variables included in the analyses were age, gender, HIV infection, HBV infection, HCV infection, route of heroin use, concomitant use of medications, benzodiazepines use frequency, amphetamine use frequency, alcohol use frequency, heroin use frequency, methadone dose and baseline QTc interval. The univariate correlations between all covariates and QTc interval change are listed in Table 4. A statistically significant association was found between frequency of use of amphetamines and QTc interval change (r = 0.214, P = 0.006). QTc interval change in participants with amphetamine co-use (any use of amphetamines in the past month) was 25.7 ms (SD = 72.3); QTc interval change in participants without ampheta-
197
mine co-use was 16.1 ms (SD = 45.5).The other factor for which a statistically significant association was found with QTc interval change was baseline QTc interval (assessed at admission) (r = −0.432, P < 0.001). No statistically significant association with QTc interval change was found between other factors and with QTc interval change. In particular, no statistically significant association was found between methadone dose and QTc interval change (r = 0.08, P = 0.285).
Discussion Methadone and increase of mean QTc/QTc interval prolongation Our study demonstrated an increase in QTc interval (of 17 ms) after six months of methadone treatment.There was a fourfold increase in the proportion that developed QTc prolongation (QTc interval prolongation increased from 2.9% at baseline to 12.4% at six months) despite the fact that the dose in this study was very low. Amphetamine use frequency was one of two risk factors associated with QTc interval change in this study. Prospective studies have reported modest increases in QTc following methadone induction with QTc interval increased from baseline to fourth week of 17.3 ms [8] and to six months of 12.4 ms [9]. The amount of QTc interval change reported in the present study lies between those reported in these previous studies [8,9]. The present finding of a statistically significant increase in QTc interval is interesting. Of perhaps greater importance is whether this is clinically significant. Although the QTc interval increased, the average increase was only about 17 ms, and average QTc scores for men and women at follow up were 420 and 425 ms, respectively. The proportion of the sample with QTc interval prolongation increased from 2.9% at baseline to 12.4% at six months. In this respect, our results are consistent with other studies. Martell et al. reported that the proportion of QTc interval prolongation at baseline was 3%; at six-month follow up, QTc interval prolongation increased to 12% (18/149) [9]. Wedam et al. also found that 23% of methadone patients with a normal QTc interval at baseline showed QTc prolongation during treatment [8]. When QTc values exceed 500 ms during treatment, this indicates a significant risk of drug-induced arrhythmia [12].We found that an observed QTc interval more than 500 ms increased from 1.2% at baseline to 2.9% at six-month follow up. The low prevalence was similar to the study of Krantz which found that QTc interval of more than 500 ms increased from 0% at baseline to 2.0% at six months [35]. However, Wedam et al. found QTc interval greater than 500 ms among 11.5% at some point during their trial [8]. © 2013 Australasian Professional Society on Alcohol and other Drugs
198
N.-C. Lin et al.
Table 3. Baseline and follow-up electrocardiographic data
Variable Heart rate, beats/min QT, milliseconds Men Women Total QTcb, milliseconds Men Women Total
Baseline (n = 170)
6 months (n = 170)
Change (from baseline to 6 months)
Mean (SD)
Mean (SD)
Mean (SD)
t
d.f.
P
75.6 (12.3)
74.2 (12.1)
−1.4 (12.6)
−1.4
169
0.15
363.9 (33.8) 360.8 (22.3) 363.7 (32.9)
381.3 (4.4) 391.9 (44.3) 382.2 (44.4)
17.4 (48.1) 31.1 (41.8) 18.6 (47.6)
4.5 2.9 5.9
154 14 169
R E V I E W
Drug and Alcohol Review (March 2014), 33, 194–201 DOI: 10.1111/dar.12099
Effect of amphetamine on corrected-QT interval change during methadone maintenance treatment in Taiwan: A prospective cohort study NI-CHI LIN1, CHIEH-LIANG HUANG2,3, CHUNG-YING CHEN1, TSANG-YAW LIN1, HSIU-YI WANG1, YE-HSU LU4*, LIN-MEI CHEN5, VINCENT CHIN-HUNG CHEN6,7 & MICHAEL GOSSOP8 1
Tsaotun Psychiatric Centre, Ministry of Health and Welfare, Nan-Tou, Taiwan, 2College of Medicine, China Medical University, Taichung, Taiwan, 3Centre for Drug Abuse & Addiction, China Medical University Hospital, Taichung, Taiwan, 4Division of Cardiology, Department of Internal Medicine, Kaohsiung Medical University Chung-Ho Memorial Hospital, Kaohsiung City, Taiwan, 5Department of Psychiatry, Taichung Veterans General Hospital, Taichung, Taiwan, 6 Department of Psychiatry, Chung Shan Medical University, Taichung, Taiwan, 7Department of Psychiatry, Chung Shan Medical University Hospital, Taichung, Taiwan, and 8National Addiction Centre, Institute of Psychiatry, King’s College London, London, UK
Abstract Introduction and Aims. Previous studies have suggested that methadone is associated with prolonged corrected-QT (QTc) interval, but published prospective research studies in this area are relatively scarce.This study investigates QTc interval change among methadone maintenance patients and possible associated risk factors. One of the aims is to explore the effect of amphetamines. Design and Methods. This prospective cohort study with six-month follow up assesses the effect of methadone on QTc interval among a sample (n = 170) of heroin users in a methadone maintenance treatment program in Taiwan. Demographic data, substance use history, medical history and laboratory studies were collected at study enrollment.Twelve-lead electrocardiograms were performed for all participants both at study enrollment and six months later. Results. The median daily methadone dose was 41 mg. A mean increase of QTc interval (17.1 ms, SD = 50.0, P < 0.001) was found at six-month follow up. QTc interval prolongation in the sample at baseline was 2.9%, and at six months was 12.4%. A positive correlation was found between comorbid amphetamine use frequency in the past month and QTc interval change. Methadone dose was not associated with QTc change. Discussion and Conclusions. An increase of mean QTc interval was found among methadone maintenance patients at six-month follow up. Electrocardiogram monitoring should be performed among patients who are at risk of frequently using amphetamines during methadone maintenance treatment. [Lin N-C, Huang C-L, Chen C-Y, Lin T-Y, Wang H-Y, Lu Y-H, Chen L-M, Chen V C-H, Gossop M. Effect of amphetamine on QTc interval change during methadone maintenance treatment in Taiwan: A prospective cohort study. Drug Alcohol Rev 2014;33:194–201] Key words: methadone, amphetamine, electrocardiogram. Introduction Methadone has long been regarded as an effective treatment for opioid dependence. However, concerns about the safety of methadone have been raised in recent years. Some case studies reported lethal arrhythmias when patients were treated with methadone [1–3]. Several cross-sectional studies also showed that metha-
done was associated with corrected-QT (QTc) interval prolongation [4–7]. Few prospective research studies in this area are available in the published literature [8,9], and only one Asian short-term prospective study is currently available [10]. QTc prolongation is associated with torsade de pointes (TdP), which is often self-terminating and may cause a syncopal episode. If TdP is more persistent,
Ni-Chi Lin BMed,Visiting Staff, Chieh-Liang Huang BMed,Visiting Staff, Chung-Ying Chen BMed,Visiting Staff,Tsang-Yaw Lin BMed,Visiting Staff, Hsiu-Yi Wang BN, Nurse,Ye-Hsu Lu BMed, Visiting Staff, Lin-Mei Chen Mpsych, Clinical Psychologist, Vincent Chin-Hung Chen PhD, Associate Professor, Michael Gossop PhD, Professor. Correspondence to Dr Vincent Chin-Hung Chen, Chung San Medical University Hospital, Department of Psychiatry, No.110, Section 1, Jianguo N. Road,Taichung City 40201,Taiwan.Tel: +886 (4) 2473 9595; Fax: +886 (4) 2471 5842; E-mail: [email protected] *Dr. Ye-Hsu Lu contributes equally to Dr. Vincent Chin-Hung Chen. Received 30 June 2013; accepted for publication 15 November 2013. © 2013 Australasian Professional Society on Alcohol and other Drugs
Amphetamine, methadone and QTc change
ventricular fibrillation leading to cardiac arrest and sudden death can result [11]. Lengthening QTc interval especially at levels greater than 500 ms increases the risk for TdP [12]. The mechanism of proarrhythmic toxicity of methadone is related to blockade of the rapidly activating delayed rectifier potassium current (IKr) to result in prolongation of ventricular repolarisation [13,14]. Levacetylmethadol (Orlaam), a derivative of methadone, was withdrawn from the European market in 2001 after the European Medicines Agency noticed 10 patients treated with Orlaam developed lifethreatening cardiac disorders such as TdP [15]. In 2006, the US Food and Drug Administration issued drug safety information for patients and health-care professionals concerning the toxic effects of methadone on the heart (QT prolongation and TdP) [16]. However, prolonged QTc interval is not only associated with methadone usage and is commonly because of multiple factors [9,17,18]. Factors predicting TdP include female sex, organic heart disease, congenital long QT syndrome, electrolyte disturbances, bradycardia, atrioventricular and sinoatrial blocks, drug-related factors (e.g. concomitant use of a medication that interferes the metabolism of a QT-prolonging drug), and hepatic impairment [19,20]. Polysubstance use is often to be observed in opioid addicts [21], and co-use of amphetamines and heroin is common [22,23]. Amphetamine-type stimulants are the second most widely used drugs globally [24]. QT interval prolongation after the use of amphetamine-type stimulants has been reported [25,26], and few studies have reported the effects of concurrent use of amphetamines and methadone on QTc change [27]. Since 2003, the number of people infected with human immunodeficiency virus (HIV) has been increasing sharply in Taiwan [28]. In response to the rapid increase in the number of HIV cases among injecting drug users (especially heroin users), the Taiwan government introduced methadone maintenance treatment (MMT) since 2005. Methamphetamine and heroin are two of the most commonly consumed illegal drugs in Taiwan [29], and methamphetamine use has been increasing since the end of the 1980s [30]. In Taiwan, methamphetamine was the most commonly used concomitant illicit drug, and approximately 85% of heroin abusers had tried methamphetamine at least once [31]. The present study used a prospective follow-up design to investigate the effects of QTc interval change among opiate-dependent patients in a methadone treatment program and to identify possible associated risk factors. One of the aims was to explore the effect of amphetamines.
195
Methods Participants and procedures The sample comprised 170 heroin-dependent patients recruited at an MMT program (outpatient clinic) at Tsaotun Psychiatric Centre in Nantou County in Middle Taiwan. All attendees were interviewed by a senior psychiatrist to determine the diagnosis of heroin dependence. A trained psychiatric nurse interviewed all attendees to make diagnosis of comorbid-psychiatric disorder by Mini-International Neuropsychiatric Interview [32] which had been validated in Taiwan [31]. On the first day of the MMT program, all participants completed a set of self-reported questionnaires (demographic data, heroin and other substancesuse history, and psychosocial information before treatment) administered by a trained psychiatric nurse. Self-reported drug use (tobacco, amphetamines, benzodiazepines, club drugs, alcohol and other substances) was assessed at baseline, one-month, three-month and six-month follow up. Participants were asked to report the amount of drug use within a day and number of days of drug use in the past 30 days and six months. Exclusion criteria included illiteracy, age less than 18 years, mental retardation and significant past heart disease or other medical history [except HIV, hepatitis B virus (HBV) and hepatitis C virus (HCV) infection].
Measurements Twelve-lead electrocardiograms (ECG) were performed for all participants both at the time of study enrollment and six months later. For each participant, baseline and six-month follow-up ECGs were tested at the same period in a day. Resting supine 12-lead ECGs were performed using the same machine (Kenz Cardico 1210, Suzuken Company, Nagoya City, Japan). The ECGs were recorded at a paper speed of 25 mm s−1 and an amplification of 10 mm mV−1. The QT interval was defined as the time from the beginning of the QRS complex to the end of the T wave. Heart rate affects QT length, so QT interval is corrected for heart rate using Bazett’s formula (QTc = QT interval divided by the square root of the R to R interval). Automated QTc measurements have a clear utility [33]. A senior cardiologist, blinded to the study, interpreted the ECGs. We randomly selected 24 cases from baseline ECGs. In these 24 random cases, the QTc interval in lead II was manually measured by a senior cardiologist blinded to study. We used intraclass correlation coefficient to calculate the inter-rater reliability between automated QTc interval and manually measured QTc interval. The result showed excellent agreement with the Intraclass correlation coeffi© 2013 Australasian Professional Society on Alcohol and other Drugs
196
N.-C. Lin et al.
cient = 0.90 (0.79–0.95). Because of the powerful agreement, we applied automated QTc interval for all of the analyses. QTc interval prolongation was defined as a QTc value greater than 450 ms in men and greater than 470 ms in women [34]. The threshold of a significant risk of drug-induced arrhythmia is defined as a QTc value of more than 500 ms during drug treatment [12]. Concomitant findings of ECG were also recorded. Data were also collected on demographic characteristics, drug use history, medical history and psychosocial information before treatment. Clinical laboratory test (complete blood count, electrolyte levels, glucose, cholesterol, triglyceride, liver function tests, blood urea nitrogen, creatinine, thyroxin level, anti-HCV antibody, hepatitis B surface antigen, anti-HIV antibody, treponema pallidum haemagglutination and venereal disease research laboratory) results were collected at study enrollment. Methadone was administered daily under supervision. Dosages of methadone prescription were based on self-reported amounts per day of heroin use and clinical assessment of opioid withdrawal symptoms. Statistical analysis A descriptive analysis was carried out to describe the demographic characteristics, heroin-related issues, other substance use history, laboratory data and the prevalence of QTc interval prolongation at baseline and at six-month follow up. Simple descriptive statistics such as means, standard deviations and proportions were used when appropriate. The difference was analysed through χ2-test (categorical) and t-test (or Mann–Whitney U-test) for continuous data. Change of proportion of QTc interval prolongation was calculated by McNemar Test. All analyses were conducted using spss 15.0 (SPSS Inc., Chicago, IL, USA) for Windows. Differences were considered statistically significant if the P value was less than 0.05.
Table 1. Baseline demographic and clinical data of participants (n = 170) Characteristics Age, years Onset age of substance use, years First time of heroin use, years
Mean 37.9 24.9 26.8 n
Gender Male Female Route of heroin use Smoke Injection Self-reported substance abuse historya Benzodiazepines Amphetamines Alcohol Concomitant medications (by self-report) Yes No HCV Yes No HBV Yes No HIV Yes No Hypokalaemia Yes No Hypocalcaemia Yes No Hypomagnesaemia Yes No
SD 7.7 7.4 7.1 Percentage %
155 15
91.2 8.8
32 135
19.2 80.8
59 26 57
36.0 15.8 34.5
13 157
7.6 92.3
153 16
90.5 9.5
23 147
13.5 86.5
38 132
22.4 77.6
3 147
2.0 98.0
2 109
1.8 98.2
0 150
0 100
a
Any use of the substance in the past month. HBV, hepatitis B virus; HCV, hepatitis C virus; HIV, human immunodeficiency virus.
Ethics The study was approved by the institutional review board of the Tsaotun Psychiatric Centre. All subjects provided written informed consent before participation. Results The study sample comprised 170 heroin-dependent patients receiving MMT. Baseline demographic data and clinical characteristics of the participants are listed in Table 1. The median daily methadone dose was 41 mg (full dose range 5–86 mg). The mean age of the participants at the time of study enrollment was 37.9 © 2013 Australasian Professional Society on Alcohol and other Drugs
(SD = 7.7) years. Most patients were men (91%, n = 155). The route of heroin administration was by injection (81%) and smoking/chasing the dragon (19%). Concurrent use of amphetamines was reported by 16% (n = 26) of the sample, use of benzodiazepines by 36% (n = 59) and alcohol by 34.5% (n = 57). Concomitant use of prescribed medications from another treatment facility was not common (reported by 8%, n = 13).Thirty-eight patients (22%) had HIV infection, 23 patients (14%) had HBV infection, and most of the patients had HCV infection (91%). Electrolyte imbalance was rare in participants at baseline.
Amphetamine, methadone and QTc change
Table 2. Baseline electrocardiographic characteristics (n = 170) Electrocardiographic classification
n
Percentage %
Sinus tachycardia (>100 bpm) Sinus bradycardia (450 ms for men, >470 ms for women) at baseline was 2.9% and at six-month follow up was 12.4%. The finding of a QTc interval longer than 500 ms increased from 1.2% at baseline to 2.9% at follow up. Heart rate was not significantly changed from the baseline. No TdP or sudden deaths were reported during this study. Further analyses were conducted to investigate possible associations between other factors and QTc interval change. The variables included in the analyses were age, gender, HIV infection, HBV infection, HCV infection, route of heroin use, concomitant use of medications, benzodiazepines use frequency, amphetamine use frequency, alcohol use frequency, heroin use frequency, methadone dose and baseline QTc interval. The univariate correlations between all covariates and QTc interval change are listed in Table 4. A statistically significant association was found between frequency of use of amphetamines and QTc interval change (r = 0.214, P = 0.006). QTc interval change in participants with amphetamine co-use (any use of amphetamines in the past month) was 25.7 ms (SD = 72.3); QTc interval change in participants without ampheta-
197
mine co-use was 16.1 ms (SD = 45.5).The other factor for which a statistically significant association was found with QTc interval change was baseline QTc interval (assessed at admission) (r = −0.432, P < 0.001). No statistically significant association with QTc interval change was found between other factors and with QTc interval change. In particular, no statistically significant association was found between methadone dose and QTc interval change (r = 0.08, P = 0.285).
Discussion Methadone and increase of mean QTc/QTc interval prolongation Our study demonstrated an increase in QTc interval (of 17 ms) after six months of methadone treatment.There was a fourfold increase in the proportion that developed QTc prolongation (QTc interval prolongation increased from 2.9% at baseline to 12.4% at six months) despite the fact that the dose in this study was very low. Amphetamine use frequency was one of two risk factors associated with QTc interval change in this study. Prospective studies have reported modest increases in QTc following methadone induction with QTc interval increased from baseline to fourth week of 17.3 ms [8] and to six months of 12.4 ms [9]. The amount of QTc interval change reported in the present study lies between those reported in these previous studies [8,9]. The present finding of a statistically significant increase in QTc interval is interesting. Of perhaps greater importance is whether this is clinically significant. Although the QTc interval increased, the average increase was only about 17 ms, and average QTc scores for men and women at follow up were 420 and 425 ms, respectively. The proportion of the sample with QTc interval prolongation increased from 2.9% at baseline to 12.4% at six months. In this respect, our results are consistent with other studies. Martell et al. reported that the proportion of QTc interval prolongation at baseline was 3%; at six-month follow up, QTc interval prolongation increased to 12% (18/149) [9]. Wedam et al. also found that 23% of methadone patients with a normal QTc interval at baseline showed QTc prolongation during treatment [8]. When QTc values exceed 500 ms during treatment, this indicates a significant risk of drug-induced arrhythmia [12].We found that an observed QTc interval more than 500 ms increased from 1.2% at baseline to 2.9% at six-month follow up. The low prevalence was similar to the study of Krantz which found that QTc interval of more than 500 ms increased from 0% at baseline to 2.0% at six months [35]. However, Wedam et al. found QTc interval greater than 500 ms among 11.5% at some point during their trial [8]. © 2013 Australasian Professional Society on Alcohol and other Drugs
198
N.-C. Lin et al.
Table 3. Baseline and follow-up electrocardiographic data
Variable Heart rate, beats/min QT, milliseconds Men Women Total QTcb, milliseconds Men Women Total
Baseline (n = 170)
6 months (n = 170)
Change (from baseline to 6 months)
Mean (SD)
Mean (SD)
Mean (SD)
t
d.f.
P
75.6 (12.3)
74.2 (12.1)
−1.4 (12.6)
−1.4
169
0.15
363.9 (33.8) 360.8 (22.3) 363.7 (32.9)
381.3 (4.4) 391.9 (44.3) 382.2 (44.4)
17.4 (48.1) 31.1 (41.8) 18.6 (47.6)
4.5 2.9 5.9
154 14 169
