Skip main navigation
×

Brugada Syndrome ECG Is Highly Prevalent in Schizophrenia

Originally publishedhttps://doi.org/10.1161/CIRCEP.113.000927Circulation: Arrhythmia and Electrophysiology. 2014;7:384–391

Abstract

Background—

The causes of increased risk of sudden cardiac death in schizophrenia are not resolved. We aimed to establish (1) whether ECG markers of sudden cardiac death risk, in particular Brugada-ECG pattern, are more prevalent among patients with schizophrenia, and (2) whether increased prevalence of these ECG markers in schizophrenia is explained by confounding factors, notably sodium channel–blocking medication.

Methods and Results—

In a cross-sectional study, we analyzed ECGs of a cohort of 275 patients with schizophrenia, along with medication use. We determined whether Brugada-ECG was present and assessed standard ECG measures (heart rate, PQ-, QRS-, and QT-intervals). We compared the findings with nonschizophrenic individuals of comparable age (the Netherlands Study of Depression and Anxiety [NESDA] cohort; N=179) and, to account for assumed increased aging rate in schizophrenia, with individuals 20 years older (Hoorn cohort; n=1168), using multivariate regression models. Brugada-ECG was significantly more prevalent in the schizophrenia cohort (11.6%) compared with NESDA controls (1.1%) or Hoorn controls (2.4%). Moreover, patients with schizophrenia had longer QT-intervals (410.9 versus 393.1 and 401.9 ms; both P<0.05), increased proportion of mild or severe QTc prolongation (13.1% and 5.8% versus 3.4% and 0.0% [NESDA], versus 5.1 and 2.8% [Hoorn]), and higher heart rates (80.8 versus 61.7 and 68.0 beats per minute; both P<0.05). The prevalence of Brugada-ECG was still increased (9.6%) when patients with schizophrenia without sodium channel–blocking medication were compared with either of the control cohorts.

Conclusions—

Brugada-ECG has increased prevalence among patients with schizophrenia. This association is not explained by the use of sodium channel–blocking medication.

Introduction

Patients with severe mental illness have 14 to 32 years reduced life expectancy.1 Schizophrenia is associated with increased standardized mortality ratios for all-cause death,2 cardiovascular death,3 and sudden cardiac death (SCD).4 The causes for SCD risk in schizophrenia are unresolved.5 SCD is mostly caused by lethal cardiac arrhythmias resulting from disrupted cardiac electrophysiology (depolarization or repolarization).6 Many researchers ascribe SCD in schizophrenia to antipsychotics, because antipsychotics may cause such disruptions.7 For instance, increased SCD risk during the use of first-generation or second-generation antipsychotics is commonly ascribed to their repolarization-blocking effects, signaled by QTc prolongation.8,9 However, individual susceptibility is crucial. Comorbidities that increase susceptibility may be acquired, for example, cardiovascular disease; such conditions are more prevalent among schizophrenic individuals.3 The possibility that inherited factors are also relevant has so far received less recognition.

Editorial see p 365

Clinical Perspective on p 391

These considerations prompted us to conduct the present study. We systematically compared ECGs of a cohort of patients with schizophrenia to ECGs of 2 cohorts of nonschizophrenia control subjects and took covariates for ECG abnormalities into account. Our primary aim was to establish whether ECG markers of SCD risk are more prevalent in patients with schizophrenia than in nonschizophrenic controls. This included the Brugada ECG pattern (Brugada-ECG) and QTc duration. Our secondary aim was to study whether differences in the prevalence of these ECG markers may be explained by use of sodium channel blockers or QTc-prolonging drugs and/or presence of cardiovascular risk factors.

Methods

Schizophrenia Cohort

In a cross-sectional study, all outpatients at the Department of Severe Mental Illness, Mental Health Care Center-North Holland North (n=603), typically in psychiatric care for >10 years, were asked to participate in yearly metabolic screening in February 2008 to January 2011. Among 387 patients who agreed to participate, 275 with Diagnostic and Statistical Manual of Mental Disorders–IV classification 295.xx (schizophrenia, schizoaffective disorder, schizophreniform disorder) were included in the study cohort (Schizophrenia cohort). This study was conducted according to the principles expressed in the Declaration of Helsinki. Written informed consent was obtained from all participants who underwent ajmaline testing.10 The Ethics Committee of the Academic Medical Center approved this study.

Control Cohorts

We compared the patients of the Schizophrenia cohort with age-comparable control persons from the Netherlands Study of Depression and Anxiety (NESDA), an ongoing longitudinal cohort study in the Netherlands.11 We used data from 179 individuals, selected randomly from a large sample of control volunteers in the NESDA study without psychiatric disorder, of whom ECG and cardiovascular measurements were made (cardiovascular subsample).12

Because several studies suggested that schizophrenic patients have higher biological age relative to their calendar age with commensurately increased prevalence of cardiovascular and metabolic disorders,13,14 we studied a second control cohort (the Hoorn Study cohort) ≈20 years older than the study cohort, in which participants, selected randomly from the population registry of the town of Hoorn, have been followed since 1989.15 After excluding 7 participants whose medication data were missing and 4 participants who used antipsychotics, we studied the ECGs of 1168 participants (Hoorn cohort).

ECG Analysis

To ensure consistent classification, all ECGs were analyzed by 1 cardiologist (H.L.T.), who was blinded to cohort status. Brugada-ECG was defined as type 1 or type 2/3 Brugada-ECG according to the Brugada syndrome (BrS) consensus criteria (Figure 1).10 Prolongation of QTc-interval (Bazett rate correction) was defined as mild (men, 431–450 ms; women, 451–470 ms) or severe (men, >450 ms; women >470 ms).16

Figure 1.

Figure 1. Example of Brugada-ECG and ajmaline test. Precordial ECG leads (25 mm/s; 10 mm/mV) are shown. Baseline ECG (A) shows saddleback-type ST-elevations in V1 (1 mm) and V2 (3 mm), suggestive of Brugada syndrome. After 30 mg ajmaline infusion (B), >2 mm coved-type ST-elevations occur (type 1 Brugada-ECG), fulfilling the diagnostic criteria for Brugada syndrome.10

Brugada-ECG and BrS

The diagnosis of BrS requires typical ST-segment elevations in right precordial ECG leads (type 1 Brugada-ECG) and events suggestive of cardiac arrhythmia or a family history of BrS or SCD. In most BrS patients, the baseline ECG is only suspicious for BrS (type 2/3 Brugada-ECG). To confirm or refute this suspicion, they must undergo provocation testing with a sodium channel–blocking drug (Figure 1).10 Accordingly, patients with schizophrenia with baseline type 2/3 Brugada-ECG were invited for ajmaline testing in the Academic Medical Center. All patients with type 1 Brugada-ECG (at baseline or after ajmaline testing) underwent DNA screening of SCN5A.17SCN5A encodes the α-subunit of the cardiac sodium channel,18 which drives depolarization. Moreover, family screening was offered. In the NESDA and Hoorn cohorts, ajmaline testing could not be performed. Therefore, the prevalences of baseline Brugada-ECG (type 1 or 2/3) were compared between cohorts.

Confounders

Risk factors for cardiovascular disease and SCD (previous myocardial infarction, hypertension, hypercholesterolemia, diabetes mellitus, body mass index [BMI]) were derived from patient files (Schizophrenia cohort) or from medication use and patient files (control cohorts).12,15 Drug use during ECG recording or ajmaline testing was derived from patient records (Schizophrenia cohort), questionnaire (Hoorn cohort), or drug container inspection (NESDA cohort). We assessed the use of sodium channel blockers,19 QTc-prolonging drugs,20 First-generation antipsychotics, second-generation antipsychotics, antidepressants, cardiovascular drugs (nitrates, β-blockers, calcium channel antagonists, antithrombotics), and lipid-lowering drugs.

Statistical Analysis

To analyze differences between cohorts in the prevalence of comorbidities and medication use, we used χ2 statistics (Pearson/Fisher exact where appropriate) for categorical variables and ANOVA for continuous variables. We performed multivariate logistic regression analyses to assess differences between cohorts in Brugada-ECG, correcting for sex and factors that were significantly (P<0.05) different between cohorts. Multivariate linear and logistic regression analyses were performed to investigate differences in quantitative ECG parameters, correcting for sex and factors that were significantly (P<0.05) different between cohorts. To evaluate the association between sodium channel blocker use and ECG outcomes, we compared schizophrenic patients with or without sodium channel blockers and performed separate analyses with patients from the Schizophrenia cohort who used no sodium channel blockers during ECG recording, using the statistical methods described above. All statistics were performed in SPSS (version 20.0 for Mac, Chicago, IL).

Results

Cohort Characteristics

Table 1 shows patient characteristics. Compared with NESDA controls, patients with schizophrenia were slightly younger and more often men. They had higher BMI (27.6 versus 25.4; P<0.05), higher prevalence of diabetes mellitus (8.0% versus 1.7%; P<0.05), and a trend toward higher lipid-lowering drug use (10.5% versus 5.6%; P=0.07). Antidepressants use was significantly more frequent among schizophrenia patients (tricyclic antidepressants, 6.5%; selective serotonin receptor inhibitors, 25.1%; other antidepressants, 4.7%); 1 participant of the NESDA cohort used low-dose amitriptyline (10 mg once daily) for neuralgy. Higher antidepressants and antipsychotics use caused higher use of sodium channel blockers (28.0% versus 0.6%) and QTc-prolonging drugs (64.4% versus 1.1%).

Table 1. Characteristics of Schizophrenia, the Netherlands Study of Depression and Anxiety (NESDA), and Hoorn Cohorts

Schizophrenia Cohort (N=275)NESDA Cohort (N=179)Hoorn Cohort (N=1168)
N%N%N%
Demographics
 Male sex19570.96636.9*51744.3*
 Mean age, y44.89.947.712.5*66.46.7*
Comorbidities
 Mean BMI27.65.525.44.5*26.63.5*
 Smoking18165.8NANA
 Myocardial infarction41.521.1615.9*
 Hypertension3914.52413.447240.4*
 Hypercholesterolemia4516.72212.343937.6*
 Diabetes mellitus228.231.7*11610.1
Medication use
 Sodium channel blockers7728.010.6*756.4*
 QT-interval–prolonging drugs17764.421.1*443.8*
 Cardiovascular drugs155.5158.427123.2*
 Lipid-lowering drugs2910.5105.6948.0
 Antipsychotics23685.800.0*00.0*
 Tricyclic antidepressants186.510.6*50.4*
 Selective serotonin reuptake inhibitors6925.100.0*90.8*
 Other antidepressants134.700.0*40.3*
Antipsychotic medication
 FGA only4917.8
 SGA only14753.5
 FGA and SGA4014.5
 No antipsychotics3914.2
Most commonly used antipsychotics
 Clozapine9133.1
 Olanzapine4717.1
 Aripiprazol3010.9
 Risperidon238.4
 Quetiapine165.8
 Haloperidol145.1
 Pimozide124.4

Data are number (percentage) unless otherwise indicated. Sodium channel blockers from www.brugadadrugs.org. QT-interval–prolonging drugs from www.azcert.org, list 1 or 2. Cardiovascular drugs are nitrates, β-blockers, calcium channel antagonists, and antithrombotics. Lipid-lowering drugs are ATC codes C10. FGA are haloperidol, pimozide, flufenazine, flupenthixol, bromperidol, pipamperon, levomepromazine, and zuclopenthixol. SGA are aripiprazol, clozapine, olanzapine, quetiapine, and risperidone. BMI indicates body mass index; FGA, first-generation antipsychotics; NA, not available; and SGA, second-generation antipsychotics.

*P<0.05 vs Schizophrenia cohort.

In the Schizophrenia cohort, data on all comorbidities were missing in 6 patients. In the Hoorn cohort, data on infarct status were missing in 128 patients and on diabetes mellitus status in 15.

Compared with Hoorn controls, schizophrenia patients were more often men and ≈20 years younger (by design). The prevalence of diabetes mellitus was comparable (8.2% versus 10.1%), but schizophrenia patients had lower prevalence of previous myocardial infarction, hypertension, and hypercholesterolemia (1.5%, 14.2%, and 16.7% versus 5.9%, 40.4%, and 37.6%; all P<0.05). Accordingly, they less often used cardiovascular drugs (5.5% versus 23.2%; P<0.05), but had comparable use of lipid-lowering drugs. They used antidepressants more often compared with Hoorn controls (tricyclic antidepressants, 6.5% versus 0.4%; selective serotonin reuptake inhibitors, 25.1% versus 0.8%; other, 4.7% versus 0.3%; all P<0.05). Antipsychotics were only used by schizophrenia patients. First-generation antipsychotics were used by 18%, second-generation antipsychotics by 54%, and both second- and first-generation antipsychotics by 15%; 14% used no antipsychotics. Sodium channel blockers and QTc-prolonging drugs were used more frequently by schizophrenia patients (28.0% versus 6.4% and 64.4% versus 3.8%).

Brugada-ECG in Schizophrenia Cohort

In the Schizophrenia cohort, 32 patients (11.6%) had Brugada-ECG at baseline: 1 had type 1 Brugada-ECG, whereas 31 had type 2/3 Brugada-ECG (Table 2). This was significantly more compared with NESDA or Hoorn controls, where no patient had type 1 Brugada-ECGs, whereas 2 (1.1%) and 28 (2.4%; both P<0.05 versus Schizophrenia cohort), respectively, had type 2/3 Brugada-ECG. Figure 2 shows Brugada-ECG prevalences in study cohorts and reported prevalences.21 In the Schizophrenia cohort, ajmaline testing was offered to the 31 patients with type 2/3 Brugada-ECG, accepted by 23 and found positive in 10. Thus, at least 11 patients had type 1 Brugada-ECG at baseline or after ajmaline testing (4% of all schizophrenia patients; 6 men; age, 48.1±10.2 years). One patient had a mutation in SCN5A (c.3956G>T). Family screening was offered to all 11 patients with type 1 Brugada-ECG, but only conducted in 5 relatives of 4 patients, because other relatives were unavailable or declined investigation; no investigated relative had Brugada-ECG. Table I in the Data Supplement shows ECG parameters, comorbidities, and medication use of all patients with type 1 Brugada-ECG. Four schizophrenia patients with type 2/3 Brugada-ECG who declined ajmaline testing provided additional medical and family history; none had syncope, dizziness, or palpitations. No SCD at age <45 years had occurred in the family, and no family members had schizophrenia.

Table 2. ECG Outcomes of Schizophrenia, the Netherlands Study of Depression and Anxiety (NESDA), and Hoorn Cohorts

Schizophrenia CohortNESDA Cohort (N=179)Hoorn Cohort (N=1168)
All (N=275)With Sodium Channel Blockers (N=77)Without Sodium Channel Blockers (N=198)
N%N%N%N%N%
Brugada-ECG3211.61316.9199.621.1*,282.4,§
QTc-interval prolongation
 Mild (men, 431–450 ms; women, 451–470 ms)3613.11418.22211.163.4605.1
 Severe (men, >450 ms; women, >470 ms)165.879.194.500.0332.8
ECG parametersMeanSDMeanSDMeanSDMeanSDMeanSD
 Heart rate, beats per minute80.816.982.916.080.017.261.79.8*,68.011.9,§
 QRS duration, ms91.611.894.513.690.510.8*91.010.3101.816.9,§
 PR duration, ms159.821.5165.822.7157.420.6*152.532.6174.025.3,§
 QTc duration, ms410.929.8416.331.9408.828.7393.125.1*,401.928.5

Data are expressed as number (percentage) unless otherwise indicated.

*P<0.05 between Schizophrenia and NESDA cohorts, corrected for sex and factors that were significantly (P<0.05) different between cohorts (body mass index [BMI], diabetes mellitus status, use of sodium channel blockers and QT-interval–prolonging drugs).

P<0.05 between Schizophrenia cohort without sodium channel blockers and NESDA cohort, corrected for sex and factors that were significantly (P<0.05) different between cohorts (BMI, diabetes mellitus status, and use of QT-interval–prolonging drugs).

P<0.05 between Schizophrenia and Hoorn cohort, corrected for sex and factors that were significantly (P<0.05) different between cohorts (BMI, cardiovascular medication use, use of sodium channel blockers and QT-interval–prolonging drugs).

§P<0.05 between Schizophrenia cohort without sodium channel blockers and Hoorn cohort, corrected for sex and factors that were significantly (P<0.05) different between cohorts (cardiovascular medication use and use of QT-interval–prolonging drugs).

Figure 2.

Figure 2. Prevalence of type 1 and type 2/3 Brugada-ECG at baseline in schizophrenia with antipsychotics, schizophrenia without antipsychotics, the Netherlands Study of Depression and Anxiety (NESDA), and Hoorn cohorts, compared with combined published prevalences in Europe.21 Table II in the Data Supplement provides details of prevalence per country.

Analysis of QTc-Interval and Other ECG Variables

Compared with NESDA controls, patients with schizophrenia had higher heart rate (80.8 versus 61.7 beats per minute; P<0.05) and longer QTc-interval (410.9 versus 393.1 ms; P<0.05). They also had higher proportion of mild or severe QTc prolongation (13.1% and 5.8% versus 3.4% and 0.0%), but not when corrected for relevant covariates (BMI, diabetes mellitus status, use of sodium channel blockers or QTc-prolonging drugs). Compared with Hoorn controls, patients with schizophrenia had higher heart rate (80.8 versus 68.0 beats per minute; P<0.05), but shorter QRS-interval (91.6 versus 101.8 ms; P<0.05) and PR-interval (159.8 versus 174.0 ms; P<0.05). QTc-interval was not different (410.9 versus 401.9 ms; P=0.251). The proportion of mild or severe QTc prolongation was higher in the schizophrenia cohort (13.1% and 5.8% versus 5.1% and 2.8%), but not significantly different when corrected for relevant covariates (BMI, cardiovascular medication use, use of sodium channel blockers or QTc-prolonging drugs).

Effects of Sodium Channel Blockers on ECG Parameters

To study whether the use of sodium channel blockers affected the prevalence of Brugada-ECG and ECG parameters in schizophrenia, we compared schizophrenia patients who used sodium channel blockers (n=77) to those who did not (n=198; Table 2). The groups differed in PR and QRS duration (165.8 versus 157.4 and 94.5 versus 90.5 ms; both P<0.05), but were otherwise similar. The prevalence of Brugada-ECG was not different (16.9% versus 9.6%; P=0.091).

To study whether ECG differences between the Schizophrenia and control cohorts may be attributed to the use of sodium channel blockers, we compared ECG parameters between schizophrenia patients who used no sodium channel blockers with both control cohorts. Compared with NESDA controls, this schizophrenia subset had more Brugada-ECG (9.6% versus 1.1%; P<0.05), higher heart rate (80.0 versus 61.7 beats per minute; P<0.05), and longer QTc duration (408.8 versus 393.1 ms; P<0.05). Mild or severe QTc prolongation was more prevalent in this schizophrenia subset (11.1% and 4.5% versus 3.4% and 0.0%; P<0.05), but not significantly when corrected for relevant covariates (sex, BMI, QT-prolonging drugs, and diabetes mellitus).

Similarly, compared with Hoorn controls, this schizophrenia subset had higher prevalence of Brugada-ECG (9.6% versus 2.4%; P<0.05), higher heart rate, and shorter QRS- and PR-intervals. Mild or severe QTc prolongation was more prevalent in this schizophrenia subset (11.1% and 4.5% versus 5.1% and 2.8%; P<0.05), but not significantly when corrected for relevant covariates (sex, BMI, QT-prolonging drugs, and diabetes mellitus).

Discussion

We found that Brugada-ECG has higher prevalence in schizophrenia patients than in similarly aged or ≈20 years older nonschizophrenic controls. Importantly, the prevalence was also significantly increased in patients with schizophrenia who used no sodium channel–blocking drugs (notably antipsychotics). In contrast, although we also found in accordance with previous studies7,22,23 that schizophrenia is associated with QTc prolongation, QTc prolongation was largely explained by confounding factors, including the use of QTc-prolonging (antipsychotics) drugs.

Schizophrenia and Brugada-ECG

As many as 4% of patients with schizophrenia had type 1 Brugada-ECG compared with an estimated prevalence of 0.05% in the general population.21 This suggests a higher prevalence of BrS in schizophrenia. This could partly explain the increased SCD risk in schizophrenia. Indeed, the yearly SCD incidence in our Schizophrenia cohort (19 of 8561 patient-years; 0.2%; not shown) is higher than the incidence in the general population in the Netherlands (0.1%).24 Schizophrenia patients with BrS would be vulnerable to the arrhythmia-causing effects of sodium channel–blocking medication, which include many antipsychotics,25 particularly in combination with the increased prevalence of cardiovascular risk factors that increase SCD risk. Still, it must be noted that type 1 Brugada-ECG per se is not sufficient for the diagnosis BrS if associated signs are absent. We did not find such signs in the schizophrenia cohort. However, these signs are difficult to ascertain or obtain in patients with schizophrenia. For instance, unexplained syncope (a presumed symptom of cardiac arrhythmias in BrS) is nonspecific in these patients, often resulting from the blood pressure–lowering effects of antipsychotics. Moreover, we had little opportunity to obtain a family history, because most patients had sparse contact with their relatives, and most available relatives declined investigation. Therefore, although the association between schizophrenia and Brugada-ECG suggested that BrS is more prevalent in schizophrenia, we could not prove this. It could be argued that the use of antipsychotics may provoke a Brugada-ECG, thereby facilitating easier detection by ECG analysis. This may especially apply to sodium channel blockers.25,26 Still, >50% of patients with Brugada-ECG used no sodium channel blockers (Table I in the Data Supplement), and differences remained significant when correcting for and stratifying according to the use of sodium channel blockers. Furthermore, it is unlikely that the use of sodium channel blockers alone results in Brugada-ECG when an innate factor is absent.27 Therefore, the high prevalence of Brugada-ECG found here can probably not be solely attributed to the use of sodium channel–blocking antipsychotics.

Future studies are required to establish the causes for the increased prevalence of Brugada-ECG (or BrS) in schizophrenia. Emerging evidence indicates that schizophrenia and acute psychosis may impact on cardiac electrophysiology.22 Accordingly, genetic studies suggest that the pathobiology of schizophrenia involves various voltage-gated ion channels.2830 Because these proteins also control cardiac electrophysiology, variants in their encoding genes (KCNH2, CACNA1C) may increase arrhythmia and SCD risk. We did not screen KCNH2 and CACNA1C in the patients with Brugada-ECG, but only SCN5A, because SCN5A is the only gene routinely screened at our institution in BrS patients.17 Nevertheless, our observations lend support to the more general notion that (nonstructural) brain disease and (electric) heart disease share common underlying pathomechanisms. For instance, in epilepsy, too, the increased incidence of SCD31 may stem from the expression of the same (mutant) ion channel in brain and heart.32,33 Furthermore, a recent study34 showed that Neuregulin1, related with both epilepsy and schizophrenia, is also associated with SCD. Autonomic dysregulation may also explain the association between BrS and schizophrenia, being reported in both conditions. However, although reduced vagal tone exists in patients with schizophrenia (including those not using antipsychotics),35 the increased vagal tone may unmask Brugada-ECG and cause SCD in BrS.26

Schizophrenia and SCD Risk

If proven in future studies that increased prevalence of Brugada-ECG in schizophrenia reflects increased prevalence of BrS, this finding could contribute to a better understanding of excess SCD risk in patients with schizophrenia, especially those using (antipsychotic or nonantipsychotic) medication that blocks cardiac depolarization. The risk for lethal cardiac arrhythmias in BrS is mediated by dysfunctional (impaired) depolarizing ion channels, notably the cardiac sodium channel. BrS patients, through their innately impaired cardiac sodium channels (reduced depolarization reserve), are particularly vulnerable to additional sodium channel block exerted by some antipsychotics (and other drugs). BrS patients may also be more vulnerable to the depolarization-blocking effects of concomitant conditions.36 In particular, acute myocardial ischemia/infarction (more likely to occur in schizophrenic patients, given their higher prevalence of diabetes mellitus) impairs cardiac depolarization.37 BrS patients may have particularly increased risk of lethal cardiac arrhythmias during acute myocardial ischemia/infarction. Thus, the combined effects of higher prevalence of Brugada-ECG and concomitant factors that impair cardiac depolarization, such as drug use or (risk factors for) ischemic heart disease, may partly explain the increased incidence of SCD in schizophrenia.

Similarly, we found increased prevalence of QTc prolongation in patients with schizophrenia. Although mostly mild and not hazardous per se, this QTc prolongation may identify individuals at increased risk for lethal cardiac arrhythmias and SCD if concomitant factors that cause further QTc prolongation (eg, cardiac hypertrophy or heart failure caused by hypertension or heart disease) are also present. However, QTc prolongation observed in patients with schizophrenia was largely explained by confounding factors, suggesting that QTc prolongation is not strongly associated with schizophrenia per se, in contrast with the occurrence of Brugada-ECG. We used Bazett formula for heart rate correction, because this method is most widely used and allows for easy comparison with other studies. Although Bazett formula may overestimate QTc duration at higher heart rates when compared with other rate correction methods, it is not resolved which method best captures the covariates of QT duration.38

Strengths and Limitations

A major strength of our study is that it involved a large group of patients with schizophrenia with data on medication use and relevant comorbidities during ECG recording. Although it is difficult to gather these data in patients with severe mental illness, the newly introduced yearly cardiovascular screening among schizophrenic patients enabled us to perform this study. Moreover, we compared our findings with 2 well-defined control cohorts.

Our study has some limitations. We were unable to perform ajmaline testing in both control cohorts, and 8 patients with schizophrenia declined ajmaline testing. Furthermore, data on family history of SCD in the Schizophrenia cohort were limited due to disturbed family relations, precluding a firm diagnosis of BrS. Clinical implications of type 2/3 Brugada-ECG are not well defined. Nonetheless, the difference in the prevalence of type 2/3 Brugada-ECG seems sufficient to warrant further research (eg, replication of our findings in a separate cohort of patients with schizophrenia) and prudence with drug prescription.

Conclusions

We found a strongly elevated prevalence of Brugada-ECG in patients with schizophrenia. Additional studies are required to elucidate whether this increased prevalence reflects an association of BrS and schizophrenia, and the underlying causes. If confirmed, our findings warrant ECG recording as part of periodic cardiovascular screening in patients with schizophrenia and prudent prescription of sodium channel blockers, to minimize SCD risk.39

Acknowledgments

We wish to express our gratitude to Jan Peetoom, MD, internist, who performed the initial ECG analysis in the schizophrenia cohort; Remco Boerman, BSc, nurse practitioner, who recorded all ECGs in the schizophrenia cohort; and Irene Beems who contributed to data collection. We greatly appreciate the contributions of Paulien Homma, MSc, and Loes Bekkers, MSc, for data collection and data entry; Julien Barc, PhD, and Leander Beekman, BSc, for DNA analysis; and Patrick Souverein, PhD, for his help in analyzing medication data.

Footnotes

The Data Supplement is available at http://circep.ahajournals.org/lookup/suppl/doi:10.1161/CIRCEP.113.000927/-/DC1.

Correspondence to Hanno L. Tan, MD, PhD, Academic Medical Center, PO Box 22660, 1100 DD Amsterdam, the Netherlands. E-mail

References

  • 1. Tiihonen J, Lönnqvist J, Wahlbeck K, Klaukka T, Niskanen L, Tanskanen A, Haukka J. 11-year follow-up of mortality in patients with schizophrenia: a population-based cohort study (FIN11 study).Lancet. 2009; 374:620–627.CrossrefMedlineGoogle Scholar
  • 2. Saha S, Chant D, McGRath J. A systematic review of mortality in schizophrenia.Arch Gen Psychiatry. 2007; 64:1123–1131.CrossrefMedlineGoogle Scholar
  • 3. DE Hert M, Correll CU, Bobes J, Cetkovich-Bakmas M, Cohen D, Asai I, Detraux J, Gautam S, Möller HJ, Ndetei DM, Newcomer JW, Uwakwe R, Leucht S. Physical illness in patients with severe mental disorders. I. Prevalence, impact of medications and disparities in health care.World Psychiatry. 2011; 10:52–77.CrossrefMedlineGoogle Scholar
  • 4. Appleby L, Thomas S, Ferrier N, Lewis G, Shaw J, Amos T. Sudden unexplained death in psychiatric in-patients.Br J Psychiatry. 2000; 176:405–406.CrossrefMedlineGoogle Scholar
  • 5. Koponen H, Alaräisänen A, Saari K, Pelkonen O, Huikuri H, Raatikainen MJ, Savolainen M, Isohanni M. Schizophrenia and sudden cardiac death: a review.Nord J Psychiatry. 2008; 62:342–345.CrossrefMedlineGoogle Scholar
  • 6. Myerburg RJ, Castellanos A. Emerging paradigms of the epidemiology and demographics of sudden cardiac arrest.Heart Rhythm. 2006; 3:235–239.CrossrefMedlineGoogle Scholar
  • 7. Timour Q, Frassati D, Descotes J, Chevalier P, Christé G, Chahine M. Sudden death of cardiac origin and psychotropic drugs.Front Pharmacol. 2012; 3:76.CrossrefMedlineGoogle Scholar
  • 8. Ray WA, Chung CP, Murray KT, Hall K, Stein CM. Atypical antipsychotic drugs and the risk of sudden cardiac death.N Engl J Med. 2009; 360:225–235.CrossrefMedlineGoogle Scholar
  • 9. Haddad PM, Anderson IM. Antipsychotic-related QTc prolongation, torsade de pointes and sudden death.Drugs. 2002; 62:1649–1671.CrossrefMedlineGoogle Scholar
  • 10. Wilde AA, Antzelevitch C, Borggrefe M, Brugada J, Brugada R, Brugada P, Corrado D, Hauer RN, Kass RS, Nademanee K, Priori SG, Towbin JA; Study Group on the Molecular Basis of Arrhythmias of the European Society of Cardiology. Proposed diagnostic criteria for the Brugada syndrome: consensus report.Circulation. 2002; 106:2514–2519.LinkGoogle Scholar
  • 11. Penninx BW, Beekman AT, Smit JH, Zitman FG, Nolen WA, Spinhoven P, Cuijpers P, De Jong PJ, Van Marwijk HW, Assendelft WJ, Van Der Meer K, Verhaak P, Wensing M, De Graaf R, Hoogendijk WJ, Ormel J, Van Dyck R; NESDA Research Consortium. The Netherlands Study of Depression and Anxiety (NESDA): rationale, objectives and methods.Int J Methods Psychiatr Res. 2008; 17:121–140.CrossrefMedlineGoogle Scholar
  • 12. Seldenrijk A, van Hout HP, van Marwijk HW, de Groot E, Gort J, Rustemeijer C, Diamant M, Penninx BW. Carotid atherosclerosis in depression and anxiety: associations for age of depression onset.World J Biol Psychiatry. 2011; 12:549–558.CrossrefMedlineGoogle Scholar
  • 13. Kirkpatrick B, Messias E, Harvey PD, Fernandez-Egea E, Bowie CR. Is schizophrenia a syndrome of accelerated aging?Schizophr Bull. 2008; 34:1024–1032.CrossrefMedlineGoogle Scholar
  • 14. Jeste DV, Wolkowitz OM, Palmer BW. Divergent trajectories of physical, cognitive, and psychosocial aging in schizophrenia.Schizophr Bull. 2011; 37:451–455.CrossrefMedlineGoogle Scholar
  • 15. de Vegt F, Dekker JM, Jager A, Hienkens E, Kostense PJ, Stehouwer CD, Nijpels G, Bouter LM, Heine RJ. Relation of impaired fasting and postload glucose with incident type 2 diabetes in a Dutch population: the Hoorn Study.JAMA. 2001; 285:2109–2113.CrossrefMedlineGoogle Scholar
  • 16. Committee for Proprietary Medicinal Products. Points to Consider: The Assessment of the Potential for QT Interval Prolongation by Non-Cardiovascular Medicinal Products. London: Committee for Proprietary Medicinal Products; 1997.Google Scholar
  • 17. Hofman N, Tan HL, Alders M, Kolder I, de Haij S, Mannens MM, Lombardi MP, Dit Deprez RH, van Langen I, Wilde AA. Yield of molecular and clinical testing for arrhythmia syndromes: report of 15 years’ experience.Circulation. 2013; 128:1513–1521.LinkGoogle Scholar
  • 18. Chen Q, Kirsch GE, Zhang D, Brugada R, Brugada J, Brugada P, Potenza D, Moya A, Borggrefe M, Breithardt G, Ortiz-Lopez R, Wang Z, Antzelevitch C, O’Brien RE, Schulze-Bahr E, Keating MT, Towbin JA, Wang Q. Genetic basis and molecular mechanism for idiopathic ventricular fibrillation.Nature. 1998; 392:293–296.CrossrefMedlineGoogle Scholar
  • 19. www.brugadadrugs.org. BrugadaDrugs.org Advisory Board. Accessed March 12, 2013.Google Scholar
  • 20. www.azcert.org. AZCERT. Accessed March 12, 2013.Google Scholar
  • 21. Postema PG. About Brugada syndrome and its prevalence.Europace. 2012; 14:925–928.CrossrefMedlineGoogle Scholar
  • 22. Hatta K, Takahashi T, Nakamura H, Yamashiro H, Yonezawa Y. Prolonged QT interval in acute psychotic patients.Psychiatry Res. 2000; 94:279–285.CrossrefMedlineGoogle Scholar
  • 23. Suvisaari J, Perälä J, Saarni SI, Kattainen A, Lönnqvist J, Reunanen A. Coronary heart disease and cardiac conduction abnormalities in persons with psychotic disorders in a general population.Psychiatry Res. 2010; 175:126–132.CrossrefMedlineGoogle Scholar
  • 24. de Vreede-Swagemakers JJ, Gorgels AP, Dubois-Arbouw WI, van Ree JW, Daemen MJ, Houben LG, Wellens HJ. Out-of-hospital cardiac arrest in the 1990’s: a population-based study in the Maastricht area on incidence, characteristics and survival.J Am Coll Cardiol. 1997; 30:1500–1505.CrossrefMedlineGoogle Scholar
  • 25. Postema PG, Wolpert C, Amin AS, Probst V, Borggrefe M, Roden DM, Priori SG, Tan HL, Hiraoka M, Brugada J, Wilde AA. Drugs and Brugada syndrome patients: review of the literature, recommendations, and an up-to-date website (www.brugadadrugs.org).Heart Rhythm. 2009; 6:1335–1341.CrossrefMedlineGoogle Scholar
  • 26. Meregalli PG, Wilde AA, Tan HL. Pathophysiological mechanisms of Brugada syndrome: depolarization disorder, repolarization disorder, or more?Cardiovasc Res. 2005; 67:367–378.CrossrefMedlineGoogle Scholar
  • 27. Yap YG, Behr ER, Camm AJ. Drug-induced Brugada syndrome.Europace. 2009; 11:989–994.CrossrefMedlineGoogle Scholar
  • 28. Atalar F, Acuner TT, Cine N, Oncu F, Yesilbursa D, Ozbek U, Turkcan S. Two four-marker haplotypes on 7q36.1 region indicate that the potassium channel gene HERG1 (KCNH2, Kv11.1) is related to schizophrenia: a case control study.Behav Brain Funct. 2010; 6:27.CrossrefMedlineGoogle Scholar
  • 29. Casamassima F, Hay AC, Benedetti A, Lattanzi L, Cassano GB, Perlis RH. L-type calcium channels and psychiatric disorders: a brief review.Am J Med Genet B Neuropsychiatr Genet. 2010; 153B:1373–1390.CrossrefMedlineGoogle Scholar
  • 30. Askland K, Read C, O’Connell C, Moore JH. Ion channels and schizophrenia: a gene set-based analytic approach to GWAS data for biological hypothesis testing.Hum Genet. 2012; 131:373–391.CrossrefMedlineGoogle Scholar
  • 31. Bardai A, Lamberts RJ, Blom MT, Spanjaart AM, Berdowski J, van der Staal SR, Brouwer HJ, Koster RW, Sander JW, Thijs RD, Tan HL. Epilepsy is a risk factor for sudden cardiac arrest in the general population.PLoS One. 2012; 7:e42749.CrossrefMedlineGoogle Scholar
  • 32. Surges R, Thijs RD, Tan HL, Sander JW. Sudden unexpected death in epilepsy: risk factors and potential pathomechanisms.Nat Rev Neurol. 2009; 5:492–504.CrossrefMedlineGoogle Scholar
  • 33. Johnson JN, Hofman N, Haglund CM, Cascino GD, Wilde AA, Ackerman MJ. Identification of a possible pathogenic link between congenital long QT syndrome and epilepsy.Neurology. 2009; 72:224–231.CrossrefMedlineGoogle Scholar
  • 34. Huertas-Vazquez A, Teodorescu C, Reinier K, Uy-Evanado A, Chugh H, Jerger K, Ayala J, Gunson K, Jui J, Newton-Cheh C, Albert CM, Chugh SS. A common missense variant in the neuregulin 1 gene is associated with both schizophrenia and sudden cardiac death.Heart Rhythm. 2013; 10:994–998.CrossrefMedlineGoogle Scholar
  • 35. Bär KJ, Wernich K, Boettger S, Cordes J, Boettger MK, Löffler S, Kornischka J, Agelink MW. Relationship between cardiovagal modulation and psychotic state in patients with paranoid schizophrenia.Psychiatry Res. 2008; 157:255–257.CrossrefMedlineGoogle Scholar
  • 36. Bardai A, Amin AS, Blom MT, Bezzina CR, Berdowski J, Langendijk PN, Beekman L, Klemens CA, Souverein PC, Koster RW, de Boer A, Tan HL. Sudden cardiac arrest associated with use of a non-cardiac drug that reduces cardiac excitability: evidence from bench, bedside, and community.Eur Heart J. 2013; 34:1506–1516.CrossrefMedlineGoogle Scholar
  • 37. Shaw RM, Rudy Y. Electrophysiologic effects of acute myocardial ischemia: a theoretical study of altered cell excitability and action potential duration.Cardiovasc Res. 1997; 35:256–272.CrossrefMedlineGoogle Scholar
  • 38. Malik M, Hnatkova K, Kowalski D, Keirns JJ, van Gelderen EM. QT/RR curvatures in healthy subjects: sex differences and covariates.Am J Physiol Heart Circ Physiol. 2013; 305:H1798–H1806.CrossrefMedlineGoogle Scholar
  • 39. Schneeweiss S, Avorn J. Antipsychotic agents and sudden cardiac death–how should we manage the risk?N Engl J Med. 2009; 360:294–296.CrossrefMedlineGoogle Scholar

CLINICAL PERSPECTIVE

The causes of increased risk of sudden cardiac death (SCD) in schizophrenia are unresolved. Many researchers ascribe SCD in schizophrenia to the use of antipsychotics, because these drugs affect cardiac electrophysiology, evoking lethal cardiac arrhythmias. QTc-interval–prolonging properties, in particular, have been implicated. Innate factors may also increase SCD risk, but are little studied. We aimed to establish (1) whether ECG markers of SCD risk, in particular Brugada-ECG pattern and QTc prolongation, are more prevalent in schizophrenia, and (2) whether increased prevalence of these ECG markers in schizophrenia is explained by confounding factors, notably sodium channel– or potassium channel–blocking drugs. In a cross-sectional study, we analyzed ECGs, along with drug use, of a cohort of 275 patients with schizophrenia, focusing on Brugada-ECG and QTc duration. We compared findings with a cohort of nonschizophrenic individuals of comparable age (the Netherlands Study of Depression and Anxiety [NESDA] cohort; N=179) and, to account for assumed increased aging rate in schizophrenia, with individuals 20 years older (Hoorn cohort; n=1168), using multivariate regression models. Brugada-ECG was significantly more prevalent in the Schizophrenia cohort (11.6%) compared with NESDA controls (1.1%) or Hoorn controls (2.4%). Moreover, schizophrenia patients had longer QT-intervals, increased proportion of mild/severe QTc prolongation, and higher heart rates compared with either of the control cohorts. The prevalence of Brugada-ECG remained increased (9.6%) when patients with schizophrenia without sodium channel–blocking drugs were compared with either of the control cohorts. In contrast, QTc prolongation was largely explained by confounding factors, including the use of QTc-prolonging (antipsychotic) drugs. If confirmed, our findings warrant ECG recording as part of periodic cardiovascular screening in schizophrenia patients and prudent prescription of sodium channel–blocking drugs.