Contemporary Outcomes in Patients With Long QT Syndrome

JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY

VOL. 70, NO. 4, 2017

ª 2017 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION PUBLISHED BY ELSEVIER

ISSN 0735-1097/$36.00 http://dx.doi.org/10.1016/j.jacc.2017.05.046

Contemporary Outcomes in Patients With Long QT Syndrome Ram K. Rohatgi, MD,a Alan Sugrue, MBBCH,b J. Martijn Bos, MD, PHD,a,b,c Bryan C. Cannon, MD,a,b Samuel J. Asirvatham, MD,a,b,d Christopher Moir, MD,a,e Heidi J. Owen, RN,a Katy M. Bos, APRN, CNS,a Teresa Kruisselbrink, MS, CGC,f Michael J. Ackerman, MD, PHDa,b,c

ABSTRACT BACKGROUND Long QT syndrome (LQTS) is a potentially lethal cardiac channelopathy with a 1% to 5% annual risk of LQTS-triggered syncope, aborted cardiac arrest, or sudden cardiac death. OBJECTIVES This study sought to evaluate LQTS outcomes from a single center in the contemporary era. METHODS The authors conducted a retrospective study comprising the 606 patients with LQTS (LQT1 in 47%, LQT2 in 34%, and LQT3 in 9%) who were evaluated in Mayo Clinic’s Genetic Heart Rhythm Clinic from January 1999 to December 2015. Breakthrough cardiac events (BCEs) were defined as LQTS-attributable syncope or seizures, aborted cardiac arrest, appropriate ventricular fibrillation–terminating implantable cardioverter-defibrillator shocks, and sudden cardiac death. RESULTS There were 166 (27%) patients who were symptomatic prior to their first Mayo Clinic evaluation. Median age at first symptom was 12 years. Treatment strategies included no active therapy in 47 (8%) patients, beta-blockers alone in 350 (58%) patients, implantable cardioverter-defibrillators alone in 25 (4%) patients, left cardiac sympathetic denervation alone in 18 (3%) patients, and combination therapy in 166 (27%) patients. Over a median follow-up of 6.7 (IQR: 3.9 to 9.8) years, 556 (92%) patients have not experienced an LQTS-triggered BCE. Only 8 of 440 (2%) previously asymptomatic patients have experienced a single BCE. In contrast, 42 of 166 (25%) previously symptomatic patients have experienced $1 BCE. Among the 30 patients with $2 BCEs, 2 patients have died and 3 LQT3 patients underwent cardiac transplantation. CONCLUSIONS Although outcomes have improved markedly, further optimization of treatment strategies is still needed given that 1 in 4 previously symptomatic patients experienced at least 1 subsequent, albeit nonlethal, LQTS-triggered cardiac event. (J Am Coll Cardiol 2017;70:453–62) © 2017 by the American College of Cardiology Foundation.

From the aDepartment of Pediatric and Adolescent Medicine/Division of Pediatric Cardiology, Mayo Clinic, Rochester, Minnesota; b

Department of Cardiovascular Diseases/Division of Heart Rhythm Services, Mayo Clinic, Rochester, Minnesota; cWindland Smith

Rice Sudden Death Genomics Laboratory, Department of Molecular Pharmacology & Experimental Therapeutics, Mayo Clinic Graduate School of Biomedical Sciences, Mayo Clinic, Rochester, Minnesota; dDepartment of Physiology and Biomedical Engineering, Mayo Clinic Graduate School of Biomedical Sciences, Mayo Clinic, Rochester, Minnesota; eDepartment of Surgery/ Division of Pediatric Surgery, Mayo Clinic, Rochester, Minnesota; and the fCenter of Individualized Medicine, Mayo Clinic, Rochester, Minnesota. This work was supported by the Mayo Clinic Windland Smith Rice Comprehensive Sudden Cardiac Death Program (to Dr. Ackerman). This publication was supported by Grant UL1 TR000135 from the National Center for Advancing Translational Sciences. Its contents are solely the responsibility of the authors and do not necessarily represent the official views Listen to this manuscript’s

of the National Institutes of Health. Dr. Ackerman and Mayo Clinic received royalties until 2016 from Transgenomic with respect

audio summary by

to their FAMILION-LQTS and FAMILION-CPVT genetic tests and have a licensing agreement with AliveCor. Dr. Cannon has served

JACC Editor-in-Chief

on a data safety monitoring board for Medtronic; and is on the board of trustees for Mayo Support Services Texas. Dr. Asirvatham

Dr. Valentin Fuster.

has served as a consultant for Aegis, ATP, Nevro, Sanovas, Sorin Medical, and FocusStart; has received honoraria or speaker fees from Abiomed, Atricure, Biotronik, Blackwell Futura, Boston Scientific, Medtronic, Medtelligence, Sanofi, Spectranetics, St. Jude, and Zoll; and has received royalties for work licensed through the Mayo Clinic from Nevro, Aegis, and the Phoenix Corp. Dr. Ackerman has served as a consultant for Boston Scientific, Gilead Sciences, Invitae, Medtronic, MyoKardia, and St. Jude Medical. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Manuscript received January 27, 2017; revised manuscript received May 16, 2017, accepted May 19, 2017.

454

Rohatgi et al.

JACC VOL. 70, NO. 4, 2017 JULY 25, 2017:453–62

Outcomes in LQTS

L

ABBREVIATIONS

ong QT syndrome (LQTS) is one of the

and treated for LQTS at Mayo Clinic’s Genetic Heart

most common cardiac channelopathies

Rhythm Clinic between the years of 1999 and 2015.

that predisposes patients to arrhythmo-

For all patients, the electronic medical records were

genic syncope, seizures, and sudden cardiac

reviewed for demographics, clinical symptomatology,

death (SCD) (1–3). Clinical expressivity of LQTS

family history, genetic studies, LQTS-directed ther-

ranges from a lifelong asymptomatic state

apy, and occurrence of LQTS-related BCEs. A patient

with no electrocardiographic findings (i.e., con-

was considered symptomatic if he or she had an

cealed LQTS) to frequent, recurrent LQTS-

LQTS-related cardiac symptom prior to diagnosis

triggered torsades culminating in SCD (4,5).

(fetal

sympathetic denervation

Despite numerous genotype-phenotype correla-

seizure, or cardiac arrest). Patients were included if

AND ACRONYMS ACA = aborted cardiac arrest BCE = breakthrough cardiac event

ICD = implantable cardioverter-defibrillator

IQR = interquartile range LCSD = left cardiac

arrhythmia,

arrhythmogenic

syncope

or

LQTM = multiple long QT

tion studies, the extreme variance in expressiv-

they met any of the following diagnostic criteria for

syndrome–associated

ity makes it difficult for the patient and the

LQTS using the 2013 Heart Rhythm Society guidelines

mutations

physician to accurately predict the chance of

at the time of last medical record review: presence of

LQTS = long QT syndrome

experiencing an LQTS-related breakthrough

LQTS risk score $3.5, unequivocal pathogenic muta-

SCD = sudden cardiac death

cardiac event (BCE).

tion, or QTc interval of 480 to 499 ms on repeated electrocardiograms without secondary cause (13). The

SEE PAGE 463

Current

knowledge

regarding

patient’s QTc interval was measured on their first treated

and

untreated cardiac event rates in patients with LQTS stems from initial retrospective and prospective studies from international LQTS registry data, or small, single-center studies generally evaluating a specific treatment (e.g., beta-blockers and left cardiac sympathetic denervation [LCSD]). It has been reported that untreated asymptomatic patients with genotyped LQTS have a high risk of any cardiac event and SCD (36% and 13%, respectively, in a 28-year follow-up) (6). As first shown in 1985, beta-blockers drastically reduced mortality (from event rates ranging from as high as 71% in previously untreated symptomatic patients to 6% in patients on therapy), and for these past 30 years beta-blocker therapy has been the initial treatment of choice (7). Recent studies have focused mostly on evaluating outcomes of single-treatment modalities. Beta-blockers (chiefly nadolol

and

propranolol)

reduce

mortality

to

approximately 0.5% to 2% during a follow-up of 5 to 10 years (8,9). However, despite this significant decrease in mortality, beta-blocker–treated LQTS patients continue to have an annualized BCE rate of around 3% (8,9). For those patients with recurrent BCEs despite pharmacotherapy, LCSD significantly reduces the number of BCEs (10–12). Given that our current outcome estimates are derived mostly from multicenter or registry-based studies, we sought to evaluate the outcomes of patients with congenital LQTS who were evaluated and treated in single specialty center dedicated to patients with genetic heart rhythm diseases such as LQTS.

METHODS

Mayo Clinic electrocardiogram using the Bazett formula by computer and manually verified (M.J.A.). The primary outcome evaluated was the occurrence of an LQTS-related BCE, which was defined as arrhythmogenic syncope, seizure, aborted cardiac arrest

(ACA),

an

appropriate

VF-termination

implantable cardioverter-defibrillator (ICD) shock, or SCD after their first evaluation at our specialty center. Because the vast majority of the patients did not have an internal loop recorder to confirm that the BCE of syncope or seizure was indeed an LQTS-triggered torsadogenic episode, arrhythmogenic syncope and arrhythmogenic syncope with subsequent seizures was a clinical judgment call adjudicated by a single genetic cardiologist (M.J.A.) after thorough review of the event during the patient’s face-to-face clinical evaluation or telecommunication. Arrhythmogenic syncope was defined as a sudden loss of consciousness with spontaneous recovery and excluded all events assessed to be likely vasovagal in nature (e.g., emotional reactions, in the setting of heat or dehydration, and abrupt postural changes). Arrhythmogenic seizure was defined as sudden loss of consciousness with subsequently observed generalized tonic or clonic seizure activity with spontaneous recovery. Seizures assessed to be focal, febrile, or acquired were excluded. Treated follow-up was defined as time from initial Mayo Clinic evaluation until last evaluation, provider-patient communication regarding LQTS-related events, or censored at December 1, 2015. Annual event rate was defined as the proportion of patients with at least 1 BCE divided by the number of median follow-up years. Statistical analysis was performed by using Wilcoxon test for nonparametric measures, and multiple

We performed a retrospective review of the electronic

comparisons of nonparametric measures were per-

medical records of 606 patients who were evaluated

formed using the Steel-Dwass test. For the purposes

Rohatgi et al.

JACC VOL. 70, NO. 4, 2017 JULY 25, 2017:453–62

of this study, all continuous data were described as median (interquartile range [IQR]). The Fisher exact

455

Outcomes in LQTS

T A B L E 1 Cohort Characteristics and Comparisons Between Asymptomatic Patients and

Previously Symptomatic Patients

test was used to compare cohort characteristics, where type I error was minimized using Bonferroni

Entire Cohort (N ¼ 606)

Asymptomatic (n ¼ 440, 73%)

Symptomatic (n ¼ 166, 27%)

356 (59)

249 (57)

107 (64)

NS NS

p Value

correction for multiple comparisons. Kaplan-Meier

Female

survival curves were created with censoring at first

Age of diagnosis, yrs

14.7 (6.8–31.8)

14.7 (7.1–32.5)

14.7 (5.8–29.3)

BCE or last follow-up, and used to compare outcomes

Median QTc interval, ms

465 (441–490)

459 (438–480)

484 (459–513)

based on symptomatic status and LQTS genotype.

Symptomatic

166 (27)

0

166 (100)

—

133 (80)

NA

133 (80)

NA

20 (12)

NA

20 (12)

NA

13 (8)

NA

13 (8)

NA

Age at first LQTS event, yrs

12 (4–18)

NA

12 (4–18)

NA

Patients with symptomatic presentation <1 yr

27 (16)

NA

27 (16)

NA

Both LQT4-17 and genotype-negative or phenotype-

Syncope/seizure

positive groups were removed from Kaplan-Meier

Fetal bradyarrhythmia

analysis due to small numbers. All tests were performed using JMP software version 12.0 (SAS Institute, Cary, North Carolina).

Cardiac arrest

0.0001

Family history of SCD

268 (56)

216 (49)

52 (31)

0.0001

Family history of LQTS

449 (74)

358 (82)

91 (54)

0.0001

LQT1

287 (47)

225 (51)

62 (37)

LQT2

204 (34)

139 (32)

65 (39)

LQT3

56 (9)

43 (10)

13 (8)

NS

Table 1, where the median age of diagnosis was 14.7

LQT4-17

16 (3)

14 (3)

2 (1)

NS

(IQR: 6.8 to 31.8) years of age and the median QTc

LQTM

29 (5)

12 (3)

17 (10)

0.0004*

interval was 465 (IQR: 441 to 490) ms. Genetic testing

G/Pþ

14 (2)

7 (1)

7 (4)

NS

6.7 (3.9–9.8)

6.7 (3.9–9.8)

6.8 (3.7–10.1)

NS

RESULTS The demographics of the entire LQTS cohort of 606 patients (356 female patients [59%]) are detailed in

was positive in 592 (98%) patients, of whom 287 (47%) were LQT1 (KCNQ1), 204 (34%) were LQT2 (KCNH2), 56 (9%) were LQT3 (SCN5A), 29 (5%) had multiple LQTS-associated mutations (LQTM), and 16

LQTS genotype

0.0001

Treated follow-up, yrs

0.002* NS

Values are n (%) or median (interquartile range). *Bonferroni correction was applied to reduce type I error. G/Pþ ¼ genotype negative/phenotype positive; LQTS ¼ long QT syndrome; LQTM ¼ multiple long QT syndrome–associated mutations; NA ¼ not applicable; SCD ¼ sudden cardiac death.

(3%) were LQT4-17. Overall, the median treated follow-up was 6.7 (IQR: 3.9 to 9.8) years. A detailed comparison of baseline cohort charac-

LQT2 (p ¼ 0.03). None of the patients with LQT4-17

teristics by those patients who were symptomatic

had a family history of SCD, which was significantly

versus those who were asymptomatic prior to diag-

lower when compared with the rest of the group

nosis is also given in Table 1. Importantly, the symp-

(p ¼ 0.002). Also, genotype-negative or phenotype-

tomatic group had a longer median QTc interval

positive patients were significantly less likely to

(484 ms vs. 459 ms; p ¼ 0.0001), a smaller proportion

have a family history of LQTS when compared with all

of patients with LQT1 (37% vs. 51%; p ¼ 0.002), and a

other groups (p ¼ 0.0004).

larger proportion of patients with LQTM (10% vs. 3%;

The individualized treatment programs ranged

p ¼ 0.0004) when compared with the asymptomatic

from intentional nontreatment (preventative mea-

group. However, the asymptomatic group had a larger

sures only including LQTS-related lifestyle recom-

proportion of patients who had a family history of

mendations such as avoidance of QT prolonging

SCD (49% vs. 31%; p ¼ 0.0001) and a family history of

medications, dehydration, and electrolyte imbal-

LQTS (82% vs. 54%; p ¼ 0.0001) when compared with

ances, as well as education on and a strong recom-

the symptomatic group.

mendation

to

purchase

an

automated

external

When comparing the clinical characteristics by

defibrillator) to triple therapy that involved a combi-

genotype (Table 2), patients with LQTM had a signif-

nation of pharmacotherapy, LCSD, and ICD (Table 3 and

icantly longer median QTc interval (485 [IQR: 460 to

Figure 1). LQTS-directed therapy listed, from most

502] ms) when compared with LQT1 (460 [IQR: 439 to

common to least, were: beta-blockers alone in 350

486] ms; p ¼ 0.02). Compared with the rest of the

(58%) patients; beta-blockers with ICD in 78 (13%)

LQTS genotypes, significantly fewer LQT1 patients (62

patients;

[22%]; p ¼ 0.002) and significantly more patients with

patients; beta-blockers with LCSD in 33 (5%) patients;

intentional

nontreatment

in

47

(8%)

LQTM (17 [58%]; p ¼ 0.0004) presented with LQTS

ICD alone in 25 (4%) patients; beta-blockers, LCSD, and

symptoms prior to diagnosis. Patients with LQTM had

ICD in 21 (3%) patients; and LCSD alone in 18 (3%)

a significantly earlier age of first event compared with

patients. The rest of the therapeutic combinations

patients with LQT2 (p ¼ 0.04); patients with LQT3

together accounted for the remaining 34 (6%) patients.

mutations had a significantly earlier age of first event

Though ICD monotherapy, LCSD monotherapy, or

compared with patients with LQT1 (p ¼ 0.03) and

intentional nontherapy are not typically considered,

456

Rohatgi et al.

JACC VOL. 70, NO. 4, 2017 JULY 25, 2017:453–62

Outcomes in LQTS

T A B L E 2 Cohort Characteristics and Comparison by LQTS Genotype

LQT1 (n ¼ 287)

LQT2 (n ¼ 204)

LQT3 (n ¼ 56)

LQTM (n ¼ 29)

LQT4-17 (n ¼ 16)

G/Pþ (n ¼ 14)

Female

166 (58)

118 (58)

35 (63)

15 (52)

13 (81)

9 (64)

NS

Age of diagnosis, yrs

15 (8–34)

15 (6–27)

12 (5–32)

8 (4–28)

14 (11–29)

16 (13–33)

NS

p Value

460 (439–486)† 466 (442–496) 474 (453–491) 485 (460–502)* 457 (436–470) 480 (457–491) 0.0001

Median QTc interval, ms Symptomatic

62 (22)†

65 (32)

13 (23)

17 (58)†

2 (13)

7 (50)

0.0001‡

55 (19)

55 (27)

3 (5)

15 (52)

2 (13)

3 (21)

—

Fetal bradyarrhythmia

5 (2)

6 (3)

8 (14)

1 (3)

0

0 (0)

—

Cardiac arrest

2 (1)

4 (2)

2 (4)

1 (3)

0

4 (29)

11 (5–18)§

14 (7–20)§

0 (0–13)§

4 (2–10)§

14 (4–24)

18 (2–35)

Syncope/seizure

Age at first LQTS event, yrs

— 0.001

Family history of SCA

143 (50)

79 (39)

29 (52)

11 (38)

0ǁ

6 (43)

0.0002‡

Family history of LQTS

225 (78)

157 (77)

36 (64)

19 (66)

8 (50)

4 (29)¶

0.0001‡

6.7 (4.0–9.8)

6.4 (3.6–9.5)

7.2 (5.2–10.9)

7.1 (4.3–10.1)

5.2 (3.1–6.8)

9.3 (5.0–12.4)

NS

Median treated follow-up, yrs

Values are n (%) or median (interquartile range). *Patients with LQTM had a significantly higher median QTc interval compared with patients with LQT1 (p ¼ 0.02). †Patients with LQT1 were significantly less symptomatic compared with the rest (p ¼ 0.002), whereas patients with LQTM were significantly more symptomatic (p ¼ 0.0004). ‡Bonferroni correction was applied to reduce type I error. §Patients with LQTM had a significantly earlier age of first event compared with patients with LQT2 (p ¼ 0.04), and patients with LQT3 mutations had a significantly earlier age of first event compared with patients with LQT1 (p ¼ 0.03) and LQT2 (p ¼ 0.03). ǁThe proportion of patients with a family history of SCA was significantly lower in those patients with LQT4-17 compared with all other LQTS groups (p ¼ 0.0002). ¶The G/Pþ has a significantly lower proportion of patients with a family history of LQTS when compared with all other LQTS groups (p ¼ 0.0004). Bold indicates significant findings. Abbreviations as in Table 1.

there were instances where these LQTS-directed

detailed in Tables 4 and 5. Overall, of the 606 patients,

treatment options were seen as the best approach in

only 50 (8%) patients experienced $1 LQTS-triggered

the

plan,

BCE during a median treated follow-up of 6.7 (IQR:

balancing the future risk of event, medication or

3.9 to 9.8) years (4,316 total patient-years). This

procedure side effects, and compliance. Specifically,

included 20 (3%) patients with only a single BCE,

the 47 patients on no active therapy were considered

21 (3%) patients with 2 to 5 BCEs, 6 (1%) patients with

extremely low risk based on a median older age of

6 to 10 BCEs, and 3 (<1%) patients with >10 BCEs. This

diagnosis (39 [IQR: 17 to 53] years), lower median QTc

amounted to an annual event rate of 1.2% (1.2% of

interval (445 [IQR: 427 to 474] ms), and the fact that

patients had at least 1 BCE per year). The types of BCE

almost all patients (45 [96%]) had been lifelong

observed, from most to least common, were appro-

patient’s

individualized

treatment

asymptomatic at the time of their first evaluation.

priate ventricular fibrillation–terminating ICD shock,

Most of the 58 patients with either LCSD or ICD

syncope or seizure secondary to a suspected LQTS-

monotherapy were considered moderate–high-risk

triggered arrhythmia, ICD storm, ACA, and death.

patients, but were unable to take beta-blockers due to

This data as well as the frequency of each type of BCE

significant

daily

medication–related

side

effects

observed is summarized in Online Table 1. The breakdown of event burden before and after the pa-

resulting in persistent noncompliance. Treatment outcomes of the entire cohort, compar-

tients’ Mayo Clinic evaluation and treatment are

isons between symptomatic and asymptomatic pa-

described and shown in Figure 2. As a tertiary care

tients, and comparisons by LQTS genotype are

center, 158 (26%) patients had been referred from an outside institution for further treatment guidance after the LQTS diagnosis had been determined and initial treatment for LQTS had already commenced.

T A B L E 3 LQTS Treatment Summary by Sex: Pharmacotherapy,

Among this physician-referred subset, 70 (44%) pa-

LCSD, and ICD

tients were symptomatic prior to diagnosis. Among all No Procedures

No medications BB Sodium-channel blocker BB and sodium-channel blocker

47 (8; 53)

LCSD

ICD

18 (3; 72) 25 (4; 76)

LCSD þ ICD

8 (1; 75)

350 (58; 55) 33 (5; 52) 78 (13; 77) 21 (3; 43) 6 (1; 17) 4 (<1; 75)

0

5 (1; 40)

0

1 (<1; 0) 2 (<1; 100) 8 (1; 50)

166 previously symptomatic patients, 125 (75%) realized

a

significant

decrease

in

event

burden.

Conversely, 42 (25%) of these patients experienced subsequent LQTS-triggered BCEs. Overall, the BCE-free survival was 96% at 1 year,

Values are n (% total; % female). Bold indicates significant findings. BB ¼ beta-blocker; ICD ¼ implantable cardioverter-defibrillator; LCSD ¼ left cardiac sympathetic denervation; LQTS ¼ long QT syndrome.

93% at 5 years, and 90% at 10 years for the entire cohort (Figure 3A). In fact, there were only 2 LQTSrelated deaths (w0.3% overall, 1% among previously

Rohatgi et al.

JACC VOL. 70, NO. 4, 2017 JULY 25, 2017:453–62

457

Outcomes in LQTS

symptomatic patients) (Online Table 2) and 3 (<0.5%) heart transplants in over 4,000 collective years of

F I G U R E 1 Treatment Summary

treated follow-up (14). Demographics and clinical

BB and ICD; 78; 13%

details for these 5 cases are summarized in Online Table 2. As previously shown, LQTS patients with

No Active Therapy; 47; 8%

symptomatic expression of their disease were more Other; 34; 6%

likely to have BCEs. Therefore, our initial comparison of treatment outcomes was by LQTS symptoms prior to diagnosis and is detailed in Table 4. Previously

BB LCSD; 33; 5%

symptomatic LQTS patients had a significantly higher percentage of patients who had a BCE compared with asymptomatic

LQTS

patients

(25%

vs.

ICD Alone; 25; 4%

2%; BB, LCSD, ICD; 21; 3% LCSD Alone; 18; 3%

p ¼ 0.0001), a significant difference between BCE-free survival (Figure 3B) (p ¼ 0.0001), and a significantly higher annual event rate in symptomatic patients when compared with asymptomatic patients (3.7%/ year vs. 0.3%/year; p ¼ 0.0001). The BCE-free survival

BB Alone; 350; 58%

for the asymptomatic cohort was 99.3% at 1 year, 98.7% at 5 years, and 96.9% at 10 years. In compari-

Summary of long QT syndrome–directed therapy after being evaluated and risk stratified

son, the BCE-free survival for the symptomatic group

at the Mayo Clinic. BB ¼ beta-blocker; ICD ¼ implantable cardioverter-defibrillator;

was 87.1% at 1 year, 76.9% at 5 years, and 71.6% at 10

LCSD ¼ left cardiac sympathetic denervation.

years (Figure 3B). There were no significant differences between treated follow-up years between asymptomatic and symptomatic patients.

between female (11 of 255 [4%]) and male (3 of 173

Furthermore, within the symptomatic group, there were 27 (16%) patients who experienced their sentinel event during the first year of life compared with 139 (84%) patients whose sentinel event occurred after

[2%]) patients in BCEs during follow-up. Similarly, no significant sex-associated differences

were

seen

when subdividing this group by the 3 canonical LQTS genotypes (LQT1, LQT2, LQT3) (Online Table 3).

1 year of age. When comparing treatment outcomes

A comparison of treatment outcomes by LQTS ge-

between these groups, the patients who presented

notype is detailed in Table 5. Both patients with

before 1 year of age showed significantly worse out-

LQTS4-17 or genotype negative or phenotype positive

comes as demonstrated by higher percentage of pa-

were not included in statistical analysis due to low

tients who experienced a BCE, compared with those

number of patients in these subgroups. There were a

who presented after 1 year of age (67% vs. 17%;

significantly higher proportion of patients with LQTM

p < 0.0001). Additionally, all 3 of the patients who have undergone transplant and 1 of the 2 patients who succumbed subsequently to SCD were among these 27 patients who presented before 1 year of age. As beta-blockers have been the gold standard of LQTS-directed

pharmacotherapy,

we

(8 [28%]; p ¼ 0.001) or LQT3 (11 [20%]; p ¼ 0.004) who had $1 BCE when compared with the rest of the LQT genotypes. There was a significant difference in BCE-free survival by genotype (p < 0.0001), as shown

further

analyzed if there was a sex-specific risk associated

T A B L E 4 Treatment Outcomes and Comparison by Symptomatic Status

with treatment outcomes. In our cohort of patients treated by a single LQTS specialist, we did not see

Entire Cohort (N ¼ 606)

Asymptomatic (n ¼ 440)

Symptomatic (n ¼ 166)

p Value

evidence of a sex-specific effect of beta-blockers,

Follow-up, yrs

6.7 (3.9–9.8)

6.7 (3.9–9.8)

6.8 (3.7–10.1)

NS

even

Event/total

50/606 (8)

8/440 (2)

42/166 (25)

0.0001

1.2

0.3

3.7

0.0001

0 BCE

556 (92)

431 (98)

125 (75)

1 BCE

20 (3)

8 (2)

12 (7)

2–5 BCEs

21 (3)

—

21 (13)

when

subdividing

by

genotype

(Online

Table 3). Specifically, among 497 patients on betablockers

(some

treated

with

concomitant

Na

blockers or LCSD), there was no statistical difference between female (23 of 290 [7.9%]) and male (17 of 207

Annual event rate, %/yr* Event burden

NA

[8.2%]) patients who went on to have at least 1 BCE

6–10 BCEs

6 (1)

—

6 (3)

during treated follow-up. Even when removing the

>10 BCEs

3 (<1)

—

3 (2)

potential protective effect of the either LCSD and sodium-channel blockers (n ¼ 428 patients on betablockers alone), there were no statistical differences

Values are median (interquartile range), n/N (%), or n (%) unless otherwise indicated. *The incidence of a patient with any breakthrough cardiac event (BCE) per yr.

Rohatgi et al.

JACC VOL. 70, NO. 4, 2017 JULY 25, 2017:453–62

Outcomes in LQTS

T A B L E 5 Treatment Outcomes and Comparison by LQTS Genotype

LQT1 (n ¼ 287)

LQT2 (n ¼ 204)

LQT3 (n ¼ 56)

LQTM (n ¼ 29)

LQT4-17‡ (n ¼ 16)

G/Pþ‡ (n ¼ 14)

6.7 (4.0–9.8)

6.4 (3.6–9.5)

7.2 (5.2–10.9)

7.1 (4.3–10.1)

5.2 (3.1–6.8)

9.3 (5.0–12.4)

NS

12 (4)

16 (8)

11 (20)

8 (28)

1 (6)

2 (14)

0.0001*

0.6

1.1

2.6

3.5

1.1

1.6

0.0001*

0 BCE

275 (95)

188 (93)

45 (81)

21 (73)

15 (94)

12 (86)

1 BCE

6 (2)

9 (4)

3 (5)

1 (3)

1 (6)

—

2–5 BCEs

5 (2)

7 (3)

3 (5)

5 (17)

—

1 (7)

Follow-up, yrs Patients with cardiac event Annual event rate, %/yr† Event burden

p Value

NA

6–10 BCEs

—

—

3 (5)

2 (7)

—

1 (7)

>10 BCEs

1 (<1)

—

2 (4)

—

—

—

Values are median (interquartile range) or n (%), unless otherwise indicated. *Bonferroni correction was used to reduce type I error. †The incidence of a patient with any breakthrough cardiac event (BCE) per year. ‡These patients were excluded from statistical analysis. Bold indicates significant findings. Abbreviations as in Table 1.

in Figure 3C. One-year and 10-year BCE-free survival

collected in different ways, complicating generaliz-

was highest in patients with LQT1 (98% and 95%,

ability. Given the high clinical stakes for the patient,

respectively), whereas 1-year BCE-free survival was

the marked heterogeneity in expressivity of the ge-

lowest in patients with LQTM (86% and 74%,

netic substrates, and the multitude of treatment op-

respectively). Treated follow-up duration was similar

tions (beta-blockers, sodium channel blockers, LCSD,

between all LQTS genotypes.

and ICD, to name just 4), we undertook the largest single-center outcome study of LQTS patients to date

DISCUSSION

to help better understand the expected genotype- and

Although the understanding of the pathophysiology, diagnosis, and treatment has advanced considerably since the sentinel description of this disease nearly 60 years ago, the study of its overall outcomes is sparse due to the low prevalence of disease, and has depended on retrospective analysis from multicenteror registry-based data. Often, patients included in these studies are risk stratified and treated differently from institution to institution with clinical data

phenotype-derived

incidences

of

BCEs

in

the

contemporary era. Until now, our understanding regarding untreated mortality due to SCD or cardiac arrest in LQTS patients comes primarily from 2 studies. In 1985, Schwartz (7) first showed the mortality in untreated symptomatic LQTS patients was extremely high at 71%. Later in 2003, Priori et al. (6) found that even among 647 previously asymptomatic LQTS patients, 13% experienced a sentinel event of ACA or SCD in the untreated state, over a mean follow-up of 28 years before 40

F I G U R E 2 Event Burden Before and After Mayo Clinic Evaluation and Treatment

>10

>10

6-10

6-10

ability of clinical genetic testing, we have learned

(N = 3)

(N = 5)

(N = 6)

2-5

2-5

(N = 60)

(N = 21)

1

1

0

0

(N = 91)

(N = 20)

(N = 440)

years of age. Following elucidation of the genetic underpinnings of LQTS along with increased avail-

Number of Cardiac Events After Evaluation

(N = 10)

Number of Cardiac Events Before Evaluation

458

(N = 556)

Mayo Clinic Evaluation, Risk Stratification, and Treatment

that approximately 25% of LQTS patients have concealed

(electrocardiographically

normal)

LQTS,

thereby increasing the prevalence of the disease and decreasing its global severity (4). Beta-blockers, LCSD, and ICD are the principal treatment modalities in LQTS with clear benefit in primary and secondary prevention of SCD in patients with LQTS. In 1985, Schwartz and Locati (15) first showed that betablockers can decrease the mortality from 71% to 6% after prospectively following 200 patients. In more recent studies, the reported mortality in patients

Blue arrows represent patients who remained event free or decreased in total number of

treated with beta-blockers (with a small percentage of

events before and after treatment. Red arrows represent patients who had the same or

patients with concomitant LCSD) has ranged between

increased number of cardiac events. The thickness of the arrow reflects the number of

1% and 7% within 5 to 10 years of follow-up (8,9,16).

patients (i.e., the thicker the arrow is, the larger the group of patients).

Here, in our patient-tailored treatment program, we report an extremely low mortality of 2 of 606 (0.3%)

Rohatgi et al.

JACC VOL. 70, NO. 4, 2017 JULY 25, 2017:453–62

459

Outcomes in LQTS

F I G U R E 3 BCE-Free Survival in Treated LQTS

A

B

1.0

0.8 BCE-Free Survival

BCE-Free Survival

0.8

1.0

0.6

0.4

0.2

Event-Free Survival by Yr (%) LQTS

1

5

10

96

93

90

0.6 P < 0.0001 0.4 Event-Free Survival by Yr (%) 0.2

0.0

1

5

10

Asymptomatic

99.3

98.7

96.9

Symptomatic

87.1

76.9

71.6

0.0 0

2

4

6

8

10

0

2

4

Time-to-Event (Years) No At Risk LQTS 606

515

354

C

6

8

10

139 42

80 30

Time-to-Event (Years) 177

109

No At Risk Asymp 440 Symp 166

390 125

275 79

1.0

BCE-Free Survival

0.8 P < 0.0001

0.6 Event-Free Survival by Yr (%) 0.4

0.2

1

5

10

LQT1

98

97

95

LQT2

96

93

90

LQT3

87

81

77

LQTM

86

74

74

0.0 0

2

4

6

8

10

88 60 18 5

53 37 10 4

Time-to-Event (Years) No At Risk LQT1 287 LQT2 204 LQT3 56 LQTM 29

246 176 46 21

169 117 38 14

(A) Breakthrough cardiac event (BCE)–free survival curve for entire long QT syndrome (LQTS) cohort. (B) Comparison of BCE-free survival by symptom status. The blue line represents patients who were asymptomatic at the time of diagnosis and the orange line represents the patients who were symptomatic at the time of diagnosis. (C) Comparison of BCE-free survival by LQTS genotype. LQT1 patients are represented by the blue line. LQT2 patients are represented by the orange line. LQT3 patients are represented by the gray line. Patients with multiple long QT syndrome–associated mutations (LQTM) are represented by the red line.

cases overall or 2 of 167 (1%) previously symptomatic

can still be improved. In fact, our study suggests that

cases with similar follow-up.

when managed and treated at a dedicated specialty

Though the incidence of LQTS-associated mortality

center, anticipated outcomes may exceed expecta-

in a treated cohort has decreased sharply, a signifi-

tions that were gleaned from previous studies. In our

cant number of previously symptomatic patients

cohort, there was a significantly lower incidence of

continue to have BCEs (syncope, seizure, and ICD

patients with 1 or more BCE while under treatment

shocks) despite maximal therapy. And although, like

when compared with previous studies: 8% with at

mortality, a decrease in BCEs has been observed over

least 1 BCE over nearly 7 years follow-up compared

the last few decades, risk stratification and outcomes

with a previously reported 16% to 33% over 5 treated

Rohatgi et al.

JACC VOL. 70, NO. 4, 2017 JULY 25, 2017:453–62

Outcomes in LQTS

C E N T R A L IL L U ST R A T I O N Trends in Rates of BCEs Among Treated LQTS Patients

4 3.7 3.5 3.1 3 % of Patients with BCE / Yr

460

2.5

2.5 2 1.5

1.3

1

1.3

0.69

0.5 0

0.28

0.23

0.05 1985 Schwartz, P 5-yr FU

2000 Moss et al 5-yr FU

2004 Priori et al 5-yr FU Any BCE

2009 Vincent et al 10-yr FU

2017 Mayo Clinic 6.7-yr FU

SCD

Rohatgi, R.K. et al. J Am Coll Cardiol. 2017;70(4):453–62.

Previous studies (primarily evaluating the efficacy of long QT syndrome–directed therapy such as beta-blockers) juxtaposed with Mayo Clinic results. BCE ¼ breakthrough cardiac event; FU ¼ follow-up; SCD ¼ sudden cardiac death.

years (8,16). Among both asymptomatic and previ-

appropriately risk stratified and treated LQTS patients,

ously symptomatic subsets, there have been less

even with the majority of patients being treated

BCEs than previously reported: 99% BCE-free sur-

without an ICD. Second, the incidence of LQTS-related

vival at 5 years among asymptomatic patients in this

BCEs in patients treated at a single LQTS specialty

study compared with 94% from a prior study, and

center is less than previously reported, whether look-

77% BCE-free survival at 5 years among previously

ing at the cohort as a whole, divided by symptomatic

symptomatic patients in this study compared with

status, or by genotype (Central Illustration). Third,

68% from a prior study (8).

compared with previous studies, the majority of pa-

Akin to previous observations, our cohort also

tients in our cohort were diagnosed while asymptom-

suggests strong genotype-specific risk for recurrent

atic (73% vs. 36% to 53%), which may be evidence of the

BCEs. Our data show that patients with LQT3 and

changing landscape of LQTS in the contemporary era

patients with LQTM are at highest risk of BCEs

(8,16). This finding is likely due to treated patients

when compared with LQT1 and LQT2 patients (17).

living longer, increased awareness of the importance

However, even among these genotypic higher-risk

of family screening, wider availability of comprehen-

subsets, overall improvements in BCE-free survival

sive genetic testing, and subsequent cascade testing of

when compared with historical cohorts are seen.

the proband’s relatives (18). Last, 2 potential distinct

This study sheds light on number of important fac-

subgroups within LQTS emerge. The first is a growing

ets of outcomes in patients with LQTS in the contem-

group of patients with concealed and asymptomatic

porary era. First, LQTS-associated mortality should

disease who have a very mild or no LQTS expressivity

be rare. Mortality has significantly decreased in

and may not need active therapy. The second is a group

Rohatgi et al.

JACC VOL. 70, NO. 4, 2017 JULY 25, 2017:453–62

Outcomes in LQTS

of patients with malignant LQTS (i.e., patients with

with neonatal malignant LQT3 (9 of 14 [64%]),

severe LQTS expressivity who, despite current ther-

skewing our LQT3 population to an earlier age of

apy, continue to have higher rates of BCEs). In both of

presentation. Nevertheless, when we compared the

these groups, further improvements in risk stratifica-

BCE rates between our symptomatic cohorts and

tion and treatment are necessary.

previously published symptomatic cohorts, there has

STUDY LIMITATIONS. As with clinical retrospective

been a favorable reduction in BCEs.

research, this study has limitations including referral

CONCLUSIONS

bias and retrospective data abstraction. Mayo Clinic’s Genetic Heart Rhythm Clinic receives many patients

With careful evaluation, risk stratification, and

from outside medical centers, and information

treatment almost all asymptomatic LQTS patients

regarding past medical history is limited to docu-

should remain asymptomatic. After establishing a

mentation obtained elsewhere and patient recollec-

robust, patient-tailored, LQTS-directed treatment

tion. As a single center, all patients were evaluated,

program,

risk stratified, counseled, and treated by a single

extremely low. However, although mortality is rare

the

annual

mortality

rate

should

be

genetic cardiologist (M.J.A.). LQTS is an uncommon

(<1% of patients in over 4,000 patient-years) and

disease, and cohort characteristics vary. Outcomes

outcomes have improved in comparison with previ-

have not been reported on a generalized scale;

ous studies, nearly 1 in 4 symptomatic patients still

therefore, comparisons with previous studies are

experienced at least 1 nonlethal, BCE. This indicates

imperfect and must take into consideration the study

that there is still need for improved risk stratification

institution’s cohort characteristics. In fact, the overall

and optimization of their treatment program.

decreased mortality and decreased incidence of BCEs compared with previous studies is likely due to

ADDRESS FOR CORRESPONDENCE: Dr. Michael J.

patient-tailored

LQTS-directed

Ackerman, Department of Cardiovascular Diseases/

therapy, and a slightly milder LQTS phenotype over-

Division of Heart Rhythm Services, Guggenheim 501,

all in comparison with previous registry studies.

Mayo Clinic, Rochester, Minnesota 55905. E-mail:

Although our cohort had similar proportion of LQTS

[email protected].

risk

stratification,

subtypes, in comparison with previous studies there were fewer patients who presented with symptoms (27% vs. 47% to 64%), and a shorter QTc interval (465 ms vs. 492 to 520 ms) (8,16). When comparing the patients in our symptomatic cohort with previous studies, we also observed differences regarding onset of first symptom. LQT1 became symptomatic later (45% symptomatic by 10 years of age vs. 54%, respectively), LQT2 became symptomatic earlier (57% symptomatic by 16 years of age vs. <50%, respectively), and LQT3 became symptomatic markedly earlier (75% symptomatic by 16 years of age vs. <50%, respectively)

(19).

However,

among

our

LQT3

PERSPECTIVES COMPETENCY IN MEDICAL KNOWLEDGE: Although LQTS is among the more common genetic channelopathies predisposing to SCD, contemporary management has been associated with improved clinical outcomes. TRANSLATIONAL OUTLOOK: Further study is needed to identify markers that predict risk of cardiac events in patients with LQTS and develop treatments for those who remain at risk of events despite current therapies.

patients, a large proportion of patients presented

REFERENCES 1. Jervell A, Lange-Nielsen F. Congenital deaf-mutism, functional heart disease with prolongation of the Q-T interval and sudden death. Am Heart J 1957;54:59–68.

genotype-confirmed long-QT syndrome and normal-range corrected QT intervals. J Am Coll Cardiol 2011;57:51–9.

2. Schwartz PJ, Ackerman MJ. The long QT syndrome: a transatlantic clinical approach to diagnosis and therapy. Eur Heart J 2013;34:3109–16.

Jervell and Lange-Nielsen syndrome: natural history, molecular basis, and clinical outcome. Circulation 2006;113:783–90.

3. Schwartz PJ, Stramba-Badiale M, Crotti L, et al.

6. Priori SG, Schwartz PJ, Napolitano C, et al. Risk

Prevalence of the congenital long-QT syndrome. Circulation 2009;120:1761–7.

stratification in the long-QT syndrome. N Engl J Med 2003;348:1866–74.

4. Goldenberg I, Horr S, Moss AJ, et al. Risk for life-threatening cardiac events in patients with

7. Schwartz PJ. Idiopathic long QT syndrome: progress and questions. Am Heart J 1985;109:399–411.

5. Schwartz PJ, Spazzolini C, Crotti L, et al. The

8. Moss AJ, Zareba W, Hall WJ, et al. Effectiveness and limitations of beta-blocker therapy in congenital long-QT syndrome. Circulation 2000; 101:616–23. 9. Vincent GM, Schwartz PJ, Denjoy I, et al. High efficacy of beta-blockers in long-QT syndrome type 1: contribution of noncompliance and QTprolonging drugs to the occurrence of betablocker treatment “failures.” Circulation 2009; 119:215–21. 10. Schwartz PJ. Cardiac sympathetic denervation to prevent life-threatening arrhythmias. Nat Rev Cardiol 2014;11:346–53.

461

462

Rohatgi et al.

JACC VOL. 70, NO. 4, 2017 JULY 25, 2017:453–62

Outcomes in LQTS

11. Bos JM, Bos KM, Johnson JN, Moir C, Ackerman MJ. Left cardiac sympathetic denervation in long QT syndrome: analysis of therapeutic nonresponders. Circ Arrhythm Electrophysiol 2013;6:705–11. 12. Schwartz PJ, Priori SG, Cerrone M, et al. Left cardiac sympathetic denervation in the management of high-risk patients affected by the long-QT syndrome. Circulation 2004;109:1826–33. 13. Priori SG, Wilde AA, Horie M, et al. HRS/EHRA/ APHRS expert consensus statement on the diagnosis and management of patients with inherited primary arrhythmia syndromes: document endorsed by HRS, EHRA, and APHRS in May 2013 and by ACCF, AHA, PACES, and AEPC in June 2013. Heart Rhythm 2013;10:1932–63. 14. Kelle AM, Bos JM, Etheridge SP, et al. Cardiac transplantation in children and adolescents with

long QT syndrome. Heart Rhythm 2017 Apr 14 [E-pub ahead of print]. 15. Schwartz PJ, Locati E. The idiopathic long QT syndrome: pathogenetic mechanisms and therapy. Eur Heart J 1985;6 Suppl D:103–14. 16. Priori SG, Napolitano C, Schwartz PJ, et al. Association of long QT syndrome loci and cardiac events among patients treated with beta-blockers. JAMA 2004;292:1341–4. 17. Schwartz PJ, Priori SG, Napolitano C. How really rare are rare diseases? the intriguing case of independent compound mutations in the long QT syndrome. J Cardiovasc Electrophysiol 2003;14: 1120–1.

and cardiomyopathies: this document was developed as a partnership between the Heart Rhythm Society (HRS) and the European Heart Rhythm Association (EHRA). Heart Rhythm 2011; 8:1308–39. 19. Schwartz PJ, Priori SG, Spazzolini C, et al. Genotype-phenotype correlation in the long-QT syndrome: gene-specific triggers for lifethreatening arrhythmias. Circulation 2001;103: 89–95.

KEY WORDS breakthrough cardiac events, genetics, long QT syndrome, LQTS, outcomes

18. Ackerman MJ, Priori SG, Willems S, et al. HRS/EHRA expert consensus statement on the state of genetic testing for the channelopathies

A PPE NDI X For supplemental tables, please see the online version of this article.