Different Causes of a Transient Loss of Consciousness with Convulsions between Two Young Sisters: Epilepsy and Type-2 Long QT Syndrome

Abstract

A misdiagnosis of syncope as a seizure can lead to catastrophic outcomes, and thus should be avoided. We herein present the cases of two young sisters who experienced transient loss of consciousness with convulsions and were diagnosed with epilepsy. After discovering that their asymptomatic mother had type-2 long QT syndrome (LQT2), it was found that the younger sister had LQT2, while the older sister did not. Although the 12-lead ECG of the younger sister showed notched T-waves, the corrected QT interval (422 ms) was normal, which precluded a diagnosis of LQT2. These cases provide clues to distinguish between LQT2-related arrhythmic events and epileptic events.

Introduction

Congenital long QT syndrome (LQTS) is a genetic disorder (cardiac channelopathy) typically without structural heart disease. It is characterized by QT prolongation on electrocardiogram (ECG) and polymorphic ventricular tachycardia, torsades de pointes, which leads to syncope and sudden cardiac death (SCD) (1). In contrast, epilepsy, a part of which is a genetic disorder (brain/cerebral channelopathy) without structural brain disease, typically causes seizures such as convulsions and/or transient loss of consciousness (TLOC), and refractory epilepsy confers a considerable lifetime risk of sudden unexplained death in epilepsy (2).

Patients with LQTS often experience episodes of TLOC with convulsions (seizure-like episodes), possibly due to prolonged cerebral hypoperfusion secondary to arrhythmias (3). When patients with LQTS experience episodes of TLOC with convulsions, they, especially young patients with type-2 LQTS (LQT2), can often be erroneously diagnosed with epilepsy (4-6), although there have been several reports that some patients with LQTS might also have genuine epilepsy (7-10). Therefore, it is often difficult to differentiate the causes of TLOC with convulsions.

A misdiagnosis of cardiac channelopathy as epilepsy delays the time from the onset of the first symptom to the correct diagnosis, and the use of antiepileptic drugs (AEDs) with a potential proarrhythmic effect increases the risk of sudden cardiac arrest and SCD (11,12). Therefore, it is important to avoid misdiagnosing cardiac channelopathy as epilepsy. Establishing reliable methods for making a differential diagnosis is essential for ensuring appropriate treatment.

We recently encountered two young sisters who had experienced episodes of TLOC with convulsions and were diagnosed with epilepsy. However, after discovering that their asymptomatic mother had LQT2, the younger sister was found to also have LQT2, but the older sister did not. These cases provide clues to distinguish between LQT2-related arrhythmic events and epileptic events.

Case Report

A 23-year-old female (older sister) (II-1) (Fig. 1A) had experienced multiple episodes of TLOC with convulsions since she was 7 years of age. These occurred in a variety of situations, such as walking and sleeping without any prodromes. The mother occasionally observed these episodes (Table). The convulsive episodes resembled tonic-clonic seizures, which usually lasted approximately 2-3 minutes, but it took several minutes for her to regain full consciousness due to postictal confusion. The patient had never experienced cut tongue or urinary incontinence. The patient did not recall these events. The brain magnetic resonance imaging (MRI) findings were normal, but her electroencephalogram (EEG) showed sharp waves, followed by slow waves (Fig. 2A). Her 12-lead ECG results were almost normal [heart rate (HR): 61 bpm, corrected QT (QTc): 420 ms] (Fig. 2B). Therefore, she was diagnosed with epilepsy by a neurologist and was prescribed 200 mg of carbamazepine daily. The AED therapy remarkably reduced her seizure episodes.

Figure 1.

Identification of a KCNH2 c.2028_2029delGA variant. A: Pedigree harboring the KCNH2 c.2028_2029delGA variant. Black filled symbols indicate heterozygous carriers of the KCNH2 c.2028_2029delGA variant. N/A indicates no assessment. Electropherogram of part of the KCNH2 exon 8 of the mother (I-2).

Table.

Reported and Observed Information of Transient Loss of Consciousness with Convulsions of Two Young Sisters.

	Number of episodes	Situation	Any prodromes	Duration of convulsion	Type of seizure	Duration of LOC	Postictal confusion	Cut tongue	Urinary incontinence
Older sister	Over 10	Walking, Sleeping etc.	No	2-3 mins	Tonic-clonic?	Relatively long	Yes	No	No
Younger sister	Once	Taking a nap	Feeling like being sucked into the darkness	2-3 mins	Tonic-clonic?	Relatively short	No	No	Yes

LOC: loss of consciousness

Figure 2.

Electroencephalographic and electrocardiographic findings of the older sister. A: Electroencephalogram (EEG) of the older sister (II-1) (Fig. 1A). Sharp waves are surrounded by a red dotted line. B: 12-lead electrocardiogram (ECG). C: Holter ECGs during bradycardia (upper panel) and tachycardia (lower panel).

A 19-year-old female (younger sister) (II-2) (Fig. 1A) had experienced an episode of TLOC with convulsions once while lying on the sofa and taking a nap at 18 years of age. The patient's mother observed this episode (Table). It appeared to be a tonic-clonic seizure, lasting approximately 2-3 minutes, after which she rapidly regained consciousness (faster than her older sister) without any postictal confusion. However, the patient experienced urinary incontinence. She did not bite her tongue. As her older sister regularly visited a neurologist, she visited the neurologist. Her brain MRI revealed normal findings, but her EEG showed suspected spike waves (Fig. 3A), although these waves were later thought to be artifacts because they lacked certain criteria for interictal epileptiform discharges in EEG, as defined in the EEG-glossary of the International Federation of Clinical Neurophysiology, such as a slow after-wave, disruption of the background activity, and apparent asymmetry of the waveform (13). Her ECG did not show QTc prolongation (HR: 98 bpm, QTc: 422 ms), but it did show notched T-waves (Fig. 3B). Considering that the older sister was diagnosed with epilepsy, the younger sister was also diagnosed with epilepsy and thus was prescribed 200 mg carbamazepine daily.

Figure 3.

EEG and ECG findings of the younger sister. A: EEG of the younger sister (II-2) (Fig. 1A). The suspected spike waves are surrounded by the red dotted line. B: 12-lead ECG (upper panel) and expanded II and V5 leads (lower panel). The red arrows indicate notched T-waves. C: Holter ECGs during bradycardia (upper panel) and tachycardia (lower panel).

Several months after the younger sister's episode, their mother (I-2) (Fig. 1A), who had been asymptomatic, was referred to a cardiologist for the first time because of QTc prolongation (HR: 66 bpm, QTc: 496 ms) on ECG at an annual medical check-up. Notably, the mother's ECG also showed notched T-waves (Fig. 4). After obtaining written informed consent for a genetic analysis (approved by the Gunma University Ethical Review Board for Medical Research Involving Human Subjects: HS2017-15), we performed target panel sequencing as described previously (10) and identified a novel deletion variant, KCNH2 c.2028_2029delGA / p.Q676HfsTer46 (NM_000238.4). This was validated using Sanger sequencing (Fig. 1B). Because the deletion variant causes a frameshift and results in truncation of KCNH2, this variant is thought to be pathogenic. Therefore, the patient was diagnosed with LQT2. To exclude the possibility that her daughters' events might have been due to LQT2-related arrhythmias rather than epileptic seizures, they were referred to cardiologists.

Figure 4.

ECG of the mother. 12-lead ECG (upper panel) of the mother (I-2) (Fig. 1A) and expanded leads II and V5 (lower panel). The red arrows indicate notched T-waves.

Notably, the younger sister reported that she felt like being sucked into the darkness prior to the episode (Table), which may suggest arrhythmic events rather than epileptic seizures. Transthoracic echocardiography revealed that neither the older sister nor the younger sister had any structural heart disease. The Holter ECG of the older sister did not show QTc prolongation even during tachycardia (Fig. 2C), while that of the younger sister showed QTc prolongation during tachycardia (Fig. 3C), thus suggesting LQTS. After obtaining written informed consent for a genetic analysis, we first performed target panel sequencing and found that the younger sister also harbored the same variant as her mother, which was validated using Sanger sequencing, whereas the older sister did not. Moreover, to confirm whether they harbor inherited epilepsy-related genetic variants, as listed previously (14), we outsourced whole exome sequencing to Macrogen Japan (https://macrogen-japan.co.jp/). However, we did not identify any pathogenic or likely pathogenic variants in the sisters.

Therefore, we diagnosed the younger sister to have LQT2. It is highly likely that the episode of TLOC with convulsions had thus been caused by LQT2-related arrhythmias. To confirm the relationship between cardiac arrhythmias and TLOC with convulsions, we recommended that the patient receive an implantable loop recorder, but she refused it. We prescribed 15 mg nadolol daily and discontinued carbamazepine. Subsequently, the patient remained asymptomatic for more than 2 years. Meanwhile, we diagnosed the older sister to have genuine epilepsy and decided to continue AED therapy.

Discussion

We initially diagnosed two young sisters, who had experienced TLOC with convulsions, with epilepsy. However, after discovering that their asymptomatic mother had LQT2, we found that the younger sister also had LQT2, whereas the older sister did not. Considering their precise clinical and genetic findings, we concluded that the events of the older sister had thus been caused by genuine epilepsy, while the event of the younger sister had been caused by LQT2-related arrhythmias.

There have been several reports on the association between LQT2 and epilepsy or EEG abnormalities (epileptiform activity on EEG). Seizure-like episodes and EEG abnormalities are more common in patients with LQTS than in healthy controls (15,16), and diagnoses of epilepsy and seizure-like episodes are more common in patients with LQT2 in comparison to other types of LQTS (3,17). There are several hypotheses regarding these matters: Some patients with LQTS, particularly LQT2, may also have “genuine” epilepsy. Otherwise, patients with LQT2 may have a decreased seizure threshold secondary to cerebral hypoperfusion stemming from LQT2-related arrhythmias, and thus may be misdiagnosed with epilepsy (3). Regarding the association between KCNH2 mutations and “genuine” epilepsy, some KCNH2 mutations have been reported to cause either LQT2, epilepsy, or both (9,18,19), as KCNH2 transcripts are expressed in both the heart and brain, including the hippocampus and frontal cortex. However, it remains unclear whether KCNH2 mutations are definitively associated with “genuine” epilepsy. In any case, since a misdiagnosis of LQTS-related arrhythmic events as epilepsy, particularly in LQT2, may delay the diagnosis and the use of AEDs may increase the risk of cardiac events (11,12), such a misdiagnosis should be avoided.

There may be several reasons why the younger sister was initially diagnosed with epilepsy. First, the older sister had been diagnosed with epilepsy prior to the seizure-like episode of the younger sister, which suggests that the cause of the younger sister's episode might be the same as that of the older sister. Second, the EEG of the younger sister showed suspected spike-waves. However, it was later thought that the EEG abnormality in the younger sister might have been an artifact. Third, it might have been difficult for non-cardiologists to recognize ECG abnormalities in the younger sister. It is known that there are patients with LQTS whose QTc intervals are within the normal range, and the QTc interval of the younger sister was also within the normal range, which might preclude a diagnosis of LQTS. However, a 12-lead ECG of the younger sister showed a T-wave abnormality, which was suggestive of LQT2 (20).

Notably, in LQT2 female carriers with a normal QTc interval, abnormal T-wave morphologies, including flat, notched, and negative T-waves, are associated with a higher risk of cardiac events compared with patients with normal T-waves (21). The younger sister's 12-lead ECG showed a normal QTc interval but notched T-waves, a hallmark of LQT2, which suggested that she might be at a higher risk of cardiac events. It should be noted that we should pay attention to the T-wave morphology in clinical practice, even if the QTc interval is normal. Additionally, cooperation between non-cardiologists and cardiologists is required to avoid the misdiagnosis of cardiac channelopathy as epilepsy.

Several self-reported and observed symptoms can help distinguish epilepsy from syncope (22,23). Symptoms such as a cut tongue, prodromal déjà vu or jamais vu before a spell, cyanosis, uncontrolled head turning, unusual posturing, postictal confusion, and bed wetting are likely attributable to epilepsy. In contrast, pre-event diaphoresis, remembering the event, post-exercise hypotension, palpitations before a spell, dyspnea before a spell, pallor before a spell, and preexisting heart disease are more indicative of syncope. However, it is not possible to fully obtain such information from a patient or witness. The presence of a prodrome (a feeling like being sucked into the darkness prior to the episode), the relatively rapid recovery from TLOC, and the absence of postictal confusion in the younger sister appear to be attributable to syncope, although only the younger sister experienced bed wetting/urinary incontinence, which is likely attributable to epilepsy. These facts suggest that the younger sister's event might have been caused by LQT2-related arrhythmias, rather than by epileptic seizures. Thus, both the reported and observed information should also be considered when making a differential diagnosis.

Taken together, these cases provide some clues to avoid a misdiagnosis of TLOC with convulsions. Careful interviews with patients and witnesses about episode details, as well as a thorough evaluation of examination results such as EEG and ECG, should be conducted collaboratively by non-cardiologists (physicians/neurologists) and cardiologists. Additionally, genetic testing should be performed to aid in making a differential diagnosis.

In conclusion, we encountered two young sisters who had experienced TLOC with convulsions and were initially diagnosed with epilepsy. However, the reported and observed symptoms, identification of notched T-waves on the younger sister's ECG, and genetic findings allowed us to reconsider that the cause of these events might be different: the older sister's events might have been caused by genuine epilepsy, whereas the younger sister's event might have been caused by LQT2-related arrhythmias. These cases emphasize the difficulty in making a differential diagnosis in young patients with TLOC with convulsions. They also highlighted the importance of cooperation between non-cardiologists and cardiologists as well as the use of genetic testing as a useful tool for making a differential diagnosis of the two conditions.

The authors state that they have no Conflict of Interest (COI).

Financial Support

This work was supported by a Grant-in-Aid for Scientific Research (C) from the Japan Society for the Promotion of Science (Grant Number: 23K06845 to T.N.), and by the MEXT Program for supporting the introduction of the new sharing system (Grant Number: JPMXS0420600120).

Hiroshi Hasegawa and Tadashi Nakajima contributed equally to this work.