Symptoms Before Sudden Arrhythmic Death Syndrome
Symptoms Before Sudden Arrhythmic Death Syndrome
In this study, we examined a nationwide cohort of SADS patients (n = 136) for their course prior to death, including symptoms, medical history and contact with the healthcare system. Symptoms prior to death were compared with a control group of subjects in the same age range who died in traffic accidents (n = 74) during the studied time interval. To our knowledge, this study describes the largest cohort of young SADS patients from the general population.
We found that significantly more patients with SADS had higher frequencies of cardiac symptoms prior to death than controls (P < 0.001), most commonly presyncope/syncope (17%). This could be due to underlying cardiac arrhythmias. Another important finding in our study is that a significant proportion of the patients had experienced an episode of seizures or had a diagnosis of epilepsy (18%). Linzer et al. suggested that the diagnosis of epilepsy in some cases may be wrong because primary cardiac arrhythmias can produce syncope and anoxic seizures. This highlights the fact that seizures can be difficult to distinguish from syncope induced by an arrhythmia and that patients with unexplained seizure activity should always have a cardiovascular examination, including ECG.
The literature is inconsistent regarding the notion that most SADS cases have symptoms. In contrast to our results, a recent study reported prior syncope was documented in only 4% of SADS patients, and up to 90% had no preceding symptoms or recognized risk factors before death. However, another study has shown that over half of the SADS patients had at least one antecedent cardiac symptom (including syncope in 19%) and were never diagnosed as being due to cardiac disease, and one study reported SCD was the first manifestation in 50% of cases. In contrast to our study, previous studies have not extensively studied records from GPs and hospitals. This could explain why we find a higher proportion of symptomatic patients in our study. Most importantly, previous studies are based on selected cases referred to tertiary cardiac centers and may not be representative for the general population. Finally, Danish death certificates have a large supplementary information field where cardiac symptoms are reported from interviews with relatives of the diseased and sometimes even with the GPs. Despite the comprehensive methodology using all available resources, only 22% of SADS had relevant contact with the healthcare system due to cardiac symptoms. This is somewhat lower than the rates reported by Wisten et al., who found that 46% of SCD patients had consulted a doctor before death.
We found that 29% of SADS patients had known psychiatric disorders, such as depression and schizophrenia. This is of particular interest, even though our study cannot explain the association between psychiatric diagnoses and SADS. This should be investigated in future studies—for instance, to determine if there is an association between the use of psychiatric medications and development of SADS. Previous studies have documented an increased risk of sudden death in psychiatric patients and there is an established increased risk of SCD among users of antipsychotic drugs.
There was a significant difference in previous psychiatric disorders between SADS with negative/positive toxicology and/or previous history of drug abuse. This may be explained by the overlap between the prevalence of drug abuse and psychiatric disorders. Since there were no significant differences in cardiac symptoms between SADS with negative/positive toxicology results, we believe that the inclusion of SADS patients with positive toxicology results into our study did not bias the results. Because 37% of SADS patients had a positive toxicology profile, it could be speculated that some drugs, even at therapeutic levels, led to a fatal arrhythmia (i.e., through the prolongation of the QT interval). However, it should be reemphasized that none of the 136 cases had a toxicology profile that could explain their death per the conclusions from forensic pathologists. All illegal drugs were found only in trace amounts and prescribed drugs in nonlethal concentrations.
We have previously investigated the prevalence of mutations in the 3 major genes involved in LQTS. In 44 of the 136 SADS cases a total of 5 carried a mutation in 1 of the 3 genes, corresponding to 11% of all investigated cases.
Syncope is quite common in the young population and is a major challenge for the physicians and may result from a wide variety of causes, ranging from benign conditions to life-threating diseases. Although our study is not going to change the recommended guidelines, the clinical impact of this study is that it raises awareness of symptoms prior to death in these SADS cases. Our findings emphasize the importance of detailed evaluations of patients, and in the future there should be focus on alarming symptoms such as: syncope during exertion, syncope in the lying position, family history of sudden death, and absence of external factors (such as drug abuse). To improve our knowledge of SADS, future studies should prospectively enroll SADS cases and collect standardized, detailed information regarding symptoms before death and medical history from family members, GPs and the hospitals.
This study has limitations inherent to any retrospective study. A prospective unselected analysis would be preferable for the registration of relevant data, but this is very difficult to achieve because of the low incidence of SADS in the general population. In this study the majority of SADS deaths were unwitnessed. Therefore, prodromal symptoms might be underestimated because individuals might have had cardiac symptoms before their sudden collapse. It was not possible for us to investigate the prodromal signs before the actual syncope such as dizziness and/or palpitations. It could be speculated a selection bias occurred with the control group, because we did not matched the control group with the SADS patients according to sex and age. There is also the potential for recall bias in both groups, which is a limitation in any case-control study where the exposure is ascertained after the event. Also, one could speculate that young persons in the control group who died in accidents did not consult a doctor compared with SADS cases or patients with structural heart disease. Data on family history of sudden death were lacking in the majority (90%) of patients, and therefore no conclusion can be made. Description of ECGs was based on records. All ECGs were not available, a total of 12 ECG were retrievable and none of the ECGs had a QTc-interval above 460 milliseconds or type I BrS pattern. Thus, ECGs have not in all cases been systematically examined by a cardiologist prior to death. Hence, we could not evaluate for LQTS or other primary arrhythmogenic disorders in the SADS population. One might speculate that nondetectable drugs may have caused death, but this is very unlikely as a standard toxicology examination is followed by a mass spectrometry investigation. The majority of patients in our study were Caucasians, and the conclusions for other demographic groups may be different. Therefore, we cannot generalize our findings to non-Caucasian individuals. Taking these limitations in consideration, we believe that our results are a comprehensive description of symptoms that occurred prior to SADS presentation.
Discussion
In this study, we examined a nationwide cohort of SADS patients (n = 136) for their course prior to death, including symptoms, medical history and contact with the healthcare system. Symptoms prior to death were compared with a control group of subjects in the same age range who died in traffic accidents (n = 74) during the studied time interval. To our knowledge, this study describes the largest cohort of young SADS patients from the general population.
Symptoms Prior to SADS
We found that significantly more patients with SADS had higher frequencies of cardiac symptoms prior to death than controls (P < 0.001), most commonly presyncope/syncope (17%). This could be due to underlying cardiac arrhythmias. Another important finding in our study is that a significant proportion of the patients had experienced an episode of seizures or had a diagnosis of epilepsy (18%). Linzer et al. suggested that the diagnosis of epilepsy in some cases may be wrong because primary cardiac arrhythmias can produce syncope and anoxic seizures. This highlights the fact that seizures can be difficult to distinguish from syncope induced by an arrhythmia and that patients with unexplained seizure activity should always have a cardiovascular examination, including ECG.
The literature is inconsistent regarding the notion that most SADS cases have symptoms. In contrast to our results, a recent study reported prior syncope was documented in only 4% of SADS patients, and up to 90% had no preceding symptoms or recognized risk factors before death. However, another study has shown that over half of the SADS patients had at least one antecedent cardiac symptom (including syncope in 19%) and were never diagnosed as being due to cardiac disease, and one study reported SCD was the first manifestation in 50% of cases. In contrast to our study, previous studies have not extensively studied records from GPs and hospitals. This could explain why we find a higher proportion of symptomatic patients in our study. Most importantly, previous studies are based on selected cases referred to tertiary cardiac centers and may not be representative for the general population. Finally, Danish death certificates have a large supplementary information field where cardiac symptoms are reported from interviews with relatives of the diseased and sometimes even with the GPs. Despite the comprehensive methodology using all available resources, only 22% of SADS had relevant contact with the healthcare system due to cardiac symptoms. This is somewhat lower than the rates reported by Wisten et al., who found that 46% of SCD patients had consulted a doctor before death.
Clinical Characteristics and Medical History
We found that 29% of SADS patients had known psychiatric disorders, such as depression and schizophrenia. This is of particular interest, even though our study cannot explain the association between psychiatric diagnoses and SADS. This should be investigated in future studies—for instance, to determine if there is an association between the use of psychiatric medications and development of SADS. Previous studies have documented an increased risk of sudden death in psychiatric patients and there is an established increased risk of SCD among users of antipsychotic drugs.
There was a significant difference in previous psychiatric disorders between SADS with negative/positive toxicology and/or previous history of drug abuse. This may be explained by the overlap between the prevalence of drug abuse and psychiatric disorders. Since there were no significant differences in cardiac symptoms between SADS with negative/positive toxicology results, we believe that the inclusion of SADS patients with positive toxicology results into our study did not bias the results. Because 37% of SADS patients had a positive toxicology profile, it could be speculated that some drugs, even at therapeutic levels, led to a fatal arrhythmia (i.e., through the prolongation of the QT interval). However, it should be reemphasized that none of the 136 cases had a toxicology profile that could explain their death per the conclusions from forensic pathologists. All illegal drugs were found only in trace amounts and prescribed drugs in nonlethal concentrations.
We have previously investigated the prevalence of mutations in the 3 major genes involved in LQTS. In 44 of the 136 SADS cases a total of 5 carried a mutation in 1 of the 3 genes, corresponding to 11% of all investigated cases.
Clinical Implications
Syncope is quite common in the young population and is a major challenge for the physicians and may result from a wide variety of causes, ranging from benign conditions to life-threating diseases. Although our study is not going to change the recommended guidelines, the clinical impact of this study is that it raises awareness of symptoms prior to death in these SADS cases. Our findings emphasize the importance of detailed evaluations of patients, and in the future there should be focus on alarming symptoms such as: syncope during exertion, syncope in the lying position, family history of sudden death, and absence of external factors (such as drug abuse). To improve our knowledge of SADS, future studies should prospectively enroll SADS cases and collect standardized, detailed information regarding symptoms before death and medical history from family members, GPs and the hospitals.
Strengths and Limitations
This study has limitations inherent to any retrospective study. A prospective unselected analysis would be preferable for the registration of relevant data, but this is very difficult to achieve because of the low incidence of SADS in the general population. In this study the majority of SADS deaths were unwitnessed. Therefore, prodromal symptoms might be underestimated because individuals might have had cardiac symptoms before their sudden collapse. It was not possible for us to investigate the prodromal signs before the actual syncope such as dizziness and/or palpitations. It could be speculated a selection bias occurred with the control group, because we did not matched the control group with the SADS patients according to sex and age. There is also the potential for recall bias in both groups, which is a limitation in any case-control study where the exposure is ascertained after the event. Also, one could speculate that young persons in the control group who died in accidents did not consult a doctor compared with SADS cases or patients with structural heart disease. Data on family history of sudden death were lacking in the majority (90%) of patients, and therefore no conclusion can be made. Description of ECGs was based on records. All ECGs were not available, a total of 12 ECG were retrievable and none of the ECGs had a QTc-interval above 460 milliseconds or type I BrS pattern. Thus, ECGs have not in all cases been systematically examined by a cardiologist prior to death. Hence, we could not evaluate for LQTS or other primary arrhythmogenic disorders in the SADS population. One might speculate that nondetectable drugs may have caused death, but this is very unlikely as a standard toxicology examination is followed by a mass spectrometry investigation. The majority of patients in our study were Caucasians, and the conclusions for other demographic groups may be different. Therefore, we cannot generalize our findings to non-Caucasian individuals. Taking these limitations in consideration, we believe that our results are a comprehensive description of symptoms that occurred prior to SADS presentation.