(Circulation. 2000;102:649.)
© 2000 American Heart Association, Inc.
Clinical Investigation and Reports |
From the Jesse E. Edwards Registry of Cardiovascular Disease, United Hospital, St Paul, Minn, and the University of Minnesota, Minneapolis.
Correspondence to Sumeet S. Chugh, MD, Cardiology UHN-62, Oregon Health Sciences University, 3181 SW Sam Jackson Park Rd, Portland, OR 97201. E-mail chughs{at}ohsu.edu
| Abstract |
|---|
|
|
|---|
Methods and ResultsWe identified and reexamined structurally normal hearts from a 13-year series of archived hearts of patients who had sudden cardiac death. Subsequently, for each patient with a structurally normal heart, a detailed review of the circumstances of death as well as clinical history was performed. Of 270 archived SCD hearts identified, 190 were male and 80 female (mean age 42 years); 256 (95%) had evidence of structural abnormalities and 14 (5%) were structurally normal. In the group with structurally normal hearts (mean age 35 years), SCD was the first manifestation of disease in 7 (50%) of the 14 cases. In 6 cases, substances were identified in serum at postmortem examination without evidence of drug overdose; 2 of these chemicals have known associations with SCD. On analysis of ECGs, preexcitation was found in 2 cases. Comorbid conditions identified were seizure disorder and obesity (2 cases each). In 6 cases, there were no identifiable conditions associated with SCD.
ConclusionsIn 50% of cases of SCD with structurally normal hearts, sudden death was the first manifestation of disease. An approach combining archived heart examinations with detailed review of the clinical history was effective in elucidating potential SCD mechanisms in 57% of cases.
Key Words: death, sudden pathology fibrillation
| Introduction |
|---|
|
|
|---|
Establishing structural normalcy is an essential prerequisite to making a diagnosis of this entity. However, focal manifestations of conditions such as myocarditis, cardiomyopathies, or small tumors may escape detection in survivors of SCD.8 9 10 Furthermore, nonspecific abnormalities such as interstitial fibrosis and mild myxomatous mitral valve changes are difficult diagnoses to make in the cardiac arrest survivor. The gold standard for confirming absence or presence of a structural abnormality is the pathological examination of the patients heart. The confirmation of structural normalcy at autopsy is thus the most suitable means to identify SCD patients with normal hearts.
The combination of a triggering event and a susceptible myocardium has evolved as a biological model for the initiation of lethal arrhythmia.1 Accordingly, a search was conducted for possible abnormal myocardial substrates and triggers of fatal arrhythmia in patients with normal hearts who died of SCD. A detailed review of anatomic and clinical findings was performed in SCD cases with normal hearts, identified from a 13-year autopsy series of 270 SCD patients.
| Methods |
|---|
|
|
|---|
Archived Hearts
The Jesse E. Edwards Cardiovascular Registry (St
Paul, Minn) has accessioned >14 000 archived hearts in the last 40
years. Data collected at time of original examination include a police
report, family correspondence, medical examiner report, autopsy
findings, toxicological screen, and available medical history; these
data and results of the morphological studies in the registry are
catalogued in a standardized fashion.
Referral Sources
For SCD, the major sources of referral are local medical
examiners (coroners) from Hennepin, Ramsey, and Anoka counties, which
constitute the MinneapolisSt Paul greater metropolitan area. A
significant number of local cases of sudden death at age <60 years are
examined by the county coroners. The coroners, in turn, refer all cases
attributable to cardiac causes to the Edwards Registry in a
consistent fashion.
Experimental Design
All locally referred cases of SCD, age
20 years, during a
period of 13 years (1984 to 1996) were studied. These constituted 81%
(270 cases) of the 333 cases classified as SCD in the Registry during
this period. All these cases were reviewed to distinguish structurally
abnormal hearts (group A) from structurally normal hearts (group B)
(Figure 1
). Structurally abnormal hearts
were further classified as subgroup A1 if specific pathological
findings were present or subgroup A2 if only nonspecific findings
were identified. Detailed reexamination of all hearts reported to be
structurally normal (group B) was done. In addition to repeat
pathological examination, this included consideration of clinical and
morphological data in the registry files and detailed review of the
clinical histories of the patients by analysis of all available
community medical records for each patient.
|
Detailed Method of Pathological Examination
For the duration of the 13 years of this study, all 270 SCD
hearts were examined by the same cardiovascular
pathologist (J.L.T.) in a standardized fashion. Specimens were weighed,
with normal heart weight criteria, based on body mass index. After
external analysis to ascertain size and shape of cardiac
chambers, the coronary arteries were cut into 4- to 5-mm
sections and the cross-sectional diameter measured. To ensure exclusion
of coronary artery disease in morphologically normal hearts,
the criterion for pathological diagnosis of significant
coronary artery disease was stenosis of
50%
cross-sectional diameter in at
1 major epicardial coronary
artery.11 Subsequently, the heart was excised open in a
conventional line of flow. If no gross abnormality was identified,
routine sampling included standard, full-thickness sections from the
posterobasal, mid-anterior, lateral, and septum regions, including
portions of relevant cardiac valves. From each block, hematoxylin and
eosinstained, trichrome-stained, and elastic van Giesonstained
slides were prepared and examined. In selected cases, when no other
pathological abnormality was identifiable, a conventional study of the
cardiac conduction system with sections at 4- to 5-mm intervals was
also performed. Nonspecific cardiac findings were defined as myocardial
hypertrophy (increased heart weight for body surface area
and/or >1.6-mm thickness of compact myocardium of the left
ventricle), nonspecific interstitial fibrosis, and mitral
valve prolapse. The association between mitral valve prolapse and
sudden death has been controversial, and the increase in risk of sudden
death with this condition is probably very small.12 13 14
For the purpose of the present study, when pathological criteria
for lone mitral valve prolapse were met, hearts were considered to have
a nonspecific structural abnormality.15
| Results |
|---|
|
|
|---|
|
Structurally Abnormal Hearts
Group A comprised 256 hearts (95% of total), with 180 (67% of
total) and 76 (28% of total) hearts in subgroups A1 and A2. In
subgroup A1, coronary artery disease was the most common
finding (65% of cases), followed by congenital conditions in 14%
(Figure 3
). The latter group comprised 11
patients with anomalous coronary arteries and 15 patients with
other congenital cardiac conditions (aortic valve malformations, 5;
corrected transposition, 2; atrial septal defect, 1; pulmonary
atresia, 1; cleft tricuspid/mitral valve, 2). The incidence of
myocarditis was 11%; arrhythmogenic right ventricular
dysplasia and hypertrophic cardiomyopathy were
present in
4% each; other abnormalities occurred in
2% cases
(Figure 3
). Findings in 3 cases were relatively unusual: 1
patient died of a ruptured sinus of Valsalva aneurysm, 1 had
mycotic aneurysm with left ventricular rupture, and
1 had coronary arteritis.
|
In subgroup A2 hearts, a nonspecific abnormality was present, but a definite pathological diagnosis could not be made. These nonspecific abnormalities included left ventricular hypertrophy (left ventricular wall thickness of compact myocardium >1.6 cm) found in two thirds of the cases (50 of 76 hearts), pathological criteria for mitral valve prolapse in approximately one third (28 of 76 hearts), and nonspecific interstitial fibrosis in the absence of discrete postinfarction scars identified in nearly one third of hearts (22 of 76 hearts) in this subgroup.
In the hearts with nonspecific fibrosis, the mean age of patients was identical to the average age for the entire series (42±12 versus 42±14 years). Fibrosis/left ventricular hypertrophy and fibrosis/mitral valve prolapse coexisted in 11 of 22 and 8 of 22 hearts, respectively. A special conduction system examination was performed in 9 of these hearts, and fibrosis was found to extend to the atrioventricular node, His bundle, or either bundle branch in 8 patients. Comorbid conditions included seizure disorder (1, same patient had systemic lupus erythematosus), obesity (2, 1 patient had sleep apnea syndrome), noncardiac sarcoidosis (1), and emphysema (1). Only 2 patients had evidence of mild coronary disease (40% stenosis). Two patients were known to have manifested prior arrhythmia (ventricular tachycardia in 1 and supraventricular tachycardia 1).
Structurally Normal Hearts
Group B contained 14 structurally normal hearts (5% of total).
Among these cases, in addition to existing registry records, past
medical records were available from 13 patients; 1 subject did not
appear to have had any healthcare visits. ECGs were available in 6 of
the 14. The mean age was 35±9 years (median age 33 years), and 10 of
the 14 patients were women. A significant noncardiac abnormality was
present in only 1 of the 14 cases; the patient had micronodular
cirrhosis. Repeat cardiac examination (gross morphological and
microscopic study) did not yield additional cardiac abnormalities.
Details of Clinical History
From review of the 13 medical records
(Table
), sudden death was the
first manifestation of disease in 7 cases. Six patients had a history
of prior symptoms, of which 2 had syncope, 3 had a history of
palpitations, and 1 had chest pain. Two patients had a family history
of sudden death. Obesity was a comorbid condition in 2 patients, and 2
patients had a history of a seizure disorder.
|
The circumstances of death varied. One subject (patient 2,
Table
) had been exercising vigorously, having just completed his
first jog of the spring. A 37-year-old man (patient 4, Table
>,
with autopsy findings of micronodular cirrhosis) was consuming alcohol
at a bar when he collapsed. One person was bathing, and 4 were
sedentary. Four subjects died in their sleep.
Chemical analysis of serum collected at the time of autopsy did
not show evidence of drug overdose in any of the cases; in 9, the
findings were negative. In the remaining 5 cases, however, serum tested
positive for some compound. One subject had nontoxic levels of a
cocaine metabolite (benzoylecgonine) in the blood and another had
therapeutic levels of haloperidol (Table
). The 2 patients with
seizure disorder were taking valproic acid and phenobarbital. A
37-year-old woman of Southeast Asian origin, who had been previously
healthy, had an unidentified peak on mass spectrometry of the serum.
With the exception of the patient with findings of cocaine in the
serum, no other patients had a history of drug abuse.
ECG Analysis
Of the 6 ECGs available, 3 were normal, without evidence of QT
prolongation. The other 3 were abnormal; patients 1 and 3
(Table
) had evidence of preexcitation on the ECG. An ECG from
patient 12 (Table
), taken 6 months before death, showed atrial
fibrillation with a rapid ventricular response and narrow
QRS complexes.
| Discussion |
|---|
|
|
|---|
For the patients with specific structural abnormalities, our findings are similar to published reports. In older adults (age>45 years), 80% to 90% of sudden cardiac death patients have significant coronary artery disease.16 17 Coronary artery disease is also associated with 58% to 70% of sudden cardiac deaths in the young adult population.18 19 20 In our study (mean age 42±14 years), 65% of SCD patients with specific pathological findings had significant coronary artery disease. Findings of congenital anomalies, myocarditis, arrhythmogenic right ventricular dysplasia, and hypertrophic cardiomyopathy in our study are consistent with earlier studies in the young adult population.5 21 22 23 24 25
In 5% of cases we studied, no structural abnormalities were identified after repeated pathological examinations. In an earlier 30-year, population-based study of young adults from the same region as the cases we studied, a potential cause of SCD could not be identified in 13% cases20 ; in that study, however, nonspecific findings such as left ventricular hypertrophy were not included as structural abnormalities.
The significant frequency (30%) of nonspecific structural
abnormalities in SCD patients, in the absence of other cardiac
abnormalities, has not been reported previously. Heart weight (indexed
by body surface area) is not increased in patients of comparable age to
the present study who die of causes unrelated to the
heart.26 From clinical trials, left
ventricular hypertrophy has been established as
an independent predictor of overall mortality.27 28 In
addition, in patients with left ventricular
hypertrophy, the risk of ventricular
arrhythmia is increased
2-fold.29 Observations
in animal models suggest that cardiac myocytes from hypertrophied
hearts develop abnormalities of repolarization, which could predispose
to fatal arrhythmogenesis.30 Similarly, it is conceivable
that nonspecific interstitial fibrosis could contribute to
the initiation of ventricular arrhythmias through a
functional reentrant mechanism.31 The presence of fibrosis
in some component of the conduction system in 8 of 9 hearts subjected
to a detailed conduction system examination suggests a propensity for
bradycardia in some of these patients. The nature of factors
responsible for the presence of interstitial fibrosis in
this group remains uncertain. As the mean age of patients with
nonspecific fibrosis was identical to the average age for the entire
series (42±12 versus 42±14 years), this is unlikely to be related to
the process of aging. We are not aware of autopsy series examining the
prevalence of nonspecific interstitial fibrosis in this age
range in the general population.
Possible Causes of SCD in Patients With Structurally Normal
Hearts
In 2 cases, Wolff-Parkinson-White (WPW) syndrome may have been the
abnormal substrate for SCD. Long QT syndrome cannot be excluded for
cases in which ECGs were not available; 2 of the patients had a family
history of sudden death. In addition, this syndrome may often not be
recognized on a single ECG. Rare and newly described conditions such as
the Brugada syndrome (sudden cardiac death with right bundle-branch
block and ST elevation on ECG)32 were not identified.
Subject 2 (Table
) was a 30-year-old man of Southeast Asian
origin who may have died of sudden nocturnal death
syndrome.33 Inability to define a cause in cases of SCD
may have placed several subjects in the category of idiopathic
ventricular fibrillation. In several cases, drugs may have
been triggers for SCD of susceptible individuals.
In this study, the incidence of WPW syndrome may have been underestimated. There was evidence of preexcitation on 2 of the 6 ECGs that were available for review. Both of these patients had a history of palpitations and documented tachycardia before death; supraventricular tachycardia is a risk factor for the development of ventricular fibrillation in the WPW syndrome.34
Both obesity and epilepsy were comorbid conditions in SCD patients with structurally normal hearts. The annual sudden cardiac mortality rate is reportedly increased 40-fold in morbidly obese subjects35 when compared with a matched nonobese population. Sympathovagal imbalance caused by parasympathetic withdrawal in obese subjects has been implicated in the enhanced sudden death risk in obesity.36 In a recent population-based study of young adults with epilepsy, the rate of sudden unexpected death was 24-fold higher than the general population.37 Mechanisms remain unresolved, but precipitation of fatal arrhythmia caused by autonomic dysfunction has been postulated as a cause of sudden unexplained death in epilepsy.38 39
The electrophysiological effects of cocaine have been examined in tissue preparations, animal models, and human subjects. The drug has class Itype activity, with significant effects on myocardial refractoriness.40 In addition, cocaine can have a proarrhythmic effect similar to that induced by quinidine as the result of triggered activity from early afterdepolarizations associated with a prolonged QT interval. Thus, cocaine ingestion could induce ventricular arrhythmia independent of its effects on coronary arteries41 and in the absence of toxic levels. Several antipsychotic agents, including phenothiazines, have been associated with sudden death.42 43 Increased susceptibility to polymorphic ventricular arrhythmia by haloperidol may be related to block of outward potassium currents in the cardiac myocyte with resultant prolongation of the QT interval.44 A recent study indicated that in some families, the long QT syndrome may have a very low penetrance, with family members being prone to development of torsade de pointes when exposed to cardiac or noncardiac drugs that block potassium channels, without manifestation of a long QT interval on the ECG.45 In the present study, possible contributions of drugs to development of fatal arrhythmia cannot be ruled out in at least 2 patients despite the absence of toxic levels in the serum.
Prevention of sudden death in subjects with apparently normal hearts is a major challenge because mechanisms are not well understood. Similar to coronary artery disease,1 SCD was the first manifestation of disease in 50% of patients. Identification of individuals at risk for SCD who do not have recognized structural cardiac abnormalities requires a search for substrate abnormalities, which may depend on molecular analysis, including genetic typing.46 47 48 For the investigation of SCD, confirmation of structural normalcy of the heart at the time of autopsy likely should be combined with targeted, prospective molecular analysis of tissue as the next step in identifying causes in this group of patients.
Possible Limitations of Study
Despite careful, methodical pathological examination of cardiac
structure, atrioventricular accessory connections can
be overlooked. In addition, the identification of such muscular
connections may not be confirmation for the existence of a clinically
relevant accessory electrical connection between the atria and
ventricles.49 50 In our study, detailed
histological examination of the conduction system with
serial histological sections cut at 4.5-µm intervals
was done in 6 of the 14 hearts; no significant abnormalities were
found. Plasma or tissue analyses for molecular defects such as
genetic testing for the long QT syndrome were not performed.
The referred cases were nearly all <60 years of age. For the purpose of uniformity, only locally referred cases were included, the large majority consisting of patients examined by coroners in the 3 surrounding counties. As the coroners may not have examined all local cases of sudden death, the incidence of sudden cardiac death may have been underestimated. However, there was a consistent referral pattern, ie, cases of sudden death attributable to cardiac causes and age <60 years. In addition, the sex distribution would argue against a sex bias because approximately one third of the entire group were women.
Because the major goal of our study was a detailed clinical correlation in structurally normal hearts, complete clinical correlation for hearts with nonspecific fibrosis is not available. However, we have provided all the information present in the registry records. ECGs were available in only 6 of the 14 group B cases. This is probably explained by the relatively young age of these patients as well as SCD being the first manifestation of illness in 50% patients in this group.
Conclusions
The vast majority of SCD patients had structurally abnormal
hearts, but in as many as 30%, abnormalities were nonspecific. For
patients with structurally normal hearts, the autopsyclinical history
combination uncovered potential mechanisms of SCD in 57% cases.
| Acknowledgments |
|---|
Received December 15, 1999; revision received March 1, 2000; accepted March 10, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
N. Hofman, H. L. Tan, S.-A. Clur, M. Alders, I. M. van Langen, and A. A. M. Wilde Contribution of Inherited Heart Disease to Sudden Cardiac Death in Childhood Pediatrics, October 1, 2007; 120(4): e967 - e973. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. D. Lewis, C. B. Holmes, G. Holmvang, and J. R. Butterton Case 8-2007 -- A 48-Year-Old Man with Chest Pain Followed by Cardiac Arrest N. Engl. J. Med., March 15, 2007; 356(11): 1153 - 1162. [Full Text] [PDF] |
||||
![]() |
N. Abdelmawla and A. J. Mitchell Sudden cardiac death and antipsychotics. Part 1: Risk factors and mechanisms Advan. Psychiatr. Treat., January 1, 2006; 12(1): 35 - 44. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. D. Lane, C. Laukes, F. I. Marcus, M. A. Chesney, L. Sechrest, K. Gear, C. L. Fort, S. G. Priori, P. J. Schwartz, and A. Steptoe Psychological Stress Preceding Idiopathic Ventricular Fibrillation Psychosom Med, May 1, 2005; 67(3): 359 - 365. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. S. Chugh, J. Jui, K. Gunson, E. C. Stecker, B. T. John, B. Thompson, N. Ilias, C. Vickers, V. Dogra, M. Daya, et al. Current burden of sudden cardiac death: Multiple source surveillance versus retrospective death certificate-based review in a large U.S. community J. Am. Coll. Cardiol., September 15, 2004; 44(6): 1268 - 1275. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. S. Chugh, O. Senashova, A. Watts, P. T. Tran, Z. Zhou, Q. Gong, J. L. Titus, and S. J. Hayflick Postmortem molecular screening in unexplained sudden death J. Am. Coll. Cardiol., May 5, 2004; 43(9): 1625 - 1629. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. F. D. Wever and E. O. Robles de Medina Sudden death in patients without structural heart disease J. Am. Coll. Cardiol., April 7, 2004; 43(7): 1137 - 1144. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. E. Arking, S. S. Chugh, A. Chakravarti, and P. M. Spooner Genomics in Sudden Cardiac Death Circ. Res., April 2, 2004; 94(6): 712 - 723. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. M. Spooner and D. P. Zipes Sudden Death Predictors: An Inflammatory Association Circulation, June 4, 2002; 105(22): 2574 - 2576. [Full Text] [PDF] |
||||
![]() |
B. Surawicz Brugada syndrome: manifest, concealed, "asymptomatic," suspected and simulated J. Am. Coll. Cardiol., September 1, 2001; 38(3): 775 - 777. [Full Text] [PDF] |
||||
![]() |
R. J Myerburg and P. M Spooner Opportunities for sudden death prevention: Directions for new clinical and basic research Cardiovasc Res, May 1, 2001; 50(2): 177 - 185. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Corrado, C. Basso, and G. Thiene Sudden cardiac death in young people with apparently normal heart Cardiovasc Res, May 1, 2001; 50(2): 399 - 408. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. O. Cheng, S. S. Chugh, K. L. Kelly, and J. L. Titus Sudden Cardiac Death in Mitral Valve Prolapse Response Circulation, April 24, 2001; 103 (16): e88 - e88. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2000 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |