David J. Driscoll

Sudden Unexpected Death in Children and Adolescents

To determine the incidence and clinicopathologic spectrum of sudden unexpected death, we reviewed the death certificate of all residents of Olmsted County, Minnesota who were between 1 and 22 years of age when they died during the period from January 1950 to October 1982. Of 515 death certificates reviewed, 12 (2.3%) recorded sudden unexpected death, resulting in an incidence of 1.3 per 100,000 patient-years. The subjects ranged in age from 3 to 20 years (median 13); 8 of the 12 were male. Of the 12 deaths, 4 were definitely cardiac-related and 3 were probably cardiac-related. In the five other cases, the cause of death was unknown. Three of the 12 subjects had a history of syncope; 2 of the 3 had syncope associated with exercise, and both died while exercising. The relative rarity of sudden unexpected death in children and adolescents probably precludes population screening techniques to identify subjects at risk. However, a subset of subjects with 1) exercise-associated syncope, 2) nonvasodepressor syncope, 3) a family history of sudden unexpected death, or 4) a family history of hypertrophic cardiomyopathy deserves extensive and thorough evaluation.

Late follow-up of 1095 patients undergoing operation for complex congenital heart disease utilizing pulmonary ventricle to pulmonary artery conduits

BACKGROUNDnPulmonary ventricle (PV) to pulmonary artery (PA) conduits have made possible the correction of many complex congenital cardiac anomalies.nnnMETHODSnBetween April 1964 and January 2001, 1270 patients underwent operation with conduit placement from the PV to PA. The present study evaluates late outcome of 1095 patients (612 males, 483 females) having an operation before July 1992. Mean age was 9.6 +/- 8.2 years old. Diagnoses included pulmonary atresia/tetralogy of Fallot (459), transposition of the great arteries (TGA) (232), truncus arteriosus (193), double outlet right ventricle (DORV) (121), corrected TGA (49), septated univentricular heart (36), and other (5). A porcine-valved Dacron conduit was used in 730, homograft in 239, and non-valved conduit in 126.nnnRESULTSnEarly mortality decreased from 23.5% prior to 1980 to 3.7% for the most recent decade. Mean follow-up was 10.9 years (maximum, 29 years). Actuarial survival for early survivors at 10 and 20 years was 77.0% +/- 1.5% and 59.5% +/- 2.6%. On univariate analysis, clinical and hemodynamic factors associated with late mortality were male gender, older age at operation, higher post-repair PV/systemic ventricle (SV) pressure ratio, higher distal PA pressure, and longer bypass time (p < or = 0.01 for all). On multivariate analysis, independent risk factors for late mortality were male gender, older age at operation, diagnosis of TGA, corrected TGA, truncus, or univentricular heart, and PV/SV pressure ratio > or = 0.72 (p < or = 0.03 for all). Freedom from reoperation for conduit failure at 10 and 20 years was 55.5% +/- 2.0% and 31.9% +/- 2.7%. On multivariate analysis, independent risk factors for conduit failure were homograft conduit, diagnosis of TGA, younger age at operation, and smaller conduit size (p < or = 0.007 for all). Reoperation for one conduit replacement was performed in 306 patients, two conduit replacements in 55 patients, three in 6 patients, and four in 3 patients. Overall early mortality for conduit replacement in this series was 4.9%; it was 1.7% for patients operated on from 1989 through 1992. At follow-up, 84% of survivors were in NYHA class I or II.nnnCONCLUSIONSnOperations that include conduit placement and replacement can be performed with low early mortality. Younger age at operation was associated with improved late survival. The diagnosis of TGA was associated with increased risk for conduit failure, and the durability of the homograft, in this series, was inferior to the porcine-valved Dacron conduit. Quality of life was excellent for most patients despite the need for reoperation.

Myocardial ischemia in patients with pulmonary atresia and intact ventricular septum

Children who die after operation for pulmonary atresia and intact ventricular septum may have myocardial ischemia. The relation between histologic evidence of myocardial ischemic injury and the presence of a right ventricle to coronary artery fistula, coronary artery dysplasia and operation in 17 autopsy specimens was assessed. Age at death ranged from 1 day to 16 years (median, 11 days). Of the 17 hearts, 6 (35%) had right ventricle to coronary artery fistulas, 5 of which had coronary artery dysplasia. In three cases, there was segmental or complete absence of a coronary artery. Ischemia was present in four of these six hearts, two of which had right ventricular outflow reconstruction. Six of the 11 hearts without right ventricle to coronary artery fistulas also had myocardial ischemia. Of these six cases, four had right ventricular outflow reconstruction and two had shunt operations. Death occurred from 1 to 8 days (mean 3) after operation. Hearts with pulmonary atresia and intact ventricular septum may have myocardial ischemia with or without either right ventricle to coronary artery fistulas or coronary artery dysplasia. Myocardial ischemia may occur after right ventricular outflow reconstruction or shunt operations. Thus, myocardial ischemia occurs commonly in patients with pulmonary atresia and intact ventricular septum and is not always related to coronary abnormalities or operation.

Influence of ventricular morphology on outcome after the Fontan procedure

The modified Fontan procedure has gained wide acceptance in the treatment of various congenital heart defects. Determination of risk factors for mortality remains an important issue for optimizing patient selection for the Fontan procedure. Conflicting results have been reported about whether ventricular morphology is a risk factor in these patients. Survival free of Fontan takedown or cardiac transplantation was assessed in the first 500 patients undergoing the Fontan procedure at our institution. This survival was correlated with ventricular morphology as evaluated by angiography. Both multivariate and univariate analyses indicated ventricular morphology was predictive of early survival free of Fontan takedown or cardiac transplantation following the procedure. However, there was no statistical evidence for ventricular morphology being a risk factor for mortality in patients alive 6 months after the procedure. Ventricular morphology is a risk factor for early survival in patients undergoing a Fontan procedure, with left ventricular morphology associated with a better early survival than right ventricular morphology.

Surgical repair of univentricular heart (double inlet left ventricle) with obstructed anterior subaortic outlet chamber

The results of operation in all patients with univentricular heart and an obstructed anterior subaortic outlet chamber who were operated on utilizing extracorporeal circulation at the Mayo Clinic from 1973 through 1983 were reviewed. Ten of the 18 patients died during the immediate postoperative period and there was one late death. Factors significantly related to operative and immediate postoperative mortality were age at operation, cardiothoracic ratio on X-ray examination, degree of ST depression on electrocardiogram and pressure gradient across the outlet foramen at catheterization. Autopsy in eight cases revealed significant hypertrophy of ventricular myocardium and a small outlet foramen that was considered stenotic relative to either body surface area or aortic root area. The ventricular myocardium showed histologic changes of chronic ischemia that predated the surgical procedure.

Safety and utility of endomyocardial biopsy in infants, children and adolescents: A review of 66 procedures in 53 patients

The benefits and risks of endomyocardial biopsy in infants, children and adolescents were determined by reviewing the indications for and complications and results of 66 procedures in 53 patients aged 2 months to 20 years. One patient had a pneumothorax, and three had a right ventricular perforation. Ventricular tachycardia developed in four patients; it was treated with lidocaine in three and was self-limited in one. The procedure was unsuccessful in two patients. Among 25 patients with a prebiopsy diagnosis of idiopathic dilated cardiomyopathy, microscopic features were consistent with cardiomyopathy in 24 (96%) and were normal in 1. Of nine patients with clinically suspected myocarditis, only two (22%) had microscopic evidence of inflammation, and seven had chronic nonspecific features suggestive of dilated cardiomyopathy. Of eight patients with unexplained arrhythmias, six (75%) had microscopic findings compatible with dilated cardiomyopathy and two had myocarditis. Biopsy tissue samples from seven patients with nondilated forms of cardiomyopathy (four hypertrophic, three restrictive) were consistent with the clinical diagnosis in six and were inadequate in one. Cardiac biopsies were also performed in four patients with other disorders. Among the 51 patients with adequate biopsy specimens, microscopic features were considered diagnostic in 5, confirmatory in 44 and not helpful in 2 with normal tissue. The results indicate that endomyocardial biopsy is safe in infants, children and adolescents. It is useful for the evaluation of cardiomyopathy and specific secondary forms of myocardial disease. There seems to be little correlation, however, between clinical and tissue diagnoses of myocarditis.

Functional single ventricle: Cardiorespiratory response to exercise

The cardiorespiratory response to exercise was measured in 27 children with functional single ventricle. All 27 patients had a significant reduction in exercise time, work performed, maximal exercise heart rate, maximal oxygen uptake and systemic arterial blood oxygen saturation. The reduction in exercise performance increased with increasing age of the patients. All patients ventilated excessively at rest and during exercise. This study documented the precise level of exercise intolerance in patients with functional single ventricle. The progressive deterioration in exercise performance with increasing patient age may lend credence to the concept that operation for physiologic correction of functional single ventricle should be considered during or before adolescence.

Left ventricular function in the Marfan syndrome without significant valvular regurgitation

Assessment of left ventricular (LV) size and function in many patients with the Marfan syndrome is confounded by the presence of aortic or mitral regurgitation. Therefore, we sought to determine LV size and function in a cohort of patients with the Marfan syndrome who had only mild or no aortic or mitral valve regurgitation. ••• Of the 128 patients in the Mayo Clinic Marfan database who met the Gent diagnostic criteria for the Marfan syndrome, 1 36 (23 males) had no or only mild aortic and mitral valve regurgitation noted on the first complete echocardiogram recorded at our institution. These 36 patients formed the study cohort. M-mode measurements derived from the echocardiogram for these patients included LV end-diastolic dimension (EDD), LV endsystolic dimension (ESD), and ejection fraction. Dimensions were indexed to age and body surface area to allow comparison with previously published normal values. 2,3 These measurements were recorded from the first and the most recent echocardiogram to determine any changes in LV dimensions and function that occurred over time. Associations between LVEDD and mitral valve prolapse or aortic root dilation were evaluated to determine if patients with mitral valve prolapse or aortic root dilation had higher LVEDD than those without these syndromes. The predicted normal value and the corresponding range for LVEDD and LVESD were calculated for each patient according to their age and body surface area using regression equations 2,3 previously developed for that purpose. The observed dimensions then were scaled relative to the predicted normal value, and reported as the percentage of the corresponding mean predicted dimension. The resulting values that were reported correspond to the observed dimension divided by the predicted normal value 100. The 95% confidence intervals of the averages of these rescaled dimensions were calculated using the formula: mean tn1, 0.025 SE, where tn1, 0.025 is the 2-sided critical value obtained from the t distribution with n1 degrees of freedom. For each dimension, the number of patients having an observed dimension beyond the predicted normal range was tabulated. Exact 95% confidence intervals were calculated for the percentage of patients with extreme observed dimensions using the binomial distribution. Exact binomial tests were used to compare the percentage of extreme values to the 5% extreme values that would be expected in a normal population. The association between LVEDD and mitral valve prolapse or aortic root dilation was assessed using a 2-sample t test. To compare the dimensions on the first and the last echocardiogram, values were scaled relative to the predicted normal range, given the age and body surface area, using the formula: scaled value (observed predicted normal for the given age and body surface area)/(half-width of normal range). These scaled values provide a measure of how extreme the observed dimension is relative to the entire normal predicted range. The scaled values on the first and last echocardiograms were compared using paired t tests. A 2-sided p value 0.05 was considered statistically signi ficant.

Cor triatriatum dexter: Two-dimensional echocardiography diagnosis

Cor triatriatum dexter is a malformation resulting from lack of normal regression of the embryonic right valve of the sinus venosus. In this situation, the right atrium is divided by a membrane into two chambers. Two-dimensional echocardiography was used in the antemortem diagnosis of this rare cardiac anomaly in a neonate. Associated cardiac lesions were also documented. The patient died, and findings were verified at autopsy.

Circulating Heart Autoantibodies in Familial as Compared With Nonfamilial Idiopathic Dilated Cardiomyopathy

BACKGROUNDnIdiopathic dilated cardiomyopathy (DCM) is a serious heart disease characterized by enlargement of one or both ventricles and ventricular dysfunction. Although most patients have sporadic disease, 20% have been found to have familial DCM when relatives are investigated by echocardiography. No other factors have been identified to date that consistently distinguish familial from nonfamilial DCM. Although some patients have a family history of DCM, a negative family history does not exclude familial DCM because affected family members may be presymptomatic or undiagnosed. Because some patients have life-threatening complications at the time of initial assessment of DCM, identifying a serum marker predictive of familial disease would help determine which families would most likely benefit from echocardiographic investigation.nnnOBJECTIVEnIn this study, our objective was to determine whether antiheart autoantibodies could be used to distinguish familial from nonfamilial idiopathic DCM.nnnMETHODSnWe analyzed serum specimens for antiheart antibodies from 19 patients categorized as having familial DCM and 15 classified as having nonfamilial DCM on the basis of echocardiographic investigation of first-degree relatives. The mean duration of disease in these 34 patients was 50 months at the time the serum specimens were obtained.nnnRESULTSnTiters of antibodies against the adenine nucleotide translocator, branched-chain keto acid dehydrogenase, and cardiac myosin did not distinguish between familial and nonfamilial cases of DCM.