To identify trends in neonatal autopsy rates at a tertiary care hospital during the last decade, to determine the concordance between antemortem and postmortem diagnoses, and to describe patient characteristics that influenced diagnostic yield on autopsy.
Retrospective review of medical records.
Tertiary care children's hospital.
All neonatal deaths between January 1, 1984, and December 31, 1993.
Main Outcome Measures
For those with autopsies, data included length of stay, diagnostic imaging studies, antemortem diagnoses, and autopsy findings. Autopsy diagnoses were compared with antemortem findings and classified according to concordance.
Of 487 neonatal deaths, autopsies were performed in 296 (61%). The autopsy rate declined significantly during 10 years from 71.2% (1984-1988) to 47.7% (1989-1993). New diagnoses were made at autopsy in 44% of cases. Major discordancies were identified in 35 infants (12%); minor discordancies in 95 (32%). Autopsies were more likely to reveal new diagnoses in infants born at 28 to 36 weeks' gestation and in those whose mothers had no prenatal care. Major findings at autopsy were more likely in infants whose mothers had no prenatal care and in infants who died within 6 hours of birth. There was no relationship between new findings at autopsy and age, length of hospital stay, or antemortem imaging studies.
This study reveals a significant decline in neonatal autopsies during a 10-year period. This study also demonstrates that neonatal autopsy continues to provide clinically significant data and remains a valuable tool in perinatal medicine.
AUTOPSY HAS contributed to medical care since the advent of modern medicine. Whether by establishing the cause of death, assisting in determining the manner of death, comparing the antemortem and postmortem findings, producing vital statistics, or monitoring public health, the value of autopsy has been well documented.1,2 In addition, autopsy has been a useful tool in clinical education, research, and quality monitoring of patient care. Especially in newborns, in whom clinical manifestations of many diseases may often be nonspecific, autopsy can be invaluable in determining the precise cause of death. In perinatal medicine, despite improvements in prenatal diagnosis, autopsy remains important in the confirmation and further delineation of prenatal diagnosis.3 By defining the cause of death, autopsy can reduce parental concerns and guilt over prenatal events such as maternal illness or maternal use of medications and can provide insight into the genetic implications of subsequent pregnancies.
Despite the recognized value of autopsy, general autopsy rates have declined significantly in recent years. There has been a steady decline in adult autopsy rates in the United States, from 50% in 1950 to 11.5% in 1992.4 Specifically with respect to neonatal autopsies, limited information is available concerning trends in autopsy rates or diagnostic yield. The purpose of this study was to evaluate neonatal autopsies during a 10-year period at a single tertiary care institution. The specific aims of this study were to identify trends in neonatal autopsy rates at a tertiary care hospital during the last decade, determine the concordance between antemortem and postmortem diagnoses, and describe patient characteristics, if any, that influenced diagnostic yield on autopsy.
This study was conducted at Lutheran General Children's Hospital, Park Ridge, Ill, following approval by the institutional review board. Lutheran General Children's Hospital is a tertiary care hospital with free-standing pediatric residency and neonatal/perinatal medicine fellowship programs. It has a 36-bed, level III neonatal intensive care unit (NICU) that is equipped to care for critically ill newborns using high-frequency ventilation, extracorporeal membrane oxygenation, and open-heart surgery for infants with complex congenital heart disease. The NICU is staffed by 5 full-time neonatologists and 6 fellows in neonatal-perinatal medicine. The unit is also supported by more than 40 pediatric medical and surgical subspecialists. The obstetric unit at Lutheran General Hospital performs more than 5000 deliveries per year.
The present study included all deaths in the NICU between January 1, 1984, and December 31, 1993. For all cases, demographic data were collected through the hospital's computer database and included patient age, sex, race, and insurance status. For those patients with autopsies, medical records and autopsy reports were reviewed and additional data were collected on length of stay, antemortem imaging studies, antemortem diagnoses, and autopsy findings. Antemortem imaging studies included ultrasound, echocardiography, and computed tomographic scans.
All autopsies were performed after obtaining informed consent from parents. Autopsy examinations were done at no charge to families, and included review of clinical records, anthropometric measurements, external gross examination, examination of the brain and organs of the thoracic, abdominal, and pelvic cavities, and recorded weight and microscopic examination of organs. Aerobic and anaerobic bacterial cultures and viral cultures were obtained selectively when indicated by history or findings.
Antemortem diagnoses included all diagnoses listed by the physician on the death certificate before autopsy and all diagnoses in the medical record that either had been established before death or appeared sufficiently likely for specific treatment to have been instituted before death. According to the criteria of Goldman et al,1 diagnoses were classified as either major or minor. A major diagnosis was defined as the basic, underlying disease that was the primary cause of death. The primary cause of death referred to the process that was the direct cause of death, and excluded terminal events (eg, cardiac arrest). Minor diagnoses were defined as conditions that may or may not have contributed to death, and may or may not have been related to the primary disease process.
Autopsy diagnoses, including histologic and microbiologic results, were obtained from the final autopsy report. Autopsy reports and medical records were reviewed by the first author (P.K.). Subspecialists were consulted as needed to verify the significance of autopsy findings and to assess their effect on clinical outcomes. Based on the classification scheme of Goldman et al,1 autopsy diagnoses were categorized as either major or minor, and were assigned to 1 of 5 classes used to reflect the concordance between antemortem and postmortem findings (Table 1). All cases with class I and II findings, and a representative sample of patients with class III, IV, and V findings were also reviewed by 2 of the other authors (J.T., H.H.M.). There was complete agreement about the classification of diagnoses among the 3 reviewers.
Statistical analyses were conducted using the Statistical Package for the Social Sciences (SPSS Inc, Chicago, Ill).5 For categorical variables, the proportion of events were compared using the χ2 analysis with Yates correction. Changes in autopsy rates and class V findings during the study period were evaluated with statistical process control charts using the Statistical Analysis System (SAS Institute Inc, Cary, NC).6P<.05 was considered significant.
There were a total of 487 neonatal deaths during the 10-year study period. Autopsies were performed on 296 patients (60.8%). The number of deaths and autopsies for each study year is presented in Table 2. There was a statistically significant decline in autopsy rates during the study period (Figure 1). The autopsy rate for the first 5 study years was 71.2%, compared with 47.7% during the last 5 years (P<.001). Comparing again the first and latter halves of the study period, there were no significant differences in either the number of antemortem studies (72% vs 80%) or infant age at death (52% vs 59% of infants aged ≤24 hours; 48% vs 41% of infants aged >24 hours).
Changes in neonatal deaths, autopsies, class V diagnoses, and antemortem imaging studies.
New information (classes I-IV) was obtained at autopsy in 44% of the infants. In the remaining 56% of the cases, there was complete concordance between antemortem and postmortem diagnoses. Table 2 provides the number of patients in each class of autopsy findings for each study year. The incidence of new major diagnoses (classes I and II) at autopsy remained the same (12%) during both the first and the second halves of the study.
During the 10-year study period, only 1 class I finding was identified. This female infant was born full term by emergency cesarean section owing to fetal distress. Apgar scores were 1, 1, and 0 at 1, 5, and 10 minutes, respectively. Multiple attempts to intubate the infant were unsuccessful and there was no response to bagging with a face mask and 100% oxygen. Autopsy findings revealed complete obliteration of laryngeal lumen at the level of the cricoid cartilage by firm cartilaginous tissue. Esophageal atresia with a tracheoesophageal fistula at the level of the carina was also noted.
Class II findings were observed at autopsy in a total of 34 patients (11%) and included congenital malformation (n = 21), infection (n = 8), and vascular accidents (n = 5). Class II findings could be further classified into subclass IIa (n = 3), IIb (n = 17), IIc (n = 12), and IId (n = 2). Seventy patients (24%) exhibited class III findings at autopsy and 25 patients (8%) evidenced class IV findings. Vascular accidents (intracranial and other internal hemorrhage) and congenital malformations constituted the majority of class III and IV findings.
In both preterm and full-term infants with autopsies, major antemortem diagnoses (principal underlying diseases and primary cause of death) most commonly involved disorders of the respiratory system (60% and 33%, respectively). As expected, specific etiopathologic diagnoses were different between preterm and full-term infants. In preterm infants, the most common major antemortem diagnoses were hyaline membrane disease (36%), congenital malformations (28%), and infections (11%). In full-term infants, congenital malformations were the leading major antemortem diagnoses (50%), followed by hypoxic-ischemic encephalopathy (12%) and aspiration syndrome (9%). In preterm infants, new diagnoses at autopsy most commonly involved the central nervous system (47%), followed by the respiratory system (18%). In full-term infants, new diagnoses at autopsy were more evenly distributed across organ systems, with the cardiovascular system being most commonly involved (22%).
Table 3 depicts the types of discordancies found at autopsy by patient characteristics. There were no statistically significant associations between new findings (classes I-IV) at autopsy and maternal age, infant sex, age at death, or antemortem imaging studies. However, new findings at autopsy were associated with gestational age (P = .05) and prenatal care (P<.001). Autopsies were most likely to reveal unexpected diagnoses in neonates born between 28 and 36 weeks' gestation and in those whose mothers had no prenatal care.
When data were analyzed to evaluate which variables were associated with major findings at autopsy (classes I and II vs classes III, IV, and V), prenatal care and age at death were statistically significant (P = .001 and P<.001, respectively). Major findings at autopsy were more likely in infants whose mothers had no prenatal care and in infants who died within 6 hours of birth. Twenty (22%) of 90 infants who died within 6 hours of birth had major findings at autopsy compared with 8 (10.6%) of 75 infants who died between 6 and 24 hours of birth and 7 (5.3%) of 131 infants who died after 24 hours.
The overall neonatal autopsy rate in the present study was 60.8%. The autopsy rate for the first 5 years of the study (1984-1988; 71.2%) was significantly higher than the rate noted during the last 5 years (1989-1993; 47.7%), and there was a statistically significant decline in neonatal autopsies during the 10-year study period. Other studies from comparable time periods have documented similar autopsy rates to those in the present study.7- 10 In a 1989 multicenter study, Landers and MacPherson7 found a 48.4% autopsy rate in 33 US perinatal centers. More recently, Dhar et al8 reported a neonatal autopsy rate of 62% from a regional NICU in Canada, and a study from a tertiary care center in Australia reported a perinatal autopsy rate of 59%.9 Studies done in the late 1970s and early 1980s reported neonatal autopsy rates ranging between 63% and 81%.10,11
Whether there has been a general downward trend in neonatal autopsy rates in the past decades is unclear. Because there is considerable variability in autopsy rates based on individual institutional characteristics,7,12 it is difficult to compare rates from different institutions and from different periods. To the best of our knowledge, this is the first study to evaluate neonatal autopsy rates at one tertiary care center during a 10-year period. During the study period, there were no significant changes in the administrative policies of either the hospital or the departments of pediatrics or pathology, or in the availability of various ancillary services, to explain the decline noted in our study. Unfortunately, because of the retrospective nature of our study, we were unable to evaluate family characteristics that may have influenced autopsy rates. We are continuing to monitor neonatal autopsy rates to evaluate whether the observed downward trend in neonatal autopsies continues.
Despite downward trends in autopsy rates, adult studies have documented that autopsies continue to provide valuable information.1 One of the most important benefits of autopsy is the ability to confirm or modify antemortem diagnoses. Our study confirms the valuable role of neonatal autopsy. In 44% of the autopsied infants, new information was obtained at autopsy. This observation is consistent with earlier reports on the yield of neonatal and perinatal autopsies. Recently, Dhar et al8 reported a total discordance rate of 58% in a Canadian study of 338 neonatal autopsies; as many as 18% of the cases had major discordancies (classes I and II). Earlier, Saller et al13 reported that after autopsy, a diagnosis was changed or added in 34% of 47 neonatal deaths. Tasdelen et al14 reported a total discordance rate of 55% between antemortem and postmortem findings in 301 neonates with autopsies. In 32.2%, major discrepancies were noted; in 23%, minor defects were diagnosed on autopsy. Meier et al15 examined 172 perinatal deaths and found that autopsy helped to establish the cause of death in 26% of 139 perinatal cases and provided a cause of death in 31% of 87 neonatal and early infant deaths. These studies and our own observations demonstrate that neonatal autopsy continues to provide significant new information.
To maximize the benefits of autopsy, attempts have been made to identify those cases in which autopsies would yield the most value. Studies in adults have tried to use patient antemortem characteristics to predict the yield of information at autopsy, but have been largely unsuccessful.16,17 Previous research on neonatal autopsy has not evaluated the association between patient characteristics and yield at autopsy. In the present study, prenatal care, gestational age, and infant age at death were all associated with the diagnostic yield at autopsy. Autopsies were most likely to reveal new diagnoses in cases where mothers had no prenatal care and in infants born between 28 and 36 weeks' gestational age. Major findings at autopsy (classes I and II) were more likely in infants whose mothers had no prenatal care and in infants who died within 6 hours of birth. The use of antemortem imaging studies did not influence the yield at autopsy. It is likely that the increased yield at autopsy in infants whose mothers had no prenatal care or in infants with a short admission-to-death interval was associated with a diminished opportunity to establish clinical diagnoses in these cases. Further studies should be conducted to confirm these findings and to explore relationships between infant characteristics and diagnostic yield.
The results of our study reveal that neonatal autopsy continues to provide additional information in a significant proportion (>40%) of neonatal deaths. Therefore, the observed decline in neonatal autopsies is cause for concern and requires coordinated efforts to reverse this trend. Efforts should be made to increase neonatal autopsy rates to enhance understanding of the cause of death, educate families on possible increased risks associated with future pregnancies, and monitor the quality of patient care.
Editor's Note: Finding new diagnoses at autopsy is not new, but the declining rate of autopsies underlines that we need to be reminded of their importance.—Catherine D. DeAngelis, MD
Accepted for publication June 18, 1999.
Presented in part as a poster at the Midwest Society for Pediatric Research Meeting, Chicago, Ill, September 19, 1996, and the Annual American Academy of Pediatrics Meeting, Boston, Mass, October 26, 1996.
We acknowledge the assistance of Sandy Maki in the preparation of the manuscript.
Corresponding author: Denise B. Angst, DNSc, Lutheran General Children's Hospital, 1775 Dempster St, Box 296, Park Ridge, IL 60068 (e-mail: email@example.com).
Kumar P, Angst DB, Taxy J, Mangurten HH. Neonatal AutopsiesA 10-Year Experience. Arch Pediatr Adolesc Med. 2000;154(1):38-42. doi:10-1001/pubs.Pediatr Adolesc Med.-ISSN-1072-4710-154-1-poa9041