Objectives:

To present a comprehensive analysis of autopsy findings in 206 fetuses referred to our genetic center and to assess the clinical utility of fetal autopsy in reaching a final diagnosis, which is essential for counseling regarding the risk of recurrence. We also compared the autopsy findings with prenatal ultrasound findings to evaluate the potential benefit of fetal autopsy in fetuses terminated after prenatal diagnosis of malformations.

Study design:

Retrospective review of patient records in a tertiary referral genetic center in North India during 5-year period (April 2000–March 2005). This includes 206 fetuses, 138 terminated after detecting an anomaly in ultrasonogram and 68 spontaneous fetal losses. In all cases, fetal autopsy was carried out and complimented by radiography, karyotype wherever possible and histopathological examination wherever necessary. In fetuses with prenatally diagnosed malformations, ultrasound findings were compared with autopsy findings.

Results:

Fetal autopsy was able to provide a definite final diagnosis in 59% (122/206) cases. Fetal autopsy confirmed the ultrasound findings in all cases but two. Moreover, autopsy provided additional findings in 77 cases and of these, 24 cases had a significant change of recurrence risk.

Conclusion:

This study confirms the utility of fetal autopsy in identifying the cause of fetal loss, which will help in the genetic counseling of the couple. In cases with prenatally diagnosed anomalies, the new information from fetal autopsy changes the predicted probability of recurrence in 18% cases. Even though the prenatal ultrasonogram reasonably predicts the malformations, fetal autopsy gives significant additional malformations in one-third of the cases and is essential for genetic counseling.

Introduction

Congenital malformations remain a common cause of perinatal deaths and account for 25–30% in developed countries and 10–15% in developing countries like India.^{1, 2} In all, 3% of neonates have a major congenital malformation and 0.7% have multiple congenital malformations.³ One such mishap creates anxiety in the couple for the fear of similar recurrence in future pregnancies. The recurrence risk of these disorders varies from negligibly low to 25%, depending on the genetic component in the etiology of the disorder. Every effort should be made to identify the etiology of perinatal death so that appropriate genetic counseling can be given. There are only few studies on fetal autopsy, especially concerning genetic etiologies of fetal loss.⁴

Routine anomaly scan during antenatal period has become a part of obstetric care and the best time for fetal malformations scan is at around 18 weeks. Even though ultrasonogram can give fairly accurate diagnosis, examination of the terminated fetus for associated anomalies is essential to confirm the diagnosis and look for associated malformations. This is necessary because some associated malformations can be missed or are undetectable on ultrasound.

This can help to reach correct etiological diagnosis essential for genetic counseling. The objective of the present study was to review our 5-year experience to evaluate the diagnostic yield of fetal autopsy of stillbirths/fetal losses and the potential benefit of autopsy in fetuses terminated after prenatal diagnosis of malformations.

Materials and methods

Two hundred and six consecutive fetuses, over a period of 5 years (April 2000–March 2005), referred to our department for examination were included in this study. This includes 138 therapeutic abortions after the recognition of some congenital malformation on ultrasonographic examination or of genetic disorder on invasive prenatal diagnosis. We also examined 68 fetal losses, which included spontaneous abortions, intrauterine deaths and stillbirths. Each fetus was examined according to a pre-designed protocol. This included a photograph, whole body radiograph, and external and internal examination. Histopathological examination of the relevant tissue was carried out according to the requirement in individual cases. Placenta was available in only few cases. If available, histology of placenta was carried out in cases with unexplained fetal death with or without growth retardation. A chromosomal analysis was performed whenever fresh tissue was available. All post-mortem examinations were carried out with written consent of the father or relatives who brought the fetus.

In cases terminated after prenatal diagnosis of malformation, the ultrasound diagnosis and post-mortem diagnosis were compared to look for disagreement. As the fetuses were received from other hospitals also, ultrasonographic examination was performed at different centers by ultrasonologist with varying degrees of expertise. Additional findings at post-mortem examination were considered significant if those findings affected the final diagnosis and/or brought a change in the recurrence risk. The etiological diagnosis was analyzed to assess the utility of post-mortem examination.

Results

Fetal autopsy was able to provide a definite final diagnosis in 122 (59%) cases. In four cases, fetal pathology was detected prenatally by cytogenetic and molecular methods and this was confirmed after termination by fetal autopsy. This includes chromosomal abnormalities, trisomy 21, trisomy 18 and partial monosomy 13, and a case of osteopetrosis diagnosed by prenatal molecular methods. These cases are not included in the further analysis. We were not able to reach a final diagnosis in 84 cases, which include 54 cases without any malformations, 16 cases with multiple malformations not fitting into any syndrome and 14 cases of hydrops fetalis (Table 1).

Table 1 - Utility of fetal autopsy for identification of etiological cause of perinatal death.

Full table

In 135 cases with malformations, 81 (58.6%) had isolated malformation and 54 (41.4%) had multiple malformations (Tables 2 and 3). The common isolated malformations were anencephaly and other neural tube defects. Urinary tract malformations constituted 14 cases, which include bilateral renal agenesis, multicystic dysplasia and urorectal septal defects. The fetuses with multiple malformations were classified into various groups namely chromosomal disorders, single-gene malformation syndrome, skeletal dysplasias, sporadic malformation syndromes and multiple malformations unclassified.

Table 2 - Post-mortem diagnoses of single anomalies.

Full table

Table 3 - Post-mortem diagnoses of multiple malformations.

Full table

In spontaneously aborted fetuses (68), intra-uterine deaths and stillbirths, malformations were identified in 16 (23.5%) cases. Of these, three cases had multiple malformations specific to autosomal recessive single-gene disorders that are having a recurrence risk of 25% (Table 4). In 33 cases (49%), chromosomal analysis was possible and two abnormalities were detected (Table 5). Chromosomal abnormalities account for 5% of stillbirths and intrauterine deaths. In 52 fetuses, no abnormalities were identified and the cause of fetal demise remains unexplained.

Table 4 - Utility of fetal autopsy after intrauterine death/stillbirth/neonatal death.

Full table

Table 5 - Chromosomal abnormalities in spontaneous fetal loss.

Full table

In 81 cases with isolated malformation, 38 cases (46.9%) had only internal malformations. In fetuses with renal malformations, two cases with autosomal recessive polycystic disease and two cases of bilateral renal agenesis were confirmed by histopathological examination. In fetuses with skeletal dysplasia, the final diagnosis was given on the basis of skeletal radiological examination.

In the fetuses terminated after prenatal diagnosis of malformations, fetal autopsy confirmed the ultrasound findings in all cases but two. These two cases were diagnosed to have intrauterine growth retardation, oligohydraminos and absent kidney ultrasonographically; however, at autopsy, there was no abnormality in kidney, ureter or urinary bladder. One of these had co-arctation of aorta and ventriculomegaly, whereas other case had malrotation of the gut and penile hypospadias.

Fetal autopsy provided additional finding in 77 cases. In 24 of these 77 cases, revision of diagnosis led to significant change in the risk of recurrence in subsequent pregnancies (Figure 1).

Figure 1.

Summary chart of prenatal anomaly scan findings and correlation with fetal autopsy findings.

Full figure and legend (18K)

Infarcts in the placenta or evidence of infection were identified by histology of placenta in some cases, giving clues to the cause of fetal loss. However, as the data about placental histology were small, we have not analyzed these data.

Discussion

Fetal loss is a common clinical problem and the family seeks and deserves answers regarding the cause of the loss of a baby. The future reproductive decision of the couple depends on the cause of the fetal loss, which will predict the recurrence risk and may provide options for prevention of similar mishap.⁵

The investigations of fetal loss include radiograph, chromosomal analysis, fetal autopsy, investigations for infections and genetic metabolic causes, histopathology of placenta and other fetal tissues as indicated.⁶ Chromosomal analysis should be performed not only in the setting of obvious major malformations but also in fetal hydrops, intrauterine growth retardation, oligohydraminos sequence, macerated fetus and unexplained fetal loss. Skeletal radiograph is mandatory as a part of fetal autopsy protocol, especially if skeletal dysplasia is suspected because of short limbs.^{6, 7} Even though skeletal dysplasia can be suspected by ultrasonogram, accurate diagnosis is possible only with post-mortem radiograph. This is important as the recurrence risk varies from negligible to 25%. There were eight cases of short limb lethal skeletal dysplasia on ultrasonogram, all of them accurately diagnosed by radiological examination. There were five cases of spondylocostal dysplasia diagnosed by radiography; of these, four cases were diagnosed as neural tube defect in ultrasonogram. The risk of recurrence of neural tube defect is 5%, whereas that for spondylocostal dysplasia is 25%. Photographs are essential in documenting the presence or absence of any external malformations. There is also a concept of 'limited autopsy' with a photograph and radiograph of the fetus that will help for the diagnosis if other examination is not possible.⁸ Internal examination during fetal autopsy is also important for diagnosis, as it is evident from the present study that 46.9% cases had only internal malformations.

Histopathological examination of fetal organs is of limited usefulness in fetuses with malformations.⁶ An exception to this is the evaluation of fetal renal disease. Renal cystic diseases may be difficult to define on an ultrasound scan as oligohydraminos is usually associated. Moreover, the differentiation between infantile (autosomal recessive) polycystic renal disease (recurrence risk – 25%) and cystic renal dysplasia (recurrence risk-3%) is based on histopathology.^{6, 9} We had 17 cases with oligohydraminos sequence with cystic renal diseases on ultrasonogram. Of these, two cases were diagnosed as polycystic renal disease by histopathological examination. The remaining cases were cystic renal dysplasia due to urorectal malformations and multicystic renal diseases. Histopathology, especially of placenta, is of great importance in cases without malformations and may provide cause of death in fresh stillbirth or neonatal death. In our series, placenta was available only in a small number of cases and hence the corresponding data are not discussed.

Saller et al.¹⁰ have analyzed 124 cases of perinatal deaths to assess the clinical utility of autopsy and were able to find the cause in 75.8% of all perinatal deaths and 62.3% of fetal deaths. In our study, the cause was identified in 59% cases and is comparable to the group of fetal deaths of study by Saller et al. Our department being interested in genetic disorders, the cases with malformations and possible genetic disorders is referred more commonly. As neonatal deaths are not included in our series, the data represent fetal losses but not perinatal losses. Similarly, the autopsies were carried out mainly with the aim of identifying the genetic causes, especially malformations and malformation syndrome and a clinical geneticist conducted the autopsies. Our series includes fetuses terminated after antenatal detection of anomalies and spontaneous fetal losses. The incidence of chromosomal abnormalities in stillbirths and intrauterine deaths in our series is comparable to previous literature. In 84 (41%) cases, the definitive etiological diagnosis was not identified. This includes 54 cases without any definite malformations, 16 cases with multiple malformations not fitting into any specific diagnosis and 14 cases of unexplained hydrops fetalis. In total, 23.5% cases from spontaneous loss group had malformations and the risk of recurrence varied from very minimal to 25%. Documentation of malformation helps greatly in genetic counseling and prenatal diagnosis in next pregnancy. In cases with multiple malformations without any specific diagnosis, same malformations can be looked for in subsequent pregnancies by ultrasound evaluation. In this group without malformations, intrauterine death and stillbirths, fetal autopsy helps to rule out many malformation syndromes that may have high risk of recurrence. The cause of fetal loss in these cases without malformations may be maternal causes leading to chronic placental insufficiency. In these cases, the causes like immunological, fetal infections, maternal illnesses and factor V Leiden need to be looked for. Histopathological examination of the placenta can give useful information, but it was not performed in most of the cases owing to unavailability of placenta.

Causes of non-immune hydrops fetalis are numerous, including chromosomal anomalies, hematological disorders, metabolic disorders and cardiovascular disorders.¹¹ Structural cardiac anomalies, abnormalities of the rhythm and cardiomyopathies have been reported to account for about 50% of cases.¹² The other common causes of hydrops are chromosomal (5–33%) and infections (12–16%).^{13, 14} Congenital malformations, fetal akinesia syndromes and skeletal dysplasias are also reported with fetal hydrops. Genetic hematological conditions like -thalassemia, pyruvate kinase deficiency and glucose-6-phosphate dehydrogenase deficiency are also reported. -Thalassemia accounts for 28.2% of cases of fetal hydrops in southeast Asia.¹⁴ Our study had cases of Neu-Laxova syndrome and lethal pterygium syndrome, which had generalized edema and other features of hydrops, and one case of monosomy X presenting with cystic hygroma and hydrops. In the 14 cases with fetal hydrops, no cause was obvious. This group may have genetic, metabolic and infectious causes. Metabolic causes accounts for 5–15% cases of hydrops.^{11, 15} Better laboratory services are needed to identify these causes of fetal hydrops. The combination of examination of the fetus and placenta with the results of microbiological, cytogenetic and metabolic investigations provides an etiological diagnosis for non-immune fetal hydrops in 65–85% cases in various previous studies.^{11, 12} The low diagnostic yield in hydrops fetalis in this study is possibly due to incomplete investigative workup for infective and metabolic disorders.

In 135 fetuses with malformations, 81 had isolated malformation and 54 had multiple malformations. Central nervous system malformations were the most common indication for therapeutic abortions. The most frequent anomalies were neural tube defect and most of these cases were correctly identified prenatally. Genitourinary anomalies comprised 17% of the isolated malformations. This is comparable to other similar studies.¹⁶ Identification of single-gene autosomal recessive malformation syndrome is important as the recurrence risk in this situation is 25%.¹⁷ A fetus diagnosed to have neural tube defect (recurrence risk 3–5%), by ultrasonogram, on fetal autopsy had additional findings like polydactyly and renal malformation. The revised diagnosis of Meckel syndrome is autosomal recessive inheritance with recurrence risk of 25% in subsequent pregnancies. In this study, we have identified eight autosomal recessive disorders with malformations by fetal autopsy. Sporadic malformation syndromes like limb body wall complex and disruption syndromes like amniotic band syndrome have negligibly low recurrence risk. Thus, the accurate risk of recurrence based on fetal autopsy is very useful for the family.

The group of fetuses terminated after prenatal diagnosis of malformations is very important. In this study, post-mortem examination confirmed the ultrasound finding in 98% of cases, which is comparable to previous studies.¹⁸ With the present level of ultrasonography, the false-positive diagnosis is extremely rare. Poor visibility due to oligohydramnios or obesity is an important cause of errors in ultrasonographic diagnosis, as was seen in two cases of this series. However, ultrasonography may miss associated malformations and detection of associated malformations on autopsy may lead to refinement in etiological diagnosis. This was evident in our study. Post-mortem examination provided significant additional information in 38% of cases, and the change in recurrence risk in 18% of cases. This was comparable to previous similar studies.^{4, 9, 18} In a 10-year retrospective study of autopsy after termination of pregnancy for fetal anomaly, the autopsy confirmed the suspected diagnosis in 72% cases and autopsy added information that lead to the refinement of risk of recurrence in 27% cases.⁹ Hence, risk of recurrence based on ultrasonographic diagnosis of fetus may be erroneous in significant number of cases and autopsy of the fetus after termination of pregnancy is essential for genetic counseling.

Even though autopsy is the best method to detect the cause for perinatal death, there has been a decline in the autopsy rate recently.^{9, 19} The various options available for investigations in this situation was reviewed in a recent article.¹⁹ Post-mortem MR imaging has a useful role in providing structural information of the central nervous system in fetuses and stillbirth neonates.^{20, 21} Another promising alternative approach used in adult post-mortem investigation is the use of laparoscopic autopsy, which can be tried in perinatal autopsy.²¹ These options are relatively costly.

This retrospective study of 206 cases confirms the utility of fetal autopsy in identifying the cause of fetal loss, which will help in the genetic counseling of the couple. In this study, a clinical geneticist performed the autopsies. Hence, the data show more stress on the malformations and the genetic syndromes of malformations. This type of analysis in perinatal autopsy series is an useful documentation of relative prevalence of various malformations. This study also stresses the need of fetal autopsy in providing accurate genetic counseling. Autopsy facilities are not yet commonly available in India. There is a need to educate obstetricians about the need of fetal autopsy and placental histology for genetic counseling. Our data showed that fetal autopsy results in Indian scenario are similar to that reported in the literature. Pediatricians with interest in malformation can take up the responsibility of establishment of fetal autopsy facilities in collaboration with a pathologist. Better investigative facilities for chromosomal analysis, metabolic disorders and infections will definitely increase the diagnostic yield, especially in intrauterine death and fetal hydrops. When autopsy is not possible because of ethical and religious reasons, careful examination, a photograph and a radiograph may provide diagnostic information. The uptake of perinatal autopsy services depends on the awareness among obstetricians and pediatricians about its need and utility.

References

1. Rajasekhar S, Vishnu Bhat B, Veliath AJ, Ratnakar C. Perinatal autopsy – a seven year study. Indian J Pediatr 1996; 63: 511–516. | PubMed |
2. Joshi VV, Bhakoo ON, Gopalan S, Gupta AN. Primary cause of perinatal mortality – autopsy study of 134 causes. Indian J Med Res 1979; 69: 963–971. | PubMed | ISI | ChemPort |
3. Muller RF, Young ID. Emery's Elements of Medical Genetics. Churchill Livingstone: New York, 2001; 1–193.
4. Pahi J, Phadke SR, Halder A, Gupta A, Pandey R, Agarwal SS. Does autopsy of antenatally diagnosed malformed fetuses aid genetic counselling? Nat Med J India 1998; 11: 169–170. | ISI | ChemPort |
5. Phadke SR. Perinatal pathology – a prerequisite for genetic counselling. Asian J Pediatr Pract 2004; 8: 33–39.
6. Curry CJ. Pregnancy loss, stillbirth and neonatal death. A guide for the Pediatrician. Medical Genetics-I. Pediatr Clin North Am 1992; 39: 36–52.
7. Oslen OE, Espeland A, Maartmann-Moe H, Lachman RS, Rosendahl K. Diagnostic value of radiography in cases of perinatal death: a population based study. Arch Dis Child Fetal Neonatal Ed 2003; 88: F521–F524. | Article | PubMed |
8. Sharma AK, Phadke SR, Kamboj M, Agarwal SS. The clinical value of a limited fetal autopsy as illustrated by 2 cases. Aus NZ J Obstetr Gynaecol 1994; 34: 1–3.
9. Boyd PA, Tondi F, Hicks NR, Chamberlain PF. Autopsy after termination of pregnancy for fetal anomaly: retrospective cohort study. BMJ 2004; 328: 137–140. | Article | PubMed | ChemPort |
10. Saller ND, Lesser BK, Harrel U, Rogers BB, Oyer CE. The clinical utility of the perinatal autopsy. JAMA 1995; 273: 663–665. | Article | PubMed | ISI |
11. Steiner RD. Hydrops fetalis: role of the geneticist. Semin Perinatol 1995; 19: 516–524. | PubMed | ISI | ChemPort |
12. Rodriguez MM, Bruce JH, Jimenez XF, Romaguera RL, Bancalari E, Garcia OL et al. Nonimmune hydrops fetalis in the live born: series of 32 autopsies. Pediatr Dev Pathol 2005; 8: 369–378. | Article | PubMed | ISI |
13. Lallemand AV, Doco-Fenzy M, Galliard DA. Investigation of nonimmune hydrops fetalis: multidisciplinary studies are necessary for diagnosis – review of 94 cases. Pediatr Dev Pathol 1999; 2: 432–439. | Article | PubMed | ISI | ChemPort |
14. Suwanrath-Kengpol C, Kor-anantakul O, Suntharasaj T, Leetanaporn R. Etiology and outcome of non-immune hydrops fetalis in southern Thailand. Gynecol Obstet Invest 2005; 59: 134–137. | Article | PubMed | ISI |
15. Burin MG, Scholz AP, Gus R, Sanseverino MT, Fritsh A, Magalhaes JA et al. Investigation of lysosomal storage diseases in nonimmune hydrops fetalis. Prenat Diagn 2004; 24: 653–657. | Article | PubMed | ISI |
16. Kaiser L, Vizer M, Arany A, Vezpremi B. Correlation of prenatal clinical findings with those observed in fetal autopsies: pathological approach. Prenat Diagn 2000; 20: 970–975. | Article | PubMed | ISI | ChemPort |
17. Phadke SR, Pahi J, Phadke RV, Pradhan S, agarwal SS. Importance of etiological diagnosis of hydrocephalus as illustrated by a case of Walker Warburg syndrome. Indian Pediatr 1997; 34: 1037–1038. | PubMed | ChemPort |
18. Johns N, Al-salti W, Cox P, Kilby MD. A comparative study of prenatal ultrasound findings and post-mortem examination in a tertiary referral center. Prenat Diag 2004; 24: 339–346. | Article | ISI |
19. Wright C, Lee REJ. Investigating perinatal death: a review of the options when autopsy consent is refused. Arch Dis Child Fetal Neonatal Ed 2004; 89: F285–F288. | Article | PubMed | ChemPort |
20. Griffiths PD, Variend D, Evans M, Jones A, Wilkinson ID, Paley MN et al. Postmortem MR imaging of the fetal and stillborn central nervous system. Am J Neuroradiol 2003; 24: 22–27. | PubMed | ISI |
21. Griffiths PD, Paley MNJ, Whitby EH. Post-mortem MRI as an adjunct to fetal or neonatal autopsy. Lancet 2005; 365: 1271–1273. | Article | PubMed | ISI | ChemPort |