Background
The American Congress of Obstetricians and Gynecologists places special emphasis on autopsy as one of the most important tests for evaluation of stillbirth. Despite a recommendation of an autopsy, many families will decline the autopsy based on religious/cultural beliefs, fear of additional suffering for the child, or belief that no additional information will be obtained or of value. Further, many obstetric providers express a myriad of barriers limiting their recommendation for a perinatal autopsy despite their understanding of its value. Consequently, perinatal autopsy rates have been declining. Without the information provided by an autopsy, many women are left with unanswered questions regarding cause of death for their fetus and without clear management strategies to reduce the risk of stillbirth in future pregnancies. To avoid this scenario, it is imperative that clinicians are knowledgeable about the benefit of autopsy so they can provide clear information on its diagnostic utility and decrease potential barriers; in so doing the obstetrician can ensure that each family has the necessary information to make an informed decision.
Objective
We sought to quantify the contribution of placental pathologic examination and autopsy in identifying a cause of stillbirth and to identify how often clinical management is modified due to each result.
Study Design
This is a cohort study of all cases of stillbirth from 2009 through 2013 at a single tertiary care center. Records were reviewed in a stepwise manner: first the clinical history and laboratory results, then the placental pathologic evaluation, and finally the autopsy. At each step, a cause of death and the certainty of that etiology were coded. Clinical changes that would be recommended by information available at each step were also recorded.
Results
Among the 144 cases of stillbirth examined, 104 (72%) underwent autopsy and these cases constitute the cohort of study. The clinical and laboratory information alone identified a cause of death in 35 (24%). After placental pathologic examination, 88 (61%) cases had a probable cause of death identified. The addition of autopsy resulted in 78 (74%) cases having an identifiable probable cause of death. Placental examination alone changed clinical management in 52 (36%) cases. Autopsy led to additional clinical management changes in 6 (6%) cases.
Conclusion
This stepwise assessment of the benefit of both placental pathological examination and autopsy in changing probable cause of death beyond traditional clinical history and laboratory results emphasizes the need to implement more comprehensive evaluation of all stillbirths. With the aim of providing a cause of stillbirth to the parents, and to prevent future stillbirths, it behooves health care professionals to understand the value of this more comprehensive approach and convey that information to the bereaved parents.
Introduction
Stillbirth affects 6.1 of every thousand pregnancies in the United States annually. Complete obstetric management ideally includes review of the clinical history, laboratory assessment, examination of the placenta, genetic evaluation, and fetal autopsy. The information from these studies can be used to glean information regarding antenatal growth, fetal development, congenital anomalies, and to confirm or refute the clinical diagnoses. Information from a complete postmortem examination, in turn, informs probable cause of death and potential intervention strategies and management in subsequent pregnancies.
A detailed placental pathologic examination is one critical component of stillbirth evaluation given the placenta’s essential role in maintaining a healthy pregnancy. Indeed, prior studies have shown that placental examination, in addition to review of the clinical history and laboratory assessments, can identify a cause of death in 11-65% of cases. In addition to placental pathology, American Congress of Obstetricians and Gynecologists (ACOG) places special emphasis on autopsy as one of the most important tests for evaluation of stillbirth. Despite recommendation of an autopsy, many families will decline the autopsy based on religious/cultural beliefs, fear of additional suffering for the child, or belief that no additional information will be obtained or of value. Further, many obstetric providers express a myriad of barriers limiting their recommendation for a perinatal autopsy despite their understanding of its value. Consequently, perinatal autopsy rates have been declining. Without the information provided by an autopsy, many women are left with unanswered questions regarding cause of death for their fetus and without clear management strategies to reduce the risk of stillbirth in future pregnancies. To avoid this scenario, it is imperative that clinicians are knowledgeable about the benefit of autopsy so they can provide clear information on its diagnostic utility and decrease potential barriers; in so doing the obstetrician can ensure that each family has the necessary information to make an informed decision.
Accordingly, the objectives of this study are to use a stepwise design to quantify the specific contributions of placental pathology and autopsy in identifying a cause of death and to identify how often clinical management for subsequent pregnancies should change due to each of these results.
Materials and Methods
General study design
This is a cohort study of all cases of stillbirth with delivery at ≥23 weeks at Northwestern University. To study a population of true intrauterine fetal demises and avoid misclassification of terminations of pregnancy, stillbirths <23 weeks’ gestation or intrapartum stillbirths (ie, extreme prematurity without planned neonatal resuscitation) were excluded. Cases were retrospectively collected from January 2009 through September 2011 by query of a comprehensive institutional database of placental pathologic exams. Thereafter cases were prospectively collected until August 31, 2013. Pregnancies were dated by their primary obstetrician using ACOG-accepted methods. The clinical diagnosis of stillbirth was made by ultrasonographic evidence of asystole, confirmed by 2 providers.
The clinical approach to stillbirth employed by all physicians during the study time period included the following: (1) a detailed history and physical exam of the mother along with a gross neonatal physical exam by the obstetrician at delivery; (2) laboratory testing via a standard order set in the electronic medical chart including maternal serum evaluation for infections (cytomegalovirus, parvovirus, toxoplasma, and syphilis), acquired thrombophilia (ie, antiphospholipid antibodies), and fetomaternal hemorrhage; (3) a recommendation for chromosomal analysis of the stillborn (standard karyotype at the time of the study); (4) placental pathologic examination; and (5) a recommendation for autopsy followed by a formal consent process emphasizing the potential benefits of the knowledge gained.
Placental pathologic examination
All patients underwent a detailed, systematic gross and histologic placental pathologic examination by a single attending perinatal pathologist (L.M.E.). This examination included recording of the trimmed placental weight, membrane insertion, dimensions of the placental disc, and insertion, length, diameter, and coiling pattern of the umbilical cord. Histologic samples included sections of membranes, umbilical cord, and 2-5 sections of the placental parenchyma. The histologic data were recorded and divided into 3 major pathologic categories: maternal vascular underperfusion, fetal vascular obstruction/fetal thrombotic vasculopathy, and evidence of amnionic fluid infection, using criteria defined by Redline et al. Per hospital protocol, the final placental pathologic report is completed prior to completion of the final autopsy report.
Autopsy examination
Components of the complete autopsy performed are described in Table 1 . Autopsy was performed at no charge to the family by an expert perinatal pathologist (L.M.E.).
| Autopsy consent |
| External gross examination |
| Photographs |
| Radiographs |
| Body measurements |
| Document maceration |
| Overall appearance/maturation/identify external anomalies |
| Obtain skin sample for cytogenetics, if needed |
| Internal gross examination |
| Y-incision |
| Routine cultures: blood, spleen, lung |
| In situ examination: abdomen |
| Note presence of peritoneal fluid |
| Umbilical vessels and bladder: note position, size, any abnormalities |
| Large bowel: fixation of cecum and appendix, fixation of left colon |
| Small bowel: ligament of Treitz, position and length of mesenteric root |
| Stomach: position, note abnormalities |
| Pancreas: tail should extend to spleen, rule out annular pancreas |
| Spleen: note position, rule out polysplenia or asplenia |
| Liver and gall bladder: position and shape |
| Genitalia: note position, any abnormalities |
| Kidney and ureters: note position, cystic change, dilation of ureters |
| Diaphragm: check for intactness, level of domes |
| In situ examination: neck and thorax |
| Thymus: position, shape, and size |
| Pericardial sac: fluid, look for defects |
| Pleural cavities: fluid, lobation of lungs |
| Heart and great vessels |
| Removal of organ block and dissection of individual organs |
| Removal of brain |
| All organs weighed and any anomalies recorded |
| Sampling of tissues for histologic examination |
| Recording of gross findings |
| Provisional anatomic diagnoses |
| Released to clinical team within 2 working days of autopsy dissection |
| Review of histologic slides and recording of histologic findings |
| Brain gross and histologic examination after fixation |
| Review of all ancillary studies (genetic, metabolic, microbiology) |
| Final autopsy report including final anatomic diagnoses, summary of clinical history, gross description, microscopic description, and clinicopathologic correlation released to clinical team within 60 working days of autopsy dissection. |
Stepwise analysis
Demographic characteristics, the clinical history and physical examination, and the clinically suspected cause of death were compared between women who agreed to autopsy and those who declined. For all women, records were reviewed by a maternal-fetal medicine subspecialist (E.S.M.) in a stepwise manner. Step 1: the clinical history, delivery notes, and laboratory assessments were reviewed, blinded to placental pathology and autopsy data. Step 2: placental pathologic report was reviewed, in consultation with the perinatal pathologist as needed, and in concert with the clinical data, but blinded to the autopsy data. Step 3: If applicable, the autopsy information was reviewed and that information was compiled with results from the clinical data, laboratory assessments, and placental pathologic examination ( Figure 1 ).
At each step of these stillbirth examinations, the cause of death was ascribed and the certainty of that etiology coded according to the initial causes of fetal death (INCODE) method. The INCODE system divides etiologies of stillbirth into 7 major categories: maternal medical conditions during pregnancy, obstetric complications, maternal or fetal hematologic conditions, fetal abnormalities, infectious etiologies, placental pathologic conditions, or other. Conditions listed as possible cause of death were those not believed to be a direct cause of the stillbirth, but possibly involved in the pathophysiologic sequence that led to death. Examples of “possible cause of death” include uncontrolled maternal seizures, inherited thrombophilia with a small-for-gestational-age fetus, fetal muscular dystrophy, inflammation limited to placenta, or velamentous cord insertion with no evidence of occlusion. Conditions listed as probable cause of death had a high likelihood of directly causing the fetal death. Examples of “probable cause of death” include included cholestasis of pregnancy, massive fetomaternal hemorrhage, fetal hydrops, aneuploidy, microbiologic, molecular or pathognomonic histologic evidence of fetal infection, or evidence of severe maternal or fetal vascular compromise in the placenta. The certainty of the ascribed cause of death (ie, unknown, possible, or probable) was compared between each stage of the evaluation (ie, clinical, placental pathology, and autopsy).
Clinical management recommendations for a future pregnancy as a result of the collected information were determined by a maternal-fetal medicine subspecialist (E.S.M.) and were examined sequentially. Examples of clinical management strategies to be employed in a future pregnancy included obtaining additional laboratory information (ie, neonatal alloimmune thrombocytopenia testing), genetic testing, a more detailed anatomic survey, serial growth ultrasounds, or nonstress tests. Clinical management changes informed by placental pathology alone and after autopsy were examined separately.
Statistical analysis
Statistical analyses were performed using software (Stata, Version 13.1; StataCorp, College Station, TX). Student t , χ 2 , Fisher exact, and analysis of variance tests were used, as appropriate. A P value of .05 represented statistical significance. All tests were 2-tailed.
Institutional review board approval
Approval for this study was obtained from the Northwestern University Institutional Review Board with a waiver of informed consent.
Materials and Methods
General study design
This is a cohort study of all cases of stillbirth with delivery at ≥23 weeks at Northwestern University. To study a population of true intrauterine fetal demises and avoid misclassification of terminations of pregnancy, stillbirths <23 weeks’ gestation or intrapartum stillbirths (ie, extreme prematurity without planned neonatal resuscitation) were excluded. Cases were retrospectively collected from January 2009 through September 2011 by query of a comprehensive institutional database of placental pathologic exams. Thereafter cases were prospectively collected until August 31, 2013. Pregnancies were dated by their primary obstetrician using ACOG-accepted methods. The clinical diagnosis of stillbirth was made by ultrasonographic evidence of asystole, confirmed by 2 providers.
The clinical approach to stillbirth employed by all physicians during the study time period included the following: (1) a detailed history and physical exam of the mother along with a gross neonatal physical exam by the obstetrician at delivery; (2) laboratory testing via a standard order set in the electronic medical chart including maternal serum evaluation for infections (cytomegalovirus, parvovirus, toxoplasma, and syphilis), acquired thrombophilia (ie, antiphospholipid antibodies), and fetomaternal hemorrhage; (3) a recommendation for chromosomal analysis of the stillborn (standard karyotype at the time of the study); (4) placental pathologic examination; and (5) a recommendation for autopsy followed by a formal consent process emphasizing the potential benefits of the knowledge gained.
Placental pathologic examination
All patients underwent a detailed, systematic gross and histologic placental pathologic examination by a single attending perinatal pathologist (L.M.E.). This examination included recording of the trimmed placental weight, membrane insertion, dimensions of the placental disc, and insertion, length, diameter, and coiling pattern of the umbilical cord. Histologic samples included sections of membranes, umbilical cord, and 2-5 sections of the placental parenchyma. The histologic data were recorded and divided into 3 major pathologic categories: maternal vascular underperfusion, fetal vascular obstruction/fetal thrombotic vasculopathy, and evidence of amnionic fluid infection, using criteria defined by Redline et al. Per hospital protocol, the final placental pathologic report is completed prior to completion of the final autopsy report.
Autopsy examination
Components of the complete autopsy performed are described in Table 1 . Autopsy was performed at no charge to the family by an expert perinatal pathologist (L.M.E.).
