Objective
The objective of the study was to evaluate the intra- and interobserver agreement among obstetric experts in court regarding the retrospective review of abnormal fetal heart rate tracings and obstetrical management of patients with abnormal fetal heart rate during labor.
Study Design
A total of 22 French obstetric experts in court reviewed 30 cases of term deliveries of singleton pregnancies diagnosed with at least 1 hour of abnormal fetal heart rate, including 10 cases with adverse neonatal outcome. The experts reviewed all cases twice within a 3-month interval, with the first review being blinded to neonatal outcome. For each case reviewed, the experts were provided with the obstetric data and copies of the complete fetal heart rate recording and the partogram. The experts were asked to classify the abnormal fetal heart rate tracing and to express whether they agreed with the obstetrical management performed. When they disagreed, the experts were asked whether they concluded that an error had been made and whether they considered the obstetrical management as the cause of cerebral palsy in children if any.
Results
Compared with blinded review, the experts were significantly more likely to agree with the obstetric management performed ( P < .001) and with the mode of delivery ( P < .001) when informed about the neonatal outcome and were less likely to conclude that an error had been made ( P < .001) or to establish a link with potential cerebral palsy ( P = .003). The experts’ intraobserver agreement for the review of abnormal fetal heart rate tracing and obstetrical management were both mediocre (kappa = 0.46–0.51 and kappa = 0.48–0.53, respectively). The interobserver agreement for the review of abnormal fetal heart rate tracing was low and was not improved by knowledge of the neonatal outcome (kappa = 0.11–0.18). The interobserver agreement for the interpretation of obstetrical management was also low (kappa = 0.08–0.19) but appeared to be improved by knowledge of the neonatal outcome (kappa = 0.15–0.32).
Conclusion
The intra- and interobserver agreement among obstetric experts in court for the review of abnormal fetal heart rate tracing and the appropriateness of obstetrical care is poor, suggesting a lack of objectivity of obstetrical expertise as currently performed in court.
In recent decades, the number of medicolegal claims against obstetricians has significantly increased, making obstetrics a medical specialty that is associated with the one of the highest medicolegal risks. Currently most obstetricians in the United States have faced a suit at some point in their career. In the United Kingdom, maternity services have been estimated to account for 60–70% of the $630 million paid every year for litigation to the National Health Service Litigation Authority from all the medical specialties but comprise only 26% of the workload. In France, obstetrics has been estimated to represent 10% of indemnities resulting from legal decisions for a total amount of $29 million.
Fear of litigation has widespread massive implications for obstetricians and institutions as well as insurance companies, and obstetricians now pay some of the highest premiums for malpractice insurance.
In the United States, 76% of obstetricians reported having already had a claim and for most of them after having allegedly caused cerebral palsy. In Western countries, the wide use of intrapartum fetal heart rate monitoring has helped reduce neonatal mortality at the cost of an increasing number of cesarean deliveries but without decreasing the incidence of cerebral palsy, which still involves 2.5 per 1000 births.
It is, however, noticeable that the wide majority of obstetrical claims are based on the presumption of malpractice and substandard care during labor related to the misinterpretation of intrapartum abnormal fetal heart rate tracing (AFHRT). However, the etiology of the cerebral palsy is known to be multifactorial; only 4–14.5% of these cases are estimated to have an intrapartum origin. Although the existing diagnostic criteria for intrapartum asphyxia and different guidelines have been published regarding the analysis and interpretation of AFHRT, there is still no consensus about the optimal AFHRT management. Additionally, the analysis of fetal heart rate tracing has been previously demonstrated to have inadequate intra- and interobserver variability, particularly in cases of AFHRT.
In any medical malpractice claim, despite crude variations in legal procedures worldwide, the jury usually refers to the opinion of a medical expert who has to provide a comprehensive retrospective review of the case, assuring for the best objective decision possible. In France, experts in court are independent experienced physicians registered on a national list of medical experts who are occasionally appointed by the judge because of their expertise in the field of the claim and knowledge of legal procedures.
In particular, obstetric experts in court (OECs) often have to determine whether a child’s disability resulted from substandard care during labor and whether different management and decision making could have avoided it. To do so, OECs are usually provided with the entire obstetrical file, including the complete copy of the fetal heart rate tracing, the partogram, and all data regarding the neonatal outcome. Retrospective analysis of AFHRT therefore plays a major role in malpractice claims, in which an adverse outcome is usually the rule.
However, in addition to the inadequate intra- and interobserver variability, knowledge of adverse neonatal outcome was previously shown to alter clinicians’ interpretation of fetal heart rate tracing and to lead to a more severe classification of the intrapartum AFHRT. Major concerns regarding the validity of OEC reviews as currently performed in obstetrical malpractice claims are therefore legitimate because many questions remain unanswered.
The aim this study was to evaluate the intra- and interobserver agreement among OECs when retrospectively reviewing AFHRT and obstetrical management of patients with AFHRT during labor. The secondary aim of our study was to determine whether knowledge of the neonatal outcome affects OEC judgment.
Materials and Methods
Cases
From January to June 2011, 30 study subject cases of deliveries following AFHRT in the Department of Obstetrics of 2 hospitals in Marseilles, France, were prospectively recruited and included. Cases were selected during the daily review of deliveries performed each morning at the staff meeting held from Monday to Friday in both hospitals. Final selection of the cases was performed by L.S., R.L.D., and X.C.
The inclusion criteria were singleton pregnancy in cephalic presentation, at more than 37 weeks of gestation, with an attempt of vaginal delivery diagnosed with AFHRT lasting for more than an hour before delivery, independent of the mode of delivery. AFHRT was defined by the occurrence of recurrent and persistent decelerations at least 15 beats/min below the baseline, lasting for at least 15 seconds in duration, whatever its severity; the presence of baseline alterations; and the variability, the accelerations, and the coincidence with the onset of uterine contractions.
The gestational age of each pregnancy was calculated from the measurement of the crown-rump length determined during the first-trimester ultrasound scan performed between 11 and 13+6 weeks of gestation. Patients for whom an intervention (cesarean delivery or instrumental extraction) had been performed for a reason other than AFHRT or for whom the duration of AFHRT was less than an hour and/or the monitoring of the fetal heart rate had been interrupted for more than 20 minutes during labor, were not included.
Among the 30 included cases, 10 were diagnosed with adverse neonatal outcome defined by a cord arterial pH <7.00 and/or base excess >12 and/or Apgar score ≤3 at 5 minutes, according to the criteria defined by the task force. For each case, the entire fetal heart rate tracing and the full labor curve (ie, partogram) were copied at a 1:1 scale in black and white after being scanned with an Epson V330 (Seiko Epson Corp, Amsterdam, the Netherlands).
All obstetric data and immediate and short-term neonatal data were prospectively collected. All data and copies of the fetal heart rate tracing and partogram were anonymized. Patients were approached during their postpartum stay at the maternity ward. Informed consent was obtained from each patient after they were provided with oral and written information about the purpose of the study and what their participation meant. The study protocol received approval from the Ethics Committee for Research in Obstetrics and Gynecology (CEROG-2010-OBS-07-26R1).
Obstetric experts in court
Following the case inclusion, all of the 116 OECs registered at the French court of appeal and at the committee of conciliation and compensation were contacted by mail and offered to participate in the current study. Of those, 10 refused to participate, 78 did not respond, and 28 agreed to participate; these 28 OECs (24.1%) participated in the study. The characteristics of the recruited OECs were collected using a questionnaire.
Case review
All of the 30 obstetric cases were submitted to the OEC review twice. The first review was blinded to the neonatal outcome. After a 3 month interval after returning in the survey, the same 30 cases were submitted to the experts for a review that was unblinded to the neonatal outcome. The cases were presented to the OECs in random order, and the order was changed between the 2 reviews. For each review, the OECs were provided with a summary of the patient’s characteristics and obstetric data, copies of the full partogram and the complete fetal heart rate (FHR) recording annotated with information regarding the progress of labor and the results of a fetal scalp pH monitoring if any. A sheet containing information on the neonatal outcome was added for the unblinded review.
The blinded and unblinded reviews were conducted using the exact same questionnaire. Thus, OECs were asked to rank the AFHRT observed during the last 60 and 30 minutes before delivery according to the National College of Obstetricians Gynecologists French (CNGOF) classification and according to the International Federation of Gynecology and Obstetrics (FIGO) classification ( Table 1 ).
| Classification of abnormal fetal heart rate | Features | ||
|---|---|---|---|
| NICHHD | CNGOF | FIGO | |
| I Normal | Normal | Normal | Baseline rate 110–160 beats/min Baseline variability moderate Late or variable decelerations absent Early decelerations present or absent |
| II Indeterminate | Low-rate risk of acidosis | Intermediate | All tracings not categorized as category I or III Represent many tracings that are encountered in everyday clinical practice Minimal variability Absent variability without recurrent decelerations Marked variability Absence of induced accelerations after fetal stimulation Recurrent variable decelerations with minimal or moderate variability Prolonged deceleration Recurrent late decelerations with moderate variability Variable decelerations with slow return to baseline, overshoots, or shoulders |
| Risk of acidosis | |||
| III Abnormal | Significant risk of acidosis | Abnormal Preterminal | Absent variability and any of the following: Recurrent late decelerations Recurrent variable decelerations Bradycardia Sinusoidal pattern |
| Major risk of acidosis | |||
The OECs were also asked to express whether they agreed with the obstetrical management performed and the mode of delivery. When they disagreed, the OECs were asked whether they thought that an error had been made and whether they considered the obstetrical management as a cause of cerebral palsy in any children. Immediately after they had performed their first review of the 30 cases, the OECs were asked to return the completed questionnaires and discard all of the documents regarding the cases. OECs were allowed a period of 3 months to return their reviews.
Statistical analysis
Statistical analysis was performed using SPSS for Windows version 17.0 software (SPSS Inc, Chicago, IL). The results of the blinded and unblinded reviews were compared using the χ 2 test. A McNemar test and a Cochran test were used for the intraobserver and the interobserver variability, respectively. The interpretation of the power of agreement was established with kappa values. A bilateral value of P < .05 was considered statistically significant.
Results
Characteristics of the 30 cases submitted to the expert review ( Table 2 )
Patients delivered at a median term of 40.1 (±SD, 1.3) weeks of gestation. The median duration between the beginning of AFHRT and delivery was 180 minutes (±SD, 83.7). Nineteen patients (63.3%) achieved spontaneous vaginal delivery, whereas instrumental delivery and cesarean delivery was performed in 4 (13.3%) and 7 (23.3%) cases, respectively. Details of the 10 cases diagnosed with adverse neonatal outcome were as follows: 6 (20%) pH <7.00, 8 (26.7%) base excess >12, and 4 (13.3%) Apgar score ≤3 at 5 minutes. Ventilation or intubation was required for 14 neonates (46.7%), with admission to intensive care unit in 6 cases (20%).
| Patients | |
| Median age (±SD), y | 28.5 (±6.3) |
| Gestational diabetes | 9 (30) |
| Previous cesarean delivery | 2 (6.7) |
| Preeclampsia | 3 (10) |
| Median duration of gestation (±SD), wks | 40.1 (±1.3) |
| Postdate delivery | 7 (23.3) |
| Induction of labor | 11 (36.7) |
| Median duration between the beginning of FHRA and delivery (±SD), min | 180 (±83.7) |
| Fetal scalp pH performed during labor | 11 (36.7) |
| Meconium-stained amniotic fluid | 12 (40) |
| Delivery | |
| Spontaneous vaginal delivery | 19 (63.3) |
| Instrumental delivery | 4 (13.3) |
| Emergency cesarean delivery during labor | 7 (23.3) |
| Neonates | |
| Median birthweight (±SD), g | 3390 (±491) |
| Male | 13 (43.3) |
| Adverse neonatal outcome | |
| 5 minute Apgar score <7 | 9 (30) |
| 10 minute Apgar score <7 | 4 (13.3) |
| Fetal umbilical cord arterial pH | |
| Median (±SD) | 7.08 (±0.21) |
| pH <7.00 | 6 (20) |
| pH <7.12 | 15 (50) |
| Fetal umbilical cord arterial base excess | |
| Median (±SD) | 9 (±5.06) |
| Base excess >12 | 8 (26.7) |
| Indirect markers of perinatal asphyxia a | 10 (33.3) |
| Neonatal assistance b | 14 (46.7) |
| Admission to ICU | 6 (20) |
a Adverse neonatal outcome was defined by the presence of at least one of the following surrogate markers of perinatal asphyxia: cord arterial pH <7.00 and/or base excess >12 and/or a 5 minutes Apgar score ≤3
Obstetric experts in court
Among the 28 OECs who agreed to participate, 5 did not return the questionnaires in time and 1 returned the blind review only and were therefore excluded from the study. Finally, a total of 22 OECs from 20 different centers throughout France (18.9%) participated in the current study.
The characteristics of the OECs are summarized in Table 3 . The median age of the OECs was 57 years (±SD, 6.3). Of those, 20 (91%) were academic physicians and most of them did not have any private practice (55%). Ten of these (45.4%) had more than 15 years of obstetric experience, with a median of 12.5 years (±SD, 0.21). Although 1 OEC did not use any FHR classification in his routine practice, the CNGOF’s classification was routinely used by 18 (85.7%) and the FIGO’s by 10 (47.6%). No correlation was observed between OEC age, seniority in obstetrical practice or expertise, and the type of classification they routinely used.
| Median age (±SD), y | 57 (±7.3) |
| Practice as an obstetrician | |
| Median duration (±SD), y | 30 (±7.6) |
| Consultant | 22 (100) |
| Performing standby duties | 9 (41) |
| Performing on-call duties | 14 (64) |
| Type of obstetrical practice | |
| Academic physicians | 20 (91) |
| No private practice | 12 (55) |
| Combined practice | 8 (36) |
| Practice as an obstetrical expert in courts | |
| Median duration (±SD) | 12.5 (±8.8) |
| Type of court expertise performed | |
| Court of appeal | 20 (91) |
| Regional committee of conciliation and compensation | 12 (54.5) |
| Insurance | 10 (45.4) |
| AFHR classification used | |
| Routine use of a classification | 21 (95) |
| Type of FHRA classification employed | |
| CNGOF classification | 18 (85.7) |
| FIGO classification | 10 (47.6) |
| Melchior classification | 10 (47.6) |
Case reviews ( Table 4 )
With all of the 30 obstetric cases submitted to the 22 OEC reviews twice, a total of 1320 reviews were performed: 660 blinded and 660 unblinded to neonatal outcome. When comparing the blinded and unblinded reviews of AFHRT using the CNGOF classification, no significant difference in the proportion of AFHRT at 60 and 30 minutes before delivery that were considered at risk of acidosis was observed.
| Variables | Review of all cases | Review of cases with adverse neonatal outcome a (n = 220 b ) | ||||||
|---|---|---|---|---|---|---|---|---|
| Review n (%) | P value | Intraobserver agreement kappa (95% CI) | Review n (%) | P value | Intraobserver agreement kappa (95% CI) | |||
| Blinded (n = 660 c ) | Unblinded (n = 660 c ) | Blinded (n = 220 b ) | Unblinded (n = 220 b ) | |||||
| FHRA considered at risk of fetal acidosis by experts | ||||||||
| According to the CNGOF classification | ||||||||
| During the last 60 min before delivery | 458 (69.4) | 445 (67.4) | .76 | 0.51 (0.45–0.56) | 133 (60.4) | 139 (63.2) | .64 | 0.59 (0.51–0.67) |
| During the last 30 min before delivery | 525 (79.5) | 512 (77.6) | .42 | 0.48 (0.43–0.53) | 162 (73.6) | 156 (70.9) | .21 | 0.57 (0.49–0.65) |
| According to the FIGO classification | ||||||||
| During the last 60 min before birth | 283 (42.9) | 244 (37) | .03 | 0.46 (0.40–0.52) | 90 (40.9) | 90 (40.9) | .93 | 0.57 (0.48–0.66) |
| During the last 30 min before birth | 381 (57.7) | 366 (55.5) | .44 | 0.48 (0.42–0.54) | 121 (55) | 134 (60.9) | .25 | 0.56 (0.47–0.67) |
| Experts’ review of obstetrical management | ||||||||
| Agreed with the obstetrical management | 322 (48.9) | 384 (58.2) | < .001 | 0.53 (0.47–0.59) | 79 (35.9) | 94 (42.7) | .053 | 0.50 (0.39–0.61) |
| Agreed with the mode of delivery | 434 (65.8) | 482 (73) | < .001 | 0.50 (0.43–0.57) | 133 (60.4) | 131 (59.5) | .89 | 0.49 (0.37–0.60) |
| Concluded that an error had been made | 213 (32.3) | 170 (25.8) | < .001 | 0.48 (0.41–0.55) | 95 (43.2) | 103 (46.8) | .33 | 0.52 (0.41–0.6) |
| Established a causal relationship between a hypothetical cerebral palsy case and obstetrical management | 204 (30.9) | 170 (25.8) | .003 | 0.52 (0.46–0.59) | 92 (41.8) | 101 (45.9) | .25 | 0.55 (0.44–0.66) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
