Placental abruption is potentially disastrous to the fetus, with a perinatal mortality rate as high as 60 % [1]. Although perinatal mortality includes both Intrauterine Fetal Death (IUFD) and early neonatal death, it is unclear whether there are differences in the risk factors for these outcomes in cases of placental abruption.

In the proceeding paper, we showed the prediction of fetal acidemia in placental abruption [2]. It is also necessary to identify the important risk factors concerning the neonatal outcome other than IUFD, because the umbilical artery pH data cannot be obtained from all patients due to the emergency nature of the situation.

The purpose of this study was therefore to investigate the fetal and neonatal outcomes including IUFD and to compare IUFD with liveborn infants in cases of placental abruption, by evaluating the clinical assessments, as well as the results of adjunctive laboratory tests, such as the ultrasonographic findings and FHR patterns.

The approval of the institutional review board (Tokyo Women’s Medical University) was obtained before the start of this retrospective study. All singleton births, born between 24 and 40 weeks of gestation between January, 1, 2009, and December 31, 2009 were included. The medical records of mothers and neonates in the 94 institutes, comprising the Perinatal Research Network in Japan (PRNJ) listed in the ‘Acknowledgements’ were reviewed.

Gestational age was determined based on the mother’s last menstrual period and first and second trimester obstetric ultrasonography. More than 30 variables were assessed, including pregestational and antenatal factors. The details of the diagnosis, onset place, time from onset of symptoms to admission/delivery, and the clinical management of any relevant condition were recorded. We also included background data of the institutes such as the 24-hour anesthetic availability, presence of more than two specialists, and rapid access to blood and blood products as items for the multilevel analyses (Table 1).

Table 1: Anetenatal and perinatal factors considered in this study.

The diagnosis of placental abruption was based primarily on the one or more of the following clinical manifestations: vaginal bleeding, abdominal pain, uterine tenderness, with/without a nonreassuring FHR tracing and was confirmed by the placental detachment after delivery [3]. The presence of hematoma during a cesarean section or coagulation/massive genital bleeding during vaginal delivery was considered to indicate placental detachment. Women found to have a healthy placenta, and those with chronic abruption [4], were excluded from this study. Chronic Abruption-Oligohydramnios Sequence (CAOS) was defined by the following criteria: (1) clinically significant vaginal bleeding in the absence of placenta previa or any other identifiable source of bleeding, (2) amniotic fluid volume initially documented as normal and (3) oligohydramnios (amniotic fluid index ≤ 5) eventually developing without concurrent evidence of ruptured membranes [4].

Ultrasonographic findings were reported as follows: preplacental collection under the chorionic plate, increased heterogeneous placental thickness (more than 5 cm), retroplacental collection, marginal hematoma, subchorionic hematoma or intra-amniotic hemorrhage [5]. FHR patterns were defined as abnormal when one of the following patterns was detected: persistent late decelerations, severe atypical deceleration, prolonged deceleration, or bradycardia [6]. The same patterns were also used for the assessment of preterm fetuses. A baseline FHR of less than 100 bpm was noted as bradycardia (>3 min) [6]. This included FHR detected only by hand-held Doppler fetal heart detectors or auscultation in some cases. The live fetuses were delivered by cesarean section because of abnormal FHR patterns or maternal indications, such as massive genital bleeding.

An adverse outcome of the neonate was defined as the occurrence of IUFD (Group 1, n=89), death before hospital discharge or a diagnosis of cerebral palsy (Group 2, n=28). The remaining 238 patients were considered to be controls. The results were expressed as mean ± Standard Deviation (SD), medians with range, or numbers with the percentage. Statistical analyses were performed with the Statflez 6.0 software package (Archtech Co., Ltd., Osaka, Japan, http://www.statflex.Net/) and were carried out using the chi-square test, Fisher’s exact probability test, and the Mann-Whitney U test. p values of less than 0.05 were considered to be significant. The Odds Ratio (OR) and 95% Confidence Intervals (CI) were calculated to estimate the relative risk between cases and controls with regard to potential predictors for Group 1 or Group 2. They were compared in both the univariate and multivariate analyses. Logistic regression models were used to assess confounding effects and to construct a discriminant function.

Clinical background in the present study

There were 355 patients complicated by placental abruption. The overall number of deliveries in 94 institutes was 54,628 and the rate of placental abruption was 0.65%. Eighty-nine fetuses (nearly onequarter) were dead on admission (Group 1), while the remaining 266 fetuses were alive. The mean gestational age at delivery was 34.3 ± 3.5 weeks (preterm: 266 patients, term: 89 patients).

Because the diagnosis of placental abruption in this study was based on the clinical manifestations, and the placental detachment was confirmed after delivery, the clinical background of patients in the present study may be more severe than that in some other studies [7,8]. As a result, the percentage of maternal and fetal/neonatal morbidity and mortality was high.

Abnormal ultrasonographic findings were observed in 263 patients (74.1%), mainly indicating retroplacental anechoicity (185 cases), and increased placental thickness (167 cases).

In the 266 fetuses that were alive on admission, abnormal FHR patterns were observed in 166 patients (62.4%). The main abnormal FHR results were persistent late decelerations in 51 cases and prolonged deceleration in 86 (including by auscultation) cases. A cesarean section was performed for 256 patients (96.2%).

Differences between IUFD and live-born infants

The clinical demographics of IUFD and live-born infants are given in Table 2. In a comparison of IUFD and live-born infants, statistically significant differences were observed in the maternal age, maternal transfer, on set of symptom outside obstetric facilities, abnormal ultrasonographic findings, and blood transfusion. The frequency of vaginal bleeding, gestational age at delivery and birth weight were lower in the IUFD group. There were no significant differences in the frequencies of preeclampsia, chronic hypertension, abdominal pain and cesarean section between these two groups. Using a logistic regression model, a blood transfusion (OR 2.21, 95%CI 1.02 -4.76) was the only significant factor associated with the occurrence of IUFD (Table 3).

Table 2: Results of the univariate analysis in terms of the difference between intrauterine fetal death (IUFD) and live-born infants.

Table 3: Results of the multivariate analysis in terms of risk factors for Intrauterine Fetal Death (IUFD).

Massive detachment of placenta induces not only the maternal consumptive coagulopathy, but also the decrease of fetal oxygenation. As the degree of placental separation increases, the risk of fetal death also increases [1]. In addition, the frequency of coagulopathy is much higher in abruptions in which fetal death has occurred. Because especially transfusion of fresh frozen plasma may be life saving and prevent patients to progress into the disseminated intravascular coagulation, blood transfusion itself is thought to be the result of an unstable maternal condition, which means that there is a close link between IUFD and the maternal status.

In the subgroup of cases in which the time course was confirmed, the interval from the onset of symptoms to the diagnosis (n=78, median 213, range 60 - 1020 min) was significantly longer than that for liveborn infants (n=147, median 130, range 10 - 780 min, p<0.0001).

More frequent maternal transfer, onset of symptoms outside obstetric facilities, abnormal ultrasonographic findings, less frequent vaginal bleeding, younger gestational week at delivery, and low birth weight, which were cleared by the univariate analysis, might affect the interval from the onset to the diagnosis, although they were not found to be significant by multivariate analysis. However, this might have been derived from the insufficient sample size.

Although the etiology of placental abruption is heterogeneous and speculative, and clinical abruption is the final culmination of a longstanding disease process within the placenta, controlling all of these factors may be important for reducing the incidence of IUFD in cases of placental abruption [9].

Differences between case and control in live-born infants

The clinical demographics of case (group 2) and control infants are given in Table 4. Using a logistic regression model, bradycardia (OR 28.25, 95%CI 6.1 - 130.84), late decelerations (OR 5.94, 1.02 - 34.61) and gestational age at delivery < 35 weeks (OR 5.37, 1.94 - 14.85), were all found to be associated with the occurrence of an adverse outcome (Table 5).

Table 4: Results of the univariate analysis in terms of risk factors of an adverse outcome.

Table 5: Results of the multivariate analysis in terms of risk factors of adverse outcome.

In the proceeding paper, the potential predictors for fetal acidemia (umbilical artery pH less than 7.0) in cases of placental abruption were bradycardia and late decelerations. For an adverse outcome, ‘gestational week at delivery <35 weeks’ was added as a significant factor. Allred and Batton previously reported a study of the short-term outcome of preterm infants (23 to 32 weeks) born from placental abruption, and concluded that abruption was not an independent risk factor for a poor outcome among infants born between 23 and 32 weeks gestation, but that the preterm delivery was the main determinant of outcome [10].

For the purpose of predicting an adverse neonatal outcome, the statistically significant factors identified by the multiple logistic regression analysis were subjected to stepwise regression analysis to construct a linear discriminant function: A+2B+3C+5D, where A was abdominal pain (0, no; 1, yes), B was gestational age less than 35 weeks (0, no; 1, yes), C was late decelerations (0, no; 1, yes), and D was bradycardia (0, no; 1, yes), because we could not obtain the pH data from all patients due to the emergency nature of the situation. This discriminant function was called ‘Abruption Prognosis Score’ (APS). A logistic regression analysis was performed to make clear the relationship between the APS and the probability of an adverse outcome in Figure 1. The probability of this cut-off point of APS was calculated to be approximately 0.1. When this score was 8, the probability of an adverse outcome was almost 0.5.

Figure 1: Relationship between the Abruption Prognosis Score (APS) and the probability of adverse outcome in the cases of placental abruption.

We had established the score by using the above-mentioned concept and named it the ‘Severe Abruption Score (SAS)’ that could be used to predict the occurrence of fetal acidemia in cases of placental abruption [2]. Similar to the ‘SAS’, the amount of vaginal bleeding correlates poorly with the degree of placental separation and does not serve as a useful marker of an impending adverse outcome. On the other hand, different from SAS score, abnormal ultrasonographic findings were not an important item for the prediction of adverse outcomes in cases of placental abruption.

There are several limitations to the present study. First, this study was done in a retrospective fashion; therefore, further studies are warranted to confirm the usefulness of this score prospectively. Second, since the abruption prognosis score was based only on cases where a diagnosis of abruption was confirmed according to placental appearance just after delivery and was designed to be used immediately after delivery, this score should be used with caution.

In conclusion, the significant factors associated with IUFD in cases of placental abruption were the interval to the diagnosis and the needs for blood transfusion. Adverse outcomes other than IUFD were linked to the gestational age, bradycardia, or late decelerations, regardless of the presence of genital bleeding or abnormal ultrasonographic findings.

We thank Mr Sugimoto for kindly providing analyses of the database.

We wish to thank the institutions and representative physicians enrolled in the database for Perinatal Research Network in Japan, which include:

Aichi Medical University: S Kinoshita; Akita University: A Sato; Asahi-chuo- Hospital: H Udagawa, A Kurihara; Asahikawa Medical University: K Nishino; Ashikaga Red Cross Hospital: Y Kasuga, T Hirao; Ehime Prefectural Central Hospital: K Noda; Fukuchiyama City Hospital: T Okuda; Fukuoka University: T Yoshisato; Fukushima Medical University: H Takahashi; Gifu University: H Toyoki; Haga Red Cross Hospital: A Ohkuchi; Hamamatsu University School of Medicine: K Suzuki; Hirosaki University: T Higuchi; Hiroshima City Hospital : O Ishida; Hiroshima General Hospital: Y Nakanishi; Hiroshima University: Y Mukai; Hokkaido University: S Yamada; Hyogo College of Medicine: T Takenobu; Hyogo Prefectural Kobe Children’s Hospital: T Funakoshi; Japanese Red Cross Fukuoka Hospital: M Nishida; Japanese Red Cross Kyoto Daiichi Hospital: H Yamamoto; Jichi Medical University: S Matsubara, R Usui; Juntendo University Urayasu Hospital: K Yoshida, A Tajima; Kagawa University: H Tanaka; Kagoshima City Hospital : M Kamitomo; Kagosima University: Y Yonehara; Kameda Medical Center: M Suzuki, H Takaya; Kanagawa Children’s Medical Center: H Ishikawa; Kanazawa Medical University: T Fujita; Kinki University: M Shiota, M Tsuritani; Kitasato University: S Kawano; Kobe University: K Tanimura; Kumamoto City Hospital: J Ishimatsu, K Aikou; Kurashiki Medical Center: F Yamazaki; Kurume University: D Hori, R Hayashi; Kyoto Prefectural University of Medicine: T Okubo, S Fujisawa; Kyoto University: J Hamanishi; Kyushu University: K Fukushima; Maternal & Child Health Center AIIKU HOSPITAL: T Adachi, Y Kawana; Mie University: T Sugiyama; Miyazaki University: S Furukawa; Nagasaki Municipal Hospital: K Kotera; Nagasaki University: S Yoshimura; Nagoya Daini Red Cross Hospital: N Kato; Nagoya University : T Kotani; Nara Medical University: T Sado; National Center for Child Health and Development: H Sagou, H Aoki; National Center for Global Health and Medicine: J Kakogawa; National Defense Medical College: Y Hasegawa; National Hospital Organization East Saga Hospital: M Nomiyama; National Hospital Organization Nagasaki Medical Center: I Yasuhi, M Fukuda; National Hospital Organization Nishisaitama Chuo National Hospital: A Yoshida; National Hospital Organization Okayama Medical Center: Y Tateishi; National Hospital Organization Takasaki General Medical Center: I Ito; National Hospital Organization Yokohama Medical Center: A Nakamura; Niigata University: T Serikawa; Nippon Medical School Tama Nagayama Hospital: I Kawabata; Nippon Medical School: M Satomi; Oita Prefectural Hospital: S Sato; Oita University: Y Nishida; Okayama University: T Segawa; Osaka City University: D Tachibana, M Tsukioka; Osaka Medical Center and Research Institute for Maternal and Child Health: N Mitsuda, A Sasahaea; Osaka University: S Fujita; Saga University: M Muro; Saiseikai Yokohamashi Tobu Hospital: Y Konishi, Y Sakakibara; Shiga University of Medical Science: T Ono; Shimane University: S Aoki; Shinshu University: N Kikuchi; Showa University: R Matsuoka; St. Marianna University School of Medicine, Yokohama City Seibu Hospital: J Saito,T NaKo; Takatsuki General Hospital: S Nakago; Teikyo University Hospital: T Ayabe, K Kido; The Japan Baptist Hospital: H Egawa, S Suzuki; The Jikei University: S Wada; The University of Tokyo: Y Kamei; Toho University Omori Medical Center: C Aoki; Tohoku University: J Sugawara; Tokai University: H Ishimoto, K Mituzuka; Tokyo Medical University Hachioji Medical Center: T Nohira; Tokyo Medical and Dental University, University Hospital of Medicine: Y Momohara; Tokyo Women’s Medical University: Y Matsuda, Y Makino; Tottori University: T Harada; University of Occupational and Environmental Health, Japan: K Yoshimura; University of Toyama: S Saito, A Shiozaki; University of the Ryukyus: K Sakumoto; Wakayama Medical University: S Yagi; Yamagata University: S Tsutsumi; Yamaguchi Red Cross Hospital: H Takahashi; Yodogawa Christian Hospital: C Mikami; Yokohama City University Medical center: M Okuda; Yokohama Minami Kyosai Hospital: H Nagase; Yokohama Rosai Hospital: M Nakayama.