Fetal macrosomia: risk factors and outcome

European Journal of Obstetrics & Gynecology and Reproductive Biology 77 (1998) 51–59 Fetal macrosomia: risk factors and outcome A study of the outcome concerning 100 cases.4500 g ´ ` J.C. Monnier, F. Puech J. Berard, P. Dufour*, D. Vinatier, D. Subtil, S. Vanderstichele, ˆ Departement of Obstetrics and Gynecology. Hopital Jeanne de Flandre, 59037 Lille cedex, France Received 20 April 1997; accepted 26 September 1997 Abstract Objective: Because difficult vaginal delivery is more frequent with macrosomic fetuses, some authors recommend routine caesarean section for the delivery of fetuses.4500 g. The purpose of this study was to evaluate the appropriateness of this recommendation, in particular, to analyze maternal and fetal complications according to the mode of delivery. Method: Maternal and neonatal records of 100 infants with weights of at least 4500 g were identified retrospectively from January 1991 to December 1996. Outcome variables included the mode of delivery and the incidence of maternal and perinatal complications. Results: The study sample consisted of 100 infant and mother pairs. Macrosomic fetuses represented 0.95% of all deliveries during this period and only ten were .5000 g. Mean birth weight was 4730 g (maximum, 5780 g). Gestational diabetes was present in nineteen patients. Diabetes was present in three patients. A trial of labour was allowed in 87 women, and elective caesarean delivery was performed in thirteen patients. The overall cesarean rate, including elective caesarean delivery and failed trial of labour, was 36%. Of those undergoing a trial of labour, 73% (64 / 87) delivered vaginally. Shoulder dystocia occurred fourteen times (22% of vaginal deliveries) and it was the most frequent complication in our series. There were five cases of Erb’s palsy, one of which was associated with humeral fracture, and four cases of clavicular fracture. By three months of age, all affected infants were without sequelae. There was no related perinatal mortality and only two cases of birth asphyxia. Maternal complications with vaginal delivery of macrosomic infants included a high incidence of lacerations requiring repair (eleven cases). No complications were noticed in the patients who had a caesarean section. Conclusion: Vaginal delivery is a reasonable alternative to elective cesarean section for infants with estimated birth weights of less than 5000 g and a trial of labour can be offered. For the fetuses with estimated birth weight .5000 g, an elective caesarean section should be recommended, especially in primiparous women.  1998 Elsevier Science Ireland Ltd. Keywords: Fetal macrosomia; Shoulder dystocia; Caesarean section 1. Introduction Perinatologists have reviewed many of the obstetric problems that could lead to infant morbidity and / or mortality but, in modern perinatal care, the small fetus has attracted more interest from practitioners, giving rise to a significantly improved outcome. The macrosomic fetus has not been given the same attention, even though increased perinatal mortality and morbidity rates have been reported in most studies. During labour, cephalo–pelvic dispropor*Corresponding author. tion can result in fetal distress and difficult deliveries are more frequent with macrosomic fetuses. For this reason, it is generally accepted that breech fetuses that are either known or suspected to be large are best delivered by caesarean section (CS). For macrosomic fetuses in cephalic presentation, cephalopelvic disproportion, difficult instrumental delivery and shoulder dystocia, amongst other factors, have lead to the recommendation of routine CS. Infants with a birth weight of at least 4500 g are at an increased risk of birth asphyxia and traumatic injury at delivery, particularly as a consequence of shoulder dystocia, with possible long-term sequelae (brachial plexus 0301-2115 / 98 / $19.00  1998 Elsevier Science Ireland Ltd. All rights reserved. PII S0301-2115( 97 )00242-X 52 ´ J. Berard et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 77 (1998) 51 – 59 injury) and even fetal death [1]. The fact that these fetuses usually are normal and healthy at the onset of labour makes a disastrous outcome particularly tragic. In addition, maternal complications may occur due to difficult labour and delivery [2]. However, there is growing criticism of these recommendations [3,4]. Performing CS in all cases seems unreasonable because of our inability to diagnose macrosomia reliably and to predict accurately those infants that are at risk of shoulder dystocia. The aim of this study was to evaluate the efficacy of current methods in identifying fetuses with a weight .4500 g and to study perinatal mortality and morbidity in relation to the mode of delivery. 2. Materials and methods A retrospective analysis of macrosomic deliveries at ˆ Pavillon Olivier Maternity and Hopital Jeanne de Flandre (Lille, France) from January 1991 to December 1996 was performed. One hundred maternal and neonatal records of infants with birth weights of at least 4500 g were identified during this period. The total macrosomic group represented 0.95% of the infants delivered (100 of 10,500) during the study. The highest birthweight was 5780 g. Charts were reviewed for demographic and medical characteristics, labour and delivery events, Apgar scores, arterial cord pH, and maternal and perinatal complications. Neonatal records were reviewed also if shoulder dystocia, birth trauma or birth asphyxia were documented. Types of birth trauma seen included Erb’s palsy and clavicular and humeral fractures. Birth asphyxia was defined as a 1 min Apgar score of less than five and an umbilical artery pH of less than 7.0. We systematically screened all patients with a 1-h 50 g glucose tolerance test (50 g GTT: O’Sullivan test) between 24 and 28 weeks. Those with a reduced 50 g GTT were then given a 3-h 100 g GTT (100 g GTT). In this way, three groups of pregnant women were defined: A normal group (O’Sullivan test ,1.40 g / l); a glucose-intolerant group (one abnormal 3 h GTT value) and a third group with gestational diabetes (two abnormal values). Primary outcome variables included the mode of delivery, and the incidence of neonatal and maternal complications. Maternal complications evaluated were lacerations requiring repair, haemorrhage and infection. The neonatal complications evaluated were shoulder dystocia, and associated birth trauma and asphyxia. After delivery of the fetal head, if delivery required further intervention other than downward traction and episiotomy, shoulder dystocia was considered to have occurred. We also compared labour, delivery and neonatal outcome in two groups: Mild macrosomia (4500 to 4999 g) and severe macrosomia (.5000 g). 3. Results Of the 100 babies with birth weights $4500 g, 99 were cephalic presentations and one was a breech. This woman was 33 years old, para 3, and weighed 90 kg (120) at delivery. A CS was performed for failure of engagement. She delivered a 4900-g, Apgar 10 / 10 infant. Most of the mothers (78%) were multiparous. Maternal median weight gain was 15 kg (1–47 kg). Of the women, 52% were obese at delivery (.85 kg). The average gestational age at delivery was 40 weeks12 days (37–42.5 weeks). Postdate deliveries (.41 weeks) occurred in 17% of the women. In this study, there were no fetal or neonatal deaths. Fetal macrosomia was suspected in 76%, on the basis of clinical examination, sonography or detection of the following risks factors, present either singularly or in association: Previous delivery of an infant .4000 g (48%), previous history of gestational diabetes or glucose intolerance (19%), an increase in weight of more than 15 kg during pregnancy (51%) and obesity (.80 kg) before the onset of pregnancy (25%). The average birth weight was 4730 g, the largest infant weighing 5780 g. Eighty-seven women were allowed a trial of labour and elective caesarean delivery was performed in thirteen patients. In eight of these cases, the indication for elective CS was fetal macrosomia and previous elective caesarean section. The overall caesarean rate, including elective caesarean delivery and failed trial of labour, was 36%. Of those undergoing a trial of labour, 64 (73%) delivered vaginally and 23 delivered by CS. The overall incidence of A1 gestational diabetes was 19%. Three patients had pre-gestational diabetes. Of the glucose-intolerant patients, 14% were controlled by diet alone. Each diabetic patient with pre-gestational diabetes and those patients in which we found gestational diabetes were hospitalized on an out-patient basis. Each patient was consulted by an endocrinologist. They were explained how to give themselves insulin (except for one patient, controlled by diet alone), to survey their blood sugar level and how to follow a diabetic dietary regime. Depending on the cases, the patients were treated on a weekly or bi-weekly out-patient basis and were hospitalized if necessary. We compared the median Apgar scores at 1 and 5 min in two groups of patients delivering vaginally: Those where shoulder dystocia did not occur, and those where shoulder dystocia did occur. The median scores in the group with shoulder dystocia (n514), 8.5 / 9.5, respectively, were significantly lower than the scores in the group without shoulder dystocia (n550), 9.9 and 9.9, respectively. Only five of the fourteen infants who experienced shoulder dystocia required instrumental delivery (one mid forceps, two ventouse and two rotation ventouse with mid ´ J. Berard et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 77 (1998) 51 – 59 53 Table 1 Macrosomic newborns requiring admission to the neonatal intensive care unit Case Parity Mode of delivery Complications of labour or delivery Weight (g) Apgar score pH Reasons for admission 1 4 Ventouse labour: 8 h 45 min Shoulder dystocia, supra-pubic pressure 4660 7/9 7.1 Fracture of humerus, brachial palsy 2 5 Vaginal delivery labour: 7 h 35 min Meconium, shoulder dystocia, rotations 4600 10 / 10 7.28 Meconium aspiration, severe asphyxia forceps). Eleven of the infants with shoulder dystocia required obstetrical manoeuvres after delivery of the fetal head (single or associated): Rotations in 10 times, McRoberts associated with supra-pubic pressure and rotations on one patient. Apgar scores ,5 at 1 min (birth asphyxia) were documented in one vaginal delivery and in one CS. The lowest Apgar scores for vaginal delivery (3 / 7) occurred in an infant who also experienced a left sided Erb’s palsy. The mother was 30 years old, para 7, with a previous history of macrosomic infants. She gained 10 kg during pregnancy (85–95 kg). The infant weighed 5780 g. With physiotherapy, full recovery was achieved in ten days. The lowest Apgar scores for an emergency CS (3 / 8) occurred in a para 2 patient with previous CS for preeclampsia. After 11 h 25 min of trial of labour, a CS was performed under general anaesthesia for acute fetal distress (bradycardia, acidosis, meconium stained haemorrhagic liquor). The birth weight was 4600 g and the arterial cord pH was 7.2. There were no postnatal complications. Of fifteen women who had previously undergone CS, eleven opted for elective CS, three had CS after failed trial of labour and one achieved vaginal delivery. The patient achieving vaginal delivery was a 24-year-old, para 2 patient, weighing 106 kg. Indications for previous CS were breech presentation and macrosomia (4250 g). On this occasion, she delivered a 4770-g infant at 39 weeks with mid forceps, after 9 h of labour. The three cases of emergency CS in mothers with previous section show a prolonged labour, 12 h 20 min, 11 h 40 min and 11 h 25 min, before section. Indications for CS were cervical dystocia, cervical dystocia and fetal distress (see above), respectively. Table 1 shows the two cases of infants transferred to neonatal intensive care. In both cases, infants experienced shoulder dystocia with obstetric manoeuvres. The first suffered a fractured humerus and brachial plexus injury, due to difficulty in delivering the anterior shoulder, and was transferred immediately for paediatric surgery. The second was transferred for severe asphyxia requiring several days of intubation, after aspiration of meconium. Table 2 shows fetal injuries associated with macrosomic vaginal delivery (no fetal traumas were related to CS deliveries). There were five cases of Erb’s palsy, one of which was associated with humeral fracture, and four cases of clavicular fracture. All of these injuries were found to have resolved on follow-up, except in the case of the infant with brachial plexus injury in association with humeral fracture, who required transfer for surgery (case no. 1, Table 1). The reasons for emergency CS are shown in Table 3. No Table 2 Fetal injuries Case number Trauma MW Parity GD Delivery BW Apgar (19 / 59) Immediate complications 1 2 3 4 5 6 7 8 9 Clavicular fracture Clavicular fracture Clavicular fracture Clavicular fracture Erb’s palsy Erb’s palsy Erb’s palsy Erb’s palsy Erb’s palsy, humeral fracture 75(116) 85(120) 86(115) 96(122) 74(118) 110(120) 118(16) 95(110) 89(19) 5 3 2 2 2 6 2 7 4 No No No No No No Yes ? No VD VD. eng. ventouse VD. eng. ventouse VD VD VD. rotations VD VD. rotations VD. ventouse 4520 45600 4680 4680 4650 5580 5050 5780 4660 10 / 10 10 / 10 10 / 10 10 / 10 10 / 10 4/8 7/9 3/7 7/9 2 2 2 2 Hypotonia Physio. Physio. Physio. Paediatric surgery BW5Birth weight. GD5Gestational diabetes. VD5Vaginal delivery. MW5Maternal weight, (1X): weight gain in pregnancy. Physio.5Physiotherapy. Eng. ventouse5Engagement ventouse. 54 ´ J. Berard et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 77 (1998) 51 – 59 Table 3 Indications for caesarean section after trial of labour Causes Number Average weight at delivery (g) Average Apgar score Average pH Outcome Cervical dystocia Engagement dystocia Rotation ventouse failure Acute fetal distress Induction failure (Pg) Brow presentation 10 7 1 3 1 1 4772 4880 4700 4603 4870 4800 9.5 / 9.8 9.6 / 9.8 10 / 10 4.3 / 9 10 / 10 10 / 10 7.21 7.24 2 7.12 7.28 2 48 h surveillance (1 / 10) 2 2 48 h surveillance (2 / 3) (VSD) 2 Total 23 VSD5Ventricular septal defect. Pg5Prostaglandin. complications due to epidural or general anaesthesia were found. Five patients required general anaesthesia, one of which was for cord prolapse. Table 4 compares maternal risk factors, mode of delivery and the incidence of fetal injury in mild macrosomic fetuses (4500–4999 g) and severe macrosomic fetuses (.5000 g). We also noted perineal complications associated with macrosomic vaginal deliveries (64 VD). We recorded 31 episiotomies (eight in primiparous women and 23 in multiparous women), seven tears requiring simple repair and four incidences of third degree tear. No fourth degree tear was found and no complications occurred after episiotomy or repair of tears. No significant haemorrhage (.1000 ml) or infection that could not be controlled by a normal antibiotic regimen were noted. 4. Discussion 4.1. Definition and incidence A major problem resides in the lack of a universally accepted definition of macrosomia. Perhaps the most commonly used definition is that of a birth weight equal to or exceeding 4500 g. Although an absolute weight is easy to determine, it fails to taking into account the influence of gestational age on birth weight. Defining macrosomia as large size for gestational age (LGA) (e.g., greater than or equal to the 90th percentile birth weight for gestational age [5,6]) provides a partial solution to this problem. However, because normal birth weights for gestational ages may vary with geography and ethnicity [7], an infant that is LGA using a set of normative data derived from one population may not meet the criteria using a set of normative data derived from another population. The ponderal index, the ratio between 100 times the weight and the cube of the length of a given infant, is a measurement that is relatively free of the influence of race, gender and gestational age [2,7]. An alternative measurement is the birth symmetry index, defined as the ratio between weight and length, each having been divided by the weight and length respectively at the 50th percentile [7]. Although, these indices are cumbersome to calculate, they appear to offer a rational means of comparing birth weight data derived from different populations. However, usually, fetal macrosomia is arbitrarily defined as a birth weight above the 90th percentile for gestational age or greater than 4500 g at term. The most recent statistics quote an incidence of about 1.5% for birth weight greater than 4500 g [8]. Spellacy et al. [2] reported a frequency of macrosomia of 1.7%, which Table 4 Comparison between two groups of macrosomic fetuses ,5000 g and .5000 g Weight (g) 4500–4999 .5000 Number Maternal age (years) Average parity Prevalence of gestational diabetes Maternal weight at delivery Gestational age at delivery (weeks) Duration of labour Prevalence of CS Prevalence of instrumental delivery 90 30 3 14.5% 88 40 7 h 30 min 33% 26.5% 10 31 4.7 60% 97.2 40 9 h 50 min 60% 0% (4 VD a) (Parity →6-2-7-7) 75% (three traumas / four VD a) Fetal traumas (VD) VD5Vaginal delivery. 9.4% (six traumas / 64 VD*) ´ J. Berard et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 77 (1998) 51 – 59 was similar to the rates noted in other studies [9–11]. In Doc’s series [12], 0.8% weighed more than 4500 g. In our series, 0.95% weighed more than 4500 g. 55 the incidence of fetal macrosomia remains open for investigation. 4.2.2. Prepartal risk factors 4.2. Factors associated with fetal macrosomia Difficulty has been encountered in identifying historical, clinical and laboratory factors that are predictive of macrosomia. Previous observers have recognized factors associated with an oversized fetus, including prior delivery of a large infant and excessive maternal weight gain. Also, women delivering macrosomic infants tend to be older and have an increased parity. More characteristically, they had one of three problems, namely, obesity [1,13], diabetes [2,13] and postmaturity [1]. It was demonstrated that with this triad of maternal complications, the risk of macrosomia ranged from 5 to 14%. It is also interesting that the macrosomic infant is more often male. After those cases attributed to large maternal stature or excessive weight gain are eliminated, at least one third of cases of macrosomia remain unexplained. The value of ascertaining maternal birth weight has recently been reconfirmed. Klebanoff et al. [14] demonstrated that a mother’s birth weight has a strong influence on her child’s birth weight. However, the predictive potential of these parameters was not sufficient to identify individual cases. 4.2.1. Pre-pregnancy risk factors 4.2.1.1. Multiparity: 78% in our series Sack [15] noted the high frequency of multiparity amongst mothers of large infants. In the study of Doc et al. [12], the multiparity rate was 70% and they did not note a particular age group in which macrosomia was more prevalent. 4.2.1.2. Diabetes Macrosomia would therefore be expected in obese and diabetic pregnant women because the principal substrate for fetal growth is glucose, which is elevated in these two conditions [2]. Posner et al. [9] found a tenfold increase in diabetes in the macrosomic group compared to the general obstetric population, and similar findings were noted by others [7,9,12,13]. In the study by Doc et al. [12], the prevalence of diabetes was 6.8% (13% of pre-gestational diabetes in our series), compared to 1.7% in their general obstetric population. The relationship between maternal glucose concentration and fetal macrosomia has been explored in several publications. Some ( [5,7]) have failed to demonstrate any significant relationship. Others ( [16– 18]) have found a positive relationship between maternal glucose levels and fetal macrosomia. Analyses of this relationship must consider the influence of all factors that may affect birth weight. Thus, the quantitative nature of the relationship between levels of maternal glycaemia and 4.2.2.1. Gestational diabetes Gestational diabetes accounts for 10% of macrosomic infants. Gestational diabetes or pregnancy-induced glucose intolerance, which includes one or two abnormal values on a 3-h GTT, is a risk factor for macrosomia and shoulder dystocia. All pregnant women should be screened for pregnancy-induced glucose intolerance in the second trimester. Those patients at high risk but with negative test results should be rescreened in the third trimester. Of patients with pregnancy-induced glucose intolerance and insulin dependence, 5–10% will have infants who show signs of macrosomia. 4.2.2.2. Excessive maternal weight gain Maternal weight gain ranks high as a factor leading to macrosomia. Boyd et al. [1] found that a weight gain of 20 kg is additive to the risk factors of obesity and postdate gestations of seven days. The risk of excess weight gain has been detailed by Doc et al. [12], but was questioned by Parks and Ziel [13]. Excessive weight gain increases the incidence of macrosomia from 1.4 to 15.2%. 4.2.2.3. Postdate gestation Postdate gestation is a very important risk factor for macrosomia and is one of the ominous components of a triad, the other components being obesity and diabetes. Boyd et al. [1] reported a 21% incidence of macrosomia with infants delivered at 42 weeks, but only 12% with those delivered at 40 weeks and suggested induction of labour as a means to prevent the complications of vaginal delivery. In the study of Spellacy et al. [2], the frequency of postdate pregnancy was 10.8% among the infants with macrosomia. In our series, 17% of the patients delivered after term. 4.2.3. Intrapartal risk factors Doc et al. [12] found certain features to be more common in mothers of macrosomic babies, including prolonged second stage of labour and a higher prevalence of shoulder dystocia and CS and they reported a 46% incidence of failure to progress in labour in a group of macrosomic infants, while Benedetti and Gabbe [19] described a 21% incidence of failure to progress in a similar group of patients. Doc et al. [12] reported that the size of the fetus does not have a significant effect on the length of the first stage of labour. However, the second stage might be prolonged if the baby is large. This finding has prognostic importance in regard to the development of shoulder dystocia [19]. In the study by Doc et al. [12], the prevalence of shoulder dystocia in the macrosomic group increased from 1%, when the second stage was less than 9 56 ´ J. Berard et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 77 (1998) 51 – 59 min, to 13.3%, when the second stage was greater than 70 min. 4.3. Prenatal diagnosis of macrosomia The prenatal diagnosis of macrosomia continues to challenge the obstetrician and antenatal suspicion of macrosomia requires careful clinical and ultrasonic evaluation. 4.3.1. Clinical evaluation Posner et al. [9] stated that macrosomia should be suspected if, at term, the distance from the superior surface of the symphysis pubis to the fundus exceeds 40 cm. The usefulness of the symphysis-fundal height measurements for identifying macrosomic infants has previously been questioned [20]. Fundal height measurement is dependent upon examiner skill and patient size and is, therefore, generally a poor indicator of fetal weight. 4.3.1.1. Ultrasonic evaluation The value of ultrasound scans for detecting macrosomic fetuses is certain. However, estimation of fetal weight by ultrasound carries a 10–15% margin of error [10]. However, it is possible that the wider use of various ultrasound parameters (transthoracic diameter, abdominal circumference, total intrauterine volume), in addition to biparietal diameter, in a serial manner will lead to better prediction of fetal size. Studies have shown that normal fetal growth is linear, while the macrosomic fetus has accelerated growth towards term [21]. So, ultrasonic investigations would therefore be of value only if performed after 38 gestational weeks but are still of limited value for identifying macrosomic fetuses, as errors in estimation of fetal weight occur in up to 15% of cases, even under optimal conditions [21]. In 1992, Balouet et al. [22] suggested a new method based on ultrasound fetal fat measurements on a transverse section of the fetal thigh, by the sartorius muscle. They found an overall accuracy (the median of departure from the means) of 6.3%, and 6.9% in their small fetuses group (1000–1500 g). This model could be interesting in the screening of small-for-date fetuses. However, so far, no single method or variable has been found to be predictive of macrosomia in the antenatal period with acceptable accuracy. 4.4. Fetal macrosomia prevention Fetal macrosomia and its prevention have been a focus of interest for some time. Different prophylactic and therapeutic modalities have been evaluated with regard to their impact on fetal macrosomia. Alternative therapies aimed at preventing macrosomia have achieved varying degrees of success. A good stability of gestational diabetes or diabetes mellitus must be obtained by dietary prescrip- tion and insulin therapy. However, despite insulin treatment, obese women had a higher incidence of macrosomic babies, even when the data were stratified by maternal glucose levels [3,7]. 4.5. Morbidity associated with fetal macrosomia Most observers have recognized that both the macrosomic infant and its mother are at high risk of injury [23–25]. Spellacy et al. [2], like other authors, demonstrated that macrosomia is a significant risk factor for the infant in terms of birth trauma, including shoulder dystocia and associated injuries [2,13], fetal death [2] and depression of neonatal Apgar scores. Shoulder dystocia is a serious potential complication of fetal macrosomia. The overall incidence of shoulder dystocia varies between 20– 30% with fetal weights of more than 4500 g [10,13], but many authors find that some 50–60% of shoulder dystocias occur in infants weighing less than 4000 g. In addition, women having a macrosomic infant are more likely to have their delivery by CS [2,13]. Because of these poor outcomes, it is important to try to identify women at risk of having a macrosomic infant, so the delivery method can be evaluated. In the study by Doc et al. [12], the total prevalence of shoulder dystocia in the macrosomic group was 3.8%, in agreement with other studies, and all of the shoulder dystocia cases were associated with non-operative spontaneous vaginal deliveries. McFarland et al. [26], in a study of 106 cases of Erb’s / Duchenne’s palsy, estimated the risk of Erb’s / Duchenne’s palsy, for infants greater than 4500 g, to be increased (OR521). According to Benedetti and Gabbe [19], the decreased incidence of shoulder dystocia at Los Angeles County Hospital, from 0.3 to 0.13%, is a reflection of an increased incidence of CS for macrosomia and prolonged second stage of labour. Lipscomb et al. [10] found an incidence of shoulder dystocia of 18.5%, a rate within the range of other published reports [4,27,28], but the most impressive finding of these authors (and for us, with a rate of 14%) was that no permanent neurological sequelae were identified. Moreover, the resolution of all injuries occurred by two months of age. This would appear to be the most important conclusion in legitimizing the option of a trial of labour. Other authors who followed infants with shoulder dystocia reported minimal permanent injury or disability [29–31]. Lipscomb et al. [10] did not find any cases of birth asphyxia or severely depressed Apgar scores, as have been demonstrated in other studies [26,28]. 4.6. Vaginal delivery or cesarean section? Macrosomic infants are at increased risk for birth trauma, including shoulder dystocia and associated injuries [23–25]. However, there is no consensus regarding the ´ J. Berard et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 77 (1998) 51 – 59 management of macrosomic infants. Both physician and patient preferences can influence the management of suspected macrosomic infants. Many patients are multiparous, and want to avoid caesarean delivery, especially those who have experienced a previous uncomplicated vaginal delivery for a macrosomic infant. The obstetrician obviously has limited means to identify, with acceptable accuracy, the macrosomic fetus in the antenatal period and those at risk of shoulder dystocia in labour. Consequently, recommendation for management of pregnancies with macrosomic fetuses will, to some extent, be based on speculation. The rationale for allowing a trial of labour exists not only because of a patient’s preference, but also because of what has been published in the literature. When fetal macrosomia is suspected, the mother should be evaluated for CS. While the method of delivery should be individually considered in each case, we must take into account the high prevalence of fetal morbidity associated with the vaginal delivery of the macrosomic infant. One must be prepared to deal effectively with shoulder dystocia and post-partum haemorrhage, if vaginal delivery is chosen. With this information, the obstetrician can better detect and plan for the management of this problem. With the use of ultrasound scans, it is possible to predict fetal weight [2]. Elliott et al. [32] described an ultrasonographic macrosomia index. In this study, CS for delivery of all fetuses with chest–biparietal diameter .1.4 cm would have reduced the incidence of traumatic morbidity from 27 to 9%. However, physical and sonographic estimates of excessive fetal weight carry substantial margins of errors [33,34]. Furthermore, cesarean delivery introduces a significant risk of maternal morbidity. Therefore, it would seem prudent to scan all women in labour who are obese (over 90 kg), have diabetes or are post-mature, to determine an estimated fetal weight. Cesarean section has been prescribed for the delivery of the macrosomic fetus [4,13,35] in order to avoid fetal death or damage as a result of mechanical difficulties at vaginal delivery. Mechanical difficulties can occur with the head or the shoulders. Excessive efforts to extract the head with instruments can lead to skull fractures or intracranial haemorrhage, but these bad outcomes are avoidable if the obstetrician is prepared to turn to CS when extraction is difficult. Nowadays, the principal worry is shoulder dystocia, because it occurs in 10 to 15% of vaginal deliveries of babies over 4500 g [2,13,36,37] and because, in rare cases, it can be disastrous for the baby [2,36,38]. For Menticoglou et al. [4], the results of their study do not justify a policy of routine CS for all macrosomic babies to prevent mechanical difficulties at delivery. Instead, they feel that a prudent supervised trial of vaginal delivery is the preferred approach. Menticoglou et al. [4] said that most large babies are delivered without shoulder dystocia, that if shoulder dystocia develops, it is usually easily resolved by the experienced obstetrician without trauma to the baby, that if trauma develops it is usually 57 temporary and that, in desperate circumstances, abdominal delivery is still available to rescue the baby [39–41]. Infants estimated to be .4500 g should be delivered by primary CS according to O’Leary and Leonetti [39]. Spellacy et al. [2] recommend abdominal delivery for a weight of 5000 g, while Doc et al. [12] suggest an upper ¨ and limit of 4000 g. The results of the study by Rydhstrom Ingemarsson [21] indicate a very poor outcome for the extremely large fetus (.5700 g), if born vaginally, and they therefore advise abdominal delivery for extremely large fetuses. This recommendation can probably be approximated to fetuses with a birthweight exceeding 5000 g. Thus, a recommendation to deliver all women carrying a fetus with an estimated birthweight .4500 g abdominally is questionable. Sandmire and O’Hallion [29] estimated that, out of three babies born by CS due to an estimated birthweight .4500 g, only one actually fulfils that criterion. The same authors also calculated that 978 CSs would be required to avoid one case of persistent mild arm weakness as a complication of shoulder dystocia, assuming that most brachial plexus injuries are mild. If this is so, the maternal complications associated with CS would far outweigh any possible fetal benefits. Menticoglou et al. [4] reported that, during a ten-year period, no baby died or was permanently damaged as a consequence of mechanical difficulties at delivery in their hospital and that, if a policy of routine CS had been in place for the delivery of fetuses thought to be macrosomic, about 1000 additional CSs would have been done (most of the 590 vaginal deliveries of infants whose weights were correctly estimated to be .4500 g plus some 10% of the infants weighing ,4500 g whose weights were incorrectly estimated). Even though macrosomia has been clearly identified as a risk factor for shoulder dystocia, some 50–60% of shoulder dystocias occur in infants weighing less than 4000 g. This statistic, coupled with the inability to predict reliably which infants will experience permanent or significant injury, makes a blanket policy of elective CS for these patients seem unwarranted. Identification of specific intrapartum risk factors has further justified selected trial of labour in patients with suspected macrosomia. Benedetti and Gabbe [19] determined that a prolonged second stage with a mid-pelvic delivery was the most consistent intrapartum risk factor for shoulder dystocia in macrosomic infants. Acker et al. [28] further identified arrest disorders as predictive for infants weighing 4500 g or more. The objective of a trial of labour is to attain a safe vaginal delivery and avoid the complications associated with CS. If the objective is to prevent persistent infant morbidity, performing cesarean delivery on every potential patient at risk for shoulder dystocia, to avoid the rare case of permanent neurological sequelae, seems a high price for women and the health care system to pay. This data would suggest that, where patients are carefully selected and well-informed, shoulder dystocia is immediately recognized and skilfully managed. 58 ´ J. Berard et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 77 (1998) 51 – 59 5. Conclusion It must be acknowledged that, very rarely, a macrosomic fetus may die or be injured during delivery (particularly from shoulder dystocia), where CS would have prevented these sequelae. But many CSs would have to be performed to prevent one such case. Since the maternal mortality from elective CS is very low, however, performing more CSs to reduce the incidence of infant injury by one may be acceptable. This is the obstetrician’s dilemma; to balance a low fetal risk from vaginal delivery against a low maternal risk from CS. Because there is an error of about 10% in this estimate, it would seem reasonable to deliver all infants estimated to weigh more than 5000 g by CS to avoid fetal trauma. This practice could reduce the incidence of perinatal injury to macrosomic infants without greatly increasing the number of cesarean deliveries. From previous reports in the literature, we suggest the following as factors associated with shoulder dystocia: Multiparity, postdate delivery, previous large-for-dates fetus, maternal diabetes, excessive weight gain in pregnancy, palpably large fetus, advanced maternal age, fundal height greater than 40 cm and poor progress in the second stage of labour. A high level of suspicion that an infant weighs .4500 g is a reasonable indication for CS, if founded on an association with other known risk factors. The patients who fit the mnemonic DOPE (Diabetes, Obesity, Postdatism, Excessive fetal weight or maternal weight gain) are at greatest risk. Meticulous antenatal care consisting of frequent ultrasound examinations, glucose screening and dietary advice should help to reduce the incidence of macrosomia and to identify correctly the women at greatest risk. Researchers performing anthropometric studies postpartum on macrosomic infants described them as having a greater shoulder–head and chest–head disproportion, regardless of birth weight. Other authors, using ultrasonography on antepartum macrosomic infants, recommended the use of an ultrasound index to detect macrosomia. There are several problems regarding fetal macrosomia. First, a universally acceptable definition of fetal macrosomia must be adopted. Second, future studies must include the reporting and analysis of all those variables that may influence birth weight. [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] Acknowledgements We would like to thank Ms. C. Wattebot O’Brien (UMDS-London). [22] [23] References [1] Boyd ME, Usher RH, McLean FH. Fetal macrosomia: prediction, risks, proposed management. Obstet Gynecol 1983;61:715. [2] Spellacy WN, Miller S, Winegar A, Peterson PQ. Macrosomia — [24] [25] Maternal characteristics and infant complications. Obstet Gynecol 1985;66:158–61. Langer O, Berkus MD, Huff RW, Samueloff A. Shoulder dystocia: Should the fetus weighing $4000 grams be delivered by cesarean section?. Am J Obstet Gynecol 1991;165:831–7. Menticoglou SM, Manning FA, Morrison I, Harman CR. Must macrosomic fetuses be delivered by a caesarean section? A review of outcome for 786 babies .4500 g . Aust NZ J Obstet Gynaecol 1992;32:100–3. Jacobson JD, Cousins L. A population-based study of maternal and perinatal outcome in patients with gestational diabetes. Am J Obstet Gynecol 1989;161:981–6. Phillipou G. Relationship between normal oral glucose tolerance test in women at risk for gestational diabetes and large for gestational age infants. Diabetes Care 1991;14:1092–7. Sacks DA. Fetal macrosomia and gestational diabetes: what’s the problem?. Obstet Gynecol 1993;81:775–81. National Center for Health Statistics. Advanced report of final natality statistics 1992. Monthly Vital Statistic report. Vol 3. no. 5. Hyattsville, MD: Public Health Service, 1994. Posner AC, Friedman S, Posner LB. The large fetus. A study of 547 cases. Obstet Gynecol 1955;5:268–78. Lipscomb et al KR. The outcome of macrosomic infants weighing at least 4500 grams: Los Angeles County 1 University of southern California experience. Obstet Gynecol 1995;85:558–64. Sanderson DA, Wilcox MA, Johnson IR. Relative macrosomia identified by the individualization birthweight ratio. Acta Obstet Gynecol Scand 1994;73:246–9. Doc N, Mosberg H, Stern W, Jagani N, Schulman H. Complications in fetal macrosomia. NY State J Med 1984;84:302–5. Parks DG, Ziel HK. Macrosomia: A proposed indication for primary cesarean section. Obstet Gynecol 1978;52:407–9. Klebanoff MA, Mills JL, Berendes HW. Mother’s birth weight as a predictor of macrosomia. Am J Obstet Gynecol 1985;153:253. Sack RA. The large infant. A study of maternal, obstetric, fetal and newborn characteristics, including a long-term pediatric follow-up. Am J Obstet Gynecol 1969;104:195–204. Forchsbach G, Contreras-Soto JJ, Fong G, Flores G, Moreno O. Prevalence of gestational diabetes and macrosomic newborns in a Mexican population. Diabetes Care 1988;11:235–8. Madsen H, Ditzel J. The influence of maternal weight, smoking, vascular complications and glucose regulation on the birth weight of infants of type 1 diabetic women. Eur J Obstet Gynecol Reprod Biol 1991;39:175–9. Hod M, Merlob P, Friedman S, Schoenfeld A, Ovadia J. Gestational diabetes mellitus. A survey of perinatal complications in the 1980’s. Diabetes 1991;40:74–8. Benedetti TJ, Gabbe SG. Shoulder dystocia. A complication of fetal macrosomia and prolonged second stage of labour with midpelvic delivery. Obstet Gynecol 1978;52:526–9. Persson B, Stangenberg M, Lunell NO, et al. Prediction of size of infants at birth by measurement of symphysis fundus height. Br J Obstet Gynaecol 1986;93:206–11. ¨ H, Ingemarsson I. The extremely large fetus — antenatal Rydhstrom identification risks and proposed management. Acta Obstet Gynecol Scand 1989;68:59–63. ´ Balouet P, Speckel D, Herlicoviez M. Estimation echographique du ´ ˆ de la mesure de la graisse des membres. J poids foetal. Interet Gynecol Obstet Biol Reprod 1992;21:795–802. Iffy L, Varadi V, Jakobovitz A. Common intrapartum denominators of shoulder dystocia related birth injuries. Zentralbl Gynakol 1994;116:33–7. Mikulandra F, Stojnic E, Perisa M, Merlak I, Sikic D, Zenic N. Fetal macrosomia, pregnancy and delivery . Zentralbl Gynakol 1993;115:553–61. Rasmussen BR, Mosgaard KE. Macrosomia: diagnosis, delivery and complications . Ugeskr Laeger 1993;155:3185–90. ´ J. Berard et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 77 (1998) 51 – 59 [26] McFarland LV, Raskin M, Daling JR, Benedetti TJ. Erb / Duchenne’s palsy: A consequence of fetal macrosomia and method of delivery. Obstet Gynecol 1986;68:784–8. [27] El Madany AA, Jallad KB, Radi FA, Elhamdam H, O’Deh HM. Shoulder dystocia: anticipation and outcome. Int J Gynaecol Obstet 1990;34:7–12. [28] Acker DB, Sachs BP, Friedman EA. Risks factors for shoulder dystocia. Obstet Gynecol 1985;66:762–8. [29] Sandmire MF, O’Hallion TJ. Shoulder dystocia: Its incidence and associated risks factors. Int J Gynaecol Obstet 1988;26:65–73. [30] Nocon JJ, McKenzie DK, Thomas LJ, Hansell RS. Shoulder dystocia: An analysis of risks and obstetric maneuvers. Am J Obstet Gynecol 1993;168:1732–9. [31] Tan KL. Brachial palsy . Br J Obstet Gynaecol 1973;80:60–2. [32] Elliott JP, Garite TJ, Freeman RJ, McQuown DS, Patel JM. Ultrasonic prediction of fetal macrosomia in diabetic patients. Obstet Gynecol 1982;60:159–62. [33] Delpapa EH, Mueller-Heubach E. Pregnancy outcome following ultrasound diagnosis of macrosomia. Obstet Gynecol 1991;78:340–3. 59 [34] Levine AB, Lockwood CJ, Brown B, Lapinski R, Berkowitz RL. Sonographic diagnosis of the large for gestational age fetus at term: does it make a difference?. Obstet Gynecol 1992;79:55–8. [35] Gross TL, Sokol RJ, Williams T, Thompson K. Shoulder dystocia: a fetal–physician risk. Am J Obstet Gynecol 1987;156:1408–18. [36] Jaschevatzky OE, Mor G, Miller et al M. Risk in the vaginal delivery of the large fetus. Aust NZ J Obstet Gynaecol 1984;24:178–81. [37] Hassan AA. Shoulder dystocia: risk factors and prevention . Aust NZ J Obstet Gynaecol 1988;28:107–9. [38] Schramm M. Impacted shoulders: a personal experience . Aust NZ J Obstet Gynaecol 1983;23:28–31. [39] O’Leary JA, Leonetti HB. Shoulder dystocia: Prevention and treatment. Am J Obstet Gynecol 1990;162:5–9. [40] Sandberg EC. The zavanelli maneuver: a potentially revolutionary method for the resolution of shoulder dystocia. Am J Obstet Gynecol 1985;152:479–88. [41] Dimitry ES. Cephalic replacement: a desperate solution for shoulder dystocia. J Obstet Gynaecol 1989;10:49–50.