Aust. N.Z.J. Obsiei. Gynaec. (1979) 19: 28
Shoulder Dystocia: A Study of 47 Cases
Neil R. Johnstone’ Royal Women’s Hospital, Melbourne
Summary: Shoulder dystocia at vaginal delivery occurred in 0.2% of cases. Antenatal prediction of this complication was very difficult. In primigravidae in labour, delay late in the first stage was a warning sign; induction of labour, the use of regional analgesia, and forceps delivery for delay in the second stage of labour were associated with subsequent shoulder dystocia, although a cause-effect relationship was not established.
The immediate maternal and fetal morbidity were high, nearly 30% of the babies suffering from a severe neural or bony injury. Forty-four per cent of the babies were beyond 41 weeks of gestation, and 70% weighed over 4,000 g. The avoidance of postmaturity by elective induction of labour at 41 weeks of gestation would significantly reduce the incidence of shoulder impaction at delivery.
Impaction of the fetal shoulders at vaginal delivery is a frightening and potentially lethal complication which requires immediate action. A classic description of the clinical picture has been given by Morris (1955), and the subject was recently reviewed by Dignam (1976). Because of the rarity of this complication, few individuals have had significant experience with it. This report presents the results of a retrospective study of 47 cases of shoulder dystocia occurring during 22,663 deliveries at The Royal Women’s Hospital, Melbourne, between 1973 and 1977. RESULTS
Maternal Data The ratio of 18 primigravidae to 29 multigravidae, along with the nationality, average height and average booking weight of the patients was the same as the remainder of our hospital population. Where the 1. Assistant Obstetrician.
patient had previously delivered a baby of 37 weeks of gestation or more, the average weight of the babies delivered was 3,3998. This is not significantly difFerent from our general multigravid hospital population. Because of the tendency to late booking amongst our patients, maternal weight gain during pregnancy was found to be of no value in predicting shoulder dystocia. In only 5 cases was some comment made in the patient’s notes about the large size of the baby. Only 2 of the multiparae had a past history of shoulder dystocia at a previous delivery. Seven patients had reduced diameters on X-ray pelvimetry performed before or after labour. However, all but one of these delivered a baby weighing 4,OOOg. or more. There were 3 patients in this study who were known to have diabetes mellitus. One further patient with a normal glucose tolerance curve, was labelled as “prediabetic” on clinical grounds. In addition, 5 patients had a family history of diabetes. Only 9 glu-
NEILR. JOHNSTONE
Table 1. Manipulations Used to Deliver Babies with Impacted Shoulders No. of patients
14 21 12
Additional traction on fetal head Suprapubic pressure and episiotomy Delivery of posterior arm and episiotomy
Table 2. Analgesia Used During Manipulations to Deliver Babies with Impacted Shoulders Nitrous oxide/oxygen alone Perineal infiltration with local anaesthetic Bilateral pudendal nerve block Epidural analgesia General anaesthesia Combinations of above
8 6 9 12 2 10
cose tolerance tests were recorded in the case notes either before or after delivery and these were all within normal limits. Labour was induced in 10 patients, significantly more than expected (15% amongst standard ward patients). Epidural analgesia was administered to 11 (22%) patients compared with a 6% utilization rate in the hospital population as a whole. Twelve of the 18 primigravid patients ( 6 7 % ) laboured for more than 12 hours, compared with 36% of the normal population. When further analysed, prolonged labours in primigravidae whose babies suffered shoulder dystocia were mainly due to slowing at the end of the first stage. The longest labour lasted 42 hours in a primigravida originally planned for Caesarean section. Five multigravid patients laboured for more than 12 hours. This figure would be expected in our multigravid hospital population. The second stage of labour lasted more than 1 hour in 8 primigravidae and more than half an hour in 11 multigravidae. The duration of the second stage of labour in both primigravidae and multigravidae was significantly longer than in our general hospital population. Twenty-five patients were delivered with forceps, twice the expected rate; in 18, the indication for forceps was delay in the second stage of labour and in 13 of these rotation of the fetal head from the occipito-posterior position was required. Normally,
29
posterior position of the occiput occurs in about 4 to 5 % of our general hospital population. The manoeuvres required to deliver babies with impacted shoulders are shown in table 1. An episiotomy was cut in 10 of the 14 patients delivered by additional traction on the fetal head. In the other 4, the perineum remained intact in 2 and a perineal tear occurred in 2. Overall, an episiotomy had been made in 43 of the 47 patients. The analgesia used during manipulations to deliver the shoulders is shown in table 2. Some form of conduction analgesia was used in 3 0 of the 47 patients. The commonest combination was nitrous oxide/ oxygen together with perineal infiltration with local anaesthetic. The status of the operator at the time of shoulder impaction and the status of the person subsequently summoned to deal with the problem are shown in table 3. Fetal Data The weights of the babies are shown in figure 1. The average weight was 4,3068. There were 26 boys (average weight 4,435g.), and 2 1 girls, (average weight 4,146g.). Thirty-three of the babies weighed over 4,000 g, and only 3 weighed less than 3,500 g. There were 2 patients in this study where impaction of the shoulders occurred at successive deliveries. These patients were also responsible for the largest (5,390g.) and the second largest (5,2508.) babies in the series. The largest baby in the series
20
17
m15W
-
m a m
11
$lo-
U
W
m Table 3. Status of Operator Delivering Babies with Impacted Shoulders Most senior person present at the time of shoulder impaction
Student midwife or student doctor Trained midwife Obstetric resident medical officer Obstetric registrar Consultant
Most senior person present at the time of delivery of the shoulders
16 0
0 2
25 6 0
29 11 5
z 3 z
5-
3
C3500
1
' 3501 - ' 4001 - ' 4501 - '> 5000 4000
4500
5000
BIRTHWEIGHT (9)
Figure 1. Birthweights of Babies whose shoulders impacted at delivery.
30
AUST.AND N.Z. JOURNAL OF OBSTETRICS AND GYNAECOLOCY
Table 4. Gestation of Babies Whose Shoulders Impacted at Deliverv Gestation in weeks and days Less than 39e 40 to 40a 41 to 41' 42 to 42' 43 or more
No. of babies 6 20
10 6 5
was delivered, apparently unscathed, through a pelvis which was found to have a true conjugate of only 10.6 cm on subsequent X-ray pelvimetry. The gestation of babies at the time of shoulder impaction is shown in table 4. Forty-one of the babies were overdue on dates, including 5 beyond 4 3 weeks of gestation.
Maternal Morbidity There were no maternal deaths in the series. Eleven patients (23 % ) had a postpartum haemorrhage (a blood loss of 600ml. or more), compared with the hospital rate of 4% ; another 6 patients lost between 500 and 600ml. Eight patients required blood transfusion; included in this group were 3 patients with extensive lateral vaginal wall tears (one of these developed a broad ligament haematoma). There were also 2 second degree perineal tears. Of the 25 patients delivered with forceps, 9 had a postpartum haemorrhage. Thirteen patients required antimicrobial chemotherapy in the puerperium because of genital or urinary tract infection. One patient showed transient signs of sacral nerve root compression after delivery. Fetal Morbidity Three babies in this study were stillborn, but all 3 had been dead for more than 24 hours before delivery. All were macroscopically normal, and post mortem reports revealed intrauterine hypoxia. Six babies sustained fractured clavicles, one bilateral. Three others sustained a fractured humerus, and in each of these a loud "crack" was heard by the attendants. There were 6 babies with Erb's palsy, all resolving at the time of discharge. Three further babies suffered injuries related to forceps delivery: one had a 6th cranial nerve palsy, one a 7th cranial nerve palsy and the third a serve laceration over the maxillary area in the region of application of the forcep blade. There were 32 babies ( 6 8 % ) whose Apgar score at 1 minute after birth was 7 or less. This is significantly different from the general hospital population where only 33% of babies have an Apgar score of 7 or less at 1 minute after birth. Five minutes after birth, only 3 babies still had Apgar scores of 7 or less; this figure is no different from the general hospital population.
DISCUSSION
In this series shoulder dystocia occurred in 0.2% of deliveries. This agrees closely with most other reports in the literature (Dignam, 1976). Since approximately 7 % of babies born in the hospital weigh over 4,OOOg., the chance of such a baby impacting its shoulders at delivery is approximately 1 in 50. Prediction of shoulder dystocia antenatally was uncommon. Many of the accepted criteria, such as previous large babies, excessive maternal weight gain, and palpably large fetus (Bolton, 1959; Dignam, 1976), proved unreliable. However, in primigravidae in labour, slow progress at the end of first stage warned of trouble ahead. This was especially so when coupled with failure of the fetal head to descend in the second stage. Such patients need careful reappraisal before vaginal delivery is contemplated. Maternal diabetes mellitus was not a major feature of this series, but during the period in question most diabetic patients were induced well before term (Jeffery et al., 1977). With the recent overseas trend in inducing diabetic patients closer to term (Essex, 1976), mechanical difficulties at delivery may again become a problem. In the present study, no case of maternal diabetes mellitus was diagnosed as a result of a glucose tolerance test performed because of impacted shoulders. Shoulder dystocia at delivery was often preceded by induction of labour, the use of epidural analgesia and forceps delivery for delay in the second stage of labour. Considering the size of the babies and the length of the labours (especially in primigravidae), this is not surprising. However, some authorities have suggested that induction of labour and the use of regional analgesia may cause shoulder dystocia (Seigworth, 1966). In this situation it is almost impossible to establish the correct sequence of events. One usually accepts the delivery of a dead baby as a fairly straightforward procedure, but in this series 3 such babies developed shoulder dystocia. This phenomenon, although recorded (McCalI, 1962; Seigworth, 1966), is not sufficiently emphasized in the literature. As would be expected in a large obstetric training hospita1, over 60% of the deliveries were handled by resident medical staff although there was always a consultant on call. Travelling time in the city is such that a consultant was able to be present at the delivery in only 5 patients. This emphasizes the fact that every person practising obstetrics should have a plan of action which is immediately instituted if shoulder impaction occurs. The present results show that cutting an adequate episiotomy is imperative. A severe fetal injury resulted in 3 of the 4 patients where this was not
NEILR. JOHNSTONE
done. Cutting or extending the episiotomy may in itself result in the delivery of the baby, but if not one is faced with a very difficult situation. Many manoeuvres have been described to deal with the problem, including rotation of the shoulders through 180 degrees (Woods, 1943), adduction of the shoulders (Rubin, 1964), and forming a channel with the fingers for the anterior shoulder (Heery, 1963). In practice, the space available for such manoeuvres is extremely limited, and many of them have not found favour in our institution. If cutting the episiotomy alone is not enough, we favour suprapubic pressure as the next step. It is less traumatic to the mother than bringing down the posterior arm and suprapubic pressure requires minimal analgesia in a situation where time is important. At all times, one must guard against excessive angulation of the fetal neck (Swartz, 1960). If the above manipulations fail, and provided there is adequate analgesia and a wide episiotomy, the posterior arm of the fetus should be brought down (Schwartz and Dixon, 1958). Elective cleidotomy on a live baby (McCall, 1962) is not recommended for fear of injury to deeper structures. The immediate maternal and fetal morbidity of shoulder impaction are clearly illustrated in this study. Twenty-three per cent of the mothers had a postpartum haemorrhage, including 17% who required blood transfusion. Twenty-seven per cent needed puerperal antibiotics. Of the babies, 28% sustained a severe neural or bony injury. It is noteworthy that an audible “crack” at such a delivery usually heralded a fractured humeius and not a fractured clavicle : radiographs should include the fetal arms. Lesser degrees of fetal damage may be reflected in the low Apgar scores 1 minute after birth, although most of the babies responded well to resuscitation. The nature of this study does not
31
permit long term follow-up, but McCall (1962) has suggested that permanent CNS impairment may occur in babies whose shoulders impact at birth. In the present series, a major factor in the development of shoulder dystocia at birth was the size of the baby. This, in turn, was partly related to the gestation period. Forty-four per cent of the babies were beyond 41 weeks of gestation, and 70% weighed over 4,000 g. The size of the maternal pelvis was less important. Women known to have contracted pelves do not impact smaller babies. This has been previously reported (Seigworth, 1966), and is due to the discrepancy between trunk girth and head size in large babies. The avoidance of postmaturity by elective induction of labour at 41 weeks of gestation would reduce the incidence of shoulder dystocia by nearly onehalf.
Acknowledgements I wish to thank the members of the medical staff of The Royal Women’s Hospital for access to the records of the patients under their care. All figures for the general hospital population have been taken from computer summary sheets and clinical records of the Hospital during the years 1973 to 1977 unless otherwise stated.
References Bolton, R. N. (1959), Amer. J . Obstet. Gynec., 77: 118. Dignam, W. J. (1976), Clin. Obstet. Gynec., 19: 577. Essex, N. (1976), Brit. J . hosp. Med., 15: 333. Heery, R. D. (1963), Obstet. and Gynec., 22: 360. Jeffery, P., Martin, F. I. R., et al. (1977), Med. J . Aust., 2: 41. McCall, J. 0. (1962), Amer. J . Obstet. Gynec., 83: 1486. Morris, W. I. C . (1955), J . Obstet. Gynaec. Brit. Emp., 62: 302. Rubin, A. (1964), J . Amer. med. Ass., 189: 835. Schwartz, B. C., and Dixon, D. M. (1958), Obstet. and Gynec., 11: 468. Seigworth, G. R. (1966), Obstet. and Gynec., 28: 764. Swartz, D. P. (1960), Obstet. and Gynec., 15: 194. Woods, C. E. (1943), Amer. J . Obstet. Gynec., 45: 796.
Shoulder Dystocia: A Study of 47 Cases
Neil R. Johnstone’ Royal Women’s Hospital, Melbourne
Summary: Shoulder dystocia at vaginal delivery occurred in 0.2% of cases. Antenatal prediction of this complication was very difficult. In primigravidae in labour, delay late in the first stage was a warning sign; induction of labour, the use of regional analgesia, and forceps delivery for delay in the second stage of labour were associated with subsequent shoulder dystocia, although a cause-effect relationship was not established.
The immediate maternal and fetal morbidity were high, nearly 30% of the babies suffering from a severe neural or bony injury. Forty-four per cent of the babies were beyond 41 weeks of gestation, and 70% weighed over 4,000 g. The avoidance of postmaturity by elective induction of labour at 41 weeks of gestation would significantly reduce the incidence of shoulder impaction at delivery.
Impaction of the fetal shoulders at vaginal delivery is a frightening and potentially lethal complication which requires immediate action. A classic description of the clinical picture has been given by Morris (1955), and the subject was recently reviewed by Dignam (1976). Because of the rarity of this complication, few individuals have had significant experience with it. This report presents the results of a retrospective study of 47 cases of shoulder dystocia occurring during 22,663 deliveries at The Royal Women’s Hospital, Melbourne, between 1973 and 1977. RESULTS
Maternal Data The ratio of 18 primigravidae to 29 multigravidae, along with the nationality, average height and average booking weight of the patients was the same as the remainder of our hospital population. Where the 1. Assistant Obstetrician.
patient had previously delivered a baby of 37 weeks of gestation or more, the average weight of the babies delivered was 3,3998. This is not significantly difFerent from our general multigravid hospital population. Because of the tendency to late booking amongst our patients, maternal weight gain during pregnancy was found to be of no value in predicting shoulder dystocia. In only 5 cases was some comment made in the patient’s notes about the large size of the baby. Only 2 of the multiparae had a past history of shoulder dystocia at a previous delivery. Seven patients had reduced diameters on X-ray pelvimetry performed before or after labour. However, all but one of these delivered a baby weighing 4,OOOg. or more. There were 3 patients in this study who were known to have diabetes mellitus. One further patient with a normal glucose tolerance curve, was labelled as “prediabetic” on clinical grounds. In addition, 5 patients had a family history of diabetes. Only 9 glu-
NEILR. JOHNSTONE
Table 1. Manipulations Used to Deliver Babies with Impacted Shoulders No. of patients
14 21 12
Additional traction on fetal head Suprapubic pressure and episiotomy Delivery of posterior arm and episiotomy
Table 2. Analgesia Used During Manipulations to Deliver Babies with Impacted Shoulders Nitrous oxide/oxygen alone Perineal infiltration with local anaesthetic Bilateral pudendal nerve block Epidural analgesia General anaesthesia Combinations of above
8 6 9 12 2 10
cose tolerance tests were recorded in the case notes either before or after delivery and these were all within normal limits. Labour was induced in 10 patients, significantly more than expected (15% amongst standard ward patients). Epidural analgesia was administered to 11 (22%) patients compared with a 6% utilization rate in the hospital population as a whole. Twelve of the 18 primigravid patients ( 6 7 % ) laboured for more than 12 hours, compared with 36% of the normal population. When further analysed, prolonged labours in primigravidae whose babies suffered shoulder dystocia were mainly due to slowing at the end of the first stage. The longest labour lasted 42 hours in a primigravida originally planned for Caesarean section. Five multigravid patients laboured for more than 12 hours. This figure would be expected in our multigravid hospital population. The second stage of labour lasted more than 1 hour in 8 primigravidae and more than half an hour in 11 multigravidae. The duration of the second stage of labour in both primigravidae and multigravidae was significantly longer than in our general hospital population. Twenty-five patients were delivered with forceps, twice the expected rate; in 18, the indication for forceps was delay in the second stage of labour and in 13 of these rotation of the fetal head from the occipito-posterior position was required. Normally,
29
posterior position of the occiput occurs in about 4 to 5 % of our general hospital population. The manoeuvres required to deliver babies with impacted shoulders are shown in table 1. An episiotomy was cut in 10 of the 14 patients delivered by additional traction on the fetal head. In the other 4, the perineum remained intact in 2 and a perineal tear occurred in 2. Overall, an episiotomy had been made in 43 of the 47 patients. The analgesia used during manipulations to deliver the shoulders is shown in table 2. Some form of conduction analgesia was used in 3 0 of the 47 patients. The commonest combination was nitrous oxide/ oxygen together with perineal infiltration with local anaesthetic. The status of the operator at the time of shoulder impaction and the status of the person subsequently summoned to deal with the problem are shown in table 3. Fetal Data The weights of the babies are shown in figure 1. The average weight was 4,3068. There were 26 boys (average weight 4,435g.), and 2 1 girls, (average weight 4,146g.). Thirty-three of the babies weighed over 4,000 g, and only 3 weighed less than 3,500 g. There were 2 patients in this study where impaction of the shoulders occurred at successive deliveries. These patients were also responsible for the largest (5,390g.) and the second largest (5,2508.) babies in the series. The largest baby in the series
20
17
m15W
-
m a m
11
$lo-
U
W
m Table 3. Status of Operator Delivering Babies with Impacted Shoulders Most senior person present at the time of shoulder impaction
Student midwife or student doctor Trained midwife Obstetric resident medical officer Obstetric registrar Consultant
Most senior person present at the time of delivery of the shoulders
16 0
0 2
25 6 0
29 11 5
z 3 z
5-
3
C3500
1
' 3501 - ' 4001 - ' 4501 - '> 5000 4000
4500
5000
BIRTHWEIGHT (9)
Figure 1. Birthweights of Babies whose shoulders impacted at delivery.
30
AUST.AND N.Z. JOURNAL OF OBSTETRICS AND GYNAECOLOCY
Table 4. Gestation of Babies Whose Shoulders Impacted at Deliverv Gestation in weeks and days Less than 39e 40 to 40a 41 to 41' 42 to 42' 43 or more
No. of babies 6 20
10 6 5
was delivered, apparently unscathed, through a pelvis which was found to have a true conjugate of only 10.6 cm on subsequent X-ray pelvimetry. The gestation of babies at the time of shoulder impaction is shown in table 4. Forty-one of the babies were overdue on dates, including 5 beyond 4 3 weeks of gestation.
Maternal Morbidity There were no maternal deaths in the series. Eleven patients (23 % ) had a postpartum haemorrhage (a blood loss of 600ml. or more), compared with the hospital rate of 4% ; another 6 patients lost between 500 and 600ml. Eight patients required blood transfusion; included in this group were 3 patients with extensive lateral vaginal wall tears (one of these developed a broad ligament haematoma). There were also 2 second degree perineal tears. Of the 25 patients delivered with forceps, 9 had a postpartum haemorrhage. Thirteen patients required antimicrobial chemotherapy in the puerperium because of genital or urinary tract infection. One patient showed transient signs of sacral nerve root compression after delivery. Fetal Morbidity Three babies in this study were stillborn, but all 3 had been dead for more than 24 hours before delivery. All were macroscopically normal, and post mortem reports revealed intrauterine hypoxia. Six babies sustained fractured clavicles, one bilateral. Three others sustained a fractured humerus, and in each of these a loud "crack" was heard by the attendants. There were 6 babies with Erb's palsy, all resolving at the time of discharge. Three further babies suffered injuries related to forceps delivery: one had a 6th cranial nerve palsy, one a 7th cranial nerve palsy and the third a serve laceration over the maxillary area in the region of application of the forcep blade. There were 32 babies ( 6 8 % ) whose Apgar score at 1 minute after birth was 7 or less. This is significantly different from the general hospital population where only 33% of babies have an Apgar score of 7 or less at 1 minute after birth. Five minutes after birth, only 3 babies still had Apgar scores of 7 or less; this figure is no different from the general hospital population.
DISCUSSION
In this series shoulder dystocia occurred in 0.2% of deliveries. This agrees closely with most other reports in the literature (Dignam, 1976). Since approximately 7 % of babies born in the hospital weigh over 4,OOOg., the chance of such a baby impacting its shoulders at delivery is approximately 1 in 50. Prediction of shoulder dystocia antenatally was uncommon. Many of the accepted criteria, such as previous large babies, excessive maternal weight gain, and palpably large fetus (Bolton, 1959; Dignam, 1976), proved unreliable. However, in primigravidae in labour, slow progress at the end of first stage warned of trouble ahead. This was especially so when coupled with failure of the fetal head to descend in the second stage. Such patients need careful reappraisal before vaginal delivery is contemplated. Maternal diabetes mellitus was not a major feature of this series, but during the period in question most diabetic patients were induced well before term (Jeffery et al., 1977). With the recent overseas trend in inducing diabetic patients closer to term (Essex, 1976), mechanical difficulties at delivery may again become a problem. In the present study, no case of maternal diabetes mellitus was diagnosed as a result of a glucose tolerance test performed because of impacted shoulders. Shoulder dystocia at delivery was often preceded by induction of labour, the use of epidural analgesia and forceps delivery for delay in the second stage of labour. Considering the size of the babies and the length of the labours (especially in primigravidae), this is not surprising. However, some authorities have suggested that induction of labour and the use of regional analgesia may cause shoulder dystocia (Seigworth, 1966). In this situation it is almost impossible to establish the correct sequence of events. One usually accepts the delivery of a dead baby as a fairly straightforward procedure, but in this series 3 such babies developed shoulder dystocia. This phenomenon, although recorded (McCalI, 1962; Seigworth, 1966), is not sufficiently emphasized in the literature. As would be expected in a large obstetric training hospita1, over 60% of the deliveries were handled by resident medical staff although there was always a consultant on call. Travelling time in the city is such that a consultant was able to be present at the delivery in only 5 patients. This emphasizes the fact that every person practising obstetrics should have a plan of action which is immediately instituted if shoulder impaction occurs. The present results show that cutting an adequate episiotomy is imperative. A severe fetal injury resulted in 3 of the 4 patients where this was not
NEILR. JOHNSTONE
done. Cutting or extending the episiotomy may in itself result in the delivery of the baby, but if not one is faced with a very difficult situation. Many manoeuvres have been described to deal with the problem, including rotation of the shoulders through 180 degrees (Woods, 1943), adduction of the shoulders (Rubin, 1964), and forming a channel with the fingers for the anterior shoulder (Heery, 1963). In practice, the space available for such manoeuvres is extremely limited, and many of them have not found favour in our institution. If cutting the episiotomy alone is not enough, we favour suprapubic pressure as the next step. It is less traumatic to the mother than bringing down the posterior arm and suprapubic pressure requires minimal analgesia in a situation where time is important. At all times, one must guard against excessive angulation of the fetal neck (Swartz, 1960). If the above manipulations fail, and provided there is adequate analgesia and a wide episiotomy, the posterior arm of the fetus should be brought down (Schwartz and Dixon, 1958). Elective cleidotomy on a live baby (McCall, 1962) is not recommended for fear of injury to deeper structures. The immediate maternal and fetal morbidity of shoulder impaction are clearly illustrated in this study. Twenty-three per cent of the mothers had a postpartum haemorrhage, including 17% who required blood transfusion. Twenty-seven per cent needed puerperal antibiotics. Of the babies, 28% sustained a severe neural or bony injury. It is noteworthy that an audible “crack” at such a delivery usually heralded a fractured humeius and not a fractured clavicle : radiographs should include the fetal arms. Lesser degrees of fetal damage may be reflected in the low Apgar scores 1 minute after birth, although most of the babies responded well to resuscitation. The nature of this study does not
31
permit long term follow-up, but McCall (1962) has suggested that permanent CNS impairment may occur in babies whose shoulders impact at birth. In the present series, a major factor in the development of shoulder dystocia at birth was the size of the baby. This, in turn, was partly related to the gestation period. Forty-four per cent of the babies were beyond 41 weeks of gestation, and 70% weighed over 4,000 g. The size of the maternal pelvis was less important. Women known to have contracted pelves do not impact smaller babies. This has been previously reported (Seigworth, 1966), and is due to the discrepancy between trunk girth and head size in large babies. The avoidance of postmaturity by elective induction of labour at 41 weeks of gestation would reduce the incidence of shoulder dystocia by nearly onehalf.
Acknowledgements I wish to thank the members of the medical staff of The Royal Women’s Hospital for access to the records of the patients under their care. All figures for the general hospital population have been taken from computer summary sheets and clinical records of the Hospital during the years 1973 to 1977 unless otherwise stated.
References Bolton, R. N. (1959), Amer. J . Obstet. Gynec., 77: 118. Dignam, W. J. (1976), Clin. Obstet. Gynec., 19: 577. Essex, N. (1976), Brit. J . hosp. Med., 15: 333. Heery, R. D. (1963), Obstet. and Gynec., 22: 360. Jeffery, P., Martin, F. I. R., et al. (1977), Med. J . Aust., 2: 41. McCall, J. 0. (1962), Amer. J . Obstet. Gynec., 83: 1486. Morris, W. I. C . (1955), J . Obstet. Gynaec. Brit. Emp., 62: 302. Rubin, A. (1964), J . Amer. med. Ass., 189: 835. Schwartz, B. C., and Dixon, D. M. (1958), Obstet. and Gynec., 11: 468. Seigworth, G. R. (1966), Obstet. and Gynec., 28: 764. Swartz, D. P. (1960), Obstet. and Gynec., 15: 194. Woods, C. E. (1943), Amer. J . Obstet. Gynec., 45: 796.
