Internal podalic version for neglected shoulder presentation with fetal demise


Dr NN Mahajan, 31, Dhanvantri Nagar, Sevagram, Wardha, Maharashtra, 442102 India. Email


In modern obstetrics, the role of internal podalic version (IPV) is limited to delivery of the second twin. A retrospective study was conducted to assess the efficacy of IPV in singleton neglected shoulder presentation with fetal demise. Women with live fetuses, previous CS or contracted pelvis were excluded. The procedure involved repositioning the prolapsed hand under anaesthetic followed by breech extraction. 12 women were identified over a 19 month period and all underwent successful IPV. One woman had a postpartum haemorrhage. We conclude that, in singleton pregnancies with a transverse lie, IPV has a role to play in the delivery of dead fetuses.


Neglected shoulder presentation or transverse lie generally refers to the series of complications that arise out of a shoulder presentation which has remained untreated for many hours of active labour. In this situation, the shoulder becomes wedged and impacted into the pelvis and the arm frequently prolapses through the vagina, becoming swollen and cyanosed. A premature fetus may still go on to deliver spontaneously (fetus condulicatus), but with larger babies this is not possible and with time the uterus becomes atonic and septic. In some women, the uterus continues to contract strongly and the lower segment finally ruptures leading to dehydration, keto-acidosis, shock and sepsis. With proper intrapartum care, this condition is completely avoidable. Unfortunately, however, it is still observed in many developing countries, especially in rural areas.

In modern obstetrics, caesarean section (CS) is the method of choice for the delivery of babies in a transverse lie. The role of internal podalic version (IPV) followed by breech extraction is usually limited to malpresentation or abnormal lie of the second twin in the presence of an experienced operator. Traditional teaching advises against IPV for neglected shoulder presentation because of a high risk of uterine rupture; CS or decapitation is instead advised in this situation.1 However, performing an unnecessary CS for a dead baby is not without risks: in the short term there is a high risk of haemorrhage and infection, and in subsequent pregnancies there is a high risk of scar rupture due to the absence of proper transport and referral services. In our hospital, therefore, we offer IPV to women for the management of neglected shoulder presentation with fetal demise. Women with a previous CS, live fetuses or clinically contracted pelvises are not offered this management.

The present study was conducted in Padhar Hospital, a tertiary referral centre in Betul, an underdeveloped district of Madhyapradesh state in India. Betul has a population of approximately 1 million, comprising of mainly socio-economically disadvantaged people. For health services, the population is dependent on one District Hospital (the only place where CS can be performed) and primary health centres (PHC). In the district, there are one or two non-specialist doctors per 20 000–30 000 people and one nurse–midwife per 3000 people working at PHC. In addition, unqualified practitioners frequently provide medical treatment in the villages. Being a poor area, most deliveries in this area are conducted by indigenous untrained midwives (Dai) at home.


In this retrospective study, cases were identified by searching through delivery records and case files. All cases where IPV was performed for neglected shoulder presentation with intrauterine fetal demise (IUFD) at Padhar Hospital over a 19-month period from March 2006 to September 2007 were identified. Permission to both conduct IPVs in these women (Padhar Hospital/MSO/68-2006) and to analyse the data (Padhar Hospital/MSO/247-2007) was granted by the hospital ethics committee. Details of the cases were collated manually and are outlined below. Total delivery numbers were taken from routine hospital statistics.

For the procedure, intravenous ketamine hydrochloride 30 mg was used, with additional intravenous nitroglycerin (IV-NTG, 50 microgram boluses every 2–3 minutes) to provide uterine relaxation. However, if the systolic blood pressure (SBP) was ≤100 mmHg, a standard general anaesthesia (GA) using rapid intubation with a cuffed oro-tracheal tube and halothane, oxygen and nitrous oxide maintenance was used instead.

All IPVs were performed by two obstetricians (NM and NG) using a standard method. After emptying the bladder using a rubber catheter, the prolapsed hand was gently repositioned inside the relaxed uterus with the aid of an episiotomy where needed. If the fetal legs were on the right side of the mother, the right hand was introduced in a cone shaped manner keeping the back of the hand against the uterine wall until it reached the podalic pole. The left hand was used for legs to the left. The foot was grasped (confirmed by palpating the fetal heel), brought down by steady traction and the delivery was completed by breech extraction. During this procedure, the external hand was sometimes used on the abdominal wall to make the leg more accessible. Routine uterine exploration was performed after placental delivery to rule out uterine rupture. Trans-abdominal massage of the uterus and an intravenous infusion of 10 units of oxytocin in 500 ml physiological saline over 4 hours were given after placental delivery in all patients. All women received broad spectrum antibiotics on admission and for 5 days thereafter.


There were 1012 deliveries in the hospital over the data collection period. Twelve women who underwent IPV for neglected shoulder presentation with IUFD were identified. In addition, during this period there were two other women where a CS was performed as the fetus was still alive. There were none with a previous CS or contracted pelvis. In all cases, a bedside ultrasound was initially conducted to confirm IUFD and to exclude uterine rupture. However, none of the 14 women admitted with uterine rupture during the study period had a neglected shoulder presentation as a cause for their rupture.

Internal podalic version was successful (defined as vaginal delivery without traumatic uterine rupture) in all 12 women and their details are summarized in Table 1. All were from the rural population and none were booked. Three were referred by the District Hospital, one woman was self-referred and 11 women were brought to the hospital by their Dai when they developed hand or cord prolapse. The mean (and standard deviation) duration of time that had elapsed from arm prolapse to hospital admission was 5 ± 1.7 hours. Six women were admitted after ≥5 hours of arm prolapse and a retraction ring was present in all of them. All were illiterate, and they could neither recall their last menstrual period nor had they undergone ultrasound during their pregnancy. Their estimated gestational age was calculated from the local festivals near to their last menses. On this basis, only two women were <36 weeks of gestation.

Table 1.   Patients’ profile and management
No.Parity and gestationDuration hand prolapse (hours)Cord prolapseDehydrationSepsisHaematuriaBP α (mmHg)Retraction ringCervical dilatation (cm)AnaesthesiaProcedure time (min)Fetal weight (kg)Complications
  1. PPH, postpartum haemorrhage; IPV, internal podalic version; BT, blood transfusion. *Blackish discolouration of anterior vaginal wall; **episiotomy; α on admission.

 1P4; 37 wk4+++90/608GA202.7
 2P3; 38 wk6++++94/608GA232.8PPH
 3P3; 36 wk5+++86/64+8GA182.6
 4P2; 40 wk3+100/627GA261.75
 5P3; 39 wk7+++94/648GA243.5
 6P3; 35 wk6++++90/60+7GA151.6
 7P1; 40 wk4120/80+8Ketamine+ NTG52.310 Perineal tear
 8P3; 38 wk5++100/60+8GA242.0
 9P0; 39 wk3+92/628**GA142.4
10P0; 40 wk8+++*100/64+8**GA182.5Retention of urine
11P1; 38 wk2130/808Ketamine+ NTG6320 Perineal tear
12P1; 34 wk5+++90/70+7GA151.5

All women had cervical dilatation of 7–9 cm — none were fully dilated at the time of IPV. Eight women had a blood leucocyte count of >18 000/mm3 and leaking of amniotic fluid for more than 24 hours. None could recollect the exact duration of labour; but for the majority it was more than 24 hours. Eight women had a cord prolapse with no cord pulsation and seven had features of dehydration and exhaustion.

Oligohydramnios and vaginal dryness frequently made it difficult to manipulate the arm back into the uterus. In seven this simply required the use of lubrication with liquid paraffin, but in one case (#5), the prolapsed arm had to be amputated at the shoulder joint to enable IPV.

Only one woman (# 2) had a postpartum haemorrhage from a torn vessel high in the cervix. There was no evidence of a uterine rupture and haemostasis was achieved with a surgical suture inserted vaginally. One unit of blood was transfused and she remained stable in the postpartum period. None had uterine ruptures or cervical tears, although two women had small perineal tears.

Hematuria was present in four women at the time of admission, but cleared postpartum in all. One woman (# 10) had blackish discolouration of the anterior vaginal wall because of prolonged obstruction. She had retention of urine when catheter was removed after 3 days so the catheter was reinserted. She regained the control of urine after 14 days of catheterisation and was discharged. Her long-term outcome is not known.

Standard GA was given in ten women and Ketamine in two. In these two women (#7 and #11), uterine relaxation was achieved by three boluses of IV-NTG each. Hypotension occurred 1 minute after NTG injection in both women, which was normalized with rapid infusion of crystalloids. Mean procedure time (excluding anaesthesia) was 19 minutes (range 14–26) in women where GA was used and 5 and 6 minutes where Ketamine and IV-NTG were used.


In the present study, IPV was successfully performed for women with IUFD and neglected transverse lies using either halothane or IV-NTG for uterine relaxation. Although none of the women’s cervices were fully dilated, the cervical application to the presenting part was so loose that it did not impede the delivery.

In the past, IPV was not uncommon in this situation, although classic texts warn of the high risk of uterine rupture when intrauterine manipulation is attempted in cases with a thin and stretched lower uterine segment.1 An alternative for small fetuses is expectant management awaiting spontaneous delivery. But with improved facilities for CS and neonatal care, the use of CS has become the norm where available. However, this may be dangerous for women in rural settings who have high risk of puerperal infection and uterine rupture in subsequent pregnancies. If the obstetrician is conversant with destructive operations, decapitation or evisceration is another option. All fetuses in our series weighed over 1.5 kg and all women had been in labour for many hours, having been referred in from outside. Spontaneous delivery was therefore thought to be unlikely.

The international literature on IPV in this situation is limited to only three reports.2–4 Mahendru et al2 successfully performed IPV in 41 women with a neglected transverse lie and Randrianantoanina et al3 did the same in three. They concluded that IPV has a role to play provided there is no oligohydramnios and that it is performed by a competent and experienced practitioner. In contrast, Andriamady et al.4 reported 177 IPV (85 in singleton pregnancies) and reported postpartum haemorrhages in 73%, uterine ruptures in 8% and cervical tears in 5% women. In our series, we had no difficulty in mobilising fetal parts during the procedure, even though these women had almost no liquor left because of prolonged rupture of membranes. If uterine relaxation is good, the accoucheur’s hand can be inserted easily even up to the uterine fundus and mobilising fetal parts becomes easy.

Nitroglycerin is a potent uterine relaxant. We used a total of 150 microgram NTG for each woman and found it to be highly effective. The ease with which IPV was carried out with IV-NTG was impressive, when compared to GA, although both experienced mild hypotension. Other side effects such as headache, prolonged uterine relaxation and palpitations were not observed. Its efficacy and safety profile give it advantages over amyl nitrate, magnesium sulphate or terbutaline.5,6 Nifedipine could provide an alternative, but its use has never been reported for this indication.

Neglected shoulder dystocia is of historical interest only in many parts of the world today. It remains a reality, however, where there are limited health resources, economic constraints and underutilised healthcare facilities. These problems result in a high incidence of undiagnosed malpresentations in labour. The move to assign skilled birth attendants to village-based health facilities will help overcome problems of distance and transport. Traditional birth attendants (TBA) can also provide culturally appropriate nurturing in the community setting and offer a first-line link with the formal healthcare system.


We conclude that IPV has a role to play in the delivery of dead fetuses in a neglected shoulder presentation, especially in developing countries. Good uterine relaxation rather than amount of liquor is a key to successful procedure of IPV, and this can be achieved through the use of IV-NTG or general anaesthesia. Reports of this technique are, however, few and further clinical studies are needed to fully evaluate its efficacy.

Disclosure of interests

All authors have no financial, personal, political, intellectual or religious interests to be disclosed.

Contribution to authorship

Niraj N. Mahajan wrote the manuscript and was instrumental in planning the study. He also performed nine out of 12 procedures. Nilima L. Gaikwad helped in finding the references and performed three IPVs. Meenakshi V. Solomon helped in editing the paper. Kshitija N. Mahajan gave anesthesia to most women and and along with Nupur P. Kothari helped with editing a paper.

Details of ethics approval

Ethics committee headed by Dr. V.K. Solomon, Medical Superintendent, Padhar Hospital. (1) Ethical committee approval for performing this procedure of IPV in dead foetuses in neglected transverse lie was obtained on 02/03/06. Ref No. is ‘Padhar Hospital/MSO/68-2006’. (2) Ethical committee approval for analysing the data and publishing was obtained on 10/10/07. Ref No. is ‘Padhar Hospital/MSO/247-2007’.


No assistance for funding was taken from any source.


We are thankful to Padhar Hospital administration for allowing us to conduct this study and also to the patients and their spouses who gave consent for the procedure.