A retrospective review of birth injuries that had occurred during caesarean section over a period of one year was conducted. Sentinel events register was searched for neonatal birth injuries sustained at LSCS from June 1^st 2015- May 31^st 2016. The register is maintained by the charge nurse in the operation theatre. A total of six cases were found, and these charts were retrieved.

Maternal characteristics and details of LSCS, including indication and intraoperative findings were obtained from the medical records of pregnant women. Neonatal characteristics such as gestational age, gender, birth weight, type of injury, diagnosis and treatment given were retrieved from neonatal records. As per hospital protocol, all babies were examined by a neonatologist at birth. Based on the type of injury sustained, relevant speciality consultation was obtained. Presence of fractures was confirmed on radiographs. Suturing of lacerations was carried out in all cases by paediatric surgeons.

For the study, data from the medical records was entered into clinical proforma and analysed using SPSS software (IBM, version 23). The retrospective study was approved by Institutional Review Board and Ethics Committee of Christian Medical College and Hospital: IRB, Min No: 10922(Retro) dated 25.10.2017.

Of the total 4375 Caesarean deliveries over a one year period, from June 1, 2015 to May 31, 2016, there were six cases of major neonatal birth injuries, incidence being 0.13%. Four babies (66.6%) had lacerations that required suturing. The sites of injury were scalp, left hemithorax, abdominal wall and right ear. These sites coincided with the incision site on uterus in all cases. Two babies had fracture of femur.

Parity and Gestational age: Four (66.6%) of the pregnant women were primigravidas and two (33.3%) were multigravidas [Table/Fig-1]. Half of these pregnancies were preterm and the rest were term. Of the preterm pregnancies, two were late preterm at 36 weeks, while one was at 32+4 weeks.

Maternal Body Mass Index (BMI): According to World Health Organisation (WHO) classification of obesity, three patients had a normal BMI, two were overweight and one had class II obesity [Table/Fig-1]. The latter’s baby suffered from a fracture of the femur.

Lie/Presentation and Technique of Delivery: Four babies (66.6%) were in cephalic presentation, one baby (16.6%) was breech and one (16.6%) in transverse lie [Table/Fig-1]. Of the four babies in cephalic presentation, three were delivered as cephalic and suffered scalp lacerations requiring suturing. One baby, although in cephalic presentation, had a deeply impacted head at delivery , therefore had to be delivered by reverse breech technique and sustained femoral fracture. Another baby in breech presentation had a laceration on trunk at the time of delivery. Yet another baby with prematurity and in transverse lie was delivered as breech and had a femoral fracture. This baby weighed only 960 grams at birth.

Indications for LSCS: One LSCS (16.6%) was posted as category I – done for doubtful scar integrity in a patient with previous LSCS. Two LSCS (33.3%) were done for non-reassuring foetal status as category II. The remaining LSCS (50%) were posted as category III [Table/Fig-1]. The indications for doing Category III Caesarean were: failed induction, foetal growth restriction with abnormal umbilical artery Doppler and breech in labour. All LSCS were done in labouring women and urgency of doing LSCS was based on modified Lucas Classification [11].

Stage of labour/ Trial of vaginal instrumental delivery prior to LSCS: Two patients had LSCS in the second stage of labour after full dilatation of cervix. One had fetal distress; the other was done for breech in labour. Neither of the two patients in second stage had a trial of vaginal instrumental delivery. Four women had LSCS in the first stage of labour [Table/Fig-1].

Intraoperative Factors: Three intraoperative risk factors were associated with neonatal birth injuries. The presence of intra-abdominal adhesions due to previous surgeries: like LSCS or myomectomy; lack of amniotic fluid; which influences the ease of delivery and the presence of the fibroids distorting the uterine anatomy. One patient had a previous LSCS as well as a previous myomectomy with dense intra-abdominal adhesions, making surgery technically difficult. Another patient had a large 6x7 cm lateral intramural fibroid, along with many other seedling fibroids. Four patients (66.6%) had scanty amniotic fluid at delivery [Table/Fig-1].

Type of Uterine Incision: We conventionally use the Kerr’s incision, low transverse incision to deliver the foetus. Out of six cases, one patient needed J shaped incision, while others were delivered by Kerr’s incision [Table/Fig-1]. Another patient undergoing LSCS at 32+4 weeks had dense adhesions from previous LSCS, obscuring the lower segment. Hence transverse incision was made in the upper segment.

Incision to Delivery Interval: The incision to delivery interval in all six cases was less than five minutes [Table/Fig-1]. Two babies were delivered within two minutes. Even though nearly half the LSCS were done as Category III, the workload of managing a busy labour room prompted the surgeons to be quick in delivering the babies.

Expertise of Surgeon: The expertise of surgeons was classified into three levels based on their years of experience: Level 1: Post graduates in their first or second year of training, Level 2: Senior Residents who are diploma holders in Obstetrics and Gynaecology who have minimum three years’ experience and Level 3: Junior Consultants who had completed their Diploma as well as Masters in Obstetrics and Gynaecology and had minimum of four years’ experience.

All the cases in this cohort were performed by Level 2 Surgeons and were assisted by either interns or house surgeons [Table/Fig-1]. The obstetricians performing the LSCS were also in charge of managing the labour room.

Timing of LSCS: Most of the LSCS were performed by the doctors posted in the labour ward ,and were also responsible for managing the labour ward. Five out of six LSCS (83.33%) were carried out between 12 midnight and 8 am.

Half of the babies having major trauma were preterm <37 weeks gestational age at birth. None of the babies in our cohort were either very low birth weight or macrosomic. The mean birth weight was 2.35 Kg ranging from 0.9 kg to 3.56 Kg. Four babies (66.66%) were females and rest were male [Table/Fig-1].

Caesarean section is the most common surgical procedure performed in obstetrics. There has been an escalation in the rates of LSCS globally in the last few decades. There are known risks to both mother and her baby during LSCS. Although a number of studies examine maternal safety associated with LSCS, there are very few studies that address the neonatal safety [3].

Birth injuries are defined by the National Vital Statistics report as “an impairment to infant’s body structure or function due to adverse influences that occurred at birth” [12]. In the United States, the incidence of birth trauma varies from 0.2 to 37 per 1000 live births [13]. Studies done in India from tertiary care centres, having similar setup as ours have found an overall incidence of birth trauma ranging from 3.2 to 15.4 /1000 live births [7,9].

During the period of June 1^st 2015 to May 31^st 2016, the incidence of major neonatal birth injuries at caesarean sections in our hospital was 0.13%. Baskett TF et al., found that the incidence of major birth trauma at caesarean delivery was low (0.04% vs 0.02% with and without labour) [1]. Alexander JM et al., however, found an incidence of 1.1% [5].

Only one study has described the type of birth injury in relation to the mode of delivery. These authors found that injuries like fracture of clavicle; injury to brachial plexus; cephalohaematoma, and caput succedaneum were all much lower among babies born by LSCS in comparison to those born by vaginal delivery. Injuries like lacerations, sub-galeal bleed; subdural or intracerebral haemorrhage; injuries to skeleton and facial nerve injuries, on the other hand, were similar in both the groups [3].

The most common birth injury noted in our cohort was lacerations (four patients, 66.6%), with an incidence of 0.0009% [Table/Fig-2]. Alexander JM et al., found an incidence of lacerations of 0.7% [5]. The lower incidence of lacerations noted in our cohort could be explained by the fact that we took into consideration only those lacerations which required suturing. Femoral fractures occurred in two patients, 33.3%, with an incidence of 0.0004% livebirths. This is similar to studies done by Morris S et al., and Basha A et al., where the incidence was 0.17/1000 livebirths and 0.13/1000 livebirths respectively [14,15].

Factors predisposing to neonatal birth trauma, traditionally described in medical literature include: primigravida mothers; maternal obesity (body mass index >40 kg/m²); cephalopelvic disproportion; small maternal stature; maternal pelvic abnormalities; prolonged labour; fetal malpresentation; operative vaginal delivery (suction cup/forceps); very low birth weight or extreme prematurity and fetal macrosomia [16-18]. However, most of these risk factors have been implicated in neonatal injuries following delivery via vaginal route. None of the studies have looked at perinatal risk factors associated with birth injuries specifically in relation to LSCS. This makes our study unique since we have tried to identify risk factors for birth injuries specific to LSCS. The presence of large uterine fibroids, intraoperative adhesions and oligohydramnios are factors that could lead to intraoperative difficulty and contribute to the occurrence of fetal birth injuries during LSCS.

All of the six patients in our cohort, whose babies sustained major birth injury, were in labour when posted for LSCS. Similar observations were made by Baskett TF et al., who found that women in labour, at the time of LSCS, had greater chances of having major and minor neonatal birth injuries in comparison to those not in labour [1]. Alexander JM et al., also found that the incidence of neonatal birth injuries was very low when LSCS was done prior to onset of labour [5]. Moreover, they also found that women undergoing trial of instrumental delivery prior to LSCS, had almost 7% incidence of birth injuries [5]. In our study, however, there was no history of failed instrumentation prior to caesarean section [Table/Fig-2].

In our study, the incision to delivery interval in all cases were <5 minutes. This observation is similar to the findings by Alexander JM et al., who found a higher incidence of birth injuries in association with an incision to delivery interval of <3 minutes [Table/Fig-2] [1,4,5]. Although most of the cases were Category III LSCS, which did not warrant emergent delivery, the clinical workload of managing a busy labour room prompted the surgeons to hasten the delivery of babies.

In contrast to findings by Alexander JM et al., who found higher incidence of birth injuries when T or J shaped incisions were used in comparison to vertical and low transverse incision, we had majority of birth injuries with low transverse Kerr’s incision [5]. Only two patients in our cohort received T or J shaped uterotomy incision. Presence of oligohydramnios intraoperatively also seemed to increase the risk for neonatal birth trauma [Table/Fig-2].

Delivery of baby as breech, especially in association with factors such as transverse lie of the fetus, prematurity and deeply impacted head during second stage caesarean, was found to be associated with femoral fracture. Basha A et al., also found that emergency caesarean, prematurity, malpresentation, abnormal lie and multiple pregnancies carried a risk of long bone fractures [15]. Alexander JM et al, however, did not find an association between prematurity and neonatal injury [Table/Fig-2] [5].

BMI of the mother in our cohort did not seem to play a role, a finding consistent with other studies [5]. However, many studies have found maternal BMI> 40 kg/m² as a risk for neonatal birth trauma [17,19]. None of the women in our cohort had a BMI>40 kg/m².

All the operations in our cohort were performed by surgeons with mid-level expertise, this is in contrast to the study by Baskett TF et al., where all deliveries studied were attended to by senior obstetricians [Table/Fig-2] [1].

Most of the cases of birth injuries were encountered in cases of LSCS done outside the routine hours, between 12 midnight to 8 am. (n=5; 83.3%). This is similar to the findings by Ray S et al., who found that the delivery between 2:00 am to 8:00 am, carried a higher risk of birth injury [Table/Fig-2] [4]. This could be attributed to fatigue of surgical staff leading to errors.

Foetal macrosomia (birth weight >4kg) has been found to increase the risk of birth injuries. In one study, the risk of birth injuries was found to increase two fold in infants weighing 4000-4499 g and threefold in those weighing 4500-4999 g in comparison to their normosomic counterparts [19]. In another study, the incidence of foetal injuries was 7.7 percent in infants with birth weights greater than 4500 g [17]. However, none of the babies in our cohort were macrosomic.

Majority of the babies in our cohort were females. This is in contrast to observations by Warke C et al., who found a male preponderance in their study [9]. Another study from India also found a stronger association of birth injuries with female gender of infants than male [Table/Fig-2] [4].

The main drawback of our study is that we could not measure the exact correlation of these risk factors with birth injuries due to the small numbers.

Risk of major neonatal birth injuries was increased when LSCS was done on labouring women especially between 12 midnight to 8am by obstetricians with mid-level expertise. Maternal risk factors that seemed to play a role included primiparity, presence of intraoperative adhesions or fibroids and oligohydramnios. Incision to delivery interval <5minutes also seems to increase the risk. Fetal risk factors that could play a role include prematurity and female gender.