Increased maternal body mass index is associated with prolonged anaesthetic and surgical times for caesarean delivery but is partially offset by clinician seniority and established epidural analgesia

Background Obesity is associated with higher surgical and anaesthetic morbidity and difficulties. Aims We aimed to investigate associations between maternal body mass index (BMI) and the in‐theatre time taken to produce an anaesthetised state or to perform surgery for caesarean delivery. Materials and Methods Using the Strengthening the Reporting of Observational Studies in Epidemiology guidelines, we identified all women who underwent caesarean section at a single institution (2009–2015). The prospectively collected data arising from antenatal and peripartum care were analysed. Generalised linear regression was used to examine associations between maternal BMI and the time taken to anaesthetise the mother and the duration of surgery. Results Of a total of 24 761 caesarean deliveries, 5607 (22.7%) women were obese at antenatal registration. In‐theatre anaesthetic preparation (18 vs 32 min, P < 0.001) and surgical duration (38 vs 52 min, P < 0.001) were longer in women with BMI ≥50 kg/m2 (BMI‐50) than those with normal BMI (BMI‐N). This difference remained significant after controlling for antenatal, intra‐operative and immediate postoperative variables. Modifiable variables were identified that may mitigate the effects of severe obesity. Senior obstetric and anaesthetic care were both independently associated with a significant reduction in mean in‐theatre anaesthetic preparation time and surgical duration, by 11 and three minutes respectively (P < 0.001), while epidural top‐up significantly lessened mean anaesthetic in‐theatre preparation duration by seven minutes (P < 0.001). Conclusions Obese women had greater anaesthesia and surgery time, but the effect may potentially be mitigated by provision of care by experienced staff and prior establishment of epidural analgesia.


INTRODUCTION
Obesity is increasingly prevalent (39% of all adults globally) and brings with it an increased burden of disease. 1 Caesarean section under neuraxial anaesthesia is a relatively common, standardised operation that is performed on a population of patients with generally low co-morbidity and for which there is routine prospective collection of associated data. It is therefore an ideal operation to study for the purpose of examining the relationship between severe morbid obesity and difficulties in generating neuraxial anaesthesia or performing intrabdominal surgery.
In 2017, 20% of Australian women who gave birth were obese at the start of antenatal care, 2 with a body mass index (BMI) ≥50 kg/m 2 occurring in 0.2% of all mothers. 3 Studies identifying obesity as a contributor to difficulties in providing neuraxial anaesthesia in the birthing suite have generally grouped all obese (BMI ≥ 30 kg/m 2 ) women together and compared them with normal BMI (18.5-24.9 kg/m 2 ) cohorts. [4][5][6] Obese women have also been shown to have higher rates of labour induction, failed induction of labour, deliver by caesarean section, have longer surgical durations, and be at increased risk of maternal and neonatal complications. 2,3,[7][8][9][10][11][12][13][14] Results of an adequately powered investigation of a bariatric subset with regard to outcomes of caesarean delivery is yet to be added to our knowledge base. [3][4][5][15][16][17][18] This study uses routinely collected, institutional data pertaining to all women delivering by caesarean section and their infants to determine if a woman's BMI is an independent variable influencing the time taken to generate anaesthesia for caesarean section and to perform the surgery.

MATERIALS AND METHODS
The Mater Mothers' Hospital and Mater Mothers' Private Hospital are co-located facilities within the Mater Health, South Brisbane campus. This is a maternity service with the highest capability ranking. 19 In 2017, 3% of all Australian neonates were delivered on this campus. 2,20 Our maternity service delivery is streamed into two colocated care models; publicly funded, multidisciplinary health care and private/health insurance funded, specialist obstetrician-led care.

Antenatal variables
Collected antenatal data included maternal height, pre-pregnancy weight, age at delivery, residential postcode, smoking status, illicit substance use, healthcare model, plurality of pregnancy and gestational age at booking. Experienced senior anaesthetic and obstetric clinicians were always the primary care providers for caesarean section of women in the private healthcare model, whereas in the multidisciplinary care model there was a mix of senior and lesser experienced clinicians. Socio-economic status was derived from maternal residential postcode, 21 then divided into quintiles for analysis.
Calculated BMI was classified according to American Heart Association recommendations specific for the surgical setting. 22 This system builds upon the World Health Organization BMI classification framework 23   Degree of urgency was recorded as a dichotomous elective/emergency variable and using the Royal Australian and New Zealand College of Obstetricians and Gynaecologists promulgated 1-4 categorisation. 24 Caesarean section as a surgical procedure is classified according to the site of uterine incision and denoted as lower segment or classical. 25 In-theatre anaesthetic preparation duration was deemed to be from the anaesthetic start time to the surgical start time.
Anaesthetic start time was noted as soon as the patient had both entered the theatre suite and had an anaesthetist in attendance. Surgical duration began at procedure start time and ended at procedure end time. Procedure start time marks both the end of anaesthetic preparation and the beginning of surgical duration. Procedure end time was noted at application of the wound dressing or alternatively, initiation of a subsequent surgical procedure. Post-caesarean in-theatre duration, total in-theatre anaesthetic care duration and duration of post-anaesthesia care unit care was also analysed. Anaesthetic start time was classified dichotomously into sociable or unsociable hours, where unsociable hours were daily from 23:00 to 7:00 hours. 26 Immediate maternal postoperative intensive care unit (ICU) admission, length of maternal inpatient admission as compared to industry funding allocation were collected to indicate maternal postoperative course. Suspected fetal problem and neonatal Apgar score at five minutes were captured; as well, neonatal nursery admission from theatre was recorded as metrics of infant well-being.

Statistical analysis
Generalised linear models with the identity link function were used to assess the association of BMI with the duration of in-theatre anaesthesia preparation and of caesarean section procedure both in univariate and multivariable analyses. All variables with origin in the antenatal period were included in the modelling.
Complete-case analysis was used for instances of missing data.
A two-tailed P-value of < 0.05 was considered statistically significant. SAS version 9.4 (SAS Institute Inc., Cary, NC, USA) was used for linear modelling. R version 3.6.1 with RStudio version 1.2.5019 was used for Pearson's χ 2 test with and without Yates' continuity correction, Fisher's exact test and Welch two-sample t-testing.

RESULTS
Of the 24 980 caesarean sections with live-born infants in the total dataset, 219 (0.9%) were removed due to lack of data on maternal height or weight, resulting in a final dataset of 24 761 records. At the time of booking, 22.7% of women (5607) were classified as obese (BMI ≥ 30 kg/m 2 ) and 108 (0.4%) had a BMI ≥ 50 kg/m 2 .
Demographic, antenatal and perinatal characteristics are summarised in Table 1. For demographic comparison of means against the BMI 18.5-24.9 kg/m 2 category (BMI-N), we combined obesity classes IV and V into BMI ≥ 50 kg/m 2 (BMI-50).

Demographic characteristics
The mean maternal age at delivery, singleton pregnancy rate and use of illicit substances did not differ across the BMI range. BMI-50 women were significantly more likely to be from a lower socioeconomic quintile (P < 0.001) and utilise publicly funded multidisciplinary health care (68% vs 41%, P < 0.001) as compared with BMI-N women.

Antenatal characteristics
The mean duration of antenatal care was shorter for BMI-50 women when compared to BMI-N women (14 vs 19 weeks, P < 0.001). They tended to book later (23 vs 19 weeks, P < 0.001) and were more likely to have their care transferred to the study hospital for intrapartum care (7% vs 2%, P < 0.001). BMI-50 women were significantly more likely to have a preterm delivery (30% vs 14%, P < 0.001) but the difference to BMI-N women in relation to very preterm delivery was not statistically significant (7% vs 4%, P = 0.074). The overall effect of booking and delivery differences was a significantly shorter duration of antenatal care (14 vs 19 weeks, P < 0.001) and this persisted after removal of the women transferred solely for intrapartum management (15 vs 19 weeks, P < 0.001).

Peripartum characteristics
It was more complicated to anaesthetise BMI-50 women: there was no difference from BMI-N with regards to epidural top-up but they were more likely to have a general anaesthetic (10% vs 5%; P = 0.015) or a CSE (39% vs 15%; P < 0.001) with a commensurate decrease in spinal anaesthesia rates (35% vs 61%; P = 0.004).
Surgical characteristics were similar between BMI-50 and BMI-N. There was no significant difference with reference to urgency classifications, surgical procedure or delivery during unsociable hours. Postoperatively BMI-50 differed from BMI-N in that a greater percentage were transferred from theatre to ICU (7% vs 1%, P < 0.001) and were also more likely to remain as an inpatient beyond the standardised length of stay (28% vs 15%, P < 0.001).
Health outcomes of babies from the BMI-50 subgroup were less assured. The fetuses of BMI-50 mothers as compared to those of the BMI-N were more likely to be allocated an alert regarding a potential problem (53% vs 36%, P = 0.026), register an Apgar score of less than seven at five minutes (2.4% vs 8.3%, P = 0.008) and be admitted to a nursery (32% vs 18%, P = 0.005).
More time was used in the theatre suite at every stage of a caesarean section for the BMI-50 subgroup. Greater obesity classification as compared to BMI-N was significantly associated with longer in-theatre anaesthetic preparation duration (Table 2 and Fig. 1). The mean in-theatre anaesthetic preparation duration for BMI-50 women was 32 min as compared with BMI-N women at 18 min (P < 0.001). The increase in in-theatre anaesthetic preparation duration was modest initially as maternal BMI increased but escalated as obesity rose beyond 40 kg/m 2 . The probability distribution for anaesthetic preparation also altered as obesity became more severe indicating increased variation of anaesthetic efficiency in the obese Class IV and V BMI groups (Figs 1 and 2).
Similarly, greater obesity classification as compared to BMI-N was associated with a longer time taken to perform a caesarean section ( Table 2 and Fig. 2). The mean surgical duration was 38 min at the nadir for the BMI-N group and rose to 52 min (P < 0.001) for BMI-50.
The mean in-theatre post-caesarean section duration was significantly longer for BMI-50 as compared with BMI-N women (17 vs 11 min, P = 0.002). Total in-theatre anaesthetic care was also prolonged (101 vs 67 min, P < 0.001). Time spent in the post-anaesthesia care unit also followed this trend (74 vs 56 min, P = 0.001). The association between in-theatre anaesthetic preparation duration or surgical duration and BMI remained significant even after controlling for all collected temporal covariates ( Table 2).
Multivariate linear regression analysis identified factors that were associated with shorter in-theatre anaesthetic preparation dura- The AMOSS super-obesity cohort study 3 was designed to determine prevalence and characteristics relevant to pregnancy outcomes of women with a BMI > 50 kg/m 2 or weight > 140 kg.  F I G U R E 1 Violin plot of in-theatre anaesthetic preparation duration by maternal body mass index (BMI). Violin plot shape illustrates probability density (wider = higher probability). White bars with black lines delineate median and interquartile range. Reference red line is at zero minutes. Note the change in shape with maternal BMI ≥ 50 kg/m 2 (Obese -Classes IV and V) as well as rising median values. The pilot study from a nearby institution in Brisbane analysed a cohort of women with a BMI of ≥40 kg/m 2 . 27 Our results confirm their findings with regard to rates of category one urgency caesarean section and surgical duration. In contrast to our results, they reported a significant difference with regard to the ability to convert labour epidural analgesia into epidural anaesthesia.
Our study also confirms the findings of a study investigating the time taken to access the neuraxis (n = 427). The authors reported that increasing BMI strongly predicted more difficulty in palpation of bony landmarks and less flexion of the lumbar spine. 16 This may, in part explain the increased in-theatre anaesthetic preparation time observed in higher BMI groups of our study.
We gathered no information on prior non-obstetric or obstetric surgery, and it may have confounded the observed association of increased surgical duration with increased maternal age. 28 Other limitations include the poor recording rate of smoking status for women in the specialist obstetrician healthcare stream as noted by previous authors 29

FUNDING
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

AVAILABILITY OF DATA AND MATERIALS
The data, 'The influence of maternal body mass index on anaes-