Skin Incision Selection in Superobese Women Undergoing Cesarean Birth

Abstract

Objective

The aim of this study was to examine whether vertical versus transverse skin incision is associated with increased wound complications in superobese women undergoing cesarean.

Study Design

This is a secondary analysis of a retrospective cohort study that included women with a body mass index (BMI) ≥ 50 kg/m² and a cesarean birth with documented skin incision type from 1/1/2008 to 12/31/2015 at a single academic medical center. The primary outcome was a composite of wound complications: infection requiring antibiotics including superficial cellulitis, deep and organ space infections requiring packing, vacuum placement or exploration and debridement in the operating room. Secondary outcomes included estimated blood loss (EBL), time from skin incision to delivery, need for classical or T-hysterotomy, prolonged hospital admission (>4 days), and a composite of adverse neonatal outcomes. The primary exposure was skin incision type, transverse or vertical. Modified Poisson regression variance was used to adjust for differences in baseline characteristics.

Results

During the study period, 298 women underwent a cesarean with a known skin incision type. Vertical skin incision occurred in 25.8%. Women with a vertical skin incision were younger, had a higher BMI at delivery, had less weight gain in pregnancy, and were less likely to have labored prior to cesarean. Wound complications were not significantly more common in women with a vertical skin incision after adjusting for covariates (vertical 48.1 vs. transverse 29.4%, adjusted relative risk (aRR): 1.31, 95% confidence interval [CI]: 0.92–1.86). Compared with a transverse skin incision, vertical skin incision was associated with an increased risk for classical hysterotomy (67 vs. 17%, aRR: 2.96, 95% CI: 2.12–4.14), higher EBL, prolonged hospital stay, and composite neonatal morbidity. There were no statistically significant differences in the time from skin incision to delivery.

Conclusion

In superobese women, vertical skin incision was not associated with increased wound complications, but was associated with increased risk for classical hysterotomy.

In the United States, the prevalence of superobesity, as defined by the National Institutes of Health Consensus Development Panel as a body mass index (BMI) of 50 or greater, is estimated to be 2% in the obstetric population and has increased steadily in recent years.¹ The cesarean birth rate among superobese women approaches 50%.² The risk of surgical-site infections increases with BMI in a dose-dependent fashion. Almost one-third of superobese women who deliver via cesarean birth will develop a wound complication, with up to a quarter of complications requiring readmission.^3–5 Despite this high rate of wound complications, optimal operative technique and perioperative care for superobese women have not been established. To our knowledge, few studies have looked at the impact of skin incision type on wound complications in women with BMI≥50.

Skin preparation types, additional antibiotic prophylaxis at the time of cesarean, supplemental oxygen, and prophylactic negative pressure wound therapy after cesarean have not proven clearly effective at decreasing wound complication rates in this high-risk population.^6–8 Few studies have examined skin incision approach for cesarean birth in superobese women, leaving room for a variety of approaches.⁵ A transverse skin incision made in the lower pelvis (i.e., Pfannenstiel or Joel-Cohen) is the most common approach used at cesarean delivery due to decreased healing time, decreased postoperative pain, shorter operative times, and better cosmetic result.⁹ However, in obese patients, an incision under the pannus can be difficult to access, clean, and keep dry.¹⁰ Additionally, case reports document associated respiratory compromise, venocaval compression, and fat necrosis from retraction of the pannus during cesarean delivery.¹¹ Vertical skin incisions avoid the difficulties of manipulating the pannus but make accessing the lower uterine segment more difficult.^4,10,12 Vertical incisions are also associated with increases in blood loss, time from incision to delivery, and total operative time.^10,12 Without clear consensus, surgeon discretion often determines skin incision approach and results in considerable practice variation,³ such as at our institution where no standard protocol for managing superobese patients has been developed.

This study compares wound complication frequencies between transverse versus vertical skin incisions among superobese women undergoing cesarean birth. In addition, we consider both maternal and neonatal risks associated with either approach.

Materials and Methods

This is a secondary analysis of a retrospective cohort study of all women with a BMI of 50 or greater at delivery from January 1, 2008 to December 31, 2015 at a single academic medical center.¹³ All women 18 to 45 years old who underwent a cesarean birth at 24 weeks gestation or greater with documented skin incision type were included for this analysis. The original study identified eligible women by extracting height and weight data from the last prenatal visit or, if available, upon admission to labor and delivery from the electronic health record. For women with multiple pregnancies during the study period, only the first pregnancy was included. Approval from Vanderbilt University Medical Center Institutional Review Board was obtained prior to initiating the study.

The primary outcome was a composite of wound complications: infection requiring antibiotics including cellulitis, wound separation after delivery requiring packing or vacuum placement, or need for surgical exploration in the operating room. These outcomes were ascertained for the duration of the 6-week postpartum period. Secondary maternal outcomes included the individual components of the primary outcome, as well as estimated blood loss (EBL) during cesarean, time from skin incision to delivery, need for classical or T-hysterotomy, endometritis (temperature greater than 38°C with fundal tenderness and antibiotic administration), and prolonged postpartum length of stay (defined as greater than 4 days).

We also examined a composite of neonatal morbidity: neonatal death, hypoxic–ischemic encephalopathy (defined as an umbilical cord pH less than 7.0 and seizures or evidence of end-organ damage), neonatal intensive care unit (NICU) stay greater than 72 hours, sepsis, intraventricular hemorrhage class III or IV, respiratory distress syndrome, umbilical artery pH less than 7.10, and 5-minute Apgar score less than 7. Admission to the NICU and umbilical artery pH less than 7.10 were also examined individually.

The standard surgical technique for a cesarean section at our institution includes a Pfannenstiel or Joel-Cohen skin incision, with blunt entry into the peritoneal cavity. A bladder flap typically is not made and the hysterotomy is commonly closed in two layers. The peritoneum and rectus typically are not closed. The usual antibiotic regimen for patients undergoing cesarean includes 3 g cefazolin and 1 g azithromycin for patients weighing more than 120 kg. Vaginal prep is completed on all scheduled and nonurgent cesarean deliveries. Our institution does not use routine prophylactic negative pressure wound systems on women delivering via cesarean with a BMI ≥ 50.

The primary exposure was skin incision type, categorized as either transverse or vertical. While physicians perform both supra and subpannicular transverse skin incisions at our institution, the height of the incision is inconsistently and imprecisely documented in the operative note. Additional covariates were selected a priori and included maternal age, BMI at delivery, gestational age at delivery, race, education, insurance status, smoking during pregnancy, total weight gain in pregnancy (weight at delivery—self-reported pre-pregnancy weight), prior cesarean birth, attempted trial of labor, and emergent delivery. Emergent delivery was designated by the surgeon and indicated that delivery needed to occur within 30 minutes. Maternal comorbidities were studied including chronic hypertension, pregestational diabetes, and gestational diabetes.

Demographic and clinical data were compared between women with a transverse versus vertical skin incision using Student’s t-test, Wilcoxon rank-sum test, and chi-squared test as appropriate. Tests were two-tailed and assumed a 5% level of statistical significance. A modified Poisson approach with robust error variance was used to estimate unadjusted and adjusted relative risks (RR) with 95% confidence intervals for binary outcomes. Multivariable linear regression was used to adjust for covariates. The final adjusted models included skin incision (transverse versus vertical) BMI at delivery, gestational age at delivery, and labor preceding delivery. Weight gain in pregnancy was not retained in the model because it closely correlated with BMI at delivery. All analyses were conducted using Stata version 15.1 (StataCorp, College Station, TX).

Results

During the study period, 300 cesarean deliveries occurred in women with a BMI of 50 or greater. Two women were excluded due to missing data on skin incision type resulting in final sample size of 298 women. Of these women, 25.8% (n = 77) underwent a vertical skin incision. Women with a vertical skin incision were younger, had a higher BMI at delivery, had less weight gain in pregnancy, and were less likely to have labored prior to cesarean (Table 1).

Table 1.

Demographic and clinical characteristics

Variable	Transverse n=221	Vertical n=77	Relative risk (95% confidence interval)	p-Value
Maternal age	31.2 ± 5.0	29.3 ± 5.1		0.003
Body mass index at delivery	53.2 (51.2–56.0)	58.7 (53.8–62.3)		<0.001
Gestational age at delivery (weeks)	38.5 (37.0–39.2)	38.0 (36.0–39.1)		0.06
Race
Non-Hispanic white	128/221 (57.9)	52/77 (67.5)	Reference
Non-Hispanic black	89/221 (40.3)	20/77 (26.0)	0.64 (0.43–0.97)	0.05
Other	4/221 (1.81)	5/77 (6.49)	3.59 (0.99–13.05)	0.04
High school education or less	136/219 (62.1)	46/75 (61.3)	0.99 (0.80–1.22)	0.91
Public health insurance	145/220 (65.9)	53/77 (68.8)	1.04 (0.87–1.25)	0.63
Smoking during pregnancy	49/221 (22.1)	12/76 (15.8)	0.71 (0.40–1.27)	0.25
Pregestational diabetes	29/221 (13.1)	12/77 (15.6)	1.19 (0.64–2.21)	0.59
Chronic hypertension	82/221 (37.1)	34/76 (44.7)	1.26 (0.86–1.86)	0.24
Gestational diabetes	40/221 (18.1)	18/77 (23.4)	1.29 (0.79–2.11)	0.31
Weight gain in pregnancy (kg)	11.0 (4.7–17.7)	8.6 (3.0–13.2)		0.03
Prior cesarean	106/221 (48.0)	42/77 (54.6)	1.14 (0.89–1.46)	0.31
Trial of labor	93/221 (42.1)	21/77 (27.3)	0.65 (0.44–0.96)	0.03
Emergent delivery	13/220 (5.9)	6/77 (7.8)	1.31 (0.52–3.35)	0.55

Abbreviation: SD, standard deviation.

All data presented as n (%), mean + SD, or median (interquartile range)

Overall, 34.2% (n = 102) of women experienced the composite primary outcome of wound complication, which was driven mostly by diagnoses of cellulitis requiring either intravenous or oral antibiotics. Wound complications were more common in women with a vertical skin incision, but this difference was not statistically significant after adjusting for maternal age, BMI at delivery, gestational age at delivery, race, prior cesarean delivery, and whether labor preceded cesarean (vertical 48.1 vs. transverse 29.4%, unadjusted RR: 1.63, 95% CI: 1.20–2.23, adjusted RR: 1.31, 95% CI: 0.92–1.86). A post-hoc power analysis revealed that our study had 91% power to detect a significant difference in the primary outcome between the two groups assuming an α of 5%.

Compared with a transverse skin incision, vertical skin incision was associated with higher intraoperative EBL, and higher rates of classical hysterotomy, postpartum hospital admission greater than 4 days, and composite neonatal morbidity. There were no statistically significant differences in the time from skin incision to delivery or in the risk of an umbilical artery cord gas less than 7.10. Endometritis was less frequent in patients with a vertical incision; however, this study was not powered to examine this outcome independently (Table 2).

Table 2.

Association between vertical skin incision and obstetric outcomes

Outcome	Transverse n = 221	Vertical n = 77	Relative risk or p-Value (95% CI)	Adjusted RRa (95% CI)
Wound complication	65 (29.4)	37 (48.1)	1.63 (1.20–2.23)	1.41 (0.99–2.00)
Cellulitis requiring antibiotics	61 (27.6)	32 (41.6)	1.51 (1.07–2.12)	1.37 (0.93–2.02)
Wound separation with packing or vacuum	32 (14.5)	14 (18.2)	1.26 (0.71–2.23)	1.23 (0.63–2.40)
Wound exploration in-operating room	5 (2.3)	1 (1.3)	0.57 (0.07–4.85)	0.40 (0.02–6.48)
Classical hysterotomy	38 (17.2)	52 (67.5)	3.93 (2.83–5.46)	3.07 (2.12–4.32)
Neonatal morbidityb	58 (26.2)	36 (46.8)	1.78 (1.29–2.47)	1.54 (1.10–2.14)
NICU admission	62 (28.1)	38 (49.4)	1.76 (1.29–2.40)	1.56 (1.14–2.14)
Postpartum admission >4 days	9 (4.1)	16 (20.8)	5.10 (2.35–11.08)	4.18 (1.88–9.28)
Estimated blood loss (per 100 mL)	800 (700–1000)	900 (800–1200)	p = 0.008	235.67 (20.26–451.10)c
Time from skin incision to delivery (per 1 minute)	18.0 (13.0–26.0)	17.5 (12.0–25.0)	p = 0.83	−2.02 (−5.27–1.23)c
Endometritis	12 (5.4)	1 (1.3)	0.24 (0.03–1.82)	0.34 (0.04–2.65)
Umbilical artery pH <7.1	20 (9.1)	8 (10.4)	1.15 (0.53–2.50)	1.02 (0.41–2.53)

Abbreviations: CI, confidence interval; NICU, neonatal intensive care unit; RR, relative risk.

All data are n (%) or median (interquartile range), referent group is transverse skin incision.

^aAdjusted for body mass index at delivery, gestational age at delivery, and labor preceding delivery.

^bIncludes neonatal death, NICU admission >72 hours, neonatal sepsis, grade III/IV intraventricular hemorrhage, hypoxic ischemic encephalopathy, respiratory distress syndrome, seizures, necrotizing enterocolitis or cardiopulmonary resuscitation.

^cBeta coefficients adjusted for body mass index at delivery, gestational age at delivery, and labor preceding delivery.

Discussion

This study compares the frequency of wound complications after cesarean birth performed via transverse versus vertical skin incision in women with superobesity (BMI: 50 kg/m² or greater). We found an overall high frequency (34%) of wound complications among women with superobesity; however, there was no statistically significant difference in wound complication rates between incision types after accounting for baseline differences in the study population. Vertical skin incision was not associated with decreased operative time from skin incision to delivery but was associated with higher frequencies of classical hysterotomy, prolonged postpartum hospital stay, and composite neonatal morbidity. Vertical skin incision was also associated with higher intraoperative EBL, but the absolute difference of 100 mL may not be clinically significant.

Our findings are consistent with the current literature in obese women with a lower BMI, which documents high rates of wound complications and no improvement in perinatal outcomes with vertical skin incisions.^{4,5,10,12,14–19} The largest study to date of 23,277 women with obesity (BMI >30 kg/m²) undergoing cesarean birth included an analysis of wound complications by incision type in a subgroup of 2,411 women with a BMI >45 kg/m².⁴ Women with a vertical skin incision were more likely to experience the composite outcome of wound complication compared with women who underwent a transverse skin incision.⁵ However, these subgroup data were not adjusted for baseline differences in the populations, which attenuated findings in our study and others.

Other researchers have specifically studied incision approach in women with lesser classes of obesity with conflicting findings. Wall et al found that vertical skin incisions were associated with a 12-fold increased adjusted odds for wound complications compared with transverse skin incisions among 239 women with a BMI >35 kg/m².¹⁰ Brocato et al found no statistically significant difference between the groups after adjusting for maternal covariates among 133 women with a BMI >40 kg/m².¹²

The finding that a vertical skin incision is associated with increased risk for classical hysterotomy deserves particular attention. Classical hysterotomy may be associated with increased blood loss and infectious morbidity as well as prolonged hospital stay during the index pregnancy.¹⁶ But also notable are the effects on subsequent pregnancies, specifically increased risk of uterine rupture, including prelabor uterine rupture, mandating delivery at earlier gestational ages.^16,17 While rare, uterine rupture is potentially catastrophic especially in this patient population that is likely to have other comorbidities. Furthermore, a history of a prior classical cesarean escalates the level of complexity of the patient for all subsequent pregnancies.

Eighty-four percent of physicians state that they prefer a Pfannenstiel approach with taping of the pannus when performing a cesarean on morbidly obese patients (BMI ≥40 kg/m²).³ However, multiple studies document that increasing obesity is associated with an increasing risk for a vertical skin incision.^4,5,17 This suggests that physicians likely consider there to be some benefit to a vertical approach, whether to allow for better surgical exposure or improve perinatal outcomes with increasing level of obesity. Our study finds only harms associated with a vertical incision (classical hysterotomy, increased blood loss) without apparent improvement in neonatal outcomes, incision to delivery times, or wound complications.

We recommend caution interpreting our findings of increased composite neonatal morbidity and NICU admission in infants born to women undergoing a vertical skin incision. These findings may represent confounding by indication; physicians selected vertical skin incisions for the purpose of perceived expeditious delivery in the setting of more fetal morbidity or non-reassuring status. It is notable that a decreased incision to delivery time was not observed in women undergoing vertical incision. Unfortunately, we did not have information on the type of transverse incision performed (supra- vs. subpannicular) or the distribution of adiposity in individual women, which can affect both incision selection and wound morbidity. Finally, baseline characteristics differed among women who underwent vertical versus transverse skin incision. On average, women with vertical incisions were younger with higher BMI, earlier gestational age at delivery and less likely to have attempted atrial of labor before cesarean delivery. Although we attempted to adjust for these differences, residual confounding is possible, but would need to be considerable to reverse the direction of the observed point estimates. Finally, the retrospective study design inherently limits the precision of data collected such as the location of the transverse incision relative to the pannus and complete ascertainment of the composite wound complication outcome.

To summarize, women with superobesity have high rates of wound complications following cesarean delivery. Vertical skin incisions do not decrease the risk of wound complications, but are associated with longer operative time, higher EBL, and possibly increased risks of neonatal morbidity. Taking into consideration patient and provider preference as well as safety, transverse skin incisions should continue to be the standard incision for cesarean delivery on patients with superobesity. Our findings add to the growing literature on incision approach in this population, which may require meta-analyses to adequately study less common outcomes especially in the absence of adequately powered randomized controlled trials.

Key Points.

• Vertical skin incision was not associated with a higher risk for composite wound morbidity after adjusting for covariates.

• Vertical skin incision was significantly associated with classical hysterotomy without associated decrease in incision to delivery time or neonatal morbidity.

• When selecting a skin incision approach in superobese women, clinicians should consider whether potential benefits outweigh known risks.