Abdominal surgical incisions and perioperative morbidity among morbidly obese women undergoing cesarean delivery☆

Article Outline

• Abstract
• 1. Introduction
• 2. Materials and methods
• 3. Results
• 4. Comments
• References
• Copyright

Abstract 

Objective

To test the hypothesis that there is no difference in perioperative morbidity and the type of uterine incisions between vertical skin incisions (VSI) and low transverse skin incisions (LTSI) at the time of cesarean delivery in morbidly obese women.

Study design

Retrospective cohort study of morbidly obese women (BMI>35kg/m²) who underwent cesarean delivery between June 2004 and December 2006.

Results

During the study, 424 morbidly obese women underwent cesarean section. Patients with VSI were older (31.0±6.2 years vs. 26.7±5.8 years), heavier (48.2±9.1kg/m² vs. 41.7±6.7kg/m²), and more likely to have a classical than a low transverse uterine incision (65.9% vs. 7.3%), p<0.001. After controlling for confounders, women with VSI did not have an increase in perioperative morbidity, but underwent more vertical uterine incisions (adjusted odds ratio=18.49, 95% CI: 6.44, 53.07).

Conclusion

VSI and LTSI are safe in morbidly obese patients undergoing cesarean section, but there is a tendency for increased vertical uterine incisions in those who underwent VSI.

Keywords: Pregnancy, Morbid obesity, Cesarean delivery, Classical uterine incision, Vertical skin incision

Back to Article Outline

1. Introduction 

Available evidence suggests that obesity has increased to pandemic levels. In 2005, nearly 1.6 billion individuals were reported to be overweight worldwide, with at least 400 million classified as obese [1]. In the United States, it is estimated that one-third of adult women are obese [2]. Though there is controversy as to which measure of obesity should be used, body mass index (BMI) is often utilized to define obesity. An individual with a BMI of 30kg/m² or greater is obese by the World Health Organization (WHO) definition [1].

Because many obese women are in the reproductive age group, obesity has had a dramatic impact on pregnancy outcome. It is well known that obese women have an increased risk of pregnancy complications such as anemia, hypertension, pre-eclampsia, preterm delivery, emergency cesarean section, and gestational diabetes [3], [4]. Weiss et al. [5] demonstrated that obesity is significantly associated with an increased rate of cesarean delivery, 20.7% in normal weight control groups, 33.8% in obese women, and 47.4% in morbidly obese women (BMI>35kg/m²). Furthermore, it is well documented that surgery on the morbidly obese patient poses many surgical and logistical difficulties. Rodriguez et al. [6] found that compared with patients having a BMI of 30–39.99kg/m², those with a BMI>40kg/m² had an increase in total operative time and time from skin incision to delivery.

Modern women tend to be more aware of their body image and the use of the Pfannenstiel incision, or low transverse skin incision (LTSI), for cesarean section has become the norm. In the setting of morbid obesity, however, the use of such an incision in the moist region below the pannus remains a debated issue. Obtaining adequate exposure in the obese pregnant patient also poses significant difficulties. Gallup [7] suggested that for obese gynecological patients, a VSI be made, and if a transverse incision is employed, the surgeon should shy away from the subpannicular fold. However, VSI compared with LTSI is associated with increased post-operative pain, post-operative atelectasis, and superficial wound and fascial dehiscence [8]. In addition, when VSI are made in morbidly obese patients undergoing cesarean section, the incision often overlies the uterine fundus, potentially limiting access to the lower segment, which may necessitate performing a vertical uterine incision. Vertical uterine incision is associated with significant morbidity in subsequent pregnancies [9]. The prevalence of vertical uterine incision in morbidly obese patients in whom a VSI is performed is not known.

Our primary objective was to test the hypothesis that in morbidly obese patients, the type of skin incision made at cesarean section does not influence the choice of uterine incision. Our secondary objective was to compare perioperative morbidity among morbidly obese women undergoing VSI compared with LTSI at the time of cesarean section.

Back to Article Outline

2. Materials and methods 

Upon approval by the Medical College of Georgia (MCG) institutional review board, we conducted a cohort study to recruit all morbidly obese women delivered by cesarean section at our institution from June 2004 to December 2006. For the purpose of our study we defined morbid obesity as a BMI greater than 35kg/m² [2], [5]. Although the WHO considers 40kg/m² to be morbidly obese (class III obesity) we chose to evaluate women with a BMI greater than 35kg/m² because this cutoff has been used to characterize severe obesity (greater than the 95th percentile for women 20–29 years of age) [10], including in a large multi-center trial in the United States [5]. BMI was calculated by using the height and weight recorded on the intraoperative anesthesia forms. We chose these parameters as they represented the BMI that would be encountered at the time of delivery. Of the 1272 women who had a cesarean delivery during the study period, 424 met the inclusion criterion. For these patients, demographics were abstracted from medical records including gravidity, parity, age at, and indication for, delivery, as well as the number of previous cesarean deliveries, type of skin and uterine incision made, operating time, estimated blood loss, and pre- and post-operative hemoglobin and hematocrit. Wound infection and/or breakdown were defined by clinical criteria based on the finding of wound separation and/or purulent material draining from the wound, with or without associated cellulitis or fever.

For the purpose of this study, a VSI was defined as a longitudinal incision irrespective of whether it was infra- or supraumbilical. Likewise, an LTSI was an incision made across the lower abdomen irrespective of its distance from the superior border of the pubic bone. Such incisions were often made after pulling up the pannus, and at variable distances from but not along the anaerobic environment of the subpannicular fold. Classical cesarean deliveries were defined as any cesarean section that involved a vertical incision in the uterus. There were 10 cases that were recorded as low vertical uterine incisions that were classified as classical incisions for the purpose of this study because after review of medical records it was unclear that the incisions did not extend into the upper uterine segment.

All statistical analyses were performed using Statistical Analysis Software (SAS) Version 9.1 (SAS Institute, Inc., Cary, NC). For continuous variables, Student t-tests were performed to compare mean (SD) age, and gestational age at delivery while Mann–Whitney U-test was used to compare the median [range] gravidity and parity between the two groups and to assess statistical significance. For categorical data, percentages were calculated and the Chi-square test or Fisher's exact test were used. For all descriptive analyses, a p-value<0.05 reflected statistical significance.

Unadjusted and adjusted odds ratios (AOR) and 95% confidence intervals (CI) were computed by logistic regression for the risk of the following outcomes of interest: (a) classical uterine incision, (b) blood transfusion, and (c) wound breakdown/infection. Potential confounders were included in the full model if they were risk factors for a classical uterine incision and associated with the type of skin incision performed in the index pregnancy, based on a p-value ≤0.20. These covariates included maternal age, gravidity, race, body mass index, anesthesia type, and gestational age at delivery. Previous cesarean section was included a priori in all adjusted models. Because the decision to perform a tubal ligation at the time of cesarean section may have affected the surgeon's decision to perform a classical uterine incision, a separate logistic regression analysis was performed which excluded women that had a classical uterine incision and tubal ligation.

Back to Article Outline

3. Results 

The prevalence of morbid obesity in our obstetrical population delivered by cesarean section during the study period was 33.3%. VSI was performed in 41 (9.7%) patients while the remaining 383 (90.3%) patients underwent LTSI. The study participants were primarily non-Hispanic black (62.5%), with a mean BMI of 42.3±7.2kg/m², a median gravidity of 2 (1–8), and a median parity of 1 (0–6) (Table 1). Women with VSI were older (31.0±6.2 years vs. 26.7±5.8 years), heavier (48.2±9.1kg/m² vs. 41.7±6.7kg/m²), and more likely to have a classical compared to low transverse uterine incision (65.9% vs. 7.3%, p<0.001).

Table 1. Sociodemographic and medical characteristics of the study population, overall and stratified by type of skin incision.

Demographic variables	Overall (n=424)	Type of skin incision		p-Value
		Vertical (n=41)	Transverse (n=383)
Body mass (kg/m2, mean±SD)	42.3±7.2	48.2±9.1	41.7±6.7	<0.001
Age (years, mean±SD)	27.1±6.0	31.0±6.2	26.7±5.8	<0.001
Race and Hispanic origin (n (%))				0.09
Non-Hispanic White	133 (31.4)	8 (19.5)	125 (32.6)
Non-Hispanic Black	265 (62.5)	32 (78.1)	233 (60.9)
Other	26 (6.1)	1 (2.4)	25 (6.5)
Gravidity (median [range])	2 [1–8]	3 [1–8]	2 [1–8]	0.13
Parity (median [range])	1 [0–6]	1 [0–4]	1 [0–6]	0.23

n: numbers.

For 53% of the study population, this was a primary cesarean delivery. However, there was no statistically significant difference in the type of cesarean section (e.g. primary vs. repeat) and the type of skin incision (VSI vs. LTSI) in the index pregnancy (p=0.35) (Table 2). The top three indications for cesarean section for women with a VSI were repeat cesarean delivery (26.8%), failed induction (19.5%), and non-reassuring fetal heart tones (17.1%). The top indications for women with an LTSI were repeat cesarean (39.2%), arrest of dilation (17.8%), and non-reassuring fetal heart tones (16.2%). The indication for cesarean section was abstracted with only one indication per patient.

Table 2. Pregnancy outcomes of the study population overall, and stratified by the type of skin incision.

	Overall (n=424)	Type of Skin Incision		p-Value
		Vertical (n=41)	Transverse (n=383)
Gestational age at delivery (weeks, mean±SD)	37.7±3.4	36.9±3.4	37.7±3.3	0.12
Cesarean section				0.35
Primary	226 (53.3)	19 (46.3)	207 (54.0)
Repeat	198 (46.7)	22 (53.7)	176 (46.0)
Indication for cesarean				0.02
Arrest of dilation	72 (17.0)	4 (9.8)	68 (17.8)
Failed induction	30 (7.1)	8 (19.5)	22 (5.7)
Malpresentation	36 (8.5)	5 (12.2)	31 (8.1)
Repeat cesarean	161 (38.0)	11 (26.8)	150 (39.2)
Non-reassuring fetal heart tones	69 (16.2)	7 (17.1)	62 (16.2)
Other	56 (13.2)	6 (14.6)	50 (13.0)
Uterine incision				<0.001
Low transverse	369 (87.0)	14 (34.1)	355 (92.7)
Classical	55 (13.0)	27 (65.9)	28 (7.3)
Anesthesia given				<0.001
General	43 (10.1)	8 (19.5)	35 (9.1)
Spinal	220 (51.9)	16 (39.0)	204 (53.3)
Epidural	141 (33.3)	9 (22.0)	132 (34.5)
aOther	20 (4.7)	8 (19.5)	12 (3.1)
Tubal ligation performed	88 (20.7)	14 (34.1)	74 (19.3)	0.04
Received a blood transfusion	10 (2.4)	4 (9.8)	6 (1.6)	0.01
Wound breakdown/infection
Yes	35 (8.3)	6 (14.6)	29 (7.6)	0.03
No	347 (81.8)	35 (85.4)	312 (81.4)
Unknown	42 (9.9)	0 (0.0)	42 (11.0)
Hospitalization (days, median [range])	4 [2–45]	4 [2–45]	4 [2–25]	0.02
Operating time (min, median [range])				0.13
Vertical uterine	67 [27–195]	70 [46–122]	67 [27–195]
Low transverse	74 [30–224]	79 [37–224]	72 [30–214]

Unless otherwise stated, the results are numbers (%).

Women with a VSI were less likely to receive spinal or epidural analgesia compared to women with an LTSI (p<0.001). Women who underwent a VSI were more likely to require a blood transfusion (p=0.01) and to have a wound infection/breakdown (p=0.03). They also experienced longer hospitalizations and a non-significant increase in surgery times (Table 2). There was non-significant trend (p=0.64) towards increased hospital stay (median (range)) from 4.0 (2–25) to 3.0 (2–45) and 4.0 (2–25) days as the BMI increased from 35–39.9kg/m² to 40.0–44.9kg/m² and >44.9kg/m², respectively.

After controlling for maternal age, gravidity, BMI, previous cesarean section, anesthesia type, and gestational age at delivery, women with a VSI were not more likely to experience a wound breakdown/infection (AOR=1.91; 95% CI: 0.57, 6.44) or require a blood transfusion (AOR=2.78; 95% CI: 0.42, 18.40). Within the index pregnancy, there was a strong association between the type of skin incision and the type of uterine incision performed. Women with a VSI were significantly more likely to have a classical uterine incision, compared to those with an LTSI [unadjusted odds ratios (AOR=24.45 95% CI: 11.53, 51.84)]. After excluding VSI in which a tubal ligation was performed (n=13), the association between VSI and classical uterine incisions remained statistically significant (AOR=18.49, 95% CI: 6.44, 53.07).

Back to Article Outline

4. Comments 

We have shown that the use of a VSI in morbidly obese patients is associated with an 18-fold increased risk of classical uterine incision in our institution. Therefore we reject the null hypothesis, and state that the type of skin incision made at cesarean section does influence the type of uterine incision in morbidly obese patients. In addition, the type of skin incision does not influence the perioperative morbidity among morbidly obese women undergoing cesarean section.

The choice of skin incision in morbidly obese patients undergoing cesarean section is controversial. Obstetricians perform VSI to avoid the pannus, which may limit access to the lower uterine segment. The fact that VSI is associated with higher rate of classical uterine incision in obese women undergoing cesarean section has been reported previously. Wolfe et al. [11] reported a 14% incidence of classical uterine incision in their series of 107 obese women delivered by cesarean. This lower incidence may be due to the fact that this latter group defined morbidly obese as weight>200lb at the time of delivery instead of BMI>35kg/m² [11]. Certainly, some patients at this weight and below can be morbidly obese by our definition depending on their height, and such patients are included in our study. Thus if the definition of morbid obesity was limited to >200lb in our study, 41 (9.7%) patients with BMI>35kg/m² would have been excluded. Wall et al. [8] evaluated morbidly obese patients, using the same criterion as in our study, and reported that 23.1% of their patients with VSI had classical uterine incisions compared to 3.3% of those with LTSI (OR 8.83, CI 2.7–28.8, p<0.001). The higher rate of classical uterine incisions in our study may be due to the high prevalence of morbidly obese patients (33.3%) in our obstetric population. In addition, we included patients undergoing repeat cesarean while Wall et al. [8] analyzed only morbidly obese patients undergoing a primary cesarean delivery. As more patients in our LTSI group had undergone a previous cesarean (39.2% compared with 26.8% in the VSI group), this should have tilted the balance in favor of more classical uterine incisions with LTSI if previous cesarean was a factor in the performance of classical uterine incision in our study. Within our study population, nearly half of those patients who underwent classical cesarean delivery were having their first pregnancy and delivery, with implications for future pregnancies. Although the median operating time did not differ in relation to type of skin and uterine incisions, these times may be regarded as long for an uncomplicated cesarean section. While surgical times may vary at different institutions, or across countries and continents, our findings are consistent with those reported in similar studies conducted at similar teaching institutions in the United States [8], [12].

The risks associated with the use of classical uterine incision are well known in obstetrics. The risk of uterine rupture with a classical uterine scar is four to nine times that of a low transverse uterine incision, and importantly, in about a third of cases such ruptures occurs before labor, sometimes several weeks before term. In addition, classical uterine incisions are associated with significantly more post-operative morbidity such as increased post-operative pulmonary complications and intestinal obstruction [9], [13]. However, the tendency for more blood loss when VSI and classical uterine incisions are made [9] is not in agreement with the finding in this study (obtained after controlling for confounding variables) and that of others [8]. Additionally, others have reported a greater incidence of wound complications when VSI are made [8]; we, like others [14], did not find such an association. Similarly, Wolfe et al. [11], suggested that neither choice of skin incision nor type of anesthesia appeared to be related to operative morbidity.

Our study had several limitations. First, it was a retrospective cohort and we were unable to clearly determine the basis for clinical decision-making that underpinned the management of these women. Patients in the VSI group were older and had a higher BMI compared with those who had LTSI. Although speculative, older women are more likely to have completed childbearing, therefore, future morbidity from a classical uterine incision may be less of an issue. Furthermore, obese patients with a higher BMI are more likely to have a larger pannus that needs to be avoided, hence a propensity for a VSI. Although such potential confounders were included in the logistic regression model, including the decision to perform a tubal ligation at the time of cesarean delivery, the wide confidence interval suggest an imprecise effect. Therefore the degree to which such confounders influenced the outcome of interest (classical uterine incision) is still largely unknown and indicates that more data (i.e. future studies) are needed to get to a true measure of the effect. The distance of the LTSI from the pubic bone was not addressed, nor was the use of drains which were infrequently and inconsistently used in our institution. No distinction was made between supraumbilical, periumbilical and infraumbilical incisions. In addition, we have not addressed the individual surgeons’ experience or surgical approach preference, all of which could bias the results of our study. Also, this study is from a single institution in one geographical location, thus reports by obstetric teams from other institutions on morbidly obese women will be required in order to validate our findings. Because our institution is a referral center and the study population was from multiple providers, our information regarding wound complications was limited to inpatient records. Therefore, some difference between groups could be secondary to patients following up with their primary obstetricians rather than at our institution.

To date, there is no accepted view on the most appropriate surgical approach for skin incision in morbidly obese patients undergoing cesarean delivery. Given the potential impact of a classical uterine scar, discussion between obstetricians and patients should involve the long-term morbidity associated with a classical uterine incision and the need for repeat cesarean for future deliveries. As Tixier et al. [14] suggest, the abdominal panniculus should be evaluated prior to onset of labor to determine where the incision would be made, even in those women attempting vaginal delivery because of the possible need for intervention. Our study suggests that the type of skin incision does not influence the perioperative morbidity among morbidly obese women undergoing cesarean section. Similar conclusions were reached by two other studies in morbidly obese patients that compared supraumbilical with subumbilical transverse [14] or with low transverse [15] skin incisions.

Back to Article Outline

References 

1. World Health Organization . Obesity and overweight. 2006;http://www.who.int/mediacentre/factsheets/fs311/en/index.html
   • View In Article
2. Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in Obesity among adults, 1999-2008. JAMA. 2010;303:235–241
   • View In Article
   • CrossRef
3. ACOG Committee Opinion number 315, September 2005 . Obesity in pregnancy. Obstet Gynecol. 2005;106:671–675
   • View In Article
   • MEDLINE
   • CrossRef
4. ACOG committee opinion. Number 319, October 2005 . The role of obstetrician-gynecologist in the assessment and management of obesity. Obstet Gynecol. 2005;106:895–899
   • View In Article
   • MEDLINE
   • CrossRef
5. Weiss JL, Malone FD, Emig D, et al. Obesity, obstetric complications and cesarean delivery rate-a population-based screening study. Am J Obstet Gynecol. 2004;190:1091–1097
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (131 KB)
   • CrossRef
6. Rodriguez A, Mejia A, Gonzalez-Quintero VH. Intraoperative and postoperative complications in cesarean deliveries among obese women. Obstet Gynecol. 2008;111:37S
   • View In Article
7. Gallup DG. Modifications of celiotomy techniques to decrease morbidity in obese gynecologic patients. Am J Obstet Gynecol. 1984;150:171–178
   • View In Article
   • MEDLINE
8. Wall PD, Deucy EE, Glantz JC, Pressman EK. Vertical skin incisions and wound complications in the obese parturient. Obstet Gynecol. 2003;102:952–956
   • View In Article
   • MEDLINE
   • CrossRef
9. Lao TT, Halpern SH, Crosby ET, Huh C. Uterine incision and maternal blood loss in preterm caesarean section. Arch Gynecol Obstet. 1993;252:113–117
   • View In Article
   • MEDLINE
   • CrossRef
10. Balsiger DM, Luque-Deleon E, Sarr MG. Surgical treatment of obesity: who is an appropriate candidate?. Mayo Clin Proc. 1997;72:551–558
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (7203 KB)
11. Wolfe HM, Gross TL, Sokol RJ, Bottoms SF, Thompson KL. Determinants of morbidity in obese women delivered by cesarean. Obstet Gynecol. 1988;71:691–696
    • View In Article
    • MEDLINE
12. Perlow JH, Morgan MA. Massive maternal obesity and perioperative cesarean morbidity. Am J Obstet Gynecol. 1994;170:560–565
    • View In Article
    • Abstract
    • Full Text
13. Carswell W. The current status of classical caesarean section. Scott Med J. 1973;18:105–108
    • View In Article
    • MEDLINE
14. Tixier H, Thouvenot S, Coulange L, et al. Cesarean section in morbidly obese women: supra or subumbilical transverse incision?. Acta Obstet Gynecol Scand. 2009;88:1049–1052
    • View In Article
    • CrossRef
15. Houston MC, Raynor BD. Postoperative morbidity in the morbidly obese parturient woman: supraumbilical and low transverse abdominal approaches. Am J Obstet Gynecol. 2000;182:1033–1035
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (71 KB)
    • CrossRef