Association Between Obesity and the Trends of Routes of Hysterectomy Performed for Benign Indications

Abstract

OBJECTIVE: 

To estimate the association between obesity and the recent trends of routes chosen for hysterectomy performed for benign indications in the United States.

MATERIALS AND METHODS: 

Using the American College of Surgeons–National Surgical Quality Improvement Project's database, patients who underwent hysterectomy for benign indications from 2005 to 2011 were identified by International Classification of Diseases, 9th Revision codes and were categorized into total abdominal hysterectomy (TAH), total vaginal hysterectomy (TVH), laparoscopically assisted vaginal hysterectomy (LAVH), and total laparoscopic hysterectomy (TLH). The patients were divided into four subgroups according to body mass index (BMI) (less than 25, 25–29.9, 30–39.9, and 40 or greater). The data were analyzed using Student's t test or χ² and Fisher's exact test.

RESULTS: 

A total of 18,810 patients underwent hysterectomy for benign indications during the study period: 9,852 (52.4%) were TAH, 5,146 (27.4%) TVH, 2,296 (12.2%) LAVH, and 1,516 (8.0%) TLH. The rates of TAH increased from 45.7% in patients with ideal body weight to 62% in morbidly obese patients (P<.001). The rate of TVH and LAVH decreased from 32.7% and 13.3% in patients with ideal body weight to 17.1% and 11.7% in morbidly obese patients, respectively (P<.001 and 0.04). The rate of TLH performed was independent of BMI (P=.61). Higher BMI was associated with longer operative time (P<.001) in all routes of hysterectomy. The rates of superficial and deep wound infections were higher with increasing BMI in patients undergoing TAH (P<.001) but not with TVH (P=.26), LAVH (P=1.0), or TLH (P=.48).

CONCLUSION: 

Regarding hysterectomy performed for benign indications, increasing BMI was associated with increased rate of TAH and decreased rate of TVH and LAVH, but not the rate of TLH. Increasing BMI was associated with increased operative time for all subgroups and increased surgical site infection in the TAH group.

LEVEL OF EVIDENCE: 

II

In Brief

Increasing obesity is associated with an increased rate of abdominal hysterectomy and a decreased rate of vaginal hysterectomy but has no association with the rate of total laparoscopic hysterectomy.

Hysterectomy is one of the most frequently performed surgical procedures in the United States.¹ There are three approaches to hysterectomy; abdominal hysterectomy, vaginal hysterectomy, and laparoscopic hysterectomy with or without robotic assistance.^1,2 A minimally invasive approach for hysterectomy is more favorable as a result of its well-known benefits including less blood loss, fewer perioperative complications, less postoperative pain, shorter hospital stay, quicker recovery time, and better cosmesis.^2,3

Different factors affect the surgeon's decision on the route chosen for hysterectomy, including safety and cost-effectiveness.¹ The American College of Obstetricians and Gynecologists considers the vaginal approach to be the ideal route for performance of benign hysterectomy when feasible.¹ Additionally, the American Association of Gynecologic Laparoscopists concludes that most hysterectomies for benign disease should be performed either vaginally or laparoscopically.⁴ Because obesity is now considered a pandemic⁵ with increasing prevalence, surgeons must consider its effects on perioperative complications and surgical decision-making. It has been shown that patients who are obese experience some of the greatest differential benefit from minimally invasive techniques.⁶ It has been suggested that if the vaginal route was deemed unsuitable for any reason, the laparoscopic route is more favorable compared with the abdominal approach.⁷

The objective of this study is to estimate the association between obesity and the trends toward minimally invasive approaches of hysterectomy performed for benign indications and to describe the perioperative outcomes in each group using the American College of Surgeons–National Surgical Quality Improvement Project's database.

MATERIALS AND METHODS

After obtaining exempt status from the University of South Florida's institutional review board, the American College of Surgeons–National Surgical Quality Improvement Project database from 2005 to 2011 was queried for all patients who underwent hysterectomy using appropriate International Classification of Diseases, 9th Revision (ICD-9); codes (Appendix 1). The American College of Surgeons–National Surgical Quality Improvement Project database, which is publically available and deidentified, is a quality improvement initiative originally developed by the Veterans' Health Administration in 1991 and adopted by the American College of Surgeons in 2001.⁸ The database includes more than 300 participating hospitals nationwide, both community and academic. Data entry points include but are not limited to demographics, comorbidities, laboratory values, operative variables, and 30-day postoperative outcomes, complications, mortality, reoperation, and length of stay.⁹ Routine auditing and the use of specially trained surgical nurses to record patient variables ensure high-quality data.^8,10 The National Surgical Quality Improvement Project provides a highly reliable data system to compare risk-adjusted outcomes that surgeons can have confidence in, but it also provides robust data to allow for intensive quality improvement efforts at the local level.¹¹ A random 8-day sampling method is used to ensure a diverse range of surgical procedures is captured.¹² It is believed that a random sample is better because a sample would provide knowledge about different types of operations and their outcomes performed by the different surgical subspecialties and the surgical service as a whole.^9,11

Only those patients with a benign indication for the hysterectomy procedures were included. The list of International Classification of Diseases, 9th Revision diagnostic codes used to query the database and flow chart of the inclusion criteria is shown in Appendix 1 and Figure 1, respectively. The National Surgical Quality Improvement Project database does not identify robotically assisted laparoscopic hysterectomy separately from total laparoscopic hysterectomy (TLH); therefore, these groups are both denoted as TLH for the purposes of this study.

Fig. 1: Flowchart of inclusion criteria. ICD-9, International Classification of Diseases, 9th Revision; CPT, Current Procedural Terminology; TAH, total abdominal hysterectomy; TVH, total vaginal hysterectomy; LAVH, laparoscopically assisted vaginal hysterectomy; TLH, total laparoscopic hysterectomy.Mikhail. Obesity and Surgical Trends for Hysterectomy. Obstet Gynecol 2015.

Patients were divided into four groups according to body mass index (BMI, calculated as weight (kg)/[height (m)]²) as follows: normal and underweight (less than 25), overweight (25–29.9), obese (class I and II; 30–39.9), and morbid obesity (class III) (40 or greater). Patients were subcategorized using Current Procedural Terminology codes into the following subgroups: 1) total abdominal hysterectomy (TAH)—58150, 2) total vaginal hysterectomy (TVH)—58260, 3) TLH—58570, and 4) laparoscopically assisted vaginal hysterectomy (LAVH)—58550.

The trends of hysterectomy during the study period were determined and the proportion of patients who underwent hysterectomy through different routes each year were compared for each BMI category. For each hysterectomy route, certain perioperative outcomes such as operative time, total length of hospital stay, rates of superficial infection, wound infection, pneumonia, pulmonary embolism, urinary tract infection, peripheral nerve injury, blood transfusion, deep venous thrombosis, and reoperation were compared and analyzed for each BMI group.

Data were analyzed using Stata 13.1. For univariate normally distributed continuous data, we applied the Student's t test. For skewed data we used the Wilcoxon rank-sum test. χ² or Fisher's exact test was used for categorical data. We used the test for trend¹³ or a nonparametric k-sample test on the equality of medians to assess the differences across categories of interest as appropriate. The Strengthening the Reporting of Observational studies in Epidemiology guidelines for reporting observational studies were strictly followed.¹⁴

RESULTS

During the study period, the total number of patients who underwent hysterectomy for benign indications who met our inclusion criteria was 18,810. The number of TAH was 9,852 (52.4%), TVH was 5,146 (27.4%), LAVH was 2,296 (12.2%), and TLH was 1,516 (8.0%). The distribution of these cases during the study period is shown in Table 1. Between 2007 and 2011, there was a more than 20% decline in the proportion of TAHs performed in the United States (Table 1). Although the proportion of TVH and LAVH remained stable over the entire study period, the proportion of TLH increased by almost 10% between 2008 and 2011 (Table 1).

Table 1: Number of Different Types of Hysterectomy Performed From 2005 to 2011

Stratification of the number of different routes for hysterectomy performed by BMI during the 7-year period showed that the rates of TAH significantly increased from 45.7% in patients with ideal body weight to 62% in morbidly obese patients (P<.001). Between each BMI category, approximately 5% increase in the proportion of TAH was noted (Fig. 2; Table 2) and the overall increase in the rate of TAH between patients with normal body weight and morbid obesity was 16.3% (Table 2). On the contrary, the rates of both TVH and LAVH significantly decreased from 32.7% and 13.3% in patients with ideal body weight to 17.1% and 11.7% in morbidly obese patients, respectively (P<.001 and .04) (Table 2). For LAVH, the decline was across all BMI categories (P<.001) and was more pronounced in the obese and morbid obesity group (Fig. 2). The overall percent decline in the rate of TVH between patients with normal body weight and morbid obesity was 15.6% (Table 2). The rate of LAVH was only negatively influenced by morbid obesity when compared with patients with normal body weight (Table 2). There were no significant differences in the rate of TLH with increasing BMI (P=.61).

Trends of different route of hysterectomy performed between 2005 and 2011 stratified by body mass index. Total abdominal hysterectomy (A), total vaginal hysterectomy (B), laparoscopically assisted vaginal hysterectomy (C), total laparoscopic hysterectomy (D).Fig. 2. Mikhail. Obesity and Surgical Trends for Hysterectomy. Obstet Gynecol 2015.

Table 2: Number of Hysterectomies Performed Between 2005 and 2011 Classified by Route and Body Mass Index

In patients who underwent TAH, a statistically significant difference was found in the mean operative time and the rate of wound infection (defined as superficial and deep surgical site infections) among the different BMI groups (P<.001). It was uniformly noted that more time was needed to complete the TAH with increasing BMI. Approximately 33 more minutes were needed to complete the TAH in the class III obesity group as compared with patients with normal or underweight body weight (Table 3). The mean operative time in patients who underwent TVH, LAVH, and TLH was significantly longer with increasing BMI (P<.001). An additional 12, 24, and 32 minutes were required to complete TVH, LAVH, and TLH, respectively, in patients with morbid obesity as compared with patients with ideal body weight (Table 3). The rates of superficial and deep surgical site infections were not affected by increasing BMI in patients who underwent TVH (P=.26), LAVH (P=1.0), or TLH (P=.48) (Table 3). Other perioperative outcomes analyzed and compared in the different groups of hysterectomy routes were total length of hospital stay (days), rates of pneumonia, pulmonary embolism, urinary tract infection, peripheral nerve injury, blood transfusion, deep venous thrombosis, and reoperation. As a result of the rare occurrence of these events, the study was underpowered to find any significant difference across BMI groups.

Table 3: Selected Perioperative Outcomes of Patient Stratified by Body Mass Index and Route of Hysterectomy

DISCUSSION

In this study, it was found that minimally invasive hysterectomy procedures such as TVH, LAVH, and TLH are being performed less frequently in patients with increased BMI. Obesity is associated with increased rate of TAH despite a national trend that shows an overall decline in the rate of TAH between 2007 and 2011. As a result of the known benefits of minimally invasive gynecologic surgery and the increasing prevalence of obesity in the general U.S. population, gynecologic surgeons should be trained and encouraged to adopt minimally invasive approaches in general and specifically for this group of patients.

Evidence in the medical literature supports the opinion that, when feasible, vaginal hysterectomy is the safest and most cost-effective route by which to remove the uterus.^1,2 Some authors have proposed certain factors as a contraindication for vaginal hysterectomy including: a narrow vagina, an undescended immobile uterus, nulliparity, prior cesarean delivery, and an enlarged uterus.¹ Others have set criteria for candidacy for vaginal hysterectomy including adequate pubic arch greater than 90°, adequate operative space in the lateral pelvic floor, and uterine size less than 16 cm.¹⁵ Many factors influence the route chosen for hysterectomy. These factors include uterine size, shape of the vagina and uterus, accessibility to the uterus, the need for concurrent procedures, patient characteristics, and surgeon training and experience (surgical volume).^1,16 Obesity can make vaginal hysterectomy more technically challenging.¹⁷ This is in agreement with the findings of our data, which show a steady decline in the rate TVH performed with increasing BMI.

Obesity has been shown to increase the risk of perioperative complications after elective procedures.¹⁸ The risks associated with increased BMI were dependent on the route of surgery and were mainly seen in patients undergoing TAH.¹⁹ This observation is supported by the findings of the current study. Because obesity is an independent risk factor for perioperative complications from hysterectomy, every effort should be made to perform a less morbid procedure in these patients. Our data demonstrate that, on the contrary, as BMI increases, TAHs are more commonly performed with more reported perioperative complications.

Laparoscopic hysterectomy seems to be a safe route for obese patients. In a study by Chopin et al,²⁰ obesity did not increase the complication rate in 1,460 patients who underwent TLH. In a retrospective study by Kondo et al²¹ studying 2,271 patients, they have found that obesity does not have an adverse effect on the feasibility and safety of laparoscopic hysterectomy. Similar results were found by Heinberg et al²² and O'Hanlan et al.²³ It is important to note that obesity as evidenced by increasing BMI makes any hysterectomy procedure more challenging as demonstrated by our data. Irrespective of the surgical approach, more time is required to complete the chosen surgery as the patient's BMI increases. Given that laparoscopy is feasible in patients with a high BMI with fewer reported perioperative complications compared with open surgery, efforts should be made to streamline techniques that allow more obese patients to undergo a hysterectomy through a minimally invasive approach.

The strengths of this study include the fact that this is a multiinstitutional database, which has been verified as accurate, reproducible, and reliable. Specially trained surgical clinical nurse reviewers are responsible for data entry for the National Surgical Quality Improvement Program database.^8,10 An additional advantage and strength of using American College of Surgeons–National Surgical Quality Improvement Program database is that the selection of cases to be included is based on random sampling, which minimizes the possibility of selection bias.⁹ Weaknesses are that the patients entered into the National Surgical Quality Improvement Program database are not a complete set of patients undergoing hysterectomy. As of 2012, approximately 382 hospitals were participating in the National Surgical Quality Improvement Program, which is a significant improvement since 2006 but is still a small fraction of all hospitals in the United States. Robotic cases could not be distinguished from laparoscopic cases based on the coding of data in the National Surgical Quality Improvement Program. Confounding factors including previous surgeries, comorbidities, uterine size, and surgeon's case volume could not be controlled for given the unavailability of these variables in the National Surgical Quality Improvement Program database. It is worth noting that records with concomitant procedures were not excluded, and concomitant procedures might be a confounding factor that affects secondary outcomes, specifically operative time. Despite these aforementioned weaknesses, it is evident from this study that more open hysterectomies are being performed in obese patients despite increased complication rates. Gynecologic surgeons should be trained and encouraged to perform minimally invasive procedures in this vulnerable group of patients. If the vaginal route is found to be technically challenging in these patients, TLH should be attempted given better perioperative outcomes when a hysterectomy is successfully completed laparoscopically.