Original Article
The Impact of Obesity on the 30-Day Morbidity and Mortality After Surgery for Endometrial Cancer Haider Mahdi, MD*, Amelia M. Jernigan, MD, Qataralnada Aljebori, BA, David Lockhart, BA, and Mehdi Moslemi-Kebria, MD From the Gynecologic Oncology Division, Ob/Gyn and Women’s Health Institute, Cleveland Clinic (Drs. Mahdi, Jernigan, and Moslemi-Kebria), Cleveland State University (Dr. Aljebori), Cleveland, Ohio, and Department of Biostatistics, University of Washington, Seattle, WA (Dr. Lockhart).
ABSTRACT Study Objectives: To examine the effect of body mass index (BMI) on postoperative 30-day morbidity and mortality after surgery to treat endometrial cancer. Design: Retrospective cohort study (Canadian Task Force classification II-2). Setting: National Surgical Quality Improvement Program. Patients: Patients with endometrial cancer who underwent surgery from 2005 to 2011. Interventions: Women were grouped according to weight, as follows: normal weight (BMI 18 to ,30), obese (BMI 30 to ,40), and morbidly obese (BMI R40). Univariate and multivariable logistic regression models were analyzed. Measurements and Main Results: Of 3947 patients, 38% were of normal weight, 38% were obese, and 24% were morbidly obese. Of these, 48% underwent laparoscopy and 52% underwent laparotomy. Overall 30-day morbidity and mortality were 13% and 0.7%, respectively. Obesity and morbid obesity were associated with a higher American Society of Anesthesiologists class, diabetes, and hypertension. Preoperatively, elevated serum creatinine concentration, hypoalbuminemia, and leukocytosis were more common in morbidly obese women than those of normal weight. Laparoscopic surgery was performed less frequently in morbidly obese women than in those of normal weight (42.5% vs 50%; p 5 .001). Morbidly obese patients were more likely to develop postoperative complications (morbidly obese 16% vs normal weight 13% vs obese 11%; p 5 .001), in particular surgical (morbidly obese 14% vs normal weight 11% vs obese 9%; p , .001) and infectious complications (morbidly obese 10% vs normal weight 5% vs obese 5%; p 5 .01). After laparotomy, morbidly obese women demonstrated a higher rate of any complication (normal weight 21%, obese 18%, morbidly obese 25%; p 5 .002), surgical complications (normal weight 18%, obese 14%, morbidly obese 22%; p 5 .002) and infectious complications (normal weight 6%, obese 10%, morbidly obese 16%; p , .001). After laparoscopy there was no difference in complication rates according to BMI group. The 30-day mortality was not significantly different according to BMI. After adjusting for confounders, obesity and morbid obesity did not independently predict 30-day morbidity or mortality. Conclusions: Morbidly obese patients with endometrial cancer have more preoperative morbidities and postoperative complications, in particular surgical and infectious complications, and are less likely to undergo minimally invasive surgery. However, obesity was not an independent predictor of perioperative outcomes after controlling for other confounders. Journal of Minimally Invasive Gynecology (2015) 22, 94–102 Ó 2015 AAGL. All rights reserved. Keywords:
DISCUSS
Endometrial cancer; Morbid obesity; Morbidity; Mortality; Obesity; Surgery
You can discuss this article with its authors and with other AAGL members at http://www.AAGL.org/jmig-22-1-JMIG-D-14-00252
Use your Smartphone to scan this QR code and connect to the discussion forum for this article now* * Download a free QR Code scanner by searching for ‘‘QR scanner’’ in your smartphone’s app store or app marketplace.
Disclosures: None declared. Abstract presented as a featured poster at the 45th Annual Meeting of the Society of Gynecologic Oncologists, March 21–25, 2014, Tampa, Florida. Corresponding author: Haider Mahdi, MD, Gynecologic Oncology Division, Ob/Gyn and Women’s Health Institute, Cleveland Clinic, 9500 Euclid Ave, Cleveland, OH. 1553-4650/$ - see front matter Ó 2015 AAGL. All rights reserved. http://dx.doi.org/10.1016/j.jmig.2014.07.014
E-mail: [email protected] Submitted May 15, 2014. Accepted for publication July 18, 2014. Available at www.sciencedirect.com and www.jmig.org
Mahdi et al.
Perioperative Outcome After Endometrial Cancer Surgery
95
Obesity is one of the most common causes of morbidity and death in the United States [1]. More than a decade ago it was estimated that morbidity and complications associated with obesity cost $11 billion every year [2]. This cost has likely increased as our population has grown more obese. Experts predict that by 2030 about half of the general population will be obese, with 11% qualifying as severely obese [3]. As gynecologic oncologists, the magnitude and prevalence of this issue is amplified; more than two-thirds of women with endometrial cancer, the most common cancer we treat, are obese [4,5]. The high cost of obesity is due, at least in part, to postoperative complications associated with excess body weight. These complications demand further interventions and longer hospital stay. Higher rates of wound and surgical site infections (SSIs), hospital stay, and readmission have been associated with obesity in a variety of surgical fields [6–10]. Understanding the postoperative complications that are associated with obesity helps us to counsel patients appropriately and to search for ways to optimize their recovery and minimize complications they might experience. In addition, in the era of modern healthcare reform, understanding which complications are more likely in the obese or morbidly obese patient are of interest, given plans to modify payment for services on the basis of hospital-acquired complications [11]. With the increase in minimally invasive surgery and advances in medical and surgical management of patients, perioperative complications and death are increasingly rare. However, when complications do occur, they can delay adjuvant therapy, create substantial anxiety, and generate additional costs. Most previous studies that explored the postoperative complications associated with obesity were small and were performed at a single institution [12–16]. Larger and more generalizable studies are needed to describe the true risk of these complications in obese women compared with those of normal weight. The objectives of the present study were to analyze the association of body mass index (BMI) with 30-day morbidity and mortality after surgery to treat endometrial cancer using nationwide data from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database.
values), intraoperative variables, and 30-day postoperative morbidity and mortality outcomes for a systematic and prospective sample of patients undergoing major surgical procedures. Data are collected in a standardized fashion according to strict definitions by dedicated surgical clinical nurse reviewers. Patients are followed up throughout their hospital course and after discharge from the hospital for up to 30 days postoperatively. A site Surgical Clinical Reviewer captures these data from medical records, physician office records, 30-day follow-up telephone interview with patients, and other methods. The present study received exempt status from the institutional review board. Patients with a diagnosis of endometrial cancer were identified from the 2005–2011 ACS NSQIP participant use files, which include data collected from 258 academic and community hospitals throughout the United States using ICD-9 (International Classification of Diseases, Ninth Revision) codes. Patients with endometrial cancer were included if they had undergone hysterectomy with or without other surgeries, according to CPT (Current Procedural Terminology) codes. For the purposes of the present study, 3 clinically relevant subgroups were generated for comparison. The normal weight group comprised patients with BMI .18 but ,30, the obese group included patients with BMI 30 to ,40, and the morbidly obese group included patients with BMI R40. A separate analysis was conducted stratifying morbidly obese patients into 3 categories: morbidly obese 1 (MO1), BMI 40 to 49; morbidly obese 2 (MO2), BMI 50 to 59; and morbidly obese 3 (MO3), BMI R60.
Materials and Methods Data Source The ACS NSQIP is a risk-adjusted data collection mechanism that collects and analyzes clinical outcomes data. Participating hospitals use their collected data to develop quality initiatives that improve surgical care and to identify elements in provided health care that can be improved when compared with other institutions. The ACS NSQIP collects data on 135 variables including preoperative risk factors (patient demographic data, comorbidities, and laboratory
Risk Factors and Outcome All risk factors available in the ACS NSQIP database were compared between the 3 BMI groups (normal weight, obese, and morbidly obese). The primary end points of the study were analysis of 30-day mortality, postoperative morbidity, procedure-related complications, surgical reexploration (return to the operating room within 30 days), and length of hospital stay. The secondary end point was to perform subset analysis of open laparotomy and laparoscopy groups separately. Composite end points were created to categorize postoperative complications into a few related groups: surgical complications (all SSIs, wound disruption, bleeding requiring transfusion, and peripheral nerve injury), renal complications (progressive renal failure and acute renal failure), pulmonary complications (pneumonia, unplanned intubation, and respiratory insufficiency requiring ventilation for 48 hours), infectious complications (systematic inflammatory response syndrome, sepsis, septic shock, SSI, and pneumonia), cardiovascular complications (pulmonary embolism, stroke/cerebrovascular event, cardiac arrest, myocardial infarction, and deep vein thrombosis requiring therapy), and any nonsurgical complications (any complication except surgical complications). Patients with preoperative sepsis were excluded from the study. Patients who were
96
ventilator dependent, had renal failure, or were receiving dialysis therapy before surgery were excluded from their respective complication category. Statistical Analysis Associations between categorical covariates were assessed using c2 tests. Group differences in means of continuous measures were assessed using the Student ttest or Wilcoxon rank-sum test. Preoperative laboratory values were used as both continuous and categorical variables: serum albumin concentration (.3 vs %3 mg/dL), hematocrit (,35 vs R35), serum creatinine concentration (R2 vs ,2 mg/dL), platelet count (,350 vs R350 cell/ mL3), and white blood cell count (,11 vs R11 cells/ml3). To adjust for surgical complexity, patients who underwent surgical procedures in addition to hysterectomy were given a specific score for each procedure. Then, based on the number of procedures performed, the sum of these scores was calculated, creating a modified surgical complexity scoring system. A score of 1 was given to any of the following procedures: hysterectomy with or without salpingo-oophorectomy, lymphadenectomy, or omentectomy. A score of 2 was given to any of the following procedures: small or large bowel resection, gastrectomy, hepatectomy, splenectomy, and pancreatectomy. Multivariable logistic regression models were used to assess the association between BMI groups and 30-day postoperative mortality, and between BMI groups and any complication, while controlling for possible confounders. For creation of the models, we considered all preoperative variables available in the ACS NSQIP database, including demographic data (age and race/ethnicity), preoperative health status and comorbidities, preoperative laboratory values (serum albumin concentration, creatinine concentration, white blood cell count, platelet count, and hematocrit), and operative factors (operative time, American Society of Anesthetists class, and modified surgical complexity score). Confounders that were significantly associated with postoperative complications or death in the univariate model (p , .05) were identified and included in the multivariate models. A final logistic regression model was run using BMI groups and all confounders found in this way. All tests were considered statistically significant at p , .05, and p values were 2-tailed. Commercially available software (STATA version 10.0; StataCorp., College Station, TX) was used for analysis of the data. Results Study inclusion criteria were met by 3947 women. Of these, 1510 (38%) were of normal weight, 1496 (38%) were obese, and 941 (24%) were morbidly obese. Demographic data and clinical characteristics of each group are given in Table 1. Younger patients represented a higher proportion of the morbidly obese group than the normal weight
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015
group (p , .001). Compared with normal weight, obesity and morbid obesity were associated with a higher American Society of Anesthesiologists (ASA) class (p , .001 and p , .001 for both), diabetes (p , .001 and p , .001 for both), and hypertension (p , .001 and p , .001 for both). However, no differences in cardiac, respiratory, renal, and neurologic morbidities were noted between BMI groups (Table 1). Preoperatively, an elevated serum creatinine concentration (p 5 .003), hypoalbuminemia (p 5 .001), and leukocytosis (p , .001) were more common in morbidly obese patients than in those of normal weight; however, these differences were not observed between obese and normal weight patients (Table 1). Laparoscopic surgery was performed less frequently in morbidly obese patients than in normal weight patients (42.5% vs 50%; p 5 .001); however, no difference was noted between obese and normal weight patients (49% vs 50%; p 5 .46) (Fig. 1; Table 1). There was no difference in percentage of lymphadenectomy between normal weight and obese patients (60% vs 59%; p 5 0.3). However, morbidly obese patients were less likely to have lymphadenectomy than were normal weight patients (47% vs 60%; p , .001). Postoperative Complications The rate of any postoperative complication among the entire cohort was 13% (21% for women who underwent laparotomy and 5% for women who underwent laparoscopy). Morbidly obese patients were at a significantly higher risk of any postoperative complication than were normal weight patients (16% vs 13%; p 5 .02); however, no difference was found between obese and normal weight patients (11% vs 13%; p 5 .14) (Table 2). Morbidly obese patients had higher rate of surgical complications (14% vs 11%; p 5 .01) and infectious complications (10% vs 5%; p , .01) than did nonobese patients; however, no difference was found between obese and normal weight patients insofar as surgical complications (9% vs 11%; p 5 .10) or infectious complications (6% vs 5%; p 5 .17) (Table 2). The 3 groups did not differ insofar as risk of postoperative nonsurgical complications (normal weight, 5%; obese, 3%; and morbidly obese, 5%; p 5 .06), cardiac complications (normal weight, 2%; obese, 2%; morbidly obese, 2%; p 5 .96), pulmonary complications (normal weight, 1%; obese, 1%; morbidly obese, 2%; p 5 .32), or renal complications (normal weight, 0.5%; obese, 1%; morbidly obese, 1%; p 5 .87). Morbidly obese women had a significantly higher mean number of complications than did women of normal weight (21 per 1000 cases vs 16 per 1000 cases; p 5 .02); however, no difference was found between obese and normal weight patients (13 per 1000 cases vs 16 per 1000 cases; p 5 .06). After laparotomy, the rates of any complication (normal weight, 21%; obese, 18%; morbidly obese, 25%; p 5 .006), surgical complications (normal weight, 18%; obese, 14%; morbidly obese, 22%; p 5 .002), and infectious complications (normal weight, 6%; obese, 10%; morbidly obese, 16%; p , .001)
Mahdi et al.
Perioperative Outcome After Endometrial Cancer Surgery
97
Table 1 Demographic and clinical characteristics of patients with endometrial cancer stratified according to body mass indexa
Variable
Total cohort (n 5 3947)
b
Age, yr Age group ,60 60–69 70–79 R80 Non-white race/ethnicity ASA class 1–2 3 R4 Dependent functional status Current smoker, within 1 year Diabetes Hypertension, receiving medication Cardiac comorbidities Respiratory comorbidities Renal comorbidities Neurologic comorbidities Ascites Weight loss, 10% within 6 months before surgery Blood transfusion before surgery Steroid use Chemotherapy before surgery Radiation therapy before surgery Laparoscopic approach Emergency surgery Surgical complexity 0–2 R3 Preoperative anemiad Preoperative hypoalbuminemiae Preoperative elevated serum creatininef Preoperative leukocytosisg Preoperative thrombocytosish
Patients with normal weight (n 5 1510)
Obese patients (n 5 1496)
Morbidly obese patients (n 5 941)
p value
64.7
63.1
58.4
,.001c ,.001c
1543 (39) 1337 (34) 746 (19) 321 (8) 1081 (27)
537 (36) 440 (29) 329 (22) 207 (14) 429 (28)
529 (35) 549 (37) 319 (21) 99 (7) 410 (27)
477 (51) 348 (37) 101 (11) 15 (2) 242 (26)
2011 (51) 1812 (46) 122 (3) 87 (2) 384 (10) 766 (19) 2234 (57) 129 (3) 119 (3) 10 (0) 116 (3) 44 (1) 53 (1)
1008 (67) 473 (31) 29 (2) 39 (3) 167 (11) 148 (10) 648 (43) 52 (3) 37 (2) 5 (0) 45 (3) 22 (1) 33 (2)
764 (51) 696 (47) 34 (2) 34 (2) 126 (8) 322 (22) 905 (60) 54 (4) 47 (3) 3 (0) 53 (4) 14 (1) 12 (1)
239 (25) 643 (68) 59 (6) 14 (1) 91 (10) 296 (31) 681 (72) 23 (2) 35 (4) 2 (0) 18 (2) 8 (1) 8 (1)
10 (1) 26 (2) 6 (0) 4 (0) 761 (50) 1 (0)
6 (0) 15 (1) 2 (0) 3 (0) 733 (49) 5 (0)
4 (0) 9 (1) 0 (9) 5 (1) 400 (42.5) 5 (1)
1492 (99) 18 (1) 196 (13) 69 (7) 23 (2) 57 (4) 185 (13)
1486 (99) 10 (1) 164 (11) 78 (8) 37 (3) 85 (6) 195 (13)
937 (100) 4 (0) 132 (14) 78 (13) 33 (4) 78 (8) 142 (16)
20 (1) 50 (1) 8 (0) 12 (0) 1901 (47.9) 11 (0) 3915 (99) 32 (1) 492 (12) 225 (6) 93 (2) 220 (6) 522 (13)
.34 ,.001c
.29 .05 ,.001c ,.001c .25 .18 .74 .07 .26 .001c .57 .13 .20 .51 ,.001c .04c .09
.04c ,.001c .008c ,.001c .09
ASA 5 American Society of Anesthesiologists. a Values are given as No. (%). b Mean values are provided, and comparisons are made using the Student t-test. c Statistical significance was determined at p , .05. d Hematocrit concentration ,35%. e Serum albumin concentration %3 g/dL. f Concentration .2 mg/dL. g White blood cell count .11 ! 109/L. h Platelet count .350 ! 109/L.
were still significantly different on the basis of BMI, in favor of fewer complications in the lower BMI groups. This difference in postoperative complication rates between BMI groups was not observed after laparoscopy (Table 2). No difference was found between the 3 groups in mean length of hospital stay (normal weight, 3.4 days; obese, 2.9 days; morbidly obese, 3.6 days; p 5 .32) or need for sur-
gical re-exploration (normal weight, 2%; obese, 1%; morbidly obese, 2%; p 5 .43). Among patients who underwent laparotomy there was a trend toward higher risk of surgical re-exploration in morbidly obese patients compared with those of normal weight (normal weight. 2%; obese, 1%; morbidly obese, 4%); however, this difference was not statistically significant (p 5 .13). Operative time was
98
Fig. 1 The relationship between obesity and the performance of minimally invasive surgery for endometrial cancer.
significantly longer in the morbidly obese group (182.4 minutes; p , .001) and obese group (173.4 minutes; p 5 .007) compared with the normal weight group (165.6 minutes). This difference was significant in both laparotomy and laparoscopy groups (Table 2). The rate and difference in risk of postoperative complications between the 3 groups stratified according to surgical approach are given in Table 2. Among the surgical and infectious complications, morbidly obese patients were at significantly higher risk of SSI (8.8% vs 3.3%; p , .001) and wound disruption (2.4% vs 0.6%; p , .001) (Table 3) than were normal weight patients. However, no difference was found between obese and normal weight patients for risk of SSI (4.6% vs 3.3%; p 5 .06) or wound disruption (0.6% vs 0.6%; p 5 .98). In contrast, the risk of perioperative blood transfusion in obese patients (4.3%; p , .001) and morbidly obese patients (4.4%; p , .001) was lower than for normal weight patients (7.9%). No difference between groups was found in the rate of peripheral nerve injuries (normal weight, 0.1%; obese, 0%; morbidly obese, 0.1%; p 5 .8), septic shock (normal weight, 0.3%; obese, 0.3%; morbidly obese, 0.4%; p 5 .72), or pneumonia (normal weight, 0.7%; obese, 0.9%; morbidly obese, 1.1%; p 5 .4). The risk of sepsis was significantly lower in obese patients compared with those of normal weight (0.7% vs 1.7%; p 5 .02), with no difference between normal weight and morbidly obese patients (1.7% vs 2%; p 5 .59) (Table 3). Among morbidly obese patients, those with BMI R60 had the highest rate of any complication (MO1, 15%; MO2, 17%; MO3, 23%; p , .001), surgical (MO1, 13%; MO2, 14%; MO3, 23%; p , .001), and infectious complications (MO1 9%; MO2, 10%; MO3, 21%; p , .001). No difference was found in the rate of nonsurgical complications (MO1, 5%; MO2, 6%; MO3, 9%; p 5 .11), cardiac complications (MO1, 2%; MO2, 1%; MO3, 2%; p 5 .99), pulmonary complications (MO1, 2%; MO2, 2%; MO3, 4%; p 5 .45), and renal complications (MO1, 1%; MO2, 0%; MO3, 0%; p 5 .69). No difference was found in mean length of hospital stay (MO1, 3.3 days; MO2, 4.6 days; MO3,
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015
3.8 days; p 5 .10) and rate of surgical re-exploration (MO1, 2%; MO2, 2%; MO3, 5%; p 5 .19). Among patients with infections and surgical complications, the rate of SSI (MO1, 8%; MO2, 9%; MO3, 18%; p , .001) and wound disruption (MO1, 2%; MO2, 2%; MO3, 4%; p , .001) was highest in group MO3 (BMI R60). After running a multivariate logistic regression model with adjustment for confounders, neither obesity nor morbid obesity were significant predictors of any complications (odds ratio [OR], 0.8; 95% confidence interval [CI], 0.6– 1.0; p 5 .09 for obese, and OR, 1.0; 95% CI, 0.8–1.3; p 5 .96 for morbidly obese) when compared with women of normal weight (Table 4). Other significant predictors of adverse 30-day postoperative outcome included surgical approach, surgical complexity, ascites, neurologic comorbidities, weight loss, preoperative blood transfusion, higher ASA class, operative time .3 hours, preoperative anemia, and leukocytosis (Table 4). Postoperative Mortality Overall 30-day mortality among the entire cohort was 0.7% (1.2% for the laparotomy group and 0.3% for the laparoscopy group). No difference in 30-day mortality was found between the 3 BMI groups (normal weight, 1.0%; obese, 0.8%; morbidly obese, 0.3%; p 5 .31; Table 2). After multivariate logistic regression model with adjustment for confounders, neither obesity nor morbid obesity were significant predictors of 30-day mortality (OR, 1.0; 95% CI, 0.4–2.4; p 5 .95 for obese, and OR, 0.4; 95% CI, 0.1–1.4; p 5 .23 for morbidly obese) when compared with normal weight (Table 4). Predictors of postoperative mortality were respiratory comorbidities, ascites, preoperative acute renal failure, and hypoalbuminemia. Discussion In this large analysis of 3947 patients from the NSQIP database we detailed the comorbidities and postoperative outcomes associated with obesity, making it one of the largest and most generalizable reports of its kind. Sixtytwo percent of the women in this study had a BMI of R30, consistent with previous studies that demonstrated that almost two-thirds of endometrial cancer survivors are obese [4,5]. At univariate analysis, morbid obesity (BMI R40) was associated with higher rates of postoperative complications, in particular surgical and infectious complications. Morbidly obese women experienced more SSIs and wound disruption. There was no difference between BMI groups insofar as hospital stay, surgical reexploration, or 30-day mortality. Of note, however, at multivariate analysis, after controlling for other confounders, obesity and morbid obesity were not independent predictors of 30-day postoperative morbidity or mortality. This indicates that excess body weight itself may not convey the increased risk but that the risk may be a result of differences
Mahdi et al.
Perioperative Outcome After Endometrial Cancer Surgery
99
Table 2 Operative parameters and 30-day perioperative outcomes among patients who underwent surgery to treat endometrial cancera
Variable Mortality Overall Laparotomy Laparoscopy Complications Surgical Overall Laparotomy Laparoscopy Nonsurgical Overall Laparotomy Laparoscopy Renal Overall Laparotomy Laparoscopy Pulmonary Overall Laparotomy Laparoscopy Cardiac Overall Laparotomy Laparoscopy Infection Overall Laparotomy Laparoscopy Any complication Overall Laparotomy Laparoscopy Surgical re-exploration Overall Laparotomy Laparoscopy Length of hospital stay, dayc Overall Laparotomy Laparoscopy No. of complicationsc Overall Laparotomy Laparoscopy Operative time, minc Overall Laparotomy Laparoscopy Operative time, min
Total cohort (n 5 3947)
Patients with normal weight (n 5 1510)
Obese patients (n 5 1496)
Morbidly obese patients (n 5 941)
p value
28 (1) 23 (1) 5 (0.3)
14 (1) 12 (2) 2 (0.3)
11 (0.8) 8 (1) 3 (0.4)
3 (0.3) 3 (1) 0
426 (11) 362 (18) 64 (3)
160 (11) 134 (18) 26 (3)
132 (9) 109 (14) 23 (3)
132 (14) 117 (22) 15 (4)
,.001b .002b .90
173 (4) 130 (6) 43 (2)
69 (5) 50 (7) 19 (2)
52 (3) 38 (5) 14 (2)
51 (5) 41 (8) 10 (3)
.06 .13 .70
21 (1) 20 (1) 1 (0.05)
7 (0.5) 7 (1) 0
9 (1) 9 (1) 0
5 (1) 4 (1) 1 (0.25)
.21 .43 .20
.87 .72 .15
61 (2) 46 (2) 15 (1)
22 (1) 16 (2) 6 (1)
19 (1) 15 (2) 4 (1)
19 (2) 14 (3) 5 (1)
.32 .74 .44
74 (2) 56 (3) 18 (1)
29 (2) 24 (3) 5 (1)
27 (2) 18 (2) 9 (1)
17 (2) 13 (2) 4 (1)
.96 .53 .52
255 (5) 206 (10) 49 (3)
71 (5) 44 (6) 27 (4)
87 (6) 73 (10) 14 (2)
96 (10) 88 (16) 8 (2)
,.001b ,.001b .10
519 (13) 427 (21) 92 (5)
196 (13) 157 (21) 39 (5)
168 (11) 134 (18) 34 (5)
153 (16) 134 (25) 19 (5)
.001b .006b .90
67 (2) 46 (2) 21 (1)
25 (2) 16 (2) 9 (1)
21 (1) 11 (1) 10 (1)
21 (2) 19 (4) 2 (1)
.30 .04b .40
3.4 5.2 1.6
2.9 4.4 1.5
3.6 5.1 1.7
.26 .50 .39
.16 .26 .06
.13 .21 .05
.21 .31 .06
.002b .02b .68
165.5 79.1 185.2
173.3 72.6 193.9
182.4 78.8 201.8
,.001b ,.001b .006b (Continued )
100
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015
Table 2 Continued
Variable %60 61–120 121–180 R180 a b c
Total cohort (n 5 3947)
Patients with normal weight (n 5 1510)
Obese patients (n 5 1496)
Morbidly obese patients (n 5 941)
107 (3) 1027 (26) 1222 (31) 1591 (40)
60 (4) 446 (30) 450 (30) 554 (37)
38 (3) 374 (25) 468 (31) 616 (41)
9 (1%) 207 (22) 304 (32) 421 (45)
p value ,.001b
Values are given as No. (%) or as mean. Statistical significance was determined at p , .05. Mean values are provided and comparisons were made using the Student t-test.
in preoperative comorbidities and treatment patterns observed in the obese population. Therefore, optimizing comorbidities and surgical treatment of obese patients may be of critical importance in preventing the complications that are commonly noted in this patient population. No difference was noted in short-term postoperative outcomes between BMI groups when evaluating only minimally invasive procedures. In the present study, minimally invasive surgery was an important predictor of favorable postoperative outcome at multivariate analysis, even after controlling for other confounders. However, the likelihood of successfully undergoing a laparoscopic endometrial cancer staging procedure was substantially lower in the morbidly obese group. The LAP2 trial provided prospective, randomized, controlled data that demonstrated that laparoscopic uterine cancer staging results in fewer postoperative complications than laparotomy does [17]. However, LAP2 also demonstrated the difficulty of successfully performing laparoscopy in obese women, with conversion rates of 27% in women with BMI .35 and 57% in women with BMI .40 [17]. In a single-institution study, Bijen et al [16] reported that in obese patients with a BMI .35, laparo-
scopic hysterectomy was not cost-effective because of the high conversion rate. However, it is important to consider surgeon expertise and experience before generating such a cut-off. Perhaps one of the most common concerns about performing a minimally invasive approach is the increased operative time associated with obesity. Our results confirm that operative times are longer in morbidly obese and obese patients compared with those of normal weight. They are also consistent with both LAP2 and previous retrospective studies that demonstrated that operative time is longer when a minimally invasive approach is used [12,17,18]. However, the difference in the present study is of marginal clinical importance and does not justify foregoing a laparoscopic approach in obese or morbidly obese women. These considerations underscore the importance of referring patients to surgeons who are familiar with performing laparoscopic or robotic surgery in women with excess body weight. As their obesity becomes more extreme, the difficulty of the procedure increases, but so does the benefit of performing it minimally invasively. Some singleinstitution studies report shorter operative times and lower
Table 3 Surgical and infectious complications among patients who underwent surgery for endometrial cancera
Complication Surgical site infection Wound disruption Perioperative bleeding requiring blood transfusion Peripheral nerve injury Sepsis Septic shock Pneumonia a b
Values are given as No. (%). Statistical significance was determined at p , .05.
Total cohort (n 5 3947)
Patients with normal weight (n 5 1510)
Obese patients (n 5 1496)
Morbidly obese patients (n 5 941)
p value
202 (5) 41 (1) 224 (6)
50 (3.3) 9 (0.6) 119 (7.9)
69 (4.6) 9 (0.6) 64 (4.3)
83 (8.8) 23 (2.4) 41 (4.4)
,.001b ,.001b ,.001b
1 (0.1) 26 (1.7) 5 (0.3) 10 (0.7)
0 11 (0.7) 4 (0.3) 13 (0.9)
1 (0.1) 19 (2.0) 4 (0.4) 10 (1.1)
.49 .01b .80 .56
2 (0.1) 56 (1.4) 13 (0.3) 33 (0.8)
Mahdi et al.
Perioperative Outcome After Endometrial Cancer Surgery
101
Table 4 Multivariate regression analysis for any complication and death within 30 days after surgery to treat endometrial cancera Variable
Any complication
p value
Variable
Death
BMI 30–39 BMI R40 Laparoscopy Surgical complexity .3 Non-white race/ethnicity
0.8 (0.6–10) 1.0 (0.8–1.3) 0.2 (0.2–0.3) 4.5 (2.0–10.8) 1.1 (0.9–1.4)
.09 .96 ,.001 ,.001 .46
BMI 30–39 BMI R40 Laparoscopy Surgical complexity .3 Age, yr 60–69 70–79 R80 Diabetes Respiratory comorbidities Cardiac comorbidities Hypertension Ascites Neurologic comorbidities Weight loss, 10% within 6 months before surgery Preoperative blood transfusion Dependent functional status ASA class 3 ASA class 4–5 Operative time, hr 1–2 2–3 .3 Serum creatinine R2 mg/dL Serum albumin .3 g/dL Anemia (hematocrit R35%) WBC count .11 000/mL Platelet count R350/mL
1.0 (0.4–2.4) 0.4 (0.1–1.6) 0.4 (0.1–1.2) 1.6 (0.1–10.4)
.99 .23 .12 .65
0.8 (0.2–2.8) 1.6 (0.5–5.4) 3.0 (0.8–10.9) NA 6.5 (1.9–19.7) 3.4 (0.8–11.9) NA 6.2 (1.3–25.6) NA 1.3 (0.2–6.0)
.77 .43 .09 ,.001 .07
2.3 (0.2–14.6) 3.7 (0.9–12.3) 0.9 (0.3–2.5) 1.3 (0.3–5.4)
.43 .05 .79 .77
NA NA NA 9.5 (2.7–29.3) 0.2 (0.1–0.6) 1.3 (0.5–4.0) 1.6 (0.5–5.0) 1.1 (0.3–3.3)
,.001 ,.001 .65 .42 .85
Diabetes Respiratory comorbidities Cardiac comorbidities Hypertension Ascites Neurologic comorbidities Weight loss, 10% within 6 months before surgery Preoperative blood transfusion Dependent functional status ASA class 3 ASA class 4–5 Operative time, hr 1–2 2–3 .3 Serum creatinine R2 mg/dL Serum albumin .3 g/dL Anemia (hematocrit R35%) WBC count .11 000/mL Platelet count R350/mL
1.0 (0.8–1.3) 1.5 (0.9–2.5) 1.3 (0.8–2.0) 1.1 (0.9–1.4) 3.5 (1.7–7.1) 1.9 (1.2–3.1) 2.3 (1.2–4.4)
.78 .10 .36 .31 ,.001 .01 .01
3.8 (1.3–11.9) 1.7 (0.9–2.9) 1.3 (1.0–1.6) 2.4 (1.4–3.8)
.02 .07 .03 ,.001
1.6 (0.8–3.3) 1.9 (1.0–4.0) 3.3 (1.7–7.1) 1.1 (0.6–1.9) 0.8 (0.5–1.2) 0.4 (0.3–0.5) 1.7 (1.2–2.4) 1.1 (0.9–1.5)
.23 .08 ,.001 .78 .94 ,.001 .01 .34
p value
.02 .79
ASA 5 American Society of Anesthesiologists; BMI 5 body mass index; NA 5 not applicable; WBC 5 white blood cell count. a Values are given as median (range).
conversion rates with robotic surgery compared with laparoscopy in morbidly obese women [19,20]; however, other institutions report contrary results [21,22]. A variety of factors, such as surgeon training and experience, dictate the optimal minimally invasive approach. The propensity for obese patients to develop wound and infectious complications has been observed in other surgical specialties [6–10] and has been repeatedly reported in series of patients with endometrial cancer [13,18,23]. BakkumGamez et al [23] recently reported on risk factors for SSI. Among other risk factors, BMI R40 was associated with a 4.7-fold risk of SSI, and laparotomy was associated with a 15-fold increase in risk of SSI when compared with patients who underwent a minimally invasive procedure. Each SSI resulted in a $5447 increase in median 30-day cost, not taking into account patient expenditures or wages lost [23]. At multivariate analysis, neither obesity nor morbid obesity was associated with 30-day morbidity or mortality. This can be explained, at least in part, by the difference in
preoperative comorbidities and surgical approach. Obese and morbidly obese patients were more likely to have a higher ASA disease class, diabetes, hypertension, acute renal failure, and hypoalbuminemia. Therefore, attention should be paid to identification and optimization of these medical comorbidities in the perioperative period. The NSQIP database consists of national data that reflect general practice patterns and outcomes in a variety of clinical settings. Therefore, unlike small single-institution studies or large prospective trials, these findings are reflective of real-world practices and their associated real-world outcomes. Still, our conclusions are limited by the nature of the study design. In this retrospective cohort study using data from a large database, generalizable associations are revealed by the large sample sizes, although cause-and-effect relationships can only be hypothesized. Much of the data used require reliance on CPT and ICD-9 codes, which raises concerns about accuracy and validity of the data. Other investigators have closely scrutinized the NSQIP database and found it to be highly accurate [24]. We also lacked
102
information on patient socioeconomic status, insurance status, and surgeon or hospital volume. Our results were not adjusted for stage at cancer diagnosis, which influences treatment plans. However, we adjusted for extent of surgical treatment, which is a proxy for burden of disease. Furthermore, we lacked information on procedures that were converted from laparoscopy to laparotomy. Ultimately, morbid obesity and obesity are not independently associated with a greater risk of postoperative complications or death after endometrial cancer surgery. The propensity for obese and morbidly obese patients to experience complications, in particular infectious and wound complications, is observed only at univariate analysis and likely reflects their preoperative comorbidities and the likelihood that they will undergo an open staging procedure. In the era of modern healthcare reform, the onus is on us as gynecologic cancer treatment providers to detail complications that can be expected, to appropriately counsel women, help develop reimbursement models, and guide further research with the objective of continuing to improve health outcomes for cancer survivors. Morbidly obese women need to be counseled that they are at increased risk of complications due to their medical comorbidities, and informed that this risk may be attenuated with use of a minimally invasive approach and the optimization of their comorbidities. Obese women should be referred to providers who have more experience with minimally invasive surgery in this population, and a concerted effort should be made to identify and optimize medical comorbidities in the perioperative period. References 1. Hurt RT, Frazier TH, McClave SA, Kaplan LM. Obesity epidemic: overview, pathophysiology, and the intensive care unit conundrum. JPEN J Parenter Enteral Nutr. 2011;35(5 Suppl):4S–13S. 2. Arterburn DE, Maciejewski ML, Tsevat J. Impact of morbid obesity on medical expenditures in adults. Int J Obese (Lond). 2005;29:334–339. 3. Finkelstein EA, Khavjou OA, Thompson H, et al. Obesity and severe obesity forecasts through 2030. Am J Prev Med. 2012;42:563–570. 4. Bittoni MA, Fisher JL, Fowler JM, Maxwell GL, Paskett ED. Assessment of the effects of severe obesity and lifestyle risk factors on stage of endometrial cancer. Cancer Epidemiol Biomarkers Prev. 2013;22: 76–81. 5. von Gruenigen VE, Gil KM, Frasure HE, Jenison EL, Hopkins MP. The impact of obesity and age on quality of life in gynecologic surgery. Am J Obstet Gynecol. 2005;193:1369–1375. 6. Fischer JP, Nelson JA, Kovach SJ, Serletti JM, Wu LC, Kanchwala S. Impact of obesity on outcomes in breast reconstruction: analysis of 15,937 patients from the ACS-NSQIP datasets. J Am Coll Surg. 2013; 217:656–664. 7. Pugely AJ, Callaghan JJ, Martin CT, Cram P, Gao Y. Incidence of and risk factors for 30-day readmission following elective primary total joint arthroplasty: analysis from the ACS-NSQIP. J Arthroplasty. 2013;28:1499–1504.
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015 8. Glarner CE, Greenblatt DY, Rettammel RJ, Neuman HB, Weber SM. Wound complications after inguinal lymph node dissection for melanoma: is ACS NSQIP adequate? Ann Surg Oncol. 2013;20:2049–2055. 9. Mustain WC, Davenport DL, Hourigan JS, Vargas HD. Obesity and laparoscopic colectomy: outcomes from the ACS-NSQIP database. Dis Colon Rectum. 2012;55:429–435. 10. Kazaure HS, Roman SA, Sosa JA. Obesity is a predictor of morbidity in 1,629 patients who underwent adrenalectomy. World J Surg. 2011;35: 1287–1295. 11. Centers for Medicare and Medicaid Services. Hospital-acquired conditions (present on admission indicator). September 20, 2012. Available at: http://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/ HospitalAcqCond/index.html?redirect5/HospitalAcqCond. Accessed November 6, 2013. 12. Litta P, Fabris AM, Breda E, et al. Laparoscopic surgical staging of endometrial cancer: does obesity influence feasibility and perioperative outcome? Eur J Gynaecol Oncol. 2013;34:231–233. 13. Giugale LE, Di Santo N, Smolkin ME, Havrilesky LJ, Modesitt SC. Beyond mere obesity: effect of increasing obesity classifications on hysterectomy outcomes for uterine cancer/hyperplasia. Gynecol Oncol. 2012;127:326–331. 14. Bernardini MQ, Gien LT, Tipping H, Murphy J, Rosen BP. Surgical outcome of robotic surgery in morbidly obese patient with endometrial cancer compared to laparotomy. Int J Gynecol Cancer. 2012; 22:76–81. 15. Helm CW, Arumugam C, Gordinier ME, Metzinger DS, Pan J, Rai SN. Laparoscopic surgery for endometrial cancer: increasing body mass index does not impact postoperative complications. J Gynecol Oncol. 2011;22:168–176. 16. Bijen CB, de Bock GH, Vermeulen KM, et al. Laparoscopic hysterectomy is preferred over laparotomy in early endometrial cancer patients, however not cost effective in the very obese. Eur J Cancer. 2011;47: 2158–2165. 17. Walker JL, Piedmonte MR, Spirtos NM, et al. Laparoscopy compared with laparotomy for comprehensive surgical staging of uterine cancer: Gynecologic Oncology Group Study LAP2. J Clin Oncol. 2009;27: 5331–5336. 18. Tang KY, Gardiner SK, Gould C, Osmundsen B, Collins M, Winter WE 3rd. Robotic surgical staging for obese patients with endometrial cancer. Am J Obstet Gynecol. 2012;206:513.e1–513.e6. 19. Gehrig PA, Cantrell LA, Shafer A, Abaid LN, Mendivil A, Boggess JF. What is the optimal minimally invasive surgical procedure for endometrial cancer staging in the obese and morbidly obese woman? Gynecol Oncol. 2008;111:41–45. 20. Seamon LG, Cohn DE, Henretta MS, et al. Minimally invasive comprehensive surgical staging for endometrial cancer: robotics or laparoscopy? Gynecol Oncol. 2009;113:36–41. 21. Bell MC, Torgerson J, Seshadri-Kreaden U, Suttle AW, Hunt S. Comparison of outcomes and cost for endometrial cancer staging via traditional laparotomy, standard laparoscopy and robotic techniques. Gynecol Oncol. 2008;111:407–411. 22. Subramaniam A, Kim KH, Bryant SA, et al. A cohort study evaluating robotic versus laparotomy surgical outcomes of obese women with endometrial carcinoma. Gynecol Oncol. 2011;122:604–607. 23. Bakkum-Gamez JN, Dowdy SC, Borah BJ, et al. Predictors and costs of surgical site infections in patients with endometrial cancer. Gynecol Oncol. 2013;130:100–106. 24. Sellers MM, Merkow RP, Halverson A, et al. Validation of new readmission data in the American College of Surgeons National Surgical Quality Improvement Program. J Am Coll Surg. 2013;216: 420–427.
The Impact of Obesity on the 30-Day Morbidity and Mortality After Surgery for Endometrial Cancer Haider Mahdi, MD*, Amelia M. Jernigan, MD, Qataralnada Aljebori, BA, David Lockhart, BA, and Mehdi Moslemi-Kebria, MD From the Gynecologic Oncology Division, Ob/Gyn and Women’s Health Institute, Cleveland Clinic (Drs. Mahdi, Jernigan, and Moslemi-Kebria), Cleveland State University (Dr. Aljebori), Cleveland, Ohio, and Department of Biostatistics, University of Washington, Seattle, WA (Dr. Lockhart).
ABSTRACT Study Objectives: To examine the effect of body mass index (BMI) on postoperative 30-day morbidity and mortality after surgery to treat endometrial cancer. Design: Retrospective cohort study (Canadian Task Force classification II-2). Setting: National Surgical Quality Improvement Program. Patients: Patients with endometrial cancer who underwent surgery from 2005 to 2011. Interventions: Women were grouped according to weight, as follows: normal weight (BMI 18 to ,30), obese (BMI 30 to ,40), and morbidly obese (BMI R40). Univariate and multivariable logistic regression models were analyzed. Measurements and Main Results: Of 3947 patients, 38% were of normal weight, 38% were obese, and 24% were morbidly obese. Of these, 48% underwent laparoscopy and 52% underwent laparotomy. Overall 30-day morbidity and mortality were 13% and 0.7%, respectively. Obesity and morbid obesity were associated with a higher American Society of Anesthesiologists class, diabetes, and hypertension. Preoperatively, elevated serum creatinine concentration, hypoalbuminemia, and leukocytosis were more common in morbidly obese women than those of normal weight. Laparoscopic surgery was performed less frequently in morbidly obese women than in those of normal weight (42.5% vs 50%; p 5 .001). Morbidly obese patients were more likely to develop postoperative complications (morbidly obese 16% vs normal weight 13% vs obese 11%; p 5 .001), in particular surgical (morbidly obese 14% vs normal weight 11% vs obese 9%; p , .001) and infectious complications (morbidly obese 10% vs normal weight 5% vs obese 5%; p 5 .01). After laparotomy, morbidly obese women demonstrated a higher rate of any complication (normal weight 21%, obese 18%, morbidly obese 25%; p 5 .002), surgical complications (normal weight 18%, obese 14%, morbidly obese 22%; p 5 .002) and infectious complications (normal weight 6%, obese 10%, morbidly obese 16%; p , .001). After laparoscopy there was no difference in complication rates according to BMI group. The 30-day mortality was not significantly different according to BMI. After adjusting for confounders, obesity and morbid obesity did not independently predict 30-day morbidity or mortality. Conclusions: Morbidly obese patients with endometrial cancer have more preoperative morbidities and postoperative complications, in particular surgical and infectious complications, and are less likely to undergo minimally invasive surgery. However, obesity was not an independent predictor of perioperative outcomes after controlling for other confounders. Journal of Minimally Invasive Gynecology (2015) 22, 94–102 Ó 2015 AAGL. All rights reserved. Keywords:
DISCUSS
Endometrial cancer; Morbid obesity; Morbidity; Mortality; Obesity; Surgery
You can discuss this article with its authors and with other AAGL members at http://www.AAGL.org/jmig-22-1-JMIG-D-14-00252
Use your Smartphone to scan this QR code and connect to the discussion forum for this article now* * Download a free QR Code scanner by searching for ‘‘QR scanner’’ in your smartphone’s app store or app marketplace.
Disclosures: None declared. Abstract presented as a featured poster at the 45th Annual Meeting of the Society of Gynecologic Oncologists, March 21–25, 2014, Tampa, Florida. Corresponding author: Haider Mahdi, MD, Gynecologic Oncology Division, Ob/Gyn and Women’s Health Institute, Cleveland Clinic, 9500 Euclid Ave, Cleveland, OH. 1553-4650/$ - see front matter Ó 2015 AAGL. All rights reserved. http://dx.doi.org/10.1016/j.jmig.2014.07.014
E-mail: [email protected] Submitted May 15, 2014. Accepted for publication July 18, 2014. Available at www.sciencedirect.com and www.jmig.org
Mahdi et al.
Perioperative Outcome After Endometrial Cancer Surgery
95
Obesity is one of the most common causes of morbidity and death in the United States [1]. More than a decade ago it was estimated that morbidity and complications associated with obesity cost $11 billion every year [2]. This cost has likely increased as our population has grown more obese. Experts predict that by 2030 about half of the general population will be obese, with 11% qualifying as severely obese [3]. As gynecologic oncologists, the magnitude and prevalence of this issue is amplified; more than two-thirds of women with endometrial cancer, the most common cancer we treat, are obese [4,5]. The high cost of obesity is due, at least in part, to postoperative complications associated with excess body weight. These complications demand further interventions and longer hospital stay. Higher rates of wound and surgical site infections (SSIs), hospital stay, and readmission have been associated with obesity in a variety of surgical fields [6–10]. Understanding the postoperative complications that are associated with obesity helps us to counsel patients appropriately and to search for ways to optimize their recovery and minimize complications they might experience. In addition, in the era of modern healthcare reform, understanding which complications are more likely in the obese or morbidly obese patient are of interest, given plans to modify payment for services on the basis of hospital-acquired complications [11]. With the increase in minimally invasive surgery and advances in medical and surgical management of patients, perioperative complications and death are increasingly rare. However, when complications do occur, they can delay adjuvant therapy, create substantial anxiety, and generate additional costs. Most previous studies that explored the postoperative complications associated with obesity were small and were performed at a single institution [12–16]. Larger and more generalizable studies are needed to describe the true risk of these complications in obese women compared with those of normal weight. The objectives of the present study were to analyze the association of body mass index (BMI) with 30-day morbidity and mortality after surgery to treat endometrial cancer using nationwide data from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database.
values), intraoperative variables, and 30-day postoperative morbidity and mortality outcomes for a systematic and prospective sample of patients undergoing major surgical procedures. Data are collected in a standardized fashion according to strict definitions by dedicated surgical clinical nurse reviewers. Patients are followed up throughout their hospital course and after discharge from the hospital for up to 30 days postoperatively. A site Surgical Clinical Reviewer captures these data from medical records, physician office records, 30-day follow-up telephone interview with patients, and other methods. The present study received exempt status from the institutional review board. Patients with a diagnosis of endometrial cancer were identified from the 2005–2011 ACS NSQIP participant use files, which include data collected from 258 academic and community hospitals throughout the United States using ICD-9 (International Classification of Diseases, Ninth Revision) codes. Patients with endometrial cancer were included if they had undergone hysterectomy with or without other surgeries, according to CPT (Current Procedural Terminology) codes. For the purposes of the present study, 3 clinically relevant subgroups were generated for comparison. The normal weight group comprised patients with BMI .18 but ,30, the obese group included patients with BMI 30 to ,40, and the morbidly obese group included patients with BMI R40. A separate analysis was conducted stratifying morbidly obese patients into 3 categories: morbidly obese 1 (MO1), BMI 40 to 49; morbidly obese 2 (MO2), BMI 50 to 59; and morbidly obese 3 (MO3), BMI R60.
Materials and Methods Data Source The ACS NSQIP is a risk-adjusted data collection mechanism that collects and analyzes clinical outcomes data. Participating hospitals use their collected data to develop quality initiatives that improve surgical care and to identify elements in provided health care that can be improved when compared with other institutions. The ACS NSQIP collects data on 135 variables including preoperative risk factors (patient demographic data, comorbidities, and laboratory
Risk Factors and Outcome All risk factors available in the ACS NSQIP database were compared between the 3 BMI groups (normal weight, obese, and morbidly obese). The primary end points of the study were analysis of 30-day mortality, postoperative morbidity, procedure-related complications, surgical reexploration (return to the operating room within 30 days), and length of hospital stay. The secondary end point was to perform subset analysis of open laparotomy and laparoscopy groups separately. Composite end points were created to categorize postoperative complications into a few related groups: surgical complications (all SSIs, wound disruption, bleeding requiring transfusion, and peripheral nerve injury), renal complications (progressive renal failure and acute renal failure), pulmonary complications (pneumonia, unplanned intubation, and respiratory insufficiency requiring ventilation for 48 hours), infectious complications (systematic inflammatory response syndrome, sepsis, septic shock, SSI, and pneumonia), cardiovascular complications (pulmonary embolism, stroke/cerebrovascular event, cardiac arrest, myocardial infarction, and deep vein thrombosis requiring therapy), and any nonsurgical complications (any complication except surgical complications). Patients with preoperative sepsis were excluded from the study. Patients who were
96
ventilator dependent, had renal failure, or were receiving dialysis therapy before surgery were excluded from their respective complication category. Statistical Analysis Associations between categorical covariates were assessed using c2 tests. Group differences in means of continuous measures were assessed using the Student ttest or Wilcoxon rank-sum test. Preoperative laboratory values were used as both continuous and categorical variables: serum albumin concentration (.3 vs %3 mg/dL), hematocrit (,35 vs R35), serum creatinine concentration (R2 vs ,2 mg/dL), platelet count (,350 vs R350 cell/ mL3), and white blood cell count (,11 vs R11 cells/ml3). To adjust for surgical complexity, patients who underwent surgical procedures in addition to hysterectomy were given a specific score for each procedure. Then, based on the number of procedures performed, the sum of these scores was calculated, creating a modified surgical complexity scoring system. A score of 1 was given to any of the following procedures: hysterectomy with or without salpingo-oophorectomy, lymphadenectomy, or omentectomy. A score of 2 was given to any of the following procedures: small or large bowel resection, gastrectomy, hepatectomy, splenectomy, and pancreatectomy. Multivariable logistic regression models were used to assess the association between BMI groups and 30-day postoperative mortality, and between BMI groups and any complication, while controlling for possible confounders. For creation of the models, we considered all preoperative variables available in the ACS NSQIP database, including demographic data (age and race/ethnicity), preoperative health status and comorbidities, preoperative laboratory values (serum albumin concentration, creatinine concentration, white blood cell count, platelet count, and hematocrit), and operative factors (operative time, American Society of Anesthetists class, and modified surgical complexity score). Confounders that were significantly associated with postoperative complications or death in the univariate model (p , .05) were identified and included in the multivariate models. A final logistic regression model was run using BMI groups and all confounders found in this way. All tests were considered statistically significant at p , .05, and p values were 2-tailed. Commercially available software (STATA version 10.0; StataCorp., College Station, TX) was used for analysis of the data. Results Study inclusion criteria were met by 3947 women. Of these, 1510 (38%) were of normal weight, 1496 (38%) were obese, and 941 (24%) were morbidly obese. Demographic data and clinical characteristics of each group are given in Table 1. Younger patients represented a higher proportion of the morbidly obese group than the normal weight
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015
group (p , .001). Compared with normal weight, obesity and morbid obesity were associated with a higher American Society of Anesthesiologists (ASA) class (p , .001 and p , .001 for both), diabetes (p , .001 and p , .001 for both), and hypertension (p , .001 and p , .001 for both). However, no differences in cardiac, respiratory, renal, and neurologic morbidities were noted between BMI groups (Table 1). Preoperatively, an elevated serum creatinine concentration (p 5 .003), hypoalbuminemia (p 5 .001), and leukocytosis (p , .001) were more common in morbidly obese patients than in those of normal weight; however, these differences were not observed between obese and normal weight patients (Table 1). Laparoscopic surgery was performed less frequently in morbidly obese patients than in normal weight patients (42.5% vs 50%; p 5 .001); however, no difference was noted between obese and normal weight patients (49% vs 50%; p 5 .46) (Fig. 1; Table 1). There was no difference in percentage of lymphadenectomy between normal weight and obese patients (60% vs 59%; p 5 0.3). However, morbidly obese patients were less likely to have lymphadenectomy than were normal weight patients (47% vs 60%; p , .001). Postoperative Complications The rate of any postoperative complication among the entire cohort was 13% (21% for women who underwent laparotomy and 5% for women who underwent laparoscopy). Morbidly obese patients were at a significantly higher risk of any postoperative complication than were normal weight patients (16% vs 13%; p 5 .02); however, no difference was found between obese and normal weight patients (11% vs 13%; p 5 .14) (Table 2). Morbidly obese patients had higher rate of surgical complications (14% vs 11%; p 5 .01) and infectious complications (10% vs 5%; p , .01) than did nonobese patients; however, no difference was found between obese and normal weight patients insofar as surgical complications (9% vs 11%; p 5 .10) or infectious complications (6% vs 5%; p 5 .17) (Table 2). The 3 groups did not differ insofar as risk of postoperative nonsurgical complications (normal weight, 5%; obese, 3%; and morbidly obese, 5%; p 5 .06), cardiac complications (normal weight, 2%; obese, 2%; morbidly obese, 2%; p 5 .96), pulmonary complications (normal weight, 1%; obese, 1%; morbidly obese, 2%; p 5 .32), or renal complications (normal weight, 0.5%; obese, 1%; morbidly obese, 1%; p 5 .87). Morbidly obese women had a significantly higher mean number of complications than did women of normal weight (21 per 1000 cases vs 16 per 1000 cases; p 5 .02); however, no difference was found between obese and normal weight patients (13 per 1000 cases vs 16 per 1000 cases; p 5 .06). After laparotomy, the rates of any complication (normal weight, 21%; obese, 18%; morbidly obese, 25%; p 5 .006), surgical complications (normal weight, 18%; obese, 14%; morbidly obese, 22%; p 5 .002), and infectious complications (normal weight, 6%; obese, 10%; morbidly obese, 16%; p , .001)
Mahdi et al.
Perioperative Outcome After Endometrial Cancer Surgery
97
Table 1 Demographic and clinical characteristics of patients with endometrial cancer stratified according to body mass indexa
Variable
Total cohort (n 5 3947)
b
Age, yr Age group ,60 60–69 70–79 R80 Non-white race/ethnicity ASA class 1–2 3 R4 Dependent functional status Current smoker, within 1 year Diabetes Hypertension, receiving medication Cardiac comorbidities Respiratory comorbidities Renal comorbidities Neurologic comorbidities Ascites Weight loss, 10% within 6 months before surgery Blood transfusion before surgery Steroid use Chemotherapy before surgery Radiation therapy before surgery Laparoscopic approach Emergency surgery Surgical complexity 0–2 R3 Preoperative anemiad Preoperative hypoalbuminemiae Preoperative elevated serum creatininef Preoperative leukocytosisg Preoperative thrombocytosish
Patients with normal weight (n 5 1510)
Obese patients (n 5 1496)
Morbidly obese patients (n 5 941)
p value
64.7
63.1
58.4
,.001c ,.001c
1543 (39) 1337 (34) 746 (19) 321 (8) 1081 (27)
537 (36) 440 (29) 329 (22) 207 (14) 429 (28)
529 (35) 549 (37) 319 (21) 99 (7) 410 (27)
477 (51) 348 (37) 101 (11) 15 (2) 242 (26)
2011 (51) 1812 (46) 122 (3) 87 (2) 384 (10) 766 (19) 2234 (57) 129 (3) 119 (3) 10 (0) 116 (3) 44 (1) 53 (1)
1008 (67) 473 (31) 29 (2) 39 (3) 167 (11) 148 (10) 648 (43) 52 (3) 37 (2) 5 (0) 45 (3) 22 (1) 33 (2)
764 (51) 696 (47) 34 (2) 34 (2) 126 (8) 322 (22) 905 (60) 54 (4) 47 (3) 3 (0) 53 (4) 14 (1) 12 (1)
239 (25) 643 (68) 59 (6) 14 (1) 91 (10) 296 (31) 681 (72) 23 (2) 35 (4) 2 (0) 18 (2) 8 (1) 8 (1)
10 (1) 26 (2) 6 (0) 4 (0) 761 (50) 1 (0)
6 (0) 15 (1) 2 (0) 3 (0) 733 (49) 5 (0)
4 (0) 9 (1) 0 (9) 5 (1) 400 (42.5) 5 (1)
1492 (99) 18 (1) 196 (13) 69 (7) 23 (2) 57 (4) 185 (13)
1486 (99) 10 (1) 164 (11) 78 (8) 37 (3) 85 (6) 195 (13)
937 (100) 4 (0) 132 (14) 78 (13) 33 (4) 78 (8) 142 (16)
20 (1) 50 (1) 8 (0) 12 (0) 1901 (47.9) 11 (0) 3915 (99) 32 (1) 492 (12) 225 (6) 93 (2) 220 (6) 522 (13)
.34 ,.001c
.29 .05 ,.001c ,.001c .25 .18 .74 .07 .26 .001c .57 .13 .20 .51 ,.001c .04c .09
.04c ,.001c .008c ,.001c .09
ASA 5 American Society of Anesthesiologists. a Values are given as No. (%). b Mean values are provided, and comparisons are made using the Student t-test. c Statistical significance was determined at p , .05. d Hematocrit concentration ,35%. e Serum albumin concentration %3 g/dL. f Concentration .2 mg/dL. g White blood cell count .11 ! 109/L. h Platelet count .350 ! 109/L.
were still significantly different on the basis of BMI, in favor of fewer complications in the lower BMI groups. This difference in postoperative complication rates between BMI groups was not observed after laparoscopy (Table 2). No difference was found between the 3 groups in mean length of hospital stay (normal weight, 3.4 days; obese, 2.9 days; morbidly obese, 3.6 days; p 5 .32) or need for sur-
gical re-exploration (normal weight, 2%; obese, 1%; morbidly obese, 2%; p 5 .43). Among patients who underwent laparotomy there was a trend toward higher risk of surgical re-exploration in morbidly obese patients compared with those of normal weight (normal weight. 2%; obese, 1%; morbidly obese, 4%); however, this difference was not statistically significant (p 5 .13). Operative time was
98
Fig. 1 The relationship between obesity and the performance of minimally invasive surgery for endometrial cancer.
significantly longer in the morbidly obese group (182.4 minutes; p , .001) and obese group (173.4 minutes; p 5 .007) compared with the normal weight group (165.6 minutes). This difference was significant in both laparotomy and laparoscopy groups (Table 2). The rate and difference in risk of postoperative complications between the 3 groups stratified according to surgical approach are given in Table 2. Among the surgical and infectious complications, morbidly obese patients were at significantly higher risk of SSI (8.8% vs 3.3%; p , .001) and wound disruption (2.4% vs 0.6%; p , .001) (Table 3) than were normal weight patients. However, no difference was found between obese and normal weight patients for risk of SSI (4.6% vs 3.3%; p 5 .06) or wound disruption (0.6% vs 0.6%; p 5 .98). In contrast, the risk of perioperative blood transfusion in obese patients (4.3%; p , .001) and morbidly obese patients (4.4%; p , .001) was lower than for normal weight patients (7.9%). No difference between groups was found in the rate of peripheral nerve injuries (normal weight, 0.1%; obese, 0%; morbidly obese, 0.1%; p 5 .8), septic shock (normal weight, 0.3%; obese, 0.3%; morbidly obese, 0.4%; p 5 .72), or pneumonia (normal weight, 0.7%; obese, 0.9%; morbidly obese, 1.1%; p 5 .4). The risk of sepsis was significantly lower in obese patients compared with those of normal weight (0.7% vs 1.7%; p 5 .02), with no difference between normal weight and morbidly obese patients (1.7% vs 2%; p 5 .59) (Table 3). Among morbidly obese patients, those with BMI R60 had the highest rate of any complication (MO1, 15%; MO2, 17%; MO3, 23%; p , .001), surgical (MO1, 13%; MO2, 14%; MO3, 23%; p , .001), and infectious complications (MO1 9%; MO2, 10%; MO3, 21%; p , .001). No difference was found in the rate of nonsurgical complications (MO1, 5%; MO2, 6%; MO3, 9%; p 5 .11), cardiac complications (MO1, 2%; MO2, 1%; MO3, 2%; p 5 .99), pulmonary complications (MO1, 2%; MO2, 2%; MO3, 4%; p 5 .45), and renal complications (MO1, 1%; MO2, 0%; MO3, 0%; p 5 .69). No difference was found in mean length of hospital stay (MO1, 3.3 days; MO2, 4.6 days; MO3,
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015
3.8 days; p 5 .10) and rate of surgical re-exploration (MO1, 2%; MO2, 2%; MO3, 5%; p 5 .19). Among patients with infections and surgical complications, the rate of SSI (MO1, 8%; MO2, 9%; MO3, 18%; p , .001) and wound disruption (MO1, 2%; MO2, 2%; MO3, 4%; p , .001) was highest in group MO3 (BMI R60). After running a multivariate logistic regression model with adjustment for confounders, neither obesity nor morbid obesity were significant predictors of any complications (odds ratio [OR], 0.8; 95% confidence interval [CI], 0.6– 1.0; p 5 .09 for obese, and OR, 1.0; 95% CI, 0.8–1.3; p 5 .96 for morbidly obese) when compared with women of normal weight (Table 4). Other significant predictors of adverse 30-day postoperative outcome included surgical approach, surgical complexity, ascites, neurologic comorbidities, weight loss, preoperative blood transfusion, higher ASA class, operative time .3 hours, preoperative anemia, and leukocytosis (Table 4). Postoperative Mortality Overall 30-day mortality among the entire cohort was 0.7% (1.2% for the laparotomy group and 0.3% for the laparoscopy group). No difference in 30-day mortality was found between the 3 BMI groups (normal weight, 1.0%; obese, 0.8%; morbidly obese, 0.3%; p 5 .31; Table 2). After multivariate logistic regression model with adjustment for confounders, neither obesity nor morbid obesity were significant predictors of 30-day mortality (OR, 1.0; 95% CI, 0.4–2.4; p 5 .95 for obese, and OR, 0.4; 95% CI, 0.1–1.4; p 5 .23 for morbidly obese) when compared with normal weight (Table 4). Predictors of postoperative mortality were respiratory comorbidities, ascites, preoperative acute renal failure, and hypoalbuminemia. Discussion In this large analysis of 3947 patients from the NSQIP database we detailed the comorbidities and postoperative outcomes associated with obesity, making it one of the largest and most generalizable reports of its kind. Sixtytwo percent of the women in this study had a BMI of R30, consistent with previous studies that demonstrated that almost two-thirds of endometrial cancer survivors are obese [4,5]. At univariate analysis, morbid obesity (BMI R40) was associated with higher rates of postoperative complications, in particular surgical and infectious complications. Morbidly obese women experienced more SSIs and wound disruption. There was no difference between BMI groups insofar as hospital stay, surgical reexploration, or 30-day mortality. Of note, however, at multivariate analysis, after controlling for other confounders, obesity and morbid obesity were not independent predictors of 30-day postoperative morbidity or mortality. This indicates that excess body weight itself may not convey the increased risk but that the risk may be a result of differences
Mahdi et al.
Perioperative Outcome After Endometrial Cancer Surgery
99
Table 2 Operative parameters and 30-day perioperative outcomes among patients who underwent surgery to treat endometrial cancera
Variable Mortality Overall Laparotomy Laparoscopy Complications Surgical Overall Laparotomy Laparoscopy Nonsurgical Overall Laparotomy Laparoscopy Renal Overall Laparotomy Laparoscopy Pulmonary Overall Laparotomy Laparoscopy Cardiac Overall Laparotomy Laparoscopy Infection Overall Laparotomy Laparoscopy Any complication Overall Laparotomy Laparoscopy Surgical re-exploration Overall Laparotomy Laparoscopy Length of hospital stay, dayc Overall Laparotomy Laparoscopy No. of complicationsc Overall Laparotomy Laparoscopy Operative time, minc Overall Laparotomy Laparoscopy Operative time, min
Total cohort (n 5 3947)
Patients with normal weight (n 5 1510)
Obese patients (n 5 1496)
Morbidly obese patients (n 5 941)
p value
28 (1) 23 (1) 5 (0.3)
14 (1) 12 (2) 2 (0.3)
11 (0.8) 8 (1) 3 (0.4)
3 (0.3) 3 (1) 0
426 (11) 362 (18) 64 (3)
160 (11) 134 (18) 26 (3)
132 (9) 109 (14) 23 (3)
132 (14) 117 (22) 15 (4)
,.001b .002b .90
173 (4) 130 (6) 43 (2)
69 (5) 50 (7) 19 (2)
52 (3) 38 (5) 14 (2)
51 (5) 41 (8) 10 (3)
.06 .13 .70
21 (1) 20 (1) 1 (0.05)
7 (0.5) 7 (1) 0
9 (1) 9 (1) 0
5 (1) 4 (1) 1 (0.25)
.21 .43 .20
.87 .72 .15
61 (2) 46 (2) 15 (1)
22 (1) 16 (2) 6 (1)
19 (1) 15 (2) 4 (1)
19 (2) 14 (3) 5 (1)
.32 .74 .44
74 (2) 56 (3) 18 (1)
29 (2) 24 (3) 5 (1)
27 (2) 18 (2) 9 (1)
17 (2) 13 (2) 4 (1)
.96 .53 .52
255 (5) 206 (10) 49 (3)
71 (5) 44 (6) 27 (4)
87 (6) 73 (10) 14 (2)
96 (10) 88 (16) 8 (2)
,.001b ,.001b .10
519 (13) 427 (21) 92 (5)
196 (13) 157 (21) 39 (5)
168 (11) 134 (18) 34 (5)
153 (16) 134 (25) 19 (5)
.001b .006b .90
67 (2) 46 (2) 21 (1)
25 (2) 16 (2) 9 (1)
21 (1) 11 (1) 10 (1)
21 (2) 19 (4) 2 (1)
.30 .04b .40
3.4 5.2 1.6
2.9 4.4 1.5
3.6 5.1 1.7
.26 .50 .39
.16 .26 .06
.13 .21 .05
.21 .31 .06
.002b .02b .68
165.5 79.1 185.2
173.3 72.6 193.9
182.4 78.8 201.8
,.001b ,.001b .006b (Continued )
100
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015
Table 2 Continued
Variable %60 61–120 121–180 R180 a b c
Total cohort (n 5 3947)
Patients with normal weight (n 5 1510)
Obese patients (n 5 1496)
Morbidly obese patients (n 5 941)
107 (3) 1027 (26) 1222 (31) 1591 (40)
60 (4) 446 (30) 450 (30) 554 (37)
38 (3) 374 (25) 468 (31) 616 (41)
9 (1%) 207 (22) 304 (32) 421 (45)
p value ,.001b
Values are given as No. (%) or as mean. Statistical significance was determined at p , .05. Mean values are provided and comparisons were made using the Student t-test.
in preoperative comorbidities and treatment patterns observed in the obese population. Therefore, optimizing comorbidities and surgical treatment of obese patients may be of critical importance in preventing the complications that are commonly noted in this patient population. No difference was noted in short-term postoperative outcomes between BMI groups when evaluating only minimally invasive procedures. In the present study, minimally invasive surgery was an important predictor of favorable postoperative outcome at multivariate analysis, even after controlling for other confounders. However, the likelihood of successfully undergoing a laparoscopic endometrial cancer staging procedure was substantially lower in the morbidly obese group. The LAP2 trial provided prospective, randomized, controlled data that demonstrated that laparoscopic uterine cancer staging results in fewer postoperative complications than laparotomy does [17]. However, LAP2 also demonstrated the difficulty of successfully performing laparoscopy in obese women, with conversion rates of 27% in women with BMI .35 and 57% in women with BMI .40 [17]. In a single-institution study, Bijen et al [16] reported that in obese patients with a BMI .35, laparo-
scopic hysterectomy was not cost-effective because of the high conversion rate. However, it is important to consider surgeon expertise and experience before generating such a cut-off. Perhaps one of the most common concerns about performing a minimally invasive approach is the increased operative time associated with obesity. Our results confirm that operative times are longer in morbidly obese and obese patients compared with those of normal weight. They are also consistent with both LAP2 and previous retrospective studies that demonstrated that operative time is longer when a minimally invasive approach is used [12,17,18]. However, the difference in the present study is of marginal clinical importance and does not justify foregoing a laparoscopic approach in obese or morbidly obese women. These considerations underscore the importance of referring patients to surgeons who are familiar with performing laparoscopic or robotic surgery in women with excess body weight. As their obesity becomes more extreme, the difficulty of the procedure increases, but so does the benefit of performing it minimally invasively. Some singleinstitution studies report shorter operative times and lower
Table 3 Surgical and infectious complications among patients who underwent surgery for endometrial cancera
Complication Surgical site infection Wound disruption Perioperative bleeding requiring blood transfusion Peripheral nerve injury Sepsis Septic shock Pneumonia a b
Values are given as No. (%). Statistical significance was determined at p , .05.
Total cohort (n 5 3947)
Patients with normal weight (n 5 1510)
Obese patients (n 5 1496)
Morbidly obese patients (n 5 941)
p value
202 (5) 41 (1) 224 (6)
50 (3.3) 9 (0.6) 119 (7.9)
69 (4.6) 9 (0.6) 64 (4.3)
83 (8.8) 23 (2.4) 41 (4.4)
,.001b ,.001b ,.001b
1 (0.1) 26 (1.7) 5 (0.3) 10 (0.7)
0 11 (0.7) 4 (0.3) 13 (0.9)
1 (0.1) 19 (2.0) 4 (0.4) 10 (1.1)
.49 .01b .80 .56
2 (0.1) 56 (1.4) 13 (0.3) 33 (0.8)
Mahdi et al.
Perioperative Outcome After Endometrial Cancer Surgery
101
Table 4 Multivariate regression analysis for any complication and death within 30 days after surgery to treat endometrial cancera Variable
Any complication
p value
Variable
Death
BMI 30–39 BMI R40 Laparoscopy Surgical complexity .3 Non-white race/ethnicity
0.8 (0.6–10) 1.0 (0.8–1.3) 0.2 (0.2–0.3) 4.5 (2.0–10.8) 1.1 (0.9–1.4)
.09 .96 ,.001 ,.001 .46
BMI 30–39 BMI R40 Laparoscopy Surgical complexity .3 Age, yr 60–69 70–79 R80 Diabetes Respiratory comorbidities Cardiac comorbidities Hypertension Ascites Neurologic comorbidities Weight loss, 10% within 6 months before surgery Preoperative blood transfusion Dependent functional status ASA class 3 ASA class 4–5 Operative time, hr 1–2 2–3 .3 Serum creatinine R2 mg/dL Serum albumin .3 g/dL Anemia (hematocrit R35%) WBC count .11 000/mL Platelet count R350/mL
1.0 (0.4–2.4) 0.4 (0.1–1.6) 0.4 (0.1–1.2) 1.6 (0.1–10.4)
.99 .23 .12 .65
0.8 (0.2–2.8) 1.6 (0.5–5.4) 3.0 (0.8–10.9) NA 6.5 (1.9–19.7) 3.4 (0.8–11.9) NA 6.2 (1.3–25.6) NA 1.3 (0.2–6.0)
.77 .43 .09 ,.001 .07
2.3 (0.2–14.6) 3.7 (0.9–12.3) 0.9 (0.3–2.5) 1.3 (0.3–5.4)
.43 .05 .79 .77
NA NA NA 9.5 (2.7–29.3) 0.2 (0.1–0.6) 1.3 (0.5–4.0) 1.6 (0.5–5.0) 1.1 (0.3–3.3)
,.001 ,.001 .65 .42 .85
Diabetes Respiratory comorbidities Cardiac comorbidities Hypertension Ascites Neurologic comorbidities Weight loss, 10% within 6 months before surgery Preoperative blood transfusion Dependent functional status ASA class 3 ASA class 4–5 Operative time, hr 1–2 2–3 .3 Serum creatinine R2 mg/dL Serum albumin .3 g/dL Anemia (hematocrit R35%) WBC count .11 000/mL Platelet count R350/mL
1.0 (0.8–1.3) 1.5 (0.9–2.5) 1.3 (0.8–2.0) 1.1 (0.9–1.4) 3.5 (1.7–7.1) 1.9 (1.2–3.1) 2.3 (1.2–4.4)
.78 .10 .36 .31 ,.001 .01 .01
3.8 (1.3–11.9) 1.7 (0.9–2.9) 1.3 (1.0–1.6) 2.4 (1.4–3.8)
.02 .07 .03 ,.001
1.6 (0.8–3.3) 1.9 (1.0–4.0) 3.3 (1.7–7.1) 1.1 (0.6–1.9) 0.8 (0.5–1.2) 0.4 (0.3–0.5) 1.7 (1.2–2.4) 1.1 (0.9–1.5)
.23 .08 ,.001 .78 .94 ,.001 .01 .34
p value
.02 .79
ASA 5 American Society of Anesthesiologists; BMI 5 body mass index; NA 5 not applicable; WBC 5 white blood cell count. a Values are given as median (range).
conversion rates with robotic surgery compared with laparoscopy in morbidly obese women [19,20]; however, other institutions report contrary results [21,22]. A variety of factors, such as surgeon training and experience, dictate the optimal minimally invasive approach. The propensity for obese patients to develop wound and infectious complications has been observed in other surgical specialties [6–10] and has been repeatedly reported in series of patients with endometrial cancer [13,18,23]. BakkumGamez et al [23] recently reported on risk factors for SSI. Among other risk factors, BMI R40 was associated with a 4.7-fold risk of SSI, and laparotomy was associated with a 15-fold increase in risk of SSI when compared with patients who underwent a minimally invasive procedure. Each SSI resulted in a $5447 increase in median 30-day cost, not taking into account patient expenditures or wages lost [23]. At multivariate analysis, neither obesity nor morbid obesity was associated with 30-day morbidity or mortality. This can be explained, at least in part, by the difference in
preoperative comorbidities and surgical approach. Obese and morbidly obese patients were more likely to have a higher ASA disease class, diabetes, hypertension, acute renal failure, and hypoalbuminemia. Therefore, attention should be paid to identification and optimization of these medical comorbidities in the perioperative period. The NSQIP database consists of national data that reflect general practice patterns and outcomes in a variety of clinical settings. Therefore, unlike small single-institution studies or large prospective trials, these findings are reflective of real-world practices and their associated real-world outcomes. Still, our conclusions are limited by the nature of the study design. In this retrospective cohort study using data from a large database, generalizable associations are revealed by the large sample sizes, although cause-and-effect relationships can only be hypothesized. Much of the data used require reliance on CPT and ICD-9 codes, which raises concerns about accuracy and validity of the data. Other investigators have closely scrutinized the NSQIP database and found it to be highly accurate [24]. We also lacked
102
information on patient socioeconomic status, insurance status, and surgeon or hospital volume. Our results were not adjusted for stage at cancer diagnosis, which influences treatment plans. However, we adjusted for extent of surgical treatment, which is a proxy for burden of disease. Furthermore, we lacked information on procedures that were converted from laparoscopy to laparotomy. Ultimately, morbid obesity and obesity are not independently associated with a greater risk of postoperative complications or death after endometrial cancer surgery. The propensity for obese and morbidly obese patients to experience complications, in particular infectious and wound complications, is observed only at univariate analysis and likely reflects their preoperative comorbidities and the likelihood that they will undergo an open staging procedure. In the era of modern healthcare reform, the onus is on us as gynecologic cancer treatment providers to detail complications that can be expected, to appropriately counsel women, help develop reimbursement models, and guide further research with the objective of continuing to improve health outcomes for cancer survivors. Morbidly obese women need to be counseled that they are at increased risk of complications due to their medical comorbidities, and informed that this risk may be attenuated with use of a minimally invasive approach and the optimization of their comorbidities. Obese women should be referred to providers who have more experience with minimally invasive surgery in this population, and a concerted effort should be made to identify and optimize medical comorbidities in the perioperative period. References 1. Hurt RT, Frazier TH, McClave SA, Kaplan LM. Obesity epidemic: overview, pathophysiology, and the intensive care unit conundrum. JPEN J Parenter Enteral Nutr. 2011;35(5 Suppl):4S–13S. 2. Arterburn DE, Maciejewski ML, Tsevat J. Impact of morbid obesity on medical expenditures in adults. Int J Obese (Lond). 2005;29:334–339. 3. Finkelstein EA, Khavjou OA, Thompson H, et al. Obesity and severe obesity forecasts through 2030. Am J Prev Med. 2012;42:563–570. 4. Bittoni MA, Fisher JL, Fowler JM, Maxwell GL, Paskett ED. Assessment of the effects of severe obesity and lifestyle risk factors on stage of endometrial cancer. Cancer Epidemiol Biomarkers Prev. 2013;22: 76–81. 5. von Gruenigen VE, Gil KM, Frasure HE, Jenison EL, Hopkins MP. The impact of obesity and age on quality of life in gynecologic surgery. Am J Obstet Gynecol. 2005;193:1369–1375. 6. Fischer JP, Nelson JA, Kovach SJ, Serletti JM, Wu LC, Kanchwala S. Impact of obesity on outcomes in breast reconstruction: analysis of 15,937 patients from the ACS-NSQIP datasets. J Am Coll Surg. 2013; 217:656–664. 7. Pugely AJ, Callaghan JJ, Martin CT, Cram P, Gao Y. Incidence of and risk factors for 30-day readmission following elective primary total joint arthroplasty: analysis from the ACS-NSQIP. J Arthroplasty. 2013;28:1499–1504.
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015 8. Glarner CE, Greenblatt DY, Rettammel RJ, Neuman HB, Weber SM. Wound complications after inguinal lymph node dissection for melanoma: is ACS NSQIP adequate? Ann Surg Oncol. 2013;20:2049–2055. 9. Mustain WC, Davenport DL, Hourigan JS, Vargas HD. Obesity and laparoscopic colectomy: outcomes from the ACS-NSQIP database. Dis Colon Rectum. 2012;55:429–435. 10. Kazaure HS, Roman SA, Sosa JA. Obesity is a predictor of morbidity in 1,629 patients who underwent adrenalectomy. World J Surg. 2011;35: 1287–1295. 11. Centers for Medicare and Medicaid Services. Hospital-acquired conditions (present on admission indicator). September 20, 2012. Available at: http://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/ HospitalAcqCond/index.html?redirect5/HospitalAcqCond. Accessed November 6, 2013. 12. Litta P, Fabris AM, Breda E, et al. Laparoscopic surgical staging of endometrial cancer: does obesity influence feasibility and perioperative outcome? Eur J Gynaecol Oncol. 2013;34:231–233. 13. Giugale LE, Di Santo N, Smolkin ME, Havrilesky LJ, Modesitt SC. Beyond mere obesity: effect of increasing obesity classifications on hysterectomy outcomes for uterine cancer/hyperplasia. Gynecol Oncol. 2012;127:326–331. 14. Bernardini MQ, Gien LT, Tipping H, Murphy J, Rosen BP. Surgical outcome of robotic surgery in morbidly obese patient with endometrial cancer compared to laparotomy. Int J Gynecol Cancer. 2012; 22:76–81. 15. Helm CW, Arumugam C, Gordinier ME, Metzinger DS, Pan J, Rai SN. Laparoscopic surgery for endometrial cancer: increasing body mass index does not impact postoperative complications. J Gynecol Oncol. 2011;22:168–176. 16. Bijen CB, de Bock GH, Vermeulen KM, et al. Laparoscopic hysterectomy is preferred over laparotomy in early endometrial cancer patients, however not cost effective in the very obese. Eur J Cancer. 2011;47: 2158–2165. 17. Walker JL, Piedmonte MR, Spirtos NM, et al. Laparoscopy compared with laparotomy for comprehensive surgical staging of uterine cancer: Gynecologic Oncology Group Study LAP2. J Clin Oncol. 2009;27: 5331–5336. 18. Tang KY, Gardiner SK, Gould C, Osmundsen B, Collins M, Winter WE 3rd. Robotic surgical staging for obese patients with endometrial cancer. Am J Obstet Gynecol. 2012;206:513.e1–513.e6. 19. Gehrig PA, Cantrell LA, Shafer A, Abaid LN, Mendivil A, Boggess JF. What is the optimal minimally invasive surgical procedure for endometrial cancer staging in the obese and morbidly obese woman? Gynecol Oncol. 2008;111:41–45. 20. Seamon LG, Cohn DE, Henretta MS, et al. Minimally invasive comprehensive surgical staging for endometrial cancer: robotics or laparoscopy? Gynecol Oncol. 2009;113:36–41. 21. Bell MC, Torgerson J, Seshadri-Kreaden U, Suttle AW, Hunt S. Comparison of outcomes and cost for endometrial cancer staging via traditional laparotomy, standard laparoscopy and robotic techniques. Gynecol Oncol. 2008;111:407–411. 22. Subramaniam A, Kim KH, Bryant SA, et al. A cohort study evaluating robotic versus laparotomy surgical outcomes of obese women with endometrial carcinoma. Gynecol Oncol. 2011;122:604–607. 23. Bakkum-Gamez JN, Dowdy SC, Borah BJ, et al. Predictors and costs of surgical site infections in patients with endometrial cancer. Gynecol Oncol. 2013;130:100–106. 24. Sellers MM, Merkow RP, Halverson A, et al. Validation of new readmission data in the American College of Surgeons National Surgical Quality Improvement Program. J Am Coll Surg. 2013;216: 420–427.
