- Email: [email protected]
hyperplasia
Gynecologic Oncology 127 (2012) 326–331
Contents lists available at SciVerse ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Beyond mere obesity: Effect of increasing obesity classifications on hysterectomy outcomes for uterine cancer/hyperplasia☆ Lauren E. Giugale a, Nicola Di Santo b, Mark E. Smolkin c, Laura J. Havrilesky b, Susan C. Modesitt d,⁎ a
University of Virginia, School of Medicine, Charlottesville, VA, USA Division of Gynecologic Oncology, OB/GYN Department, Duke University Medical Center, Durham, NC, USA Public Health Sciences Department, University of Virginia, School of Medicine, Charlottesville, USA d Division of Gynecologic Oncology, OB/GYN Department, University of Virginia Health System, Charlottesville, VA, USA b c
H I G H L I G H T S ► Surgical complications significantly increased as obesity classification increased (BMI ≥ 30 vs. ≥ 40 vs. ≥ 50 kg/m2). ► Performance of lymph node dissection and number of nodes decreased with increasing obesity classification. ► Performance of minimally invasive surgery was associated with decreased complications regardless of obesity classification.
a r t i c l e
i n f o
Article history: Received 4 July 2012 Accepted 9 August 2012 Available online 19 August 2012 Keywords: Obesity Endometrial cancer Super obesity Surgical outcomes
a b s t r a c t Objective. To assess the impact of obesity severity on hysterectomy outcomes for uterine hyperplasia/ cancer. Methods. The data from women undergoing hysterectomies for endometrial hyperplasia/uterine cancer with a BMI ≥ 30 kg/m2 were abstracted from records at the University of Virginia and Duke University following IRB approval. Univariate and multivariate statistical analyses were performed. Results. Mean age of the 659 patients was 58.1 yrs; mean body mass index (BMI) was 43 kg/m2. Women were grouped based on BMI: 39.6% (261) were obese (30–39 kg/m2), 41.7% (275) were morbidly obese (40–49 kg/m2) and 18.7% (123) were super obese (≥50 kg/m2). Minimally invasive surgical procedures (MIS) were attempted in 280 patients with a conversion rate of 16.1%; BMI was higher in the converted group (47.3 vs. 40.6 kg/m2; pb 0.001). As obesity group increased, there was a decreased frequency of lymphadenectomy (63.8% vs. 37.1% vs. 20.3%; pb 0.001), increased blood loss (242 vs. 281 vs. 378 mL; pb 0.001) and fewer nodes removed (pb 0.001). On multivariate analysis, type of surgery (open vs. MIS) and obesity classification were independently and significantly associated with wound complications (p b 0.001) and the presence of postoperative complications (pb 0.001, p=0.003). Surgical staging with lymphadenectomy was significantly associated with obesity (pb 0.001) but not procedure type (p=0.11). Blood transfusion (pb 0.001), hospital readmission (p=0.025), and ileus (pb 0.001) were significantly associated with open procedures but not obesity. There were no significant differences in progression-free or disease-specific survival based on obesity group. Conclusion. Women with BMI's exceeding 40 kg/m2 have worse surgical outcomes than their less obese counterparts. © 2012 Elsevier Inc. All rights reserved.
Introduction Obesity rates and associated health problems continue to skyrocket in the United States. The data from 2001 to 2005 showed that obesity was increasing and that the subgroups of the heaviest women were most
☆ This research was presented during an oral plenary session at the Annual Meeting of Society of Gynecologic Oncology in Austin, Texas on March 25, 2012. ⁎ Corresponding author at: Gynecologic Oncology Division, Department of Obstetrics and Gynecology, Box 800712, University of Virginia Health System, Charlottesville, VA 22908, USA. E-mail address: [email protected] (S.C. Modesitt). 0090-8258/$ – see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ygyno.2012.08.014
impacted. There was a 75% increase in women with a body mass index (BMI)≥50 kg/m2 and a 52% increase in women with a BMI ≥40 kg/m2 compared with a 24% increase in women with a BMI≥30 kg/m2 [1]. One of the obesity linked health problems for women is the vastly increased risk of developing endometrial cancer as up to 40% of all endometrial cancer is due to obesity [2]; many of these women have BMIs that far exceed the obesity threshold value of 30 kg/m2. Surgery in obese women is technically difficult and gynecologic oncologists frequently operate on women within the morbidly obese (BMI ≥ 40 kg/m 2) and super obese classifications (BMI ≥ 50 kg/m 2). Studies evaluating the effect of BMI on hysterectomy outcomes have shown higher perioperative blood loss [3–7], increased infection [8],
L.E. Giugale et al. / Gynecologic Oncology 127 (2012) 326–331
increased length of hospital stay [5], increased rates of conversion to open hysterectomy [9] and increased major complications rates associated with increasing BMI [10]. Conversely, others studies in total laparoscopic hysterectomies for benign indications have not found significant associations between obesity and operating time, hemoglobin change, complication rates, or length of stay [4,11,12]. The majority of studies that have evaluated surgical outcomes based upon BMI do not distinguish among values greater than 40 kg/m 2; thus specific outcome data for the super obese population are limited. The objectives of this study were to compare surgical complications and outcomes for obese women undergoing hysterectomy specifically for uterine cancer/hyperplasia based on BMI and to compare route of hysterectomy and the impact of BMI on completion of surgical staging. Methods Institutional review board (IRB) approval was obtained from the University of Virginia Medical Center and Duke University Medical Center. The University of Virginia's clinical data repositories were searched to identify subjects meeting inclusion criteria from June 1, 2001 through June 1, 2011 and the Duke database searched from January 2007 through December 2010. Inclusion criteria were age >18 yrs and a hysterectomy performed for either uterine cancer or complex atypical endometrial hyperplasia. Women also had to be obese (BMI≥30 kg/m2) at the time of surgery as defined by a documented BMI value or a height and weight. Patients were excluded if there were incomplete surgical records or BMI could not be ascertained. Demographic, clinical, and surgical data were abstracted retrospectively. Any complication documented in the record was recorded and when possible specific wound complications were further classified (infection, superficial separation, fascial dehiscence, any wound complication requiring specific wound care). Urinary tract infections were not consistently able to be evaluated and were not included. Women were grouped according to the following obesity severity classifications: obese (BMI 30–39 kg/m 2), morbidly obese (BMI 40–49 kg/m 2) and super obese (BMI ≥ 50 kg/m 2). Surgeries were classified by original surgical intent (open vs. minimally invasive) and conversion rates calculated. Univariate statistical analyses were performed using independent t-test and ANOVA for comparison of means, chi-square for categorical variables. Survival was analyzed with Kaplan Meier and log rank statistic (SPSS Version 19.0, Chicago, IL). Multivariate analysis was performed using logistic regression analysis in which complications/outcomes were modeled against BMI group (3 levels) and procedure to obtain estimated odds ratios and 95% confidence intervals. For all tests, a p value of b0.05 was deemed significant. Results Six hundred and fifty nine women underwent hysterectomies for uterine cancer or hyperplasia at both institutions and also met inclusion criteria during the specified time periods. Mean age was 58.1 yrs (27–87 yrs), mean weight was 113.8 kg (67–215 kg), mean BMI was 43.0 kg/m 2 (30–136 kg/m2), and 81.1% were White (Table 1). At final pathology, 89.5% (590) had endometrial cancer and 10.5% (69) had hyperplasia; 80.5% had Stage I disease and median follow up time was 24 months. Of note, 39.6% (261) were obese, 41.7% (275) were morbidly obese and 18.7% (123) were super obese. Procedures performed were total abdominal hysterectomy/bilateral salpingooophorectomy (TAH/BSO) (32.3%, n = 213) and TAH/BSO with lymph node dissection (22.2%, n = 146) (Table 1); 280 minimally invasive procedures (MIS) were attempted and 235 completed (conversion rate — 16.1%). The BMI was significantly higher in the converted group (47.3 kg/m 2 vs. 40.6 kg/m 2; p b 0.001). Among 37 women with a BMI ≥ 60 kg/m 2 there was a 38.5% conversion rate
327
Table 1 Patient demographics (N = 659). Variables
Total (%)
Age at surgery in years (mean) Body mass index Obese (30–39 kg/m2) Morbidly obese (40–49 kg/m2) Super obese (≥ 50 kg/m2) Race White Black Asian/other Concomitant medical diagnoses Hypertension Diabetes Heart disease Hyperlipidemia Surgery performeda Open surgery Supracervical Hysterectomy TAH/BSO TAH/BSO/lymphadenectomy Radical hysterectomy Minimally invasive TLH/BSO TLH/BSO/lymphadenectomy Robotic hysterectomy/BSO Robotic hysterectomy/BSO/lymphadenectomy TVH Final pathologic diagnosis Endometrioid endometrial cancer Endometrial hyperplasia Carcinosarcoma Uterine papillary serous carcinoma Clear cell Uterine sarcoma Other cancer of the uterus Staging Stage I IA IB Stage II Stage III Stage IV Recurrence
58.1 yrs 261 (39.6) 275 (41.7) 123 (18.7) 534 (81.1) 118 (17.9) 7 (1.0) 446 (67.7) 242 (36.7) 95 (14.4) 215 (32.6) 379 (57.5) 16 (2.5) 213 (32.3) 146 (22.2) 4 (0.7) 280 (42.5) 88 (13.4) 75 (11.4) 46 (7.0) 51 (7.7) 20 (3.0)
515 (78.1) 69 (10.5) 26 (3.9) 6 (0.9) 3 (0.5) 14 (2.1) 471 (80.5) 405 (68.6) 66 (11.2) 31 (5.3) 68 (11.6) 15 (2.6) 60 (11.7)
a TAH = total abdominal hysterectomy; BSO = bilateral salpingo-oophorectomy; TLH = total laparoscopic hysterectomy; TVH = total vaginal hysterectomy.
for attempted minimally invasive surgeries, a 13.5% supracervical hysterectomy rate, and a 13.5% rate of nodal dissection. Women were stratified by obesity category and compared. The mean age of women significantly decreased as obesity category increased (60.0 vs. 57.3 vs. 54.0; pb 0.001) (Table 2). The percentage of MIS performed decreased with increasing obesity category (52.5% vs. 37.1% vs. 33.3%; pb 0.001). Almost 90% of the super obese had Stage I disease compared with 79.9% and 77.3% (184) of the obese and morbidly obese, respectively (p=0.027). With regard to surgical outcomes, across increasing obesity groups, there was a significant increase in estimated blood loss (EBL, 243 vs. 281 vs. 378 mL; pb 0.001), and the super obese group had significantly increased blood loss compared to both other groups (p b 0.001 and p=0.011). Decreased frequency of node dissection (63.8% vs. 37.1% vs. 20.3%; pb 0.001 — Fig. 1) was also observed with increasing obesity and there was a significant decrease in the mean number of lymph nodes removed (20.4 vs. 15.1 vs. 10.0; pb 0.001). A significant increase in length of stay was found (2.9 days vs. 3.1 days vs. 3.7 days, p=0.021) between the obese and super obese (p=0.015) as well as a significant increase in the total number of post-operative complications (0.6 vs. 0.9 vs. 1.2 complications, pb 0.001). The mean BMI was significantly higher in women who experienced the following complications: ICU admission (52.8 vs. 42.7 kg/m2, pb 0.001), any wound care requirement (47.1 vs. 42.3 kg/m2, p b 0.001), EBL ≥500 cm3 (46.1 vs. 39.3 kg/m2, pb 0.001),
328
L.E. Giugale et al. / Gynecologic Oncology 127 (2012) 326–331
Table 2 Comparison of all women based on obesity classification. Variable
Obese (n = 261) n (%)
Morbidly obese Super obese p Value (n = 123) (n = 275) n (%) n (%)
Mean age (years) Mean height (cm) Mean weight (kg) Mean BMI (kg/m2) Comorbidities Hypertension Diabetes Hyperlipidemia Heart disease Procedurea Open Open with lymphadenectomy Minimally invasive Minimally invasive with lymphadenectomy Lymph nodes Dissection performed Mean pelvic nodes Mean para-aortic nodes Total mean nodes (number) Wound Infection Superficial separation Fascial dehiscence Any wound care Wound reoperation Intraoperative factors EBL ≥ 500 EBL (cubic centimeters) Blood transfusion Procedure time (hrs:mins) Conversion to open Prolonged mechanical ventilation Bowel injury Postoperative factors ICU admission Length of stay (days) Pneumonia Ileus Hematoma Readmission to the hospital Discharged to SNF or NH Total number of complications
60.0 163.3 94.2 34.9
57.3 162.0 117.5 44.2
53.7 160.5 146.9 57.8
175 (67.0 86 (33.0) 97 (37.2) 43 (16.5)
179 (65.1) 99 (36.0) 80 (29.1) 38 (13.8)
92 57 38 14
53 (20.3) 71 (27.2)
109 (39.6) 64 (23.3)
67 (54.5) 15 (12.2)
54 (20.7) 83 (31.8)
69 (25.1) 33 (12.0)
31 (25.2) 10 (8.1)
166 (63.8) 16.5 4.6 20.4
102 (37.1) 13.1 2.2 15.5
25 (20.3) 10.0 0.15 10.0
p b 0.001 p = 0.001 p b 0.001 p = 0.001
23 (8.8) 18 (6.9) 5 (1.9) 21 (8.1) 6 (2.3)
47 (17.5) 41 (15.3) 2 (0.7) 36 (13.4) 6 (2.2)
19 (15.8) 35 (29.2) 1 (0.8) 30 (25.0) 3 (2.5)
p = 0.011 p b 0.001 p = 0.429 P b 0.001 p = 0.988
21 (8.1) 242.8 17 (6.5) 3:29
37 (13.5) 280.8 20 (7.5) 3:13
27 (22.3) 377.7 17 (14.2) 3:08
p = 0.001 p b 0.001 p = 0.035 p = 0.003
16 (11.7) 1 (0.4)
19 (18.6) 6 (2.2)
10 (24.4) 3 (2.5)
p = 0.102 p = 0.144
3 (1.1)
4 (1.5)
2 (1.6)
p = 0.919
1 (0.4) 2.87 1 (0.4) 21 (8.1) 3 (1.2) 15 (5.8)
10 (3.7) 3.13 2 (0.7) 27 (10.1) 8 (3.0) 22 (8.2)
9 (7.4) 3.66 2 (1.7) 6 (5.0) 1 (0.8) 13 (10.9)
p = 0.001 p = 0.021 p = 0.413 p = 0.243 p = 0.192 p = 0.203
3 (1.3) 0.61
5 (2.2) 0.90
1 (1.0) 1.21
p = 0.632 p b 0.001
(74.8) (46.3) (30.9) (11.4)
p b 0.001
p = 0.154 p = 0.038 p = 0.124 p = 0.388 p b 0.001
Women were also compared based on the route of the hysterectomy performed: open abdominal hysterectomy or MIS (including laparoscopic, robotic, and total vaginal hysterectomies; Table 3). The mean BMI of patients undergoing minimally invasive procedures was significantly lower than those of women undergoing open abdominal hysterectomies (41.7 kg/m 2 vs. 44.0 kg/m 2; p = 0.003). Minimally invasive procedures were more frequently performed in patients with Stage I disease (p = 0.008), and with FIGO grade 1/2 disease (p = 0.048). Compared to open surgeries, MIS procedures were associated with significantly less blood loss (173.9 vs. 365.7 mL, p b 0.001) and fewer blood transfusions (2.9% vs. 12.4%, p b 0.001). There were also fewer superficial wound separations (7.9% vs. 19.4%, p b 0.001), fascial wound separations (0.0% vs. 2.2%, p = 0.014), wound infections (4.7% vs. 20.5%, p b 0.001), requirement for any wound care (6.5% vs. 18.6%, p b 0.001) and decreased ileus percentages (3.6% vs. 11.9%, p b 0.001). Lymphadenectomy was performed more frequently during MIS procedures (55.8% vs. 44.1%, p = 0.006). Lastly, patients undergoing MIS procedures were less likely to be discharged to a skilled nursing facility (0.4% vs. 2.8%, p = 0.025) or readmitted (4.7% vs. 10.0%, p = 0.012). There was no difference in rates of bowel/bladder injury, post-operative admission to the intensive care unit, post-operative mechanical ventilation, myocardial infarction, pulmonary embolus, deep vein thrombosis, pneumonia, peri-operative death, or reoperation. Analysis was also performed after the minimally invasive category was broken down into laparoscopic and robotic hysterectomies. The only significant differences between robotic and laparoscopic procedures were an increased rate of post-operative mechanical ventilation (6.2%) with robotic procedures compared with both open (0.8%) and laparoscopic hysterectomies (0.6%) (p = 0.001). Given the significant correlation for surgical complications/outcomes with both obesity classification and route of surgery, a multivariate analysis was performed. Both an open type of surgery and increasing obesity classification were independently and significantly associated with the following: superficial wound complication (p b 0.001 for both), postoperative wound care (p b 0.001 for both), and the presence of at least one postoperative complication (p b 0.001 for procedure type; p= 0.003 for obesity classification). The following complications were significantly associated with an open procedure but not with obesity classification: post-operative blood transfusion (p b 0.001), readmission to the hospital (p =0.025), and postoperative ileus (p b 0.001). In contrast, lymphadenectomy was significantly associated with decreasing obesity classification (pb 0.001) but not with type of procedure (p= 0.11). Odds ratios were calculated to compare relative impact of both type of surgery and obesity classification on outcomes of interest (Table 4). For the performance of a node dissection, the following
a Open includes TAH, supracervical hysterectomies, and radical hysterectomies. Minimally Invasive includes TLH, robotic Hysterectomies, and TVH.
blood transfusion (46.1 vs. 42.7 kg/m2, p =0.014), bowel injury (53.4 vs. 42.9 kg/m2, p=0.001) and pneumonia (58.2 vs. 42.9 kg/m2, p= 0.001) (Supplemental Table).
Fig. 1. Lymphadenectomy by obesity group: the total number of women undergoing lymphadenectomy is listed along with percentage within each obesity group.
Table 3 Comparison of all women based on procedure performed.
Age (years) Weight (kg) BMI (kg/m2) Comorbidities Hyperlipidemia Stage I Blood EBL (mL) Transfusion Wound Infection Superficial separation Fascial separation Any wound care Ileus Nursing facility Readmission
Abdominal n (%)
Minimally invasive n (%)
p Value
57.8 116.1 44.0
58.6 110.7 41.7
p = 0.389 p = 0.004 p = 0.003
111 (29.3%) 259 (76.9%)
104 (37.1%) 215 (85.7%)
p = 0.036 p = 0.008
365.7 46 (12.4)
173.9 8 (2.9)
76 (20.5) 72 (19.4) 8 (2.2) 18 (6.5) 44 (11.9) 8 (2.8) 37 (10.0)
13 (4.7) 22 (7.9) 0 (0.0) 69 (18.6) 10 (3.6) 1 (0.4) 13 (4.7)
p b 0.001 p b 0.001 p b 0.001 P b 0.001 p = 0.014 p b 0.001 p b 0.001 p = 0.025 p = 0.012
L.E. Giugale et al. / Gynecologic Oncology 127 (2012) 326–331 Table 4 Odds ratios. Variable
Effect
Odds ratio (95% confidence interval)
Lymphadenectomy
Open surgery Minimally invasive surgery Obese Morbidly obese Super obese Open surgery Minimally invasive surgery Obese Morbidly obese Super obese Open surgery
1 (reference) 1.31 (0.94–1.83) 1 (reference) 0.35 (0.24–0.49) 0.15 (0.09–0.25) 1 (reference) 0.27 (0.13–0.55) 1 (reference) 1.07 (0.58–1.97) 0.48 (0.19–1.24) 1 (reference)
Minimally invasive surgery Obese Morbidly obese Super obese Open surgery
0.41 (0.24–0.69) 1 (reference) 2.16 (1.20–3.89) 4.91 (2.62–9.20) 1 (reference)
Minimally invasive surgery Obese Morbidly obese Super obese Open surgery Minimally invasive surgery Obese Morbidly obese Super obese Open surgery Minimally invasive surgery Obese Morbidly obese Super obese Open surgery Minimally invasive surgery Obese Morbidly obese Super obese Open surgery
0.22 (0.10–0.47) 1 (reference) 0.95 (0.48–1.87) 1.91 (0.93–3.95) 1 (reference) 0.34 (0.19–0.58) 1 (reference) 1.53 (0.86–2.72) 3.27 (1.76–6.08) 1 (reference) 0.21 (0.11–0.38) 1 (reference) 1.83 (1.06–3.16) 1.55 (0.80–3.03) 1 (reference) 0.47 (0.24–0.91) 1 (reference) 1.31 (0.66–2.61) 1.77 (0.81–3.89) 1 (reference)
Minimally invasive surgery Obese Morbidly obese Super obese
0.20 (0.14–0.28) 1 (reference) 1.32 (0.90–1.91) 2.28 (1.41–3.69)
Postoperative ileus
Superficial wound separation
Postoperative blood transfusion
Postoperative wound care
Wound infection
Readmission
Total number of complications
estimates were obtained: minimally invasive surgery (OR 1.31 95% CI 0.94–1.83) versus open procedure, and morbidly obese (OR 0.35 95% CI 0.24–0.49) and super obese (OR 0.15 95% CI 0.09–0.25) versus obese. At final follow-up, 500 patients (75.9%) were alive with no evidence of disease, 18 (2.7%) were deceased either due to surgery (1 patient due to pneumonia) or to cancer (17 patients), 41 (6.2%) were alive with cancer and 9 patients (1.4%) were dead of other causes. Ninety-one patients (13.8%) were lost to follow up. There was a recurrence rate of 11.7% (60 patients) with 13 vaginal recurrences, 14 pelvic recurrences, and 31 distant recurrences. Progression free survival (PFS) was analyzed using Kaplan Meier method and log rank statistic and found to be significantly associated with both grade and stage (Pb 0.001) but not with obesity category (p= 0.066; Fig. 2) or procedure type (p= 0.392). Both disease specific survival data and overall survival data are limited in that there were only 17 deaths due to endometrial cancer and 27 deaths from any cause in the entire cohort. Decreased disease specific survival was significantly associated with both increasing grade and stage (p b 0.001) but not with obesity classification (p= 0.993) or type of surgery performed (p= 0.21).
Discussion Although obesity is defined as a BMI exceeding 30 kg/m 2, this study highlights the fact that all obese women with endometrial
329
cancer do not have equivalent surgical risks. Most prior studies on the impact of obesity on surgical outcomes were not limited to women with an endometrial malignancy or did not stratify obesity beyond the morbidly obese or simply compared obese to non obese patients and thus have not analyzed differential outcomes for women with more extreme obesity. Our study confirmed that increasing BMI beyond mere obesity (BMI ≥ 30 kg/m 2) appears to be significantly associated with certain intraoperative and postoperative outcomes, most notably: conversion to an open procedure, ICU admission, superficial wound separation and wound care requirement. While blood transfusion, EBL greater than 500 mL, bowel injury, and pneumonia were significantly increased by obesity on univariate analyses; these effects appeared to be mitigated if minimally invasive procedures were performed. With regard to EBL during open cases, our study supports the contention that the significant increased risk occurs in the super obese category. These data corroborate Osler and colleagues' findings of increased risk of bleeding associated with obesity [7] but are in contrast to studies which have not found a relation between obesity and blood loss and which have found an association with longer operative times and infections; these studies may have lacked a sufficient quantity of super obese patients to detect any significant differences [5,7,8]. The total number of complications increased among increasing obesity groups; while it is difficult to discern a BMI breakpoint for this, our study suggests that more complications may not be seen until one reaches a BMI greater than 50 kg/m 2. Women in our study who were admitted to an intensive care unit or required postoperative wound care also had higher BMIs and as BMI surpassed 50 kg/m 2, the risk of longer hospital stays increased. One other study analyzed women with BMIs > 50 kg/m 2 and also found that these women required ventilation with higher airway pressures [13]. For very medically infirm patients, whether due to obesity or other co-morbid conditions, radiation therapy may also be a nonsurgical alternative. In addition to BMI, procedure type also influenced outcomes in our study. Minimally invasive procedures were associated with less blood loss, fewer blood transfusions, decreased wound complications, ileus, or readmission. For laparoscopic cases, previous studies have not found associations between obesity and intraoperative blood loss, postoperative infections, or length of stay and there have been varying conclusions regarding operative times [3,6,8–12,14]. Heinberg et al. did find significant associations between obesity and blood loss greater than 500 mL [6]. Regarding robotic assisted total laparoscopic hysterectomies, other studies have not reported a correlation between BMI and blood loss, postoperative complications, operative room times, or length of stay [15,16]. Our study found that robotic hysterectomy demonstrated an increased frequency of post-operative mechanical ventilation (6.5%) compared to both open and laparoscopic surgeries; this was a novel finding. While the exact mechanism is unclear, it may be the increased intra-abdominal pressure needed for insufflation and subsequent hypercarbia along with the steep Trendelenburg position are potential causative factors but further exploration is warranted. The multivariate analysis helped clarify the relative impact of obesity level compared to procedure type. Both obesity and procedure type independently affected superficial wound complications, postoperative wound care and the presence of postoperative complications whereas obesity preferentially impacted performance of lymph node dissection and procedure type impacted postoperative blood transfusion, readmission to the hospital and postoperative ileus. This underscores the importance of increasing the percentage of minimally invasive surgeries in endometrial cancer patients in order to reduce surgical complications and the aforementioned risks associated with open hysterectomies. Numerous studies have advocated the use of minimally invasive procedures for endometrial cancer surgery and demonstrated the ability to perform adequate staging in obese and morbidly obese women and these methods are being more universally
330
L.E. Giugale et al. / Gynecologic Oncology 127 (2012) 326–331
Fig. 2. Progression free survival by obesity group.
adopted in gynecologic oncology [16,18,19]. Minimally invasive procedures were more prevalent in the latter part of our study; in the first half of the study (before 2007) 18% of hysterectomies were performed via a minimally invasive approach whereas this steadily increased to over 75% by the last years studied. Another important area of impact will be with regard to the goals of surgery for endometrial cancer and realistic expectations for completion of full surgical staging. Decreased frequency of node dissection with a significant decrease in number of lymph nodes was observed in our study as obesity category increased from 64% in the obese women down to 20% of the super obese patients. However, since the significance was between the obese and the higher groups; this suggests that comprehensive surgical staging decreases around a BMI of 40 kg/m2 but does not decrease significantly with further increases in BMI. Previous authors have demonstrated the safety and feasibility of lymph node dissection in obese women, but the mean BMIs were either not reported or less than our study [8,11,14,17,18]. For example, O'Hanlan assessed the impact of obesity on outcomes of total laparoscopic hysterectomy for uterine pathology; however the average BMI was only 30 kg/m2 [11]. Similarly, Lau et al. analyzed surgical outcomes of normal weight patients compared with obese and morbidly obese women undergoing robotic hysterectomy for endometrial cancer and concluded that obesity did not impact surgical staging in this study of 108 women (23 were morbidly obese). Of note, while the number of pelvic nodes removed was equivalent across groups (range 10.4–10.8), significantly fewer of the morbidly obese women had para-aortic lymphadenectomy (42% normal weight vs. 27% obese vs. 1% morbidly obese; p =0.002) [16]. Lymphadenectomy was performed more frequently during minimally invasive procedures in our study; however, on multivariate analysis, it appears that lymphadenectomy was more feasible due to the decreased obesity in that sub-population. Thus, surgeons are foregoing lymph node dissection more frequently in the morbidly and super obese women, likely for safety or technical concerns. While debate continues on
whether all endometrial cancer patients should undergo complete pelvic and para-aortic lymphadenectomy, our study confirms that the dissection rates vary widely based on the extent of obesity. More super obese patients had apparent Stage I disease and fewer had a lymphadenectomy; however, among only the patients who underwent lymphadenectomy, the differences in the prevalence of Stage I disease were no longer significant. This suggests that while the higher prevalence of Stage I disease in the super obese may be falsely elevated, this did not appear to effect progression free or disease specific survival significantly and the disease process may be more indolent. With regard to recurrence data and survival, there were no significant differences based on obesity classifications although the potential of impact of adjuvant treatment data was not analyzed. Thus, the major impact of obesity that we were able to observe in our study was on short term outcomes as opposed to long term disease outcomes. Obesity has previously been shown to carry a vastly increased risk of endometrial cancer death that is six fold higher for women with a BMI ≥40 kg/m 2 in large population studies [20]. We may have lacked sufficient power and length of follow up to show differences in survival in this population and further long term analysis is needed to determine the relative impact of both cancer and co-morbidities. Our study does have notable limitations inherent to all retrospective studies. Our study is susceptible to selection bias by surgeons on the type of procedures performed. There is a reliance on a chart review and complications are undoubtedly underreported and only factors present in the records could be captured and analyzed. There are also differences in the numbers of open, laparoscopic, and robotic cases and more of the recent cases were performed through minimally invasive approaches. More laparoscopic and robotic case numbers would have improved our ability to detect significant differences. Lastly, BMI was used as the means to classify obesity; however BMI may not always be the best way to determine the extent of obesity. Waist-to-hip ratio might have been
L.E. Giugale et al. / Gynecologic Oncology 127 (2012) 326–331
helpful in determining the extent of abdominal/visceral obesity in particular. In conclusion, all obese women are not equal in terms of their risks for adverse peri-operative outcomes and ability to adequately perform comprehensive surgical staging. As more endometrial cancer patients have a BMI over 50 kg/m 2, both surgeons and patients need realistic estimations of expected surgical outcomes. As part of the preoperative surgical consent process, women with endometrial cancer need to understand the associations between increased BMI and surgical complications, increased conversion rates and decreased lymph node dissection. Although technically more difficult, our data show that, even in this population which is composed of almost 20% super obese women, minimally invasive procedures may offer the advantage of fewer complications than traditional open surgeries. Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.ygyno.2012.08.014. Conflict of interest statement The authors declare that there are no conflicts of interest.
References [1] Sturm R. Increases in morbid obesity in the USA: 2000–2005. Public Health 2007;121(7):492-6. [2] Bergstrom A, Pisani P, Tenet V, Wolk A, Adami HO. Overweight as an avoidable cause of cancer in Europe. Int J Cancer 2001;91:421-30. [3] Chopin N, Malaret JM, Lafay-Pillet MC, Fotso A, Foulot H, Chapron C. Total laparoscopic hysterectomy for benign uterine pathologies: obesity does not increase the risk of complications. Hum Reprod 2009;24(12):3057-62. [4] Sheth SS. Vaginal hysterectomy as a primary route for morbidly obese women. Acta Obstet Gynecol Scand 2010;89(7):971-4. [5] Rasmussen KL, Neumann G, Ljungstrom B, Hansen V, Lauszus FF. The influence of body mass index on the prevalence of complications after vaginal and abdominal hysterectomy. Acta Obstet Gynecol Scand 2004;83(1):85-8. [6] Heinberg EM, Crawford III BL, Weitzen SH, Bonilla DJ. Total laparoscopic hysterectomy in obese versus nonobese patients. Obstet Gynecol 2004;103(4):674-80.
331
[7] Osler M, Daugbjerg S, Frederiksen BL, Ottesen B. Body mass and risk of complications after hysterectomy on benign indications. Hum Reprod 2011;26(6):1512-8. [8] Santi A, Kuhn A, Gyr T, Eberhard M, Johann S, Gunthert AR, et al. Laparoscopy or laparotomy? A comparison of 240 patients with early-stage endometrial cancer. Surg Endosc 2010;24(4):939-43. [9] Ghezzi F, Cromi A, Bergamini V, Uccella S, Beretta P, Franchi M, et al. Laparoscopic management of endometrial cancer in nonobese and obese women: a consecutive series. J Minim Invasive Gynecol 2006;13(4):269-75. [10] Kodama J, Seki N, Ojima Y, Nakamura K, Hongo A, Hiramatsu Y. Risk factors for early and late postoperative complications of patients with endometrial cancer. Eur J Obstet Gynecol Reprod Biol 2006;124(2):222-6. [11] O'Hanlan KA, Dibble SL, Fisher DT. Total laparoscopic hysterectomy for uterine pathology: impact of body mass index on outcomes. Gynecol Oncol 2006;103(3): 938-41. [12] O'Hanlan KA, Lopez L, Dibble SL, Garnier AC, Huang GS, Leuchtenberger M. Total laparoscopic hysterectomy: body mass index and outcomes. Obstet Gynecol 2003;102(6):1384-92. [13] O'Gorman T, MacDonald N, Mould T, Cutner A, Hurley R, Olaitan A. Total laparoscopic hysterectomy in morbidly obese women with endometrial cancer anaesthetic and surgical complications. Eur J Gynaecol Oncol 2009;30(2): 171-3. [14] Holub Z, Bartos P, Jabor A, Eim J, Fischlova D, Kliment L. Laparoscopic surgery in obese women with endometrial cancer. J Am Assoc Gynecol Laparosc 2000;7(1): 83-8. [15] Nawfal AK, Orady M, Eisenstein D, Wegienka G. Effect of body mass index on robotic-assisted total laparoscopic hysterectomy. J Minim Invasive Gynecol 2011;18(3): 328-32. [16] Lau S, Buzaglo K, Vaknin Z, Brin S, Kaufer R, Drummond N, et al. Relationship between body mass index and robotic surgery outcomes of women diagnosed with endometrial cancer. Int J Gynecol Cancer 2011;21(4):722-9. [17] Pavelka JC, Ben-Shachar I, Fowler JM, Ramirez NC, Copeland LJ, Eaton LA, et al. Morbid obesity and endometrial cancer: surgical, clinical, and pathologic outcomes in surgically managed patients. Gynecol Oncol 2004;95(3):588-92. [18] Seamon LG, Bryant SA, Rheaume PS, Kimball KJ, Huh WK, Fowler JM, et al. Comprehensive surgical staging for endometrial cancer in obese patients: comparing robotics and laparotomy. Obstet Gynecol 2009;114(1):16-21. [19] Subramaniam A, Kim KH, Bryant SA, Zhang B, Sikes C, Kimball KJ, et al. A cohort study evaluating robotic versus laparotomy surgical outcomes of obese women with endometrial carcinoma. Gynecol Oncol 2011;122(3):604-7. [20] Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med 2003;348:1625-38.
Contents lists available at SciVerse ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Beyond mere obesity: Effect of increasing obesity classifications on hysterectomy outcomes for uterine cancer/hyperplasia☆ Lauren E. Giugale a, Nicola Di Santo b, Mark E. Smolkin c, Laura J. Havrilesky b, Susan C. Modesitt d,⁎ a
University of Virginia, School of Medicine, Charlottesville, VA, USA Division of Gynecologic Oncology, OB/GYN Department, Duke University Medical Center, Durham, NC, USA Public Health Sciences Department, University of Virginia, School of Medicine, Charlottesville, USA d Division of Gynecologic Oncology, OB/GYN Department, University of Virginia Health System, Charlottesville, VA, USA b c
H I G H L I G H T S ► Surgical complications significantly increased as obesity classification increased (BMI ≥ 30 vs. ≥ 40 vs. ≥ 50 kg/m2). ► Performance of lymph node dissection and number of nodes decreased with increasing obesity classification. ► Performance of minimally invasive surgery was associated with decreased complications regardless of obesity classification.
a r t i c l e
i n f o
Article history: Received 4 July 2012 Accepted 9 August 2012 Available online 19 August 2012 Keywords: Obesity Endometrial cancer Super obesity Surgical outcomes
a b s t r a c t Objective. To assess the impact of obesity severity on hysterectomy outcomes for uterine hyperplasia/ cancer. Methods. The data from women undergoing hysterectomies for endometrial hyperplasia/uterine cancer with a BMI ≥ 30 kg/m2 were abstracted from records at the University of Virginia and Duke University following IRB approval. Univariate and multivariate statistical analyses were performed. Results. Mean age of the 659 patients was 58.1 yrs; mean body mass index (BMI) was 43 kg/m2. Women were grouped based on BMI: 39.6% (261) were obese (30–39 kg/m2), 41.7% (275) were morbidly obese (40–49 kg/m2) and 18.7% (123) were super obese (≥50 kg/m2). Minimally invasive surgical procedures (MIS) were attempted in 280 patients with a conversion rate of 16.1%; BMI was higher in the converted group (47.3 vs. 40.6 kg/m2; pb 0.001). As obesity group increased, there was a decreased frequency of lymphadenectomy (63.8% vs. 37.1% vs. 20.3%; pb 0.001), increased blood loss (242 vs. 281 vs. 378 mL; pb 0.001) and fewer nodes removed (pb 0.001). On multivariate analysis, type of surgery (open vs. MIS) and obesity classification were independently and significantly associated with wound complications (p b 0.001) and the presence of postoperative complications (pb 0.001, p=0.003). Surgical staging with lymphadenectomy was significantly associated with obesity (pb 0.001) but not procedure type (p=0.11). Blood transfusion (pb 0.001), hospital readmission (p=0.025), and ileus (pb 0.001) were significantly associated with open procedures but not obesity. There were no significant differences in progression-free or disease-specific survival based on obesity group. Conclusion. Women with BMI's exceeding 40 kg/m2 have worse surgical outcomes than their less obese counterparts. © 2012 Elsevier Inc. All rights reserved.
Introduction Obesity rates and associated health problems continue to skyrocket in the United States. The data from 2001 to 2005 showed that obesity was increasing and that the subgroups of the heaviest women were most
☆ This research was presented during an oral plenary session at the Annual Meeting of Society of Gynecologic Oncology in Austin, Texas on March 25, 2012. ⁎ Corresponding author at: Gynecologic Oncology Division, Department of Obstetrics and Gynecology, Box 800712, University of Virginia Health System, Charlottesville, VA 22908, USA. E-mail address: [email protected] (S.C. Modesitt). 0090-8258/$ – see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ygyno.2012.08.014
impacted. There was a 75% increase in women with a body mass index (BMI)≥50 kg/m2 and a 52% increase in women with a BMI ≥40 kg/m2 compared with a 24% increase in women with a BMI≥30 kg/m2 [1]. One of the obesity linked health problems for women is the vastly increased risk of developing endometrial cancer as up to 40% of all endometrial cancer is due to obesity [2]; many of these women have BMIs that far exceed the obesity threshold value of 30 kg/m2. Surgery in obese women is technically difficult and gynecologic oncologists frequently operate on women within the morbidly obese (BMI ≥ 40 kg/m 2) and super obese classifications (BMI ≥ 50 kg/m 2). Studies evaluating the effect of BMI on hysterectomy outcomes have shown higher perioperative blood loss [3–7], increased infection [8],
L.E. Giugale et al. / Gynecologic Oncology 127 (2012) 326–331
increased length of hospital stay [5], increased rates of conversion to open hysterectomy [9] and increased major complications rates associated with increasing BMI [10]. Conversely, others studies in total laparoscopic hysterectomies for benign indications have not found significant associations between obesity and operating time, hemoglobin change, complication rates, or length of stay [4,11,12]. The majority of studies that have evaluated surgical outcomes based upon BMI do not distinguish among values greater than 40 kg/m 2; thus specific outcome data for the super obese population are limited. The objectives of this study were to compare surgical complications and outcomes for obese women undergoing hysterectomy specifically for uterine cancer/hyperplasia based on BMI and to compare route of hysterectomy and the impact of BMI on completion of surgical staging. Methods Institutional review board (IRB) approval was obtained from the University of Virginia Medical Center and Duke University Medical Center. The University of Virginia's clinical data repositories were searched to identify subjects meeting inclusion criteria from June 1, 2001 through June 1, 2011 and the Duke database searched from January 2007 through December 2010. Inclusion criteria were age >18 yrs and a hysterectomy performed for either uterine cancer or complex atypical endometrial hyperplasia. Women also had to be obese (BMI≥30 kg/m2) at the time of surgery as defined by a documented BMI value or a height and weight. Patients were excluded if there were incomplete surgical records or BMI could not be ascertained. Demographic, clinical, and surgical data were abstracted retrospectively. Any complication documented in the record was recorded and when possible specific wound complications were further classified (infection, superficial separation, fascial dehiscence, any wound complication requiring specific wound care). Urinary tract infections were not consistently able to be evaluated and were not included. Women were grouped according to the following obesity severity classifications: obese (BMI 30–39 kg/m 2), morbidly obese (BMI 40–49 kg/m 2) and super obese (BMI ≥ 50 kg/m 2). Surgeries were classified by original surgical intent (open vs. minimally invasive) and conversion rates calculated. Univariate statistical analyses were performed using independent t-test and ANOVA for comparison of means, chi-square for categorical variables. Survival was analyzed with Kaplan Meier and log rank statistic (SPSS Version 19.0, Chicago, IL). Multivariate analysis was performed using logistic regression analysis in which complications/outcomes were modeled against BMI group (3 levels) and procedure to obtain estimated odds ratios and 95% confidence intervals. For all tests, a p value of b0.05 was deemed significant. Results Six hundred and fifty nine women underwent hysterectomies for uterine cancer or hyperplasia at both institutions and also met inclusion criteria during the specified time periods. Mean age was 58.1 yrs (27–87 yrs), mean weight was 113.8 kg (67–215 kg), mean BMI was 43.0 kg/m 2 (30–136 kg/m2), and 81.1% were White (Table 1). At final pathology, 89.5% (590) had endometrial cancer and 10.5% (69) had hyperplasia; 80.5% had Stage I disease and median follow up time was 24 months. Of note, 39.6% (261) were obese, 41.7% (275) were morbidly obese and 18.7% (123) were super obese. Procedures performed were total abdominal hysterectomy/bilateral salpingooophorectomy (TAH/BSO) (32.3%, n = 213) and TAH/BSO with lymph node dissection (22.2%, n = 146) (Table 1); 280 minimally invasive procedures (MIS) were attempted and 235 completed (conversion rate — 16.1%). The BMI was significantly higher in the converted group (47.3 kg/m 2 vs. 40.6 kg/m 2; p b 0.001). Among 37 women with a BMI ≥ 60 kg/m 2 there was a 38.5% conversion rate
327
Table 1 Patient demographics (N = 659). Variables
Total (%)
Age at surgery in years (mean) Body mass index Obese (30–39 kg/m2) Morbidly obese (40–49 kg/m2) Super obese (≥ 50 kg/m2) Race White Black Asian/other Concomitant medical diagnoses Hypertension Diabetes Heart disease Hyperlipidemia Surgery performeda Open surgery Supracervical Hysterectomy TAH/BSO TAH/BSO/lymphadenectomy Radical hysterectomy Minimally invasive TLH/BSO TLH/BSO/lymphadenectomy Robotic hysterectomy/BSO Robotic hysterectomy/BSO/lymphadenectomy TVH Final pathologic diagnosis Endometrioid endometrial cancer Endometrial hyperplasia Carcinosarcoma Uterine papillary serous carcinoma Clear cell Uterine sarcoma Other cancer of the uterus Staging Stage I IA IB Stage II Stage III Stage IV Recurrence
58.1 yrs 261 (39.6) 275 (41.7) 123 (18.7) 534 (81.1) 118 (17.9) 7 (1.0) 446 (67.7) 242 (36.7) 95 (14.4) 215 (32.6) 379 (57.5) 16 (2.5) 213 (32.3) 146 (22.2) 4 (0.7) 280 (42.5) 88 (13.4) 75 (11.4) 46 (7.0) 51 (7.7) 20 (3.0)
515 (78.1) 69 (10.5) 26 (3.9) 6 (0.9) 3 (0.5) 14 (2.1) 471 (80.5) 405 (68.6) 66 (11.2) 31 (5.3) 68 (11.6) 15 (2.6) 60 (11.7)
a TAH = total abdominal hysterectomy; BSO = bilateral salpingo-oophorectomy; TLH = total laparoscopic hysterectomy; TVH = total vaginal hysterectomy.
for attempted minimally invasive surgeries, a 13.5% supracervical hysterectomy rate, and a 13.5% rate of nodal dissection. Women were stratified by obesity category and compared. The mean age of women significantly decreased as obesity category increased (60.0 vs. 57.3 vs. 54.0; pb 0.001) (Table 2). The percentage of MIS performed decreased with increasing obesity category (52.5% vs. 37.1% vs. 33.3%; pb 0.001). Almost 90% of the super obese had Stage I disease compared with 79.9% and 77.3% (184) of the obese and morbidly obese, respectively (p=0.027). With regard to surgical outcomes, across increasing obesity groups, there was a significant increase in estimated blood loss (EBL, 243 vs. 281 vs. 378 mL; pb 0.001), and the super obese group had significantly increased blood loss compared to both other groups (p b 0.001 and p=0.011). Decreased frequency of node dissection (63.8% vs. 37.1% vs. 20.3%; pb 0.001 — Fig. 1) was also observed with increasing obesity and there was a significant decrease in the mean number of lymph nodes removed (20.4 vs. 15.1 vs. 10.0; pb 0.001). A significant increase in length of stay was found (2.9 days vs. 3.1 days vs. 3.7 days, p=0.021) between the obese and super obese (p=0.015) as well as a significant increase in the total number of post-operative complications (0.6 vs. 0.9 vs. 1.2 complications, pb 0.001). The mean BMI was significantly higher in women who experienced the following complications: ICU admission (52.8 vs. 42.7 kg/m2, pb 0.001), any wound care requirement (47.1 vs. 42.3 kg/m2, p b 0.001), EBL ≥500 cm3 (46.1 vs. 39.3 kg/m2, pb 0.001),
328
L.E. Giugale et al. / Gynecologic Oncology 127 (2012) 326–331
Table 2 Comparison of all women based on obesity classification. Variable
Obese (n = 261) n (%)
Morbidly obese Super obese p Value (n = 123) (n = 275) n (%) n (%)
Mean age (years) Mean height (cm) Mean weight (kg) Mean BMI (kg/m2) Comorbidities Hypertension Diabetes Hyperlipidemia Heart disease Procedurea Open Open with lymphadenectomy Minimally invasive Minimally invasive with lymphadenectomy Lymph nodes Dissection performed Mean pelvic nodes Mean para-aortic nodes Total mean nodes (number) Wound Infection Superficial separation Fascial dehiscence Any wound care Wound reoperation Intraoperative factors EBL ≥ 500 EBL (cubic centimeters) Blood transfusion Procedure time (hrs:mins) Conversion to open Prolonged mechanical ventilation Bowel injury Postoperative factors ICU admission Length of stay (days) Pneumonia Ileus Hematoma Readmission to the hospital Discharged to SNF or NH Total number of complications
60.0 163.3 94.2 34.9
57.3 162.0 117.5 44.2
53.7 160.5 146.9 57.8
175 (67.0 86 (33.0) 97 (37.2) 43 (16.5)
179 (65.1) 99 (36.0) 80 (29.1) 38 (13.8)
92 57 38 14
53 (20.3) 71 (27.2)
109 (39.6) 64 (23.3)
67 (54.5) 15 (12.2)
54 (20.7) 83 (31.8)
69 (25.1) 33 (12.0)
31 (25.2) 10 (8.1)
166 (63.8) 16.5 4.6 20.4
102 (37.1) 13.1 2.2 15.5
25 (20.3) 10.0 0.15 10.0
p b 0.001 p = 0.001 p b 0.001 p = 0.001
23 (8.8) 18 (6.9) 5 (1.9) 21 (8.1) 6 (2.3)
47 (17.5) 41 (15.3) 2 (0.7) 36 (13.4) 6 (2.2)
19 (15.8) 35 (29.2) 1 (0.8) 30 (25.0) 3 (2.5)
p = 0.011 p b 0.001 p = 0.429 P b 0.001 p = 0.988
21 (8.1) 242.8 17 (6.5) 3:29
37 (13.5) 280.8 20 (7.5) 3:13
27 (22.3) 377.7 17 (14.2) 3:08
p = 0.001 p b 0.001 p = 0.035 p = 0.003
16 (11.7) 1 (0.4)
19 (18.6) 6 (2.2)
10 (24.4) 3 (2.5)
p = 0.102 p = 0.144
3 (1.1)
4 (1.5)
2 (1.6)
p = 0.919
1 (0.4) 2.87 1 (0.4) 21 (8.1) 3 (1.2) 15 (5.8)
10 (3.7) 3.13 2 (0.7) 27 (10.1) 8 (3.0) 22 (8.2)
9 (7.4) 3.66 2 (1.7) 6 (5.0) 1 (0.8) 13 (10.9)
p = 0.001 p = 0.021 p = 0.413 p = 0.243 p = 0.192 p = 0.203
3 (1.3) 0.61
5 (2.2) 0.90
1 (1.0) 1.21
p = 0.632 p b 0.001
(74.8) (46.3) (30.9) (11.4)
p b 0.001
p = 0.154 p = 0.038 p = 0.124 p = 0.388 p b 0.001
Women were also compared based on the route of the hysterectomy performed: open abdominal hysterectomy or MIS (including laparoscopic, robotic, and total vaginal hysterectomies; Table 3). The mean BMI of patients undergoing minimally invasive procedures was significantly lower than those of women undergoing open abdominal hysterectomies (41.7 kg/m 2 vs. 44.0 kg/m 2; p = 0.003). Minimally invasive procedures were more frequently performed in patients with Stage I disease (p = 0.008), and with FIGO grade 1/2 disease (p = 0.048). Compared to open surgeries, MIS procedures were associated with significantly less blood loss (173.9 vs. 365.7 mL, p b 0.001) and fewer blood transfusions (2.9% vs. 12.4%, p b 0.001). There were also fewer superficial wound separations (7.9% vs. 19.4%, p b 0.001), fascial wound separations (0.0% vs. 2.2%, p = 0.014), wound infections (4.7% vs. 20.5%, p b 0.001), requirement for any wound care (6.5% vs. 18.6%, p b 0.001) and decreased ileus percentages (3.6% vs. 11.9%, p b 0.001). Lymphadenectomy was performed more frequently during MIS procedures (55.8% vs. 44.1%, p = 0.006). Lastly, patients undergoing MIS procedures were less likely to be discharged to a skilled nursing facility (0.4% vs. 2.8%, p = 0.025) or readmitted (4.7% vs. 10.0%, p = 0.012). There was no difference in rates of bowel/bladder injury, post-operative admission to the intensive care unit, post-operative mechanical ventilation, myocardial infarction, pulmonary embolus, deep vein thrombosis, pneumonia, peri-operative death, or reoperation. Analysis was also performed after the minimally invasive category was broken down into laparoscopic and robotic hysterectomies. The only significant differences between robotic and laparoscopic procedures were an increased rate of post-operative mechanical ventilation (6.2%) with robotic procedures compared with both open (0.8%) and laparoscopic hysterectomies (0.6%) (p = 0.001). Given the significant correlation for surgical complications/outcomes with both obesity classification and route of surgery, a multivariate analysis was performed. Both an open type of surgery and increasing obesity classification were independently and significantly associated with the following: superficial wound complication (p b 0.001 for both), postoperative wound care (p b 0.001 for both), and the presence of at least one postoperative complication (p b 0.001 for procedure type; p= 0.003 for obesity classification). The following complications were significantly associated with an open procedure but not with obesity classification: post-operative blood transfusion (p b 0.001), readmission to the hospital (p =0.025), and postoperative ileus (p b 0.001). In contrast, lymphadenectomy was significantly associated with decreasing obesity classification (pb 0.001) but not with type of procedure (p= 0.11). Odds ratios were calculated to compare relative impact of both type of surgery and obesity classification on outcomes of interest (Table 4). For the performance of a node dissection, the following
a Open includes TAH, supracervical hysterectomies, and radical hysterectomies. Minimally Invasive includes TLH, robotic Hysterectomies, and TVH.
blood transfusion (46.1 vs. 42.7 kg/m2, p =0.014), bowel injury (53.4 vs. 42.9 kg/m2, p=0.001) and pneumonia (58.2 vs. 42.9 kg/m2, p= 0.001) (Supplemental Table).
Fig. 1. Lymphadenectomy by obesity group: the total number of women undergoing lymphadenectomy is listed along with percentage within each obesity group.
Table 3 Comparison of all women based on procedure performed.
Age (years) Weight (kg) BMI (kg/m2) Comorbidities Hyperlipidemia Stage I Blood EBL (mL) Transfusion Wound Infection Superficial separation Fascial separation Any wound care Ileus Nursing facility Readmission
Abdominal n (%)
Minimally invasive n (%)
p Value
57.8 116.1 44.0
58.6 110.7 41.7
p = 0.389 p = 0.004 p = 0.003
111 (29.3%) 259 (76.9%)
104 (37.1%) 215 (85.7%)
p = 0.036 p = 0.008
365.7 46 (12.4)
173.9 8 (2.9)
76 (20.5) 72 (19.4) 8 (2.2) 18 (6.5) 44 (11.9) 8 (2.8) 37 (10.0)
13 (4.7) 22 (7.9) 0 (0.0) 69 (18.6) 10 (3.6) 1 (0.4) 13 (4.7)
p b 0.001 p b 0.001 p b 0.001 P b 0.001 p = 0.014 p b 0.001 p b 0.001 p = 0.025 p = 0.012
L.E. Giugale et al. / Gynecologic Oncology 127 (2012) 326–331 Table 4 Odds ratios. Variable
Effect
Odds ratio (95% confidence interval)
Lymphadenectomy
Open surgery Minimally invasive surgery Obese Morbidly obese Super obese Open surgery Minimally invasive surgery Obese Morbidly obese Super obese Open surgery
1 (reference) 1.31 (0.94–1.83) 1 (reference) 0.35 (0.24–0.49) 0.15 (0.09–0.25) 1 (reference) 0.27 (0.13–0.55) 1 (reference) 1.07 (0.58–1.97) 0.48 (0.19–1.24) 1 (reference)
Minimally invasive surgery Obese Morbidly obese Super obese Open surgery
0.41 (0.24–0.69) 1 (reference) 2.16 (1.20–3.89) 4.91 (2.62–9.20) 1 (reference)
Minimally invasive surgery Obese Morbidly obese Super obese Open surgery Minimally invasive surgery Obese Morbidly obese Super obese Open surgery Minimally invasive surgery Obese Morbidly obese Super obese Open surgery Minimally invasive surgery Obese Morbidly obese Super obese Open surgery
0.22 (0.10–0.47) 1 (reference) 0.95 (0.48–1.87) 1.91 (0.93–3.95) 1 (reference) 0.34 (0.19–0.58) 1 (reference) 1.53 (0.86–2.72) 3.27 (1.76–6.08) 1 (reference) 0.21 (0.11–0.38) 1 (reference) 1.83 (1.06–3.16) 1.55 (0.80–3.03) 1 (reference) 0.47 (0.24–0.91) 1 (reference) 1.31 (0.66–2.61) 1.77 (0.81–3.89) 1 (reference)
Minimally invasive surgery Obese Morbidly obese Super obese
0.20 (0.14–0.28) 1 (reference) 1.32 (0.90–1.91) 2.28 (1.41–3.69)
Postoperative ileus
Superficial wound separation
Postoperative blood transfusion
Postoperative wound care
Wound infection
Readmission
Total number of complications
estimates were obtained: minimally invasive surgery (OR 1.31 95% CI 0.94–1.83) versus open procedure, and morbidly obese (OR 0.35 95% CI 0.24–0.49) and super obese (OR 0.15 95% CI 0.09–0.25) versus obese. At final follow-up, 500 patients (75.9%) were alive with no evidence of disease, 18 (2.7%) were deceased either due to surgery (1 patient due to pneumonia) or to cancer (17 patients), 41 (6.2%) were alive with cancer and 9 patients (1.4%) were dead of other causes. Ninety-one patients (13.8%) were lost to follow up. There was a recurrence rate of 11.7% (60 patients) with 13 vaginal recurrences, 14 pelvic recurrences, and 31 distant recurrences. Progression free survival (PFS) was analyzed using Kaplan Meier method and log rank statistic and found to be significantly associated with both grade and stage (Pb 0.001) but not with obesity category (p= 0.066; Fig. 2) or procedure type (p= 0.392). Both disease specific survival data and overall survival data are limited in that there were only 17 deaths due to endometrial cancer and 27 deaths from any cause in the entire cohort. Decreased disease specific survival was significantly associated with both increasing grade and stage (p b 0.001) but not with obesity classification (p= 0.993) or type of surgery performed (p= 0.21).
Discussion Although obesity is defined as a BMI exceeding 30 kg/m 2, this study highlights the fact that all obese women with endometrial
329
cancer do not have equivalent surgical risks. Most prior studies on the impact of obesity on surgical outcomes were not limited to women with an endometrial malignancy or did not stratify obesity beyond the morbidly obese or simply compared obese to non obese patients and thus have not analyzed differential outcomes for women with more extreme obesity. Our study confirmed that increasing BMI beyond mere obesity (BMI ≥ 30 kg/m 2) appears to be significantly associated with certain intraoperative and postoperative outcomes, most notably: conversion to an open procedure, ICU admission, superficial wound separation and wound care requirement. While blood transfusion, EBL greater than 500 mL, bowel injury, and pneumonia were significantly increased by obesity on univariate analyses; these effects appeared to be mitigated if minimally invasive procedures were performed. With regard to EBL during open cases, our study supports the contention that the significant increased risk occurs in the super obese category. These data corroborate Osler and colleagues' findings of increased risk of bleeding associated with obesity [7] but are in contrast to studies which have not found a relation between obesity and blood loss and which have found an association with longer operative times and infections; these studies may have lacked a sufficient quantity of super obese patients to detect any significant differences [5,7,8]. The total number of complications increased among increasing obesity groups; while it is difficult to discern a BMI breakpoint for this, our study suggests that more complications may not be seen until one reaches a BMI greater than 50 kg/m 2. Women in our study who were admitted to an intensive care unit or required postoperative wound care also had higher BMIs and as BMI surpassed 50 kg/m 2, the risk of longer hospital stays increased. One other study analyzed women with BMIs > 50 kg/m 2 and also found that these women required ventilation with higher airway pressures [13]. For very medically infirm patients, whether due to obesity or other co-morbid conditions, radiation therapy may also be a nonsurgical alternative. In addition to BMI, procedure type also influenced outcomes in our study. Minimally invasive procedures were associated with less blood loss, fewer blood transfusions, decreased wound complications, ileus, or readmission. For laparoscopic cases, previous studies have not found associations between obesity and intraoperative blood loss, postoperative infections, or length of stay and there have been varying conclusions regarding operative times [3,6,8–12,14]. Heinberg et al. did find significant associations between obesity and blood loss greater than 500 mL [6]. Regarding robotic assisted total laparoscopic hysterectomies, other studies have not reported a correlation between BMI and blood loss, postoperative complications, operative room times, or length of stay [15,16]. Our study found that robotic hysterectomy demonstrated an increased frequency of post-operative mechanical ventilation (6.5%) compared to both open and laparoscopic surgeries; this was a novel finding. While the exact mechanism is unclear, it may be the increased intra-abdominal pressure needed for insufflation and subsequent hypercarbia along with the steep Trendelenburg position are potential causative factors but further exploration is warranted. The multivariate analysis helped clarify the relative impact of obesity level compared to procedure type. Both obesity and procedure type independently affected superficial wound complications, postoperative wound care and the presence of postoperative complications whereas obesity preferentially impacted performance of lymph node dissection and procedure type impacted postoperative blood transfusion, readmission to the hospital and postoperative ileus. This underscores the importance of increasing the percentage of minimally invasive surgeries in endometrial cancer patients in order to reduce surgical complications and the aforementioned risks associated with open hysterectomies. Numerous studies have advocated the use of minimally invasive procedures for endometrial cancer surgery and demonstrated the ability to perform adequate staging in obese and morbidly obese women and these methods are being more universally
330
L.E. Giugale et al. / Gynecologic Oncology 127 (2012) 326–331
Fig. 2. Progression free survival by obesity group.
adopted in gynecologic oncology [16,18,19]. Minimally invasive procedures were more prevalent in the latter part of our study; in the first half of the study (before 2007) 18% of hysterectomies were performed via a minimally invasive approach whereas this steadily increased to over 75% by the last years studied. Another important area of impact will be with regard to the goals of surgery for endometrial cancer and realistic expectations for completion of full surgical staging. Decreased frequency of node dissection with a significant decrease in number of lymph nodes was observed in our study as obesity category increased from 64% in the obese women down to 20% of the super obese patients. However, since the significance was between the obese and the higher groups; this suggests that comprehensive surgical staging decreases around a BMI of 40 kg/m2 but does not decrease significantly with further increases in BMI. Previous authors have demonstrated the safety and feasibility of lymph node dissection in obese women, but the mean BMIs were either not reported or less than our study [8,11,14,17,18]. For example, O'Hanlan assessed the impact of obesity on outcomes of total laparoscopic hysterectomy for uterine pathology; however the average BMI was only 30 kg/m2 [11]. Similarly, Lau et al. analyzed surgical outcomes of normal weight patients compared with obese and morbidly obese women undergoing robotic hysterectomy for endometrial cancer and concluded that obesity did not impact surgical staging in this study of 108 women (23 were morbidly obese). Of note, while the number of pelvic nodes removed was equivalent across groups (range 10.4–10.8), significantly fewer of the morbidly obese women had para-aortic lymphadenectomy (42% normal weight vs. 27% obese vs. 1% morbidly obese; p =0.002) [16]. Lymphadenectomy was performed more frequently during minimally invasive procedures in our study; however, on multivariate analysis, it appears that lymphadenectomy was more feasible due to the decreased obesity in that sub-population. Thus, surgeons are foregoing lymph node dissection more frequently in the morbidly and super obese women, likely for safety or technical concerns. While debate continues on
whether all endometrial cancer patients should undergo complete pelvic and para-aortic lymphadenectomy, our study confirms that the dissection rates vary widely based on the extent of obesity. More super obese patients had apparent Stage I disease and fewer had a lymphadenectomy; however, among only the patients who underwent lymphadenectomy, the differences in the prevalence of Stage I disease were no longer significant. This suggests that while the higher prevalence of Stage I disease in the super obese may be falsely elevated, this did not appear to effect progression free or disease specific survival significantly and the disease process may be more indolent. With regard to recurrence data and survival, there were no significant differences based on obesity classifications although the potential of impact of adjuvant treatment data was not analyzed. Thus, the major impact of obesity that we were able to observe in our study was on short term outcomes as opposed to long term disease outcomes. Obesity has previously been shown to carry a vastly increased risk of endometrial cancer death that is six fold higher for women with a BMI ≥40 kg/m 2 in large population studies [20]. We may have lacked sufficient power and length of follow up to show differences in survival in this population and further long term analysis is needed to determine the relative impact of both cancer and co-morbidities. Our study does have notable limitations inherent to all retrospective studies. Our study is susceptible to selection bias by surgeons on the type of procedures performed. There is a reliance on a chart review and complications are undoubtedly underreported and only factors present in the records could be captured and analyzed. There are also differences in the numbers of open, laparoscopic, and robotic cases and more of the recent cases were performed through minimally invasive approaches. More laparoscopic and robotic case numbers would have improved our ability to detect significant differences. Lastly, BMI was used as the means to classify obesity; however BMI may not always be the best way to determine the extent of obesity. Waist-to-hip ratio might have been
L.E. Giugale et al. / Gynecologic Oncology 127 (2012) 326–331
helpful in determining the extent of abdominal/visceral obesity in particular. In conclusion, all obese women are not equal in terms of their risks for adverse peri-operative outcomes and ability to adequately perform comprehensive surgical staging. As more endometrial cancer patients have a BMI over 50 kg/m 2, both surgeons and patients need realistic estimations of expected surgical outcomes. As part of the preoperative surgical consent process, women with endometrial cancer need to understand the associations between increased BMI and surgical complications, increased conversion rates and decreased lymph node dissection. Although technically more difficult, our data show that, even in this population which is composed of almost 20% super obese women, minimally invasive procedures may offer the advantage of fewer complications than traditional open surgeries. Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.ygyno.2012.08.014. Conflict of interest statement The authors declare that there are no conflicts of interest.
References [1] Sturm R. Increases in morbid obesity in the USA: 2000–2005. Public Health 2007;121(7):492-6. [2] Bergstrom A, Pisani P, Tenet V, Wolk A, Adami HO. Overweight as an avoidable cause of cancer in Europe. Int J Cancer 2001;91:421-30. [3] Chopin N, Malaret JM, Lafay-Pillet MC, Fotso A, Foulot H, Chapron C. Total laparoscopic hysterectomy for benign uterine pathologies: obesity does not increase the risk of complications. Hum Reprod 2009;24(12):3057-62. [4] Sheth SS. Vaginal hysterectomy as a primary route for morbidly obese women. Acta Obstet Gynecol Scand 2010;89(7):971-4. [5] Rasmussen KL, Neumann G, Ljungstrom B, Hansen V, Lauszus FF. The influence of body mass index on the prevalence of complications after vaginal and abdominal hysterectomy. Acta Obstet Gynecol Scand 2004;83(1):85-8. [6] Heinberg EM, Crawford III BL, Weitzen SH, Bonilla DJ. Total laparoscopic hysterectomy in obese versus nonobese patients. Obstet Gynecol 2004;103(4):674-80.
331
[7] Osler M, Daugbjerg S, Frederiksen BL, Ottesen B. Body mass and risk of complications after hysterectomy on benign indications. Hum Reprod 2011;26(6):1512-8. [8] Santi A, Kuhn A, Gyr T, Eberhard M, Johann S, Gunthert AR, et al. Laparoscopy or laparotomy? A comparison of 240 patients with early-stage endometrial cancer. Surg Endosc 2010;24(4):939-43. [9] Ghezzi F, Cromi A, Bergamini V, Uccella S, Beretta P, Franchi M, et al. Laparoscopic management of endometrial cancer in nonobese and obese women: a consecutive series. J Minim Invasive Gynecol 2006;13(4):269-75. [10] Kodama J, Seki N, Ojima Y, Nakamura K, Hongo A, Hiramatsu Y. Risk factors for early and late postoperative complications of patients with endometrial cancer. Eur J Obstet Gynecol Reprod Biol 2006;124(2):222-6. [11] O'Hanlan KA, Dibble SL, Fisher DT. Total laparoscopic hysterectomy for uterine pathology: impact of body mass index on outcomes. Gynecol Oncol 2006;103(3): 938-41. [12] O'Hanlan KA, Lopez L, Dibble SL, Garnier AC, Huang GS, Leuchtenberger M. Total laparoscopic hysterectomy: body mass index and outcomes. Obstet Gynecol 2003;102(6):1384-92. [13] O'Gorman T, MacDonald N, Mould T, Cutner A, Hurley R, Olaitan A. Total laparoscopic hysterectomy in morbidly obese women with endometrial cancer anaesthetic and surgical complications. Eur J Gynaecol Oncol 2009;30(2): 171-3. [14] Holub Z, Bartos P, Jabor A, Eim J, Fischlova D, Kliment L. Laparoscopic surgery in obese women with endometrial cancer. J Am Assoc Gynecol Laparosc 2000;7(1): 83-8. [15] Nawfal AK, Orady M, Eisenstein D, Wegienka G. Effect of body mass index on robotic-assisted total laparoscopic hysterectomy. J Minim Invasive Gynecol 2011;18(3): 328-32. [16] Lau S, Buzaglo K, Vaknin Z, Brin S, Kaufer R, Drummond N, et al. Relationship between body mass index and robotic surgery outcomes of women diagnosed with endometrial cancer. Int J Gynecol Cancer 2011;21(4):722-9. [17] Pavelka JC, Ben-Shachar I, Fowler JM, Ramirez NC, Copeland LJ, Eaton LA, et al. Morbid obesity and endometrial cancer: surgical, clinical, and pathologic outcomes in surgically managed patients. Gynecol Oncol 2004;95(3):588-92. [18] Seamon LG, Bryant SA, Rheaume PS, Kimball KJ, Huh WK, Fowler JM, et al. Comprehensive surgical staging for endometrial cancer in obese patients: comparing robotics and laparotomy. Obstet Gynecol 2009;114(1):16-21. [19] Subramaniam A, Kim KH, Bryant SA, Zhang B, Sikes C, Kimball KJ, et al. A cohort study evaluating robotic versus laparotomy surgical outcomes of obese women with endometrial carcinoma. Gynecol Oncol 2011;122(3):604-7. [20] Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med 2003;348:1625-38.