j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 7 ( 2 0 1 5 ) 4 5 e4 9
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Association for Academic Surgery
Body mass index predicts operative time in elective colorectal procedures Harish Saiganesh, BS, David E. Stein, MD, and Juan L. Poggio, MD* Division of Colorectal Surgery, Department of Surgery, Drexel University College of Medicine, Philadelphia, Pennsylvania
article info
abstract
Article history:
Background: Obesity currently affects more than a third of the United States population and
Received 28 December 2014
is associated with increased surgical complications. Compared to all other subspecialties,
Received in revised form
colorectal surgery is the most affected by the increasing trend in obese surgical patients.
5 February 2015
Operative time has been found to have the greatest impact on hospital costs and physician
Accepted 27 February 2015
workload. This study was conducted to determine whether obesity has a direct impact on
Available online 6 March 2015
operative time in elective colorectal procedures using a high-powered, nationally representative patient sample.
Keywords:
Methods: A retrospective analysis was conducted on 45,362 patients who underwent open
Body mass index
and laparoscopic ileocolic resections, partial colectomies, and low pelvic anastomoses using
Obesity
American College of Surgeons National Surgical Quality Improvement Program data from
Operative time
2005e2009. Operative time was the main outcome variable, whereas body mass index (BMI)
Colorectal surgery
was the main independent variable. BMI was divided into three classes as follows: normal
American College of Surgeons
(35). A univariate linear model
National Surgical Quality
was used to analyze the relationship while controlling for confounding factors such as de-
Improvement Program
mographics and preoperative conditions. Statistical significance was established at P 0.05. Results: Morbidly obese patients were found to have longer operative times than did normal patients across each individual colorectal procedure (P < 0.001), ranging from a mean difference of 17.8 min for open ileocolic resections to 56.6 min for laparoscopic low pelvic anastomoses with colostomies. Conclusions: BMI, as an objective measure of obesity, is a direct, statistically significant independent predictor of operative time across elective colorectal procedures. ª 2015 Elsevier Inc. All rights reserved.
1.
Introduction
Obesity has been cited as one of the major health issues associated with industrialized counties such as the United States, where a third of the population is affected and one in twenty is classified as morbidly obese [1]. Obese individuals have a high
body mass index (BMI), a measurement that is partly determined by the patients’ total body fat [2]. Consequently, operating on these obese patients can be a challenge for surgeons during open or laparoscopic colorectal procedures. Various studies have established the association of obesity with higher rates of surgical site infections in colorectal procedures [1,3].
* Corresponding author. Division of Colorectal Surgery, Department of Surgery, Drexel University College of Medicine, 245 N. 15th Street, MS 413, Philadelphia, PA 19102 1192. Tel.: þ1 215 762 1750; fax: þ1 215 762 8389. E-mail address: [email protected] (J.L. Poggio). 0022-4804/$ e see front matter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2015.02.067
46
j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 7 ( 2 0 1 5 ) 4 5 e4 9
Patients undergoing colorectal procedures present a greater degree of postoperative complications when compared with patients undergoing other surgical procedures [4]. The most prevalent of these complications include surgical site infections, which have been associated with an increased postoperative length of stay, greater costs of care, and higher rates of readmission [5]. Operative time has been identified as a significant predictor of surgical site infections as per the Centers for Disease Control and Prevention National Nosocomial Infections Surveillance index [6]. In addition, studies have demonstrated the impact of longer operative times on increased hospital costs and increased surgical workloads [7e9]. Although a few studies have described an association of BMI and body surface area with operative time in specific colorectal procedures and conditions, they did not analyze other colorectal procedures or compare both laparoscopic cases and open cases [10,11]. Furthermore, there have been no prior studies on whether BMI independently predicts operative time in colorectal procedures. We have queried the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database to more accurately analyze the relationship between obesity and operative time across multiple colorectal procedures and to highlight the importance of an increasing trend in obesity on increased time of work for colorectal surgeons.
2.
Methods
2.1.
Data collection
The ACS-NSQIP database was evaluated in a retrospective outcome analysis to select patients who underwent nonemergent colorectal procedures. Using NSQIP Participant Use Files data from the years 2005e2009, 54,084 patients were selected based on Current Procedural Terminology coding criteria. Surgical procedures of interest included the following three most commonly performed procedures (open and laparoscopic procedures are listed, respectively, in parentheses): ileocolic resections (44,160; 44,205), partial
colectomies (44,140; 44,204), and low pelvic anastomoses (44,145/44,146; 44,207/44,208). BMI in kilogram per square meter was calculated using height and weight data. A new variable was created to divide BMI into three classes as follows: normal (35). Operative time was defined in minutes beginning with the initial skin incision and ending with application of postoperative dressing and served as the primary outcome variable, whereas BMI was the primary predictor variable. Results were reported as mean standard error. Exclusion criteria included both missing data (534 patients) and presence of outliers (8188 patients) in potential predictor variables as well as in the outcome variable, leading to a final sample of 45,362 patients for data analyses. Of those patients, when classified by BMI, 15,639 patients were normal, 24,146 patients were overweight and/or obese, and 5577 patients were morbidly obese. When classified by surgical procedure, there were 10,643 patients who had undergone ileocolic resections (6458 for open and 4185 for laparoscopic), 23,583 patients who had undergone partial colectomies (14,155 for open and 9428 for laparoscopic), and 11,136 patients who had undergone low pelvic anastomoses (6702 for open and 4434 for laparoscopic).
2.2.
Statistical analysis
The effect of potential predictor variables on operative time was determined via a univariate analysis of variance. The variables used in the analysis are identified (Table 1). Of these variables, BMI, surgical procedures, age, gender, race, diabetes mellitus, dyspnea, financial status before illness, financial status before surgery, ventilator dependence, ascites, esophageal varices, congestive heart failure, previous angioplasty, previous cardiac surgery, angina, rest pain, renal failure, history of transient ischemic attacks, disseminated cancer, wound infection, chronic steroid use, weight loss, chemotherapy, radiotherapy, pregnancy, and systemic sepsis were identified as statistically significant predictor variables of operative time. Discrete predictor variables significant in the univariate analysis were converted to continuous ones to be used in
Table 1 e List of potential predictor variables of operative time. BMI Dyspnea Congestive heart failure Acute renal failure Chronic steroid use Ascites
Surgical procedures
Age
Financial status before illness History of myocardial infarction Dialysis
Financial status prior to surgery Previous angioplasty
Ventilator dependence Previous cardiac surgery
COPD history
Pneumonia
Angina
Impaired sensorium Bleeding disorders
Coma Blood transfusions before surgery
Weight loss
Revascularization history
Gender
Race
Diabetes mellitus
Alcohol use
History of stroke CNS tumor
Hypertension
Esophageal varices Rest Pain
Hemiplegia
Paraplegia
Quadriplegia
Chemotherapy
Radiotherapy
Pregnancies
Wound infection Operations a month before surgery Systemic sepsis
History of transient ischemic Attacks
CNS ¼ central nervous system; COPD ¼ chronic obstructive pulmonary disease.
Cigarette smoking
Disseminated cancer
j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 7 ( 2 0 1 5 ) 4 5 e4 9
dimension reduction analysis. Age, gender, and race were reduced into a factor called demographics, financial statuses before illness and surgery were reduced into a factor called combined financial status, and all other significant predictor variables except surgical procedures and BMI were reduced into a factor called preoperative conditions. Dimension reduction was performed to account for a greater number of significant predictor variables in further analyses of variance, which places limits on the number of individual variables used as covariates. The use of more variables accounts for greater variance in operative time that can be adjusted to solely measure BMI’s effects on operative time [12]. An analysis of covariance (ANCOVA) was performed to determine the relationship between BMI and operative time adjusted for the covariates demographics, surgical procedures, and preoperative conditions. Assumptions for ANCOVA such as homogeneity of variance, normality of residuals, error term independence, and homogeneity and linearity of regression slopes were tested and met for all but one variable. Combined financial status failed the homogeneity of regression assumption and thus was not included in the final ANCOVA. An ANCOVA was also performed for each surgical procedure to further measure differences in operative time means by BMI. All analyses were performed using type 1 sum of squares to account for unequal sample sizes in each of the variable groups. A linear regression was also conducted with the same results to verify the ANCOVA results. A post hoc test for multiple comparisons was conducted and adjusted via the Bonferroni correction. Statistical significance was established at P 0.05. Statistical analyses were performed using SPSS software (version 22, Chicago, IL).
3.
Results
After a univariate analysis, BMI was found to have a statistically significant association with operative time (P < 0.001). An ANCOVA showed that BMI remained a statistically significant predictor of operative time (P < 0.001), while adjusting for variances due to surgical procedures, demographics, and preoperative conditions. As per the post hoc multiple comparisons test, differences in operative time means were statistically significant between all BMI classes (P < 0.001). The linear regression also identified that the unstandardized coefficient (B) of the BMI classes was 13.540, indicating that for every increment in BMI classification (from normal to overweight and/or obese to morbidly obese), operative time increased by 13.5 min on average (P < 0.001). The adjusted mean operative times by different BMI classes were reported across all the surgical procedures (Table 2) and across each individual surgical procedure (Table 3). Across the procedures, there was a mean difference of 27.8 min between normal weight and morbidly obese patients. For open and laparoscopic procedures, respectively, there was a mean difference of 17.8 and 22.6 min between normal weight and morbidly obese patients who underwent ileocolic resections; 25.4 and 26.9 min for those who underwent partial colectomies; 34.3 and 49.9 min for those who underwent low pelvic anastomoses without colostomies; and 26.8 and 56.6 min for those who underwent low pelvic anastomoses
47
Table 2 e Mean operative times (in minutes) versus BMI (in kg/m2) across all colorectal procedures, adjusted for effects from covariates demographics and preoperative conditions. BMI
Mean operative time
Normal weight (35)
162.4 0.7 (P < 0.001) 174.9 0.5 (P < 0.001) 190.2 1.1 (P < 0.001)
with colostomies. The greatest mean difference in operative time because of BMI exists among patients who underwent laparoscopic low pelvic anastomoses with colostomy (56.6 min), whereas the smallest mean difference exists among patients who underwent open ileocolic resection (17.8 min). On average, laparoscopic colorectal procedures had longer operative times than open ones. Ileocolic resections had the shortest operative times, followed by partial colectomies and low pelvic anastomoses without and with colostomies.
4.
Discussion
Using data from ACS-NSQIP, our study determined that BMI has a direct, statistically significant independent relationship with operative time in elective colorectal procedures. Unlike previous studies on BMI and operative time association in colorectal surgery, our study included a nationwide sample of 45,362 patients across procedures, making it a more powerful measure of the degree of association. Moreover, our study measured the association across eight different colorectal surgical procedures and used data from at least 400 to over 16,000 patients per procedure, based on the relative frequency of the individual procedure each year. With the skyrocketing incidence of obesity, it becomes necessary to identify the resources required to optimize surgical treatment in this population. This trend may have a strong impact on surgeon workload. Longer operative times have been demonstrated for obese patients undergoing general surgical procedures requiring increased work by the surgeon [13]. A US census-based population study determined that workload is expected to increase about 31.5% for general surgeons by 2020, with a 40.3% increase for gastrointestinal surgeons [14]. Furthermore, there is a projected increase of 42% in colon resections by 2020, with colon resections accounting for potentially 13.7% of the total surgical workload in 2020 [14]. Although significant, such increases were reported primarily with the knowledge of an increasing elderly population in the future and did not account for increased work required to treat obese patients [14]. There has also been evidence of decreasing reimbursement for surgeons across all fields, particularly for those performing colectomies [15]. Surgical reimbursement is currently provided by Medicare, Medicaid, and commercial insurers. Reimbursement is dependent on relative value units, which consider factors such as physician workload, overhead, and malpractice and vary for different surgical procedures [14]. Thus, future colorectal surgeons will be facing greater
48
Normal (35)
Laparoscopic Open Laparoscopic Open Laparoscopic Open Laparoscopic Open
Low pelvic anastomosis with colostomy Low pelvic anastomosis without colostomy Partial colectomy Ileocolic resection BMI
Table 3 e Mean operative times (in minutes) versus BMI (in kg/m2) by individual colorectal procedure, adjusted for effects from covariates demographics and preoperative conditions.
j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 7 ( 2 0 1 5 ) 4 5 e4 9
workloads while receiving less. With the increased time commitment required to conduct surgery for obese patients also comes an increased workload for supporting personnel such as anesthesiologists and nurses. In addition, there are reports that we are facing a national shortage of anesthesia providers. It is possible that without adjusting reimbursement for the care of these obese patients, providers may simply refuse to care for them [13,16]. In a study conducted on minority patients who underwent various inpatient surgeries, BMI has been associated with increased hospital costs due to surgical difficulties and consequent longer use of the operating room [17]. Moreover, operating duration has been associated with about 40% of total hospital costs, with an average cost of $30 per minute in colorectal surgical procedures [9,18]. Based on our results, it would thus cost an additional $834 on average to operate on a morbidly obese patient compared with a normal one for a general elective colorectal procedure based on duration alone. When considering individual procedures, it would cost an additional $600 on average for ileocolic resections; $780 on average for partial colectomies; and $1260 on average for both types of low pelvic anastomoses. With increasing procedure-based workloads, increasing timebased costs, and decreasing Medicare reimbursement for colorectal surgeons operating on an increasing obese population in the United States, we suggest that the relative value unit system of surgical reimbursement be restructured to include BMI as a predictive measure of operative costs. Furthermore, changes in demographics and other preoperative conditions in the surgical population should also be considered when restructuring reimbursement.
4.1.
Limitations
One limitation in this study is that the variables recorded in the NSQIP database may not capture other factors that can have an impact on operative time such as adhesions and surgeon experience [13,19]. Another limitation is that data on operating room cost per minute were not found for each individual colorectal surgical procedure in this study but rather for colorectal surgery in general, hence the use of the latter in our discussion [17]. In addition, the costs per minute may be higher now because of inflation. Nevertheless, in spite of these limitations, this study is supported by the use of a large-scale, nationally representative sample of elective colorectal surgery patients to determine the impact of obesity on operative time and consequent health-care costs.
5.
Conclusions
BMI, which has been documented as an objective measure of obesity, is a statistically significant, independent predictor of operative time in both open and laparoscopic elective colorectal procedures. We suggest future studies to further investigate the creation of a modifier to accurately reflect the increased time of work required to care for morbidly obese patients.
j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 7 ( 2 0 1 5 ) 4 5 e4 9
Acknowledgment Authors’ contributions: J.L.P. and H.S. contributed to the study conception and design, acquisition of data, and analysis and interpretation of data. J.L.P., H.S., and D.E.S. did the drafting of the article and the critical revision.
Disclosure All authors deny any disclosure of potential conflicts of interest, including financial interests, activities, relationships, and affiliations. The American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) and the hospitals participating in the ACS-NSQIP are the source of the data used herein; they have not verified and are not responsible for either the statistical validity of the data analysis or the conclusions derived by the authors.
references
[1] Haas EM, Aminian A, Nieto J, et al. Minimally invasive colorectal surgery in the morbidly obese: does high body mass index lead to poorer outcomes? Surg Curr Res 2013;3:149. [2] Chern H, Chou J, Donkor C, et al. Effects of obesity in rectal cancer surgery. J Am Coll Surg 2010;211:55. [3] Fujii T, Tsutsumi S, Matsumoto A, et al. Thickness of subcutaneous fat as a strong risk factor for wound infections in elective colorectal surgery: impact of prediction using preoperative CT. Dig Surg 2010;27:331. [4] Young H, Knepper B, Moore EE, et al. Surgical site infection after colon surgery: national Healthcare Safety Network risk factors and modeled rates compared with published risk factors and rates. J Am Coll Surg 2012;214:852. [5] Jones RS, Brown C, Opelka F. Surgeon compensation: “Pay for performance,” the American College of Surgeons National Surgical Quality Improvement Program, the Surgical Care Improvement Program, and other considerations. Surgery 2005;138:829.
49
[6] Mangran AJ, Horan TC, Pearlson ML, Silver LC, Jarvis WR. Guideline for prevention of surgical site infection. Infect Control Hosp Epidemiol 1999;20:247. [7] Dexter F, Blake JT, Penning DH, Sloan B, Chung P, Lubarsky DA. Use of linear programming to estimate impact of changes in a hospital’s operating room time allocation on perioperative variable costs. Anesthesiology 2002;96:718. [8] Beck DE. Reoperative surgery: what can we learn from a large randomized prospective trial? Clin Colon Rectal Surg 2006;19:247. [9] Little DC, St Peter SD, Calkins CM, et al. Relative value units correlate with pediatric surgeons’ operating time; when perceived myth becomes reality. J Pediatr Surg 2006;41:234. [10] Krane MK, Allaix ME, Zoccali M, et al. Does morbid obesity change outcomes after laparoscopic surgery for inflammatory bowel disease? Review of 626 consecutive cases. J Am Coll Surg 2013;216:986. [11] Vaccaro CA, Vaccarezza H, Rossi GL, et al. Body surface area: a new predictor factor for conversion and prolonged operative time in laparoscopic colorectal surgery. Dis Colon Rectum 2012;55:1153. [12] Field AP. Discovering statistics using SPSS. 2nd ed. London: Sage; 2005. [13] Hawn MT, Bian J, Leeth RR, et al. Impact of obesity on resource utilization for general surgical procedures. Ann Surg 2005;241:821. [14] Liu JH, Etzioni DA, O’Connell JB, Maggard MA, Ko CY. The increasing workload of general surgery. Arch Surg 2004; 139:423. [15] Russell T. Statement of the American College of Surgeons to the subcommittee on health committee on energy and commerce. US House of Representatives; 2002. [16] Spetz J, Given R. The future of the nurse shortage: will wage increases close the gap? Health Aff 2003;22:199. [17] Nafiu OO, Shanks AM, Hayanaga AJ, Tremper KK, Campbell DA Jr. The impact of body mass index on postoperative complications and resource utilization in minority patients. J Natl Med Assoc 2011;103:9. [18] Marrin CAS, Johnson LC, Beggs VL, Batalden PB. Clinical process cost analysis. Ann Thorac Surg 1997;64:690. [19] Derossis AM, Fried GM, Abrahamowicz M, Sigman HH, Barkun JS, Meakins JL. Development of a model for training and evaluation of laparoscopic skills. Am J Surg 1998;175:482.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Association for Academic Surgery
Body mass index predicts operative time in elective colorectal procedures Harish Saiganesh, BS, David E. Stein, MD, and Juan L. Poggio, MD* Division of Colorectal Surgery, Department of Surgery, Drexel University College of Medicine, Philadelphia, Pennsylvania
article info
abstract
Article history:
Background: Obesity currently affects more than a third of the United States population and
Received 28 December 2014
is associated with increased surgical complications. Compared to all other subspecialties,
Received in revised form
colorectal surgery is the most affected by the increasing trend in obese surgical patients.
5 February 2015
Operative time has been found to have the greatest impact on hospital costs and physician
Accepted 27 February 2015
workload. This study was conducted to determine whether obesity has a direct impact on
Available online 6 March 2015
operative time in elective colorectal procedures using a high-powered, nationally representative patient sample.
Keywords:
Methods: A retrospective analysis was conducted on 45,362 patients who underwent open
Body mass index
and laparoscopic ileocolic resections, partial colectomies, and low pelvic anastomoses using
Obesity
American College of Surgeons National Surgical Quality Improvement Program data from
Operative time
2005e2009. Operative time was the main outcome variable, whereas body mass index (BMI)
Colorectal surgery
was the main independent variable. BMI was divided into three classes as follows: normal
American College of Surgeons
(35). A univariate linear model
National Surgical Quality
was used to analyze the relationship while controlling for confounding factors such as de-
Improvement Program
mographics and preoperative conditions. Statistical significance was established at P 0.05. Results: Morbidly obese patients were found to have longer operative times than did normal patients across each individual colorectal procedure (P < 0.001), ranging from a mean difference of 17.8 min for open ileocolic resections to 56.6 min for laparoscopic low pelvic anastomoses with colostomies. Conclusions: BMI, as an objective measure of obesity, is a direct, statistically significant independent predictor of operative time across elective colorectal procedures. ª 2015 Elsevier Inc. All rights reserved.
1.
Introduction
Obesity has been cited as one of the major health issues associated with industrialized counties such as the United States, where a third of the population is affected and one in twenty is classified as morbidly obese [1]. Obese individuals have a high
body mass index (BMI), a measurement that is partly determined by the patients’ total body fat [2]. Consequently, operating on these obese patients can be a challenge for surgeons during open or laparoscopic colorectal procedures. Various studies have established the association of obesity with higher rates of surgical site infections in colorectal procedures [1,3].
* Corresponding author. Division of Colorectal Surgery, Department of Surgery, Drexel University College of Medicine, 245 N. 15th Street, MS 413, Philadelphia, PA 19102 1192. Tel.: þ1 215 762 1750; fax: þ1 215 762 8389. E-mail address: [email protected] (J.L. Poggio). 0022-4804/$ e see front matter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2015.02.067
46
j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 7 ( 2 0 1 5 ) 4 5 e4 9
Patients undergoing colorectal procedures present a greater degree of postoperative complications when compared with patients undergoing other surgical procedures [4]. The most prevalent of these complications include surgical site infections, which have been associated with an increased postoperative length of stay, greater costs of care, and higher rates of readmission [5]. Operative time has been identified as a significant predictor of surgical site infections as per the Centers for Disease Control and Prevention National Nosocomial Infections Surveillance index [6]. In addition, studies have demonstrated the impact of longer operative times on increased hospital costs and increased surgical workloads [7e9]. Although a few studies have described an association of BMI and body surface area with operative time in specific colorectal procedures and conditions, they did not analyze other colorectal procedures or compare both laparoscopic cases and open cases [10,11]. Furthermore, there have been no prior studies on whether BMI independently predicts operative time in colorectal procedures. We have queried the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database to more accurately analyze the relationship between obesity and operative time across multiple colorectal procedures and to highlight the importance of an increasing trend in obesity on increased time of work for colorectal surgeons.
2.
Methods
2.1.
Data collection
The ACS-NSQIP database was evaluated in a retrospective outcome analysis to select patients who underwent nonemergent colorectal procedures. Using NSQIP Participant Use Files data from the years 2005e2009, 54,084 patients were selected based on Current Procedural Terminology coding criteria. Surgical procedures of interest included the following three most commonly performed procedures (open and laparoscopic procedures are listed, respectively, in parentheses): ileocolic resections (44,160; 44,205), partial
colectomies (44,140; 44,204), and low pelvic anastomoses (44,145/44,146; 44,207/44,208). BMI in kilogram per square meter was calculated using height and weight data. A new variable was created to divide BMI into three classes as follows: normal (35). Operative time was defined in minutes beginning with the initial skin incision and ending with application of postoperative dressing and served as the primary outcome variable, whereas BMI was the primary predictor variable. Results were reported as mean standard error. Exclusion criteria included both missing data (534 patients) and presence of outliers (8188 patients) in potential predictor variables as well as in the outcome variable, leading to a final sample of 45,362 patients for data analyses. Of those patients, when classified by BMI, 15,639 patients were normal, 24,146 patients were overweight and/or obese, and 5577 patients were morbidly obese. When classified by surgical procedure, there were 10,643 patients who had undergone ileocolic resections (6458 for open and 4185 for laparoscopic), 23,583 patients who had undergone partial colectomies (14,155 for open and 9428 for laparoscopic), and 11,136 patients who had undergone low pelvic anastomoses (6702 for open and 4434 for laparoscopic).
2.2.
Statistical analysis
The effect of potential predictor variables on operative time was determined via a univariate analysis of variance. The variables used in the analysis are identified (Table 1). Of these variables, BMI, surgical procedures, age, gender, race, diabetes mellitus, dyspnea, financial status before illness, financial status before surgery, ventilator dependence, ascites, esophageal varices, congestive heart failure, previous angioplasty, previous cardiac surgery, angina, rest pain, renal failure, history of transient ischemic attacks, disseminated cancer, wound infection, chronic steroid use, weight loss, chemotherapy, radiotherapy, pregnancy, and systemic sepsis were identified as statistically significant predictor variables of operative time. Discrete predictor variables significant in the univariate analysis were converted to continuous ones to be used in
Table 1 e List of potential predictor variables of operative time. BMI Dyspnea Congestive heart failure Acute renal failure Chronic steroid use Ascites
Surgical procedures
Age
Financial status before illness History of myocardial infarction Dialysis
Financial status prior to surgery Previous angioplasty
Ventilator dependence Previous cardiac surgery
COPD history
Pneumonia
Angina
Impaired sensorium Bleeding disorders
Coma Blood transfusions before surgery
Weight loss
Revascularization history
Gender
Race
Diabetes mellitus
Alcohol use
History of stroke CNS tumor
Hypertension
Esophageal varices Rest Pain
Hemiplegia
Paraplegia
Quadriplegia
Chemotherapy
Radiotherapy
Pregnancies
Wound infection Operations a month before surgery Systemic sepsis
History of transient ischemic Attacks
CNS ¼ central nervous system; COPD ¼ chronic obstructive pulmonary disease.
Cigarette smoking
Disseminated cancer
j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 7 ( 2 0 1 5 ) 4 5 e4 9
dimension reduction analysis. Age, gender, and race were reduced into a factor called demographics, financial statuses before illness and surgery were reduced into a factor called combined financial status, and all other significant predictor variables except surgical procedures and BMI were reduced into a factor called preoperative conditions. Dimension reduction was performed to account for a greater number of significant predictor variables in further analyses of variance, which places limits on the number of individual variables used as covariates. The use of more variables accounts for greater variance in operative time that can be adjusted to solely measure BMI’s effects on operative time [12]. An analysis of covariance (ANCOVA) was performed to determine the relationship between BMI and operative time adjusted for the covariates demographics, surgical procedures, and preoperative conditions. Assumptions for ANCOVA such as homogeneity of variance, normality of residuals, error term independence, and homogeneity and linearity of regression slopes were tested and met for all but one variable. Combined financial status failed the homogeneity of regression assumption and thus was not included in the final ANCOVA. An ANCOVA was also performed for each surgical procedure to further measure differences in operative time means by BMI. All analyses were performed using type 1 sum of squares to account for unequal sample sizes in each of the variable groups. A linear regression was also conducted with the same results to verify the ANCOVA results. A post hoc test for multiple comparisons was conducted and adjusted via the Bonferroni correction. Statistical significance was established at P 0.05. Statistical analyses were performed using SPSS software (version 22, Chicago, IL).
3.
Results
After a univariate analysis, BMI was found to have a statistically significant association with operative time (P < 0.001). An ANCOVA showed that BMI remained a statistically significant predictor of operative time (P < 0.001), while adjusting for variances due to surgical procedures, demographics, and preoperative conditions. As per the post hoc multiple comparisons test, differences in operative time means were statistically significant between all BMI classes (P < 0.001). The linear regression also identified that the unstandardized coefficient (B) of the BMI classes was 13.540, indicating that for every increment in BMI classification (from normal to overweight and/or obese to morbidly obese), operative time increased by 13.5 min on average (P < 0.001). The adjusted mean operative times by different BMI classes were reported across all the surgical procedures (Table 2) and across each individual surgical procedure (Table 3). Across the procedures, there was a mean difference of 27.8 min between normal weight and morbidly obese patients. For open and laparoscopic procedures, respectively, there was a mean difference of 17.8 and 22.6 min between normal weight and morbidly obese patients who underwent ileocolic resections; 25.4 and 26.9 min for those who underwent partial colectomies; 34.3 and 49.9 min for those who underwent low pelvic anastomoses without colostomies; and 26.8 and 56.6 min for those who underwent low pelvic anastomoses
47
Table 2 e Mean operative times (in minutes) versus BMI (in kg/m2) across all colorectal procedures, adjusted for effects from covariates demographics and preoperative conditions. BMI
Mean operative time
Normal weight (35)
162.4 0.7 (P < 0.001) 174.9 0.5 (P < 0.001) 190.2 1.1 (P < 0.001)
with colostomies. The greatest mean difference in operative time because of BMI exists among patients who underwent laparoscopic low pelvic anastomoses with colostomy (56.6 min), whereas the smallest mean difference exists among patients who underwent open ileocolic resection (17.8 min). On average, laparoscopic colorectal procedures had longer operative times than open ones. Ileocolic resections had the shortest operative times, followed by partial colectomies and low pelvic anastomoses without and with colostomies.
4.
Discussion
Using data from ACS-NSQIP, our study determined that BMI has a direct, statistically significant independent relationship with operative time in elective colorectal procedures. Unlike previous studies on BMI and operative time association in colorectal surgery, our study included a nationwide sample of 45,362 patients across procedures, making it a more powerful measure of the degree of association. Moreover, our study measured the association across eight different colorectal surgical procedures and used data from at least 400 to over 16,000 patients per procedure, based on the relative frequency of the individual procedure each year. With the skyrocketing incidence of obesity, it becomes necessary to identify the resources required to optimize surgical treatment in this population. This trend may have a strong impact on surgeon workload. Longer operative times have been demonstrated for obese patients undergoing general surgical procedures requiring increased work by the surgeon [13]. A US census-based population study determined that workload is expected to increase about 31.5% for general surgeons by 2020, with a 40.3% increase for gastrointestinal surgeons [14]. Furthermore, there is a projected increase of 42% in colon resections by 2020, with colon resections accounting for potentially 13.7% of the total surgical workload in 2020 [14]. Although significant, such increases were reported primarily with the knowledge of an increasing elderly population in the future and did not account for increased work required to treat obese patients [14]. There has also been evidence of decreasing reimbursement for surgeons across all fields, particularly for those performing colectomies [15]. Surgical reimbursement is currently provided by Medicare, Medicaid, and commercial insurers. Reimbursement is dependent on relative value units, which consider factors such as physician workload, overhead, and malpractice and vary for different surgical procedures [14]. Thus, future colorectal surgeons will be facing greater
48
Normal (35)
Laparoscopic Open Laparoscopic Open Laparoscopic Open Laparoscopic Open
Low pelvic anastomosis with colostomy Low pelvic anastomosis without colostomy Partial colectomy Ileocolic resection BMI
Table 3 e Mean operative times (in minutes) versus BMI (in kg/m2) by individual colorectal procedure, adjusted for effects from covariates demographics and preoperative conditions.
j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 7 ( 2 0 1 5 ) 4 5 e4 9
workloads while receiving less. With the increased time commitment required to conduct surgery for obese patients also comes an increased workload for supporting personnel such as anesthesiologists and nurses. In addition, there are reports that we are facing a national shortage of anesthesia providers. It is possible that without adjusting reimbursement for the care of these obese patients, providers may simply refuse to care for them [13,16]. In a study conducted on minority patients who underwent various inpatient surgeries, BMI has been associated with increased hospital costs due to surgical difficulties and consequent longer use of the operating room [17]. Moreover, operating duration has been associated with about 40% of total hospital costs, with an average cost of $30 per minute in colorectal surgical procedures [9,18]. Based on our results, it would thus cost an additional $834 on average to operate on a morbidly obese patient compared with a normal one for a general elective colorectal procedure based on duration alone. When considering individual procedures, it would cost an additional $600 on average for ileocolic resections; $780 on average for partial colectomies; and $1260 on average for both types of low pelvic anastomoses. With increasing procedure-based workloads, increasing timebased costs, and decreasing Medicare reimbursement for colorectal surgeons operating on an increasing obese population in the United States, we suggest that the relative value unit system of surgical reimbursement be restructured to include BMI as a predictive measure of operative costs. Furthermore, changes in demographics and other preoperative conditions in the surgical population should also be considered when restructuring reimbursement.
4.1.
Limitations
One limitation in this study is that the variables recorded in the NSQIP database may not capture other factors that can have an impact on operative time such as adhesions and surgeon experience [13,19]. Another limitation is that data on operating room cost per minute were not found for each individual colorectal surgical procedure in this study but rather for colorectal surgery in general, hence the use of the latter in our discussion [17]. In addition, the costs per minute may be higher now because of inflation. Nevertheless, in spite of these limitations, this study is supported by the use of a large-scale, nationally representative sample of elective colorectal surgery patients to determine the impact of obesity on operative time and consequent health-care costs.
5.
Conclusions
BMI, which has been documented as an objective measure of obesity, is a statistically significant, independent predictor of operative time in both open and laparoscopic elective colorectal procedures. We suggest future studies to further investigate the creation of a modifier to accurately reflect the increased time of work required to care for morbidly obese patients.
j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 7 ( 2 0 1 5 ) 4 5 e4 9
Acknowledgment Authors’ contributions: J.L.P. and H.S. contributed to the study conception and design, acquisition of data, and analysis and interpretation of data. J.L.P., H.S., and D.E.S. did the drafting of the article and the critical revision.
Disclosure All authors deny any disclosure of potential conflicts of interest, including financial interests, activities, relationships, and affiliations. The American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) and the hospitals participating in the ACS-NSQIP are the source of the data used herein; they have not verified and are not responsible for either the statistical validity of the data analysis or the conclusions derived by the authors.
references
[1] Haas EM, Aminian A, Nieto J, et al. Minimally invasive colorectal surgery in the morbidly obese: does high body mass index lead to poorer outcomes? Surg Curr Res 2013;3:149. [2] Chern H, Chou J, Donkor C, et al. Effects of obesity in rectal cancer surgery. J Am Coll Surg 2010;211:55. [3] Fujii T, Tsutsumi S, Matsumoto A, et al. Thickness of subcutaneous fat as a strong risk factor for wound infections in elective colorectal surgery: impact of prediction using preoperative CT. Dig Surg 2010;27:331. [4] Young H, Knepper B, Moore EE, et al. Surgical site infection after colon surgery: national Healthcare Safety Network risk factors and modeled rates compared with published risk factors and rates. J Am Coll Surg 2012;214:852. [5] Jones RS, Brown C, Opelka F. Surgeon compensation: “Pay for performance,” the American College of Surgeons National Surgical Quality Improvement Program, the Surgical Care Improvement Program, and other considerations. Surgery 2005;138:829.
49
[6] Mangran AJ, Horan TC, Pearlson ML, Silver LC, Jarvis WR. Guideline for prevention of surgical site infection. Infect Control Hosp Epidemiol 1999;20:247. [7] Dexter F, Blake JT, Penning DH, Sloan B, Chung P, Lubarsky DA. Use of linear programming to estimate impact of changes in a hospital’s operating room time allocation on perioperative variable costs. Anesthesiology 2002;96:718. [8] Beck DE. Reoperative surgery: what can we learn from a large randomized prospective trial? Clin Colon Rectal Surg 2006;19:247. [9] Little DC, St Peter SD, Calkins CM, et al. Relative value units correlate with pediatric surgeons’ operating time; when perceived myth becomes reality. J Pediatr Surg 2006;41:234. [10] Krane MK, Allaix ME, Zoccali M, et al. Does morbid obesity change outcomes after laparoscopic surgery for inflammatory bowel disease? Review of 626 consecutive cases. J Am Coll Surg 2013;216:986. [11] Vaccaro CA, Vaccarezza H, Rossi GL, et al. Body surface area: a new predictor factor for conversion and prolonged operative time in laparoscopic colorectal surgery. Dis Colon Rectum 2012;55:1153. [12] Field AP. Discovering statistics using SPSS. 2nd ed. London: Sage; 2005. [13] Hawn MT, Bian J, Leeth RR, et al. Impact of obesity on resource utilization for general surgical procedures. Ann Surg 2005;241:821. [14] Liu JH, Etzioni DA, O’Connell JB, Maggard MA, Ko CY. The increasing workload of general surgery. Arch Surg 2004; 139:423. [15] Russell T. Statement of the American College of Surgeons to the subcommittee on health committee on energy and commerce. US House of Representatives; 2002. [16] Spetz J, Given R. The future of the nurse shortage: will wage increases close the gap? Health Aff 2003;22:199. [17] Nafiu OO, Shanks AM, Hayanaga AJ, Tremper KK, Campbell DA Jr. The impact of body mass index on postoperative complications and resource utilization in minority patients. J Natl Med Assoc 2011;103:9. [18] Marrin CAS, Johnson LC, Beggs VL, Batalden PB. Clinical process cost analysis. Ann Thorac Surg 1997;64:690. [19] Derossis AM, Fried GM, Abrahamowicz M, Sigman HH, Barkun JS, Meakins JL. Development of a model for training and evaluation of laparoscopic skills. Am J Surg 1998;175:482.
