- Email: [email protected]
Operative time as a marker of quality in bariatric surgery
Surgery for Obesity and Related Diseases 15 (2019) 1113–1120
Original article
Operative time as a marker of quality in bariatric surgery Colette S. Inaba, M.D.a, Christina Y. Koh, M.D.a, Sarath Sujatha-Bhaskar, M.D.a, Shea Gallagher, B.S.a, Yanjun Chen, M.S.b, Ninh T. Nguyen, M.D. F.A.C.S.a,* b
a Department of Surgery, University of California Irvine Medical Center, Orange, California Institute for Clinical and Translational Science, University of California Irvine, Irvine, California
Received 1 January 2019; accepted 6 April 2019
Abstract
Background: Few studies have examined the effect of prolonged operative time (OT) on outcomes in laparoscopic bariatric surgery. Existing studies mostly focus on 30-day complications, whereas serious complications may not occur until well after 30 days from the index operation. Objective: To determine the effect of prolonged OT on 1-year morbidity and mortality after laparoscopic Roux-en-Y gastric bypass (LRYGB) and laparoscopic sleeve gastrectomy (LSG). Setting: The Bariatric Outcomes Longitudinal Database (BOLD). Methods: Data on primary LRYGB and LSG cases performed between 2008 and 2012 in the BOLD were analyzed. Converted cases and cases concurrent with other procedures were excluded. Multivariate logistic regression was used to assess the association between OT and 1-year morbidity and mortality, with adjustment for preoperative demographic and clinical characteristics. Results: A total of 93,051 cases were examined, including 74,745 (80.3%) LRYGB and 18,306 (19.7%) LSG cases. For LRYGB, mean OT was 104 minutes (standard deviation [SD] 46.6). Every additional 10 minutes of OT was associated with increased odds of 1-year mortality (adjusted odds ratio [AOR] 1.04; P 5 .02), leak (AOR 1.07; P , .0001), and any adverse event (AOR 1.03; P , .001). For LSG, mean OT was 78 minutes (SD 37.4). Every additional 10 minutes of OT was associated with increased odds of 1-year leak (AOR 1.07; P 5 .0002). Data on patients lost to follow-up was unavailable. Conclusion: Prolonged operative time is associated with a significant increase in the odds of mortality and serious complications after laparoscopic bariatric surgery. Operative time may be a useful marker of quality in primary laparoscopic bariatric surgery. (Surg Obes Relat Dis 2019;15:1113–1120.) Ó 2019 American Society for Bariatric Surgery. Published by Elsevier Inc. All rights reserved.
Key words:
Bariatric surgery; Gastric bypass; Sleeve gastrectomy; Operative time; Mortality; Leak; Outcomes; Laparoscopic
Prolonged operative time (OT) has been associated with worse outcomes in a variety of different types of operations, including general surgery, vascular, and colorectal
This research was presented October 2017 at the American College of Surgeons Clinical Congress in San Diego, CA. * Correspondence: Ninh T. Nguyen, M.D., F.A.C.S., Department of Surgery, University of California Irvine Medical Center, 333 City Blvd West, Suite 1600, Orange, CA 92868. E-mail address: [email protected] (N.T. Nguyen).
procedures [1–4]. Several studies examining the effect of OT on bariatric surgery outcomes have also found an association between prolonged OT and increased complications [5–10]. Most of these studies are limited to analysis of 30-day outcomes, whereas a high percentage of serious complications such as pulmonary emboli (PE) and mortality may occur later than 30 days after surgery [11–13]. The objective of our study was to evaluate the effect of prolonged OT on 1-year morbidity and mortality after laparoscopic bariatric surgery.
https://doi.org/10.1016/j.soard.2019.04.010 1550-7289/Ó 2019 American Society for Bariatric Surgery. Published by Elsevier Inc. All rights reserved.
1114
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121
Methods Data source The Bariatric Outcomes Longitudinal Database (BOLD) was administered by the American Society of Metabolic and Bariatric Surgery and contains Health Insurance Portability and Accountability Act-compliant patient data prospectively collected from over 700 sites nationwide participating in the American Society of Metabolic and Bariatric Surgery Bariatric Surgery Centers of Excellence program. The strength of the BOLD database is the reporting of outcomes out to 1 year from the index bariatric operation. This study was exempt from institutional review board approval, as the BOLD contains only deidentified patient data and any specific patient intervention or interaction was not possible. Data analysis Data from all patients who underwent primary laparoscopic Roux-en-Y gastric bypass (LRYGB) or laparoscopic sleeve gastrectomy (LSG) between 2008 and 2012 were examined. Data were analyzed separately based on procedural type, either LRYGB or LSG. Cases that were converted or that included other concurrent procedures were
excluded. The primary outcomes of interest were 1-year mortality, leak, and PE. A secondary outcome of interest was 1-year any adverse event, a composite variable including any of 128 adverse events recorded in the database. Data on patients lost to follow-up was unavailable. Statistical analysis Statistical analysis was performed using SAS version 9.4 (SAS Institute, Cary, NC). Patient demographic characteristics, clinical and operative characteristics, and outcomes were examined by comparing the shortest, average, and longest operative-time terciles. Multivariate logistic regression was then used to determine adjusted odds ratios (AOR) for the association between longer OT and the outcomes of interest in a continuous manner (i.e., every 10 min). Adjustments were made for patient age, gender, race, preoperative body mass index (BMI), American Society of Anesthesiologists class, all preoperative comorbidities, and operative blood loss. Patients with missing data were excluded from the regression analysis. Two-sided tests with P , .05 were considered statistically significant. Further adjustment was made to account for multiple comparisons using the false discovery rate for P values.
Table 1 Demographic and clinical characteristics of patients undergoing laparoscopic Roux-en-Y gastric bypass Variable
Overall N 5 74,745
Shortest OT Tercile
Middle OT Tercile
Longest OT Tercile
P Value*
Age, yr, mean (SD) Female Race Caucasian Hispanic Black Other BMI, kg/m2, mean (SD) ASA Class I II III IV Comorbidities Congestive heart failure Depression Diabetes Hypertension Ischemic heart disease Liver disease Sleep apnea Peripheral vascular disease Venous thromboembolism Alcohol use Tobacco use Substance abuse Functional status—any impairment
44.7 (11.7) 57,944 (77.5)
44.5 (11.6) 19,898 (80.2)
44.8 (11.7) 19,327 (77.6)
44.9 (11.6) 18,623 (74.8)
.0003 , .0001
52,079 (69.7) 7,553 (10.1) 8,584 (11.5) 6,529 (8.7) 47.6 (8.1)
16,464 (66.3) 3,485 (14.0) 2,651 (10.7) 2,221 (9.0) 47.0 (7.7)
17,712 (71.1) 2,121 (8.5) 2,964 (11.9) 2,113 (8.5) 47.7 (8.1)
17,809 (71.6) 1,933 (7.8) 2,956 (11.9) 2,185 (8.8) 47.9 (8.4)
, .0001
2,562 (3.4) 16,450 (22.0) 52,428 (70.1) 3,305 (4.4)
1,127 (4.5) 4,413 (17.8) 18,172 (73.2) 1,109 (4.5)
659 (2.7) 5,652 (22.7) 17,385 (69.8) 1,214 (4.9)
746 (3.0) 6,365 (25.6) 16,792 (67.5) 980 (3.9)
, .0001
1,484 (2.0) 25,857 (34.8) 24,062 (32.4) 40,075 (54.0) 1,893 (2.6) 3,772 (5.1) 31,372 (42.3) 811 (1.1) 1,295 (1.7) 9,307 (12.5) 3,572 (4.8) 248 (.33) 2,143 (2.9)
493 (2.0) 8,134 (32.8) 7,827 (31.5) 13,341 (53.8) 641 (2.6) 1,159 (4.7) 9,719 (39.2) 260 (1.10) 438 (1.8) 2,909 (11.7) 1,242 (5.0) 58 (.3) 682 (2.8)
512 (2.1) 8,842 (35.5) 7,977 (32.0) 13,438 (54.0) 640 (2.6) 1,075 (4.3) 10,473 (42.0) 264 (1.10) 442 (1.8) 3,191 (12.8) 1,164 (4.7) 85 (.3) 730 (2.9)
476 (1.9) 8,840 (35.5) 8,222 (33.0) 13,226 (53.2) 609 (2.5) 1,538 (6.2) 11,139 (44.8) 287 (1.20) 414 (1.7) 3,191 (12.8) 1,154 (4.6) 94 (.4) 730 (2.9)
.4887 , .0001 .0009 .0876 .5628 , .0001 , .0001 .4634 .5633 .0001 .1287 .1204 .3265
ASA 5 American Society of Anesthesiologists; BMI 5 body mass index; OT 5 operative time; SD 5 standard deviation. Values are presented as frequencies (%) unless otherwise indicated. * P value is for comparison of all 3 terciles.
, .0001
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121
1115
Table 2 Demographic and clinical characteristics of patients undergoing laparoscopic sleeve gastrectomy Variable
Overall N 5 18,306
Shortest OT Tercile
Middle OT Tercile
Longest OT Tercile
P Value*
Age, yr, mean (SD) Female Race Caucasian Hispanic black Other BMI, kg/m2, mean (SD) ASA Class I II III IV Comorbidities Congestive heart failure Depression Diabetes Hypertension Ischemic heart disease Liver disease Sleep apnea Peripheral vascular disease Venous thromboembolism Alcohol use Tobacco use Substance abuse Functional status—any impairment
43.6 (11.3) 13,757 (75.2)
43.1 (11.2) 4,998 (80.8)
43.3 (11.3) 4,424 (74.5)
44.3 (11.3) 4,330 (70.2)
, .0001 , .0001
12,173 (66.5) 2,126 (11.6) 2,223 (12.1) 1,784 (9.8) 46.7 (8.6)
4,215 (68.1) 730 (11.8) 687 (11.1) 557 (9.0) 45.2 (7.7)
3,860 (65.0) 706 (11.9) 775 (13.1) 598 (10.1) 46.7 (8.5)
4,094 (66.3) 690 (11.2) 760 (12.3) 629 (10.2) 48.1 (9.5)
.0017
688 (3.8) 5,365 (29.3) 11,716 (64.0) 537 (2.9)
171 (2.8) 1,750 (28.3) 4,131 (66.8) 137 (2.2)
158 (2.7) 1,766 (29.7) 3,826 (64.4) 189 (3.2)
356 (5.8) 1,849 (30.0) 3,758 (60.9) 210 (3.4)
, .0001
265 (1.5) 5,448 (30.0) 4,879 (26.9) 8,973 (49.4) 337 (1.9) 772 (4.3) 7,057 (38.9) 145 (.8) 216 (1.2) 3,303 (18.2) 949 (5.2) 52 (.29) 338 (1.9)
68 (1.1) 1,846 (29.8) 1,578 (25.5) 2,934 (47.4) 85 (1.4) 160 (2.6) 2,139 (34.6) 46 (.7) 52 (.8) 1,195 (19.3) 288 (4.7) 11 (.2) 65 (1.1)
92 (1.6) 1,825 (30.7) 1,577 (26.6) 2,960 (49.8) 118 (2.0) 250 (4.2) 2,340 (39.4) 51 (.9) 82 (1.4) 1,116 (18.8) 338 (5.7) 18 (.3) 87 (1.5)
105 (1.7) 31 (28.8) 1,723 (27.9) 3,075 (49.8) 133 (2.2) 362 (5.9) 2,578 (41.8) 48 (.8) 82 (1.3) 989 (16.0) 323 (5.2) 23 (.4) 186 (3.0)
.0149 .0545 .0116 .0108 .0032 , .0001 , .0001 .7659 .0096 , .0001 .0370 .1206 , .0001
, .0001
ASA 5 American Society of Anesthesiologists; BMI 5 body mass index; OT 5 operative time; SD 5 standard deviation. Values are presented as frequencies (%) unless otherwise indicated. * P value is for comparison of all 3 terciles.
Results Patient demographic and clinical characteristics We examined 74,745 LRYGB cases. Patient demographic and clinical characteristics are listed in Table 1. Univariate comparisons among the shortest, middle, and longest OT terciles showed that longer OT was associated with older patient mean age (44.5 versus 44.8 versus 44.9 yr, respectively; P 5 .0003); higher proportion of male gender (19.8% versus 22.4% versus 25.2%, respectively; P , .0001); higher mean BMI (47.0 versus 47.7 versus 47.9
kg/m2, respectively; P , .0001); and higher rates of depression (32.8% versus 35.5% versus 35.5%, respectively; P , .0001), diabetes (31.5% versus 32.0% versus 33.0%; P 5 .0009), sleep apnea (39.2% versus 42.0% versus 44.8%, respectively; P , .0001), liver disease (4.7% versus 4.3% versus 6.2%, respectively; P , .0001), and alcohol use (11.7% versus 12.8% versus 12.8%, respectively; P 5 .0001). We examined 18,306 LSG cases. Patient demographic and clinical characteristics for LSG are listed in Table 2. Univariate comparisons among the shortest, middle, and
Table 3 Operative characteristics and outcomes for patients undergoing laparoscopic Roux-en-Y gastric bypass Variable
Overall N 5 74,745
Shortest OT Tercile
Middle OT Tercile
Longest OT Tercile
P Value*
Operative time, min, mean (SD) Estimated blood loss, mL, mean (SD) 1-yr mortality 1-yr leak 1-yr pulmonary embolus 1-yr any adverse event
104.4 (46.6) 42.8 (72.3) 151 (.20) 325 (.43) 137 (.18) 2,078 (2.8)
61.3 (10.8) 36.4 (81.6) 32 (.13) 69 (.28) 36 (.15) 557 (2.24)
95.9 (10.5) 40.2 (54.7) 57 (.23) 94 (.38) 52 (.21) 679 (2.73)
155.8 (41.3) 51.7 (76.8) 62 (.25) 161 (.65) 48 (.19) 839 (3.37)
, .0001 , .0001 .0061 , .0001 .2225 , .0001
OT 5 operative time; SD 5 standard deviation. Values are presented as frequencies (%) unless otherwise indicated. * P value is or comparison of all 3 terciles.
1116
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121 Table 4 Operative characteristics and outcomes for patients undergoing laparoscopic sleeve gastrectomy Variable
Overall N 5 18,306
Shortest OT Tercile
Middle OT Tercile
Slowest OT Tercile
P Value*
Operative time, min, mean (SD) Estimated blood loss, mL, mean (SD) 1-yr mortality 1-yr leak 1-yr pulmonary embolus 1-yr any adverse event
78.1 (37.4) 32.2 (58.3) 21 (.11) 50 (.27) 26 (.14) 244 (1.3)
46.5 (8.4) 23.7 (28.4) 4 (.06) 8 (.13) 5 (.08) 55 (.89)
70.1 (7.2) 31.3 (44.3) 6 (.10) 19 (.32) 13 (.22) 89 (1.50)
117.4 (37.7) 41.5 (85.0) 11 (.18) 23 (.37) 8 (.13) 99 (1.60)
, .0001 , .0001 .1635 .0245 .1239 .0009
OT 5 operative time; SD 5 standard deviation. Values are presented as frequencies (%) unless otherwise indicated. * P value is comparison for all 3 terciles.
longest OT terciles revealed that longer OT was associated with higher mean age (43.1 versus 43.3 versus 44.3 yr, respectively; P , .0001); higher proportion of male gender (19.2% versus 25.5% versus 29.8%, respectively; P , .0001); higher mean BMI (45.2 versus 46.7 versus 48.1 kg/m2, respectively; P , .0001); and higher rates of congestive heart failure (1.1% versus 1.6% versus 1.7%, respectively; P 5 .0149), diabetes (25.5% versus 26.6% versus 27.9%, respectively; P 5 .0116), hypertension (47.4% versus 49.8% versus 49.8%; P 5 .0108), ischemic heart disease (2.6% versus 4.2% versus 5.9%, respectively; P , .0001), sleep apnea (34.6% versus 39.4% versus 41.8%, respectively; P , .0001), and impaired functional status (i.e., being bedridden or needing an assistance device to mobilize; 1.1% versus 1.5% versus 3.0%, respectively; P , .0001) but lower rates of alcohol use (19.3% versus 18.8% versus 16%, respectively; P , .0001).
operative blood loss (36.4 versus 40.2 versus 51.7 mL, respectively; P , .0001) and higher 1-year mortality rate (.13% versus .23% versus .25%, respectively; P 5 .0061), leak rate (.28% versus .38% versus .65%, respectively; P , .0001), and any-adverse-event rate (2.24% versus 2.73% versus 3.37%, respectively; P , .0001). Table 4 lists the operative characteristics and outcomes for LSG. The overall mean OT for LSG was 78.1 (SD 37.4) minutes. The mean OT of the shortest, middle, and longest terciles was 46.5 (8.4) minutes, 70.1 (SD 7.2) minutes, and 117.4 (SD 37.7) minutes, respectively. Univariate comparisons among the shortest, middle, and longest operative-time terciles revealed that longer OT was associated with higher mean operative blood loss (23.7 versus 31.3 versus 41.5 mL, respectively; P , .0001) and higher 1-year leak rate (.13% versus .32% versus .37%, respectively; P 5 .0245) and any-adverse-event rate (.89% versus 1.50% versus 1.60%, respectively; P 5 .0009).
Operative characteristics and outcomes Table 3 lists the operative characteristics and outcomes for LRYGB. The overall mean OT for LRYGB was 104.4 (standard deviation [SD] 46.6) minutes. The mean OT of the shortest, middle, and longest terciles was 61.3 (SD 10.8) minutes, 95.9 (SD 10.5) minutes, and 155.8 (SD 41.3) minutes, respectively. Univariate comparisons among the shortest, middle, and longest operative-time terciles revealed that longer OT was associated with higher mean
Multivariate logistic regression analysis Multivariate logistic regression analysis was used to determine the association between prolonged OT and 1-year postoperative outcomes after adjustment for patient demographic characteristics and comorbidities. The results for multivariate logistic regression for association between prolonged operative time and outcome are listed in Table 5.
Table 5 Multivariate logistic regression analysis for association between every 10 minutes of prolonged operative time and outcomes after laparoscopic bariatric surgery Outcome
LRYGB
LSG
AOR (95% CI)
P Value*
AOR (95% CI)
P Value*
Mortality Leak Pulmonary embolism Any adverse event
1.04 (1.01, 1.07) 1.07 (1.06, 1.09) 1.01 (.98, 1.05) 1.03 (1.02, 1.04)
.0195 ,.0001 .5528 ,.0001
1.06 (1.00, 1.12) 1.07 (1.03, 1.10) 1.05 (.99, 1.11) 1.03 (1.00, 1.05)
.0828 .0002 .1291 .0652
AOR 5 adjusted odds ratio; CI 5 confidence interval; LRYGB 5 laparoscopic Roux-en-Y gastric bypass; LSG 5 laparoscopic sleeve gastrectomy. Model is adjusted for age, BMI, blood loss, gender, race, ASA, and all preoperative comorbidities and characteristics. * P value is adjusted for false discovery rate.
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121
1117
Figure 1. Unadjusted outcome as a function of operative time for A) laparoscopic Roux-en-Y gastric bypass and B) laparoscopic sleeve gastrectomy.
1118
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121
For LRYGB, every additional 10 minutes of OT was associated with a statistically significant increase in the odds of 1-year mortality (AOR 1.04; P 5 .0195), leak (AOR 1.07; P , .0001), and any adverse event (AOR 1.03; P , .0001). The relationships between OT and unadjusted outcomes after LRYGB are demonstrated in Fig. 1A. For LSG, every additional 10 minutes of OT was associated with a statistically significant increase in the odds of 1-year leak (AOR 1.07; P 5 .0002). There was also a trend toward an increase in the odds of 1-year mortality (AOR 1.06; P 5 .0828) and any adverse event (AOR 1.03; P 5 .0652), but neither of these were statistically significant at the 2-sided P , .05 level. The relationships between OT and unadjusted outcomes after LSG are demonstrated in Fig. 1B. Discussion Using a national database to evaluate the effect of prolonged OT on 1-year outcomes after laparoscopic bariatric surgery, we found an association between prolonged OT and higher odds of 1-year mortality, leak, and any adverse event after LRYGB and higher odds of 1-year leak after LSG. These findings suggest that OT may be used as a surrogate measurement for quality in bariatric surgery and might be useful as part of the designation criteria for Bariatric Centers of Excellence. Several studies have reported an association between prolonged OT in laparoscopic bariatric surgery and increased complications, particularly increased risk of 30-day complications [5–8], 90-day venous thromboembolism [14], and postoperative sepsis [9]. Our study also found an association between prolonged OT and higher odds of mortality after LRYGB, as well as higher odds of leak after both LRYGB and LSG. Our findings likely reflect the large number of patients required to adequately power an analysis that can detect the effect of OT for events as rare as leak or mortality, with 1-year leak rates ,.45% and 1-year mortality rates ,.20% in this study. Our study did not find a significant effect of prolonged OT on rates of PE. This is contrary to the findings of a study by Chan et al. [14] that examined 500 laparoscopic bariatric procedures and found an increased relative risk for PE of .0002 per additional minute of OT (P 5 .009). In another study evaluating PE and deep vein thrombosis as a composite measure after 16,344 LRYGB cases, the authors found the rates to be significantly higher in cases performed by surgeons who operated in the longest versus shortest OT tercile (.32% versus .22%, respectively; P , .001) [5]. There is limited literature specifically evaluating the effect of OT on LSG outcomes. In a recently published institutional review of 233 patients who underwent LSG, the authors found that longer OT was significantly associated with increased risk of overall complications (AOR 1.01 per additional minute; P 5 .029) [8]. While we found that
prolonged OT was associated with an increase in odds of leak, there was a nonsignificant trend toward an increase in odds of overall complications (AOR 1.03; P 5 .0652). Given our findings that prolonged OT is correlated with increased patient morbidity and mortality, OT may be used as a marker for quality of care in bariatric surgery. Not only is OT correlated with patient outcomes, it is also correlated with perioperative factors that may themselves affect patient outcomes in bariatric surgery, such as the surgeon’s technical skill, case complexity, resident involvement, and dynamics of the surgical team [5,6]. In a study in which bariatric surgeons ranked their peers’ technical skill during LRYGB cases, surgeons whose skill was ranked in the lowest versus highest quartile had longer OT (137 versus 98 minutes, respectively; P , .001), higher overall complication rates (14.5% versus 5.2%, respectively; P , .001), and higher patient mortality (.26% versus .05%, respectively; P 5 .01), suggesting that better technical prowess is correlated with both decreased OT and improved outcomes [10]. This corroborates findings from other studies demonstrating decreased OT and lower rates of complications after overcoming the learning curve for LRYGB [15–17]. Another factor correlated with longer OT is case complexity, such as operating on patients with higher BMI, which is associated with both longer OT and increased complications [14,18]. To reduce bias from unmeasured patient characteristics that may contribute to case complexity, one study compared outcomes based on the median OT of specific bariatric surgeons performing LRYGB, instead of evaluating specific cases [5]. The authors still found that patients of surgeons who operated in the longest versus shortest OT tercile had higher adjusted complication rates (12.9% versus 8.1%, respectively P 5 .022). Resident involvement in bariatric surgery has also been associated with longer OT, but results are mixed regarding whether resident involvement in bariatric surgery is also associated with increased postoperative complications [19–24]. However, even after adjusting for the presence of resident involvement, surgeons with longer OT may have higher rates of overall complications [5,25]. Finally, using the same operative team throughout the day has been demonstrated to decrease OT for bariatric surgery, as well as improve teamwork and the safety climate [26]. Therefore, OT may function as a surrogate marker of various preoperative and intraoperative factors that may impact patient outcomes. As such, OT might be useful as a marker of surgical quality in laparoscopic bariatric surgery, similar to institutional accreditation status and higher case volume, which have both been shown to be associated with better bariatric surgery outcomes [15,27–30]. There are several limitations to this study. First, this was a retrospective database study, and data may be subject to selection bias, missing data, and coding errors. Second,
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121
data on patients lost to follow-up, as well as other long-term outcomes such as incidence of deep vein thrombosis, are not readily available in the database, meaning true complication rates may be underestimated. Finally, the database did not include hospital-level or surgeon-level data nor any data on trainee involvement in cases; thus, we were unable to account for the effect of these factors in our analysis. Despite these limitations, ours is the first study to examine the longterm effect of OT on morbidity and mortality after laparoscopic bariatric surgery. Conclusion Prolonged OT is associated with a significant increase in odds of mortality, leak, and any adverse event after LRYGB and an increase in odds of leak after LSG. Given these associations, OT may be a useful surrogate marker for quality of care in laparoscopic bariatric surgery, similar to institutional accreditation status or case volume. Future studies should also consider adjusting for OT when examining outcomes after bariatric surgery. Acknowledgments This study was partially supported by grant UL1 TR001414 from the National Center for Advancing Translational Sciences, National Institutes of Health (NIH), through the Biostatistics, Epidemiology and Research Design Unit at the University of California Irvine. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Disclosures The American Society for Metabolic and Bariatric Surgery and the surgeons and the hospitals participating in the Centers of Excellence program from 2007 to 2012 are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors. Inaba, Koh, Sujatha-Bhaskar, Gallagher, and Nguyen have no conflicts of interest to disclose. Chen works with a statistical team that receives funding from the National Institutes of Health. References [1] Procter LD, Davenport DL, Bernard AC, Zwischenberger JB. General surgical operative duration is associated with increased risk-adjusted infectious complication rates and length of hospital stay. J Am Coll Surg 2010;210(1):60–5. [2] Daley BJ, Cecil W, Clarke PC, Cofer JB, Guillamondegui OD. How slow is too slow? Correlation of operative time to complications: an analysis from the Tennessee Surgical Quality Collaborative. J Am Coll Surg 2015;220(4):550–8. [3] Scheer A, Martel G, Moloo H, et al. Laparoscopic colon surgery: does operative time matter? Dis Colon Rectum 2009;52(10): 1746–52.
1119
[4] Bailey MB, Davenport DL, Vargas HD, Evers BM, Mckenzie SP. Longer operative time: deterioration of clinical outcomes of laparoscopic colectomy versus open colectomy. Dis Colon Rectum 2014;57(5):616–22. [5] Reames BN, Bacal D, Krell RW, Birkmeyer JD, Birkmeyer NJ, Finks JF. Influence of median surgeon operative duration on adverse outcomes in bariatric surgery. Surg Obes Relat Dis 2015;11(1): 207–13. [6] Jackson TD, Wannares JJ, Lancaster RT, Rattner DW, Hutter MM. Does speed matter? The impact of operative time on outcome in laparoscopic surgery. Surg Endosc 2011;25(7):2288–95. [7] Chen SY, Stem M, Schweitzer MA, Magnuson TH, Lidor AO. Assessment of postdischarge complications after bariatric surgery: A National Surgical Quality Improvement Program analysis. Surgery 2015;158(3):777–86. [8] Major P, Wysocki M, Pedziwiatr M, et al. Risk factors for complications of laparoscopic sleeve gastrectomy and laparoscopic Rouxen-Y gastric bypass. Int J Surg 2017;37:71–8. [9] Blair LJ, Huntington CR, Cox TC, et al. Risk factors for postoperative sepsis in laparoscopic gastric bypass. Surg Endosc 2016;30(4): 1287–93. [10] Birkmeyer JD, Finks JF, O’reilly A, et al. Surgical skill and complication rates after bariatric surgery. N Engl J Med 2013;369(15): 1434–42. [11] Celik F, Bounif F, Fliers JM, et al. The impact of surgical complications as a main risk factor for venous thromboembolism: a multicenter study. Obes Surg 2014;24(10):1603–9. [12] Steele KE, Schweitzer MA, Prokopowicz G, et al. The long-term risk of venous thromboembolism following bariatric surgery. Obes Surg 2011;21(9):1371–6. [13] Omalu BI, Ives DG, Buhari AM, et al. Death rates and causes of death after bariatric surgery for Pennsylvania residents, 1995 to 2004. Arch Surg 2007;142(10):923–8. [14] Chan MM, Hamza N, Ammori BJ. Duration of surgery independently influences risk of venous thromboembolism after laparoscopic bariatric surgery. Surg Obes Relat Dis 2013;9(1): 88–93. [15] Doumouras AG, Saleh F, Anvari S, Gmora S, Anvari M, Hong D. Mastery in bariatric surgery: the long-term surgeon learning curve of Roux-en-Y gastric bypass. Ann Surg 2018;(3):489–94. [16] De Beaux AC. Simulation-based training and learning curves in laparoscopic Roux-en-Y gastric bypass. Br J Surg 2012;99(7): 887–95. [17] Ben David M, Maler I, Kashtan H, Keidar A. Learning curve in laparoscopic Roux-en-Y gastric bypass for the treatment of morbid obesity. Harefuah 2015;154(4):254–8. [18] Schwartz ML, Drew RL, Chazin-Caldie M. Laparoscopic Roux-en-Y gastric bypass: preoperative determinants of prolonged operative times, conversion to open gastric bypasses, and postoperative complications. Obes Surg 2003;13(5):734–8. [19] Inaba CS, Koh CY, Sujatha-Bhaskar S, Lee Y, Pejcinovska M, Nguyen NT. The effect of hospital teaching status on outcomes in bariatric surgery. Surg Obes Relat Dis 2017;13(10):1723–7. [20] Davis SS Jr, Husain FA, Lin E, Nandipati KC, Perez S, Sweeney JF. Resident participation in index laparoscopic general surgical cases: impact of the learning environment on surgical outcomes. J Am Coll Surg 2013;216(1):96–104. [21] Fanous M, Carlin A. Surgical resident participation in laparoscopic Roux-en-Y bypass: is it safe? Surgery 2012;152(1):21–5. [22] Krell RW, Birkmeyer NJ, Reames BN, et al. Effects of resident involvement on complication rates after laparoscopic gastric bypass. J Am Coll Surg 2014;218(2):253–60. [23] Tseng WH, Jin L, Canter RJ, et al. Surgical resident involvement is safe for common elective general surgery procedures. J Am Coll Surg 2011;213(1):19–26.
1120
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121
[24] Doyon L, Moreno-Koehler A, Ricciardi R, Nepomnayshy D. Resident participation in laparoscopic Roux-en-Y gastric bypass: a comparison of outcomes from the ACS-NSQIP database. Surg Endosc 2016;30(8):3216–24. [25] Major P, Wysocki M, Dworak J, Pedziwiatr M, Malczak P, Budzynski A. Are bariatric operations performed by residents safe and efficient? Surg Obes Relat Dis 2017;13(4):614–21. [26] Stepaniak PS, Heij C, Buise MP, Mannaerts GH, Smulders JF, Nienhuijs SW. Bariatric surgery with operating room teams that stayed fixed during the day: a multicenter study analyzing the effects on patient outcomes, teamwork and safety climate, and procedure duration. Anesth Analg 2012;115(6):1384–92.
[27] Nguyen NT, Nguyen B, Nguyen VQ, Ziogas A, Hohmann S, Stamos MJ. Outcomes of bariatric surgery performed at accredited vs nonaccredited centers. J Am Coll Surg 2012;215(4):467–74. [28] Young MT, Jafari MD, Gebhart A, Phelan MJ, Nguyen NT. A decade analysis of trends and outcomes of bariatric surgery in Medicare beneficiaries. J Am Coll Surg 2014;219(3): 480–8. [29] Morton JM, Garg T, Nguyen N. Does hospital accreditation impact bariatric surgery safety? Ann Surg 2014;260(3):504–8. [30] Jafari MD, Jafari F, Young MT, Smith BR, Phalen MJ, Nguyen NT. Volume and outcome relationship in bariatric surgery in the laparoscopic era. Surg Endosc 2013;27(12):4539–46.
Editorial comment
Comment on: Operative time as a marker of quality in bariatric surgery Does operative time (OT) matter and should it serve as a quality metric? Does shorter OT indicate superior and efficient surgical skills, or does it contribute to less attention to detail and more mistakes? Does longer OT suggest a safe and considered approach, or inefficiency, poor skills, and inexperience? OT results from the amalgamation of surgical skill, patient factors, case complexity, surgical trainee involvement, surgical team, equipment, and infrastructure. It is difficult to determine the exact magnitude of contribution of each factor to OT, but it is intuitive that a surgeon’s technical skill has great impact on OT and outcome. Inaba et al. [1] have used the Bariatric Outcomes Longitudinal Database to show an association between prolonged OT and increased risk of complications, such as leak and mortality, within 1 year after laparoscopic Roux-en-Y gastric bypass (LRYGB) and laparoscopic sleeve gastrectomy [1]. Their findings add to a number of studies that have consistently shown an association between increased OT and greater odds of early morbidity and mortality after laparoscopic bariatric and general surgery. The use of retrospective Bariatric Outcomes Longitudinal Database data, however, has limitations, such as possible selection bias, coding error, and loss to follow-up. Furthermore, it does not include surgeon- and hospital-level data, surgical trainee involvement, specific cause of death, and case complexity, such as intraoperative bleeding or leak. Ideally, a comparison of fast and slow surgeons may yield more useful information about the impact of surgeonspecific characteristics on OT than does the comparison of fast and slow cases. Regardless, it is safe to assume that poor surgical skill contributes to prolonged OT and increases complications. When blinded peer surgeons rated individual surgeons’ technical skill for LRYGB, those with low technical skill had longer OT and higher complication, reoperation, readmission, and mortality rates [2]. Several studies have shown that surgical resident and fellow participation in laparoscopic bariatric surgery
increases OT, but the impact on adverse outcome has been mixed. A study on the impact of resident involvement in bariatric surgery at 1 institution revealed longer OT, without an increase in the risk of reoperation, intraoperative adverse events, or surgical complications [3]. By contrast, Aminian et al. [4] found that fellow involvement increased OT and the odds of worse early postoperative outcomes after LRYGB but not after laparoscopic sleeve gastrectomy. Finally, a study by Reames et al. [5] showed that slow surgeons had higher rates of complication after LRYGB even when adjusting for trainee involvement. OT should be used as a quality metric because multiple studies have shown a relationship between prolonged OT and increased risk of morbidity and mortality. To improve OT, the following steps must be considered. Hospitals should provide consistent operating room teams and necessary equipment and infrastructure for bariatric surgery. We must continue to emphasize maximum technical proficiency for surgical trainees. The popularity of laparoscopic sleeve gastrectomy has deprived trainees of the LRYGB skill set. Simulation training serves a role in training, but actual operative experience is irreplaceable. We should involve fellows in as many advanced minimally invasive and bariatric cases as possible, with gradual participation tailored to their skill level. Young attending surgeons should have strong mentorship, and all of us should be receptive to blinded peer review to identify and improve deficiencies in our technical skill to improve outcome and quality. Saber Ghiassi, M.P.H., M.D., F.A.C.S., F.A.S.M.B.S Department of Surgery, Yale School of Medicine New Haven, Connecticut References [1] Inaba CS, Koh CY, Sujatha-Bhaskar S, et al. Operative time as a marker of quality in bariatric surgery. Surg Obes Relat Dis 2019;15(7):1113–21.
Original article
Operative time as a marker of quality in bariatric surgery Colette S. Inaba, M.D.a, Christina Y. Koh, M.D.a, Sarath Sujatha-Bhaskar, M.D.a, Shea Gallagher, B.S.a, Yanjun Chen, M.S.b, Ninh T. Nguyen, M.D. F.A.C.S.a,* b
a Department of Surgery, University of California Irvine Medical Center, Orange, California Institute for Clinical and Translational Science, University of California Irvine, Irvine, California
Received 1 January 2019; accepted 6 April 2019
Abstract
Background: Few studies have examined the effect of prolonged operative time (OT) on outcomes in laparoscopic bariatric surgery. Existing studies mostly focus on 30-day complications, whereas serious complications may not occur until well after 30 days from the index operation. Objective: To determine the effect of prolonged OT on 1-year morbidity and mortality after laparoscopic Roux-en-Y gastric bypass (LRYGB) and laparoscopic sleeve gastrectomy (LSG). Setting: The Bariatric Outcomes Longitudinal Database (BOLD). Methods: Data on primary LRYGB and LSG cases performed between 2008 and 2012 in the BOLD were analyzed. Converted cases and cases concurrent with other procedures were excluded. Multivariate logistic regression was used to assess the association between OT and 1-year morbidity and mortality, with adjustment for preoperative demographic and clinical characteristics. Results: A total of 93,051 cases were examined, including 74,745 (80.3%) LRYGB and 18,306 (19.7%) LSG cases. For LRYGB, mean OT was 104 minutes (standard deviation [SD] 46.6). Every additional 10 minutes of OT was associated with increased odds of 1-year mortality (adjusted odds ratio [AOR] 1.04; P 5 .02), leak (AOR 1.07; P , .0001), and any adverse event (AOR 1.03; P , .001). For LSG, mean OT was 78 minutes (SD 37.4). Every additional 10 minutes of OT was associated with increased odds of 1-year leak (AOR 1.07; P 5 .0002). Data on patients lost to follow-up was unavailable. Conclusion: Prolonged operative time is associated with a significant increase in the odds of mortality and serious complications after laparoscopic bariatric surgery. Operative time may be a useful marker of quality in primary laparoscopic bariatric surgery. (Surg Obes Relat Dis 2019;15:1113–1120.) Ó 2019 American Society for Bariatric Surgery. Published by Elsevier Inc. All rights reserved.
Key words:
Bariatric surgery; Gastric bypass; Sleeve gastrectomy; Operative time; Mortality; Leak; Outcomes; Laparoscopic
Prolonged operative time (OT) has been associated with worse outcomes in a variety of different types of operations, including general surgery, vascular, and colorectal
This research was presented October 2017 at the American College of Surgeons Clinical Congress in San Diego, CA. * Correspondence: Ninh T. Nguyen, M.D., F.A.C.S., Department of Surgery, University of California Irvine Medical Center, 333 City Blvd West, Suite 1600, Orange, CA 92868. E-mail address: [email protected] (N.T. Nguyen).
procedures [1–4]. Several studies examining the effect of OT on bariatric surgery outcomes have also found an association between prolonged OT and increased complications [5–10]. Most of these studies are limited to analysis of 30-day outcomes, whereas a high percentage of serious complications such as pulmonary emboli (PE) and mortality may occur later than 30 days after surgery [11–13]. The objective of our study was to evaluate the effect of prolonged OT on 1-year morbidity and mortality after laparoscopic bariatric surgery.
https://doi.org/10.1016/j.soard.2019.04.010 1550-7289/Ó 2019 American Society for Bariatric Surgery. Published by Elsevier Inc. All rights reserved.
1114
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121
Methods Data source The Bariatric Outcomes Longitudinal Database (BOLD) was administered by the American Society of Metabolic and Bariatric Surgery and contains Health Insurance Portability and Accountability Act-compliant patient data prospectively collected from over 700 sites nationwide participating in the American Society of Metabolic and Bariatric Surgery Bariatric Surgery Centers of Excellence program. The strength of the BOLD database is the reporting of outcomes out to 1 year from the index bariatric operation. This study was exempt from institutional review board approval, as the BOLD contains only deidentified patient data and any specific patient intervention or interaction was not possible. Data analysis Data from all patients who underwent primary laparoscopic Roux-en-Y gastric bypass (LRYGB) or laparoscopic sleeve gastrectomy (LSG) between 2008 and 2012 were examined. Data were analyzed separately based on procedural type, either LRYGB or LSG. Cases that were converted or that included other concurrent procedures were
excluded. The primary outcomes of interest were 1-year mortality, leak, and PE. A secondary outcome of interest was 1-year any adverse event, a composite variable including any of 128 adverse events recorded in the database. Data on patients lost to follow-up was unavailable. Statistical analysis Statistical analysis was performed using SAS version 9.4 (SAS Institute, Cary, NC). Patient demographic characteristics, clinical and operative characteristics, and outcomes were examined by comparing the shortest, average, and longest operative-time terciles. Multivariate logistic regression was then used to determine adjusted odds ratios (AOR) for the association between longer OT and the outcomes of interest in a continuous manner (i.e., every 10 min). Adjustments were made for patient age, gender, race, preoperative body mass index (BMI), American Society of Anesthesiologists class, all preoperative comorbidities, and operative blood loss. Patients with missing data were excluded from the regression analysis. Two-sided tests with P , .05 were considered statistically significant. Further adjustment was made to account for multiple comparisons using the false discovery rate for P values.
Table 1 Demographic and clinical characteristics of patients undergoing laparoscopic Roux-en-Y gastric bypass Variable
Overall N 5 74,745
Shortest OT Tercile
Middle OT Tercile
Longest OT Tercile
P Value*
Age, yr, mean (SD) Female Race Caucasian Hispanic Black Other BMI, kg/m2, mean (SD) ASA Class I II III IV Comorbidities Congestive heart failure Depression Diabetes Hypertension Ischemic heart disease Liver disease Sleep apnea Peripheral vascular disease Venous thromboembolism Alcohol use Tobacco use Substance abuse Functional status—any impairment
44.7 (11.7) 57,944 (77.5)
44.5 (11.6) 19,898 (80.2)
44.8 (11.7) 19,327 (77.6)
44.9 (11.6) 18,623 (74.8)
.0003 , .0001
52,079 (69.7) 7,553 (10.1) 8,584 (11.5) 6,529 (8.7) 47.6 (8.1)
16,464 (66.3) 3,485 (14.0) 2,651 (10.7) 2,221 (9.0) 47.0 (7.7)
17,712 (71.1) 2,121 (8.5) 2,964 (11.9) 2,113 (8.5) 47.7 (8.1)
17,809 (71.6) 1,933 (7.8) 2,956 (11.9) 2,185 (8.8) 47.9 (8.4)
, .0001
2,562 (3.4) 16,450 (22.0) 52,428 (70.1) 3,305 (4.4)
1,127 (4.5) 4,413 (17.8) 18,172 (73.2) 1,109 (4.5)
659 (2.7) 5,652 (22.7) 17,385 (69.8) 1,214 (4.9)
746 (3.0) 6,365 (25.6) 16,792 (67.5) 980 (3.9)
, .0001
1,484 (2.0) 25,857 (34.8) 24,062 (32.4) 40,075 (54.0) 1,893 (2.6) 3,772 (5.1) 31,372 (42.3) 811 (1.1) 1,295 (1.7) 9,307 (12.5) 3,572 (4.8) 248 (.33) 2,143 (2.9)
493 (2.0) 8,134 (32.8) 7,827 (31.5) 13,341 (53.8) 641 (2.6) 1,159 (4.7) 9,719 (39.2) 260 (1.10) 438 (1.8) 2,909 (11.7) 1,242 (5.0) 58 (.3) 682 (2.8)
512 (2.1) 8,842 (35.5) 7,977 (32.0) 13,438 (54.0) 640 (2.6) 1,075 (4.3) 10,473 (42.0) 264 (1.10) 442 (1.8) 3,191 (12.8) 1,164 (4.7) 85 (.3) 730 (2.9)
476 (1.9) 8,840 (35.5) 8,222 (33.0) 13,226 (53.2) 609 (2.5) 1,538 (6.2) 11,139 (44.8) 287 (1.20) 414 (1.7) 3,191 (12.8) 1,154 (4.6) 94 (.4) 730 (2.9)
.4887 , .0001 .0009 .0876 .5628 , .0001 , .0001 .4634 .5633 .0001 .1287 .1204 .3265
ASA 5 American Society of Anesthesiologists; BMI 5 body mass index; OT 5 operative time; SD 5 standard deviation. Values are presented as frequencies (%) unless otherwise indicated. * P value is for comparison of all 3 terciles.
, .0001
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121
1115
Table 2 Demographic and clinical characteristics of patients undergoing laparoscopic sleeve gastrectomy Variable
Overall N 5 18,306
Shortest OT Tercile
Middle OT Tercile
Longest OT Tercile
P Value*
Age, yr, mean (SD) Female Race Caucasian Hispanic black Other BMI, kg/m2, mean (SD) ASA Class I II III IV Comorbidities Congestive heart failure Depression Diabetes Hypertension Ischemic heart disease Liver disease Sleep apnea Peripheral vascular disease Venous thromboembolism Alcohol use Tobacco use Substance abuse Functional status—any impairment
43.6 (11.3) 13,757 (75.2)
43.1 (11.2) 4,998 (80.8)
43.3 (11.3) 4,424 (74.5)
44.3 (11.3) 4,330 (70.2)
, .0001 , .0001
12,173 (66.5) 2,126 (11.6) 2,223 (12.1) 1,784 (9.8) 46.7 (8.6)
4,215 (68.1) 730 (11.8) 687 (11.1) 557 (9.0) 45.2 (7.7)
3,860 (65.0) 706 (11.9) 775 (13.1) 598 (10.1) 46.7 (8.5)
4,094 (66.3) 690 (11.2) 760 (12.3) 629 (10.2) 48.1 (9.5)
.0017
688 (3.8) 5,365 (29.3) 11,716 (64.0) 537 (2.9)
171 (2.8) 1,750 (28.3) 4,131 (66.8) 137 (2.2)
158 (2.7) 1,766 (29.7) 3,826 (64.4) 189 (3.2)
356 (5.8) 1,849 (30.0) 3,758 (60.9) 210 (3.4)
, .0001
265 (1.5) 5,448 (30.0) 4,879 (26.9) 8,973 (49.4) 337 (1.9) 772 (4.3) 7,057 (38.9) 145 (.8) 216 (1.2) 3,303 (18.2) 949 (5.2) 52 (.29) 338 (1.9)
68 (1.1) 1,846 (29.8) 1,578 (25.5) 2,934 (47.4) 85 (1.4) 160 (2.6) 2,139 (34.6) 46 (.7) 52 (.8) 1,195 (19.3) 288 (4.7) 11 (.2) 65 (1.1)
92 (1.6) 1,825 (30.7) 1,577 (26.6) 2,960 (49.8) 118 (2.0) 250 (4.2) 2,340 (39.4) 51 (.9) 82 (1.4) 1,116 (18.8) 338 (5.7) 18 (.3) 87 (1.5)
105 (1.7) 31 (28.8) 1,723 (27.9) 3,075 (49.8) 133 (2.2) 362 (5.9) 2,578 (41.8) 48 (.8) 82 (1.3) 989 (16.0) 323 (5.2) 23 (.4) 186 (3.0)
.0149 .0545 .0116 .0108 .0032 , .0001 , .0001 .7659 .0096 , .0001 .0370 .1206 , .0001
, .0001
ASA 5 American Society of Anesthesiologists; BMI 5 body mass index; OT 5 operative time; SD 5 standard deviation. Values are presented as frequencies (%) unless otherwise indicated. * P value is for comparison of all 3 terciles.
Results Patient demographic and clinical characteristics We examined 74,745 LRYGB cases. Patient demographic and clinical characteristics are listed in Table 1. Univariate comparisons among the shortest, middle, and longest OT terciles showed that longer OT was associated with older patient mean age (44.5 versus 44.8 versus 44.9 yr, respectively; P 5 .0003); higher proportion of male gender (19.8% versus 22.4% versus 25.2%, respectively; P , .0001); higher mean BMI (47.0 versus 47.7 versus 47.9
kg/m2, respectively; P , .0001); and higher rates of depression (32.8% versus 35.5% versus 35.5%, respectively; P , .0001), diabetes (31.5% versus 32.0% versus 33.0%; P 5 .0009), sleep apnea (39.2% versus 42.0% versus 44.8%, respectively; P , .0001), liver disease (4.7% versus 4.3% versus 6.2%, respectively; P , .0001), and alcohol use (11.7% versus 12.8% versus 12.8%, respectively; P 5 .0001). We examined 18,306 LSG cases. Patient demographic and clinical characteristics for LSG are listed in Table 2. Univariate comparisons among the shortest, middle, and
Table 3 Operative characteristics and outcomes for patients undergoing laparoscopic Roux-en-Y gastric bypass Variable
Overall N 5 74,745
Shortest OT Tercile
Middle OT Tercile
Longest OT Tercile
P Value*
Operative time, min, mean (SD) Estimated blood loss, mL, mean (SD) 1-yr mortality 1-yr leak 1-yr pulmonary embolus 1-yr any adverse event
104.4 (46.6) 42.8 (72.3) 151 (.20) 325 (.43) 137 (.18) 2,078 (2.8)
61.3 (10.8) 36.4 (81.6) 32 (.13) 69 (.28) 36 (.15) 557 (2.24)
95.9 (10.5) 40.2 (54.7) 57 (.23) 94 (.38) 52 (.21) 679 (2.73)
155.8 (41.3) 51.7 (76.8) 62 (.25) 161 (.65) 48 (.19) 839 (3.37)
, .0001 , .0001 .0061 , .0001 .2225 , .0001
OT 5 operative time; SD 5 standard deviation. Values are presented as frequencies (%) unless otherwise indicated. * P value is or comparison of all 3 terciles.
1116
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121 Table 4 Operative characteristics and outcomes for patients undergoing laparoscopic sleeve gastrectomy Variable
Overall N 5 18,306
Shortest OT Tercile
Middle OT Tercile
Slowest OT Tercile
P Value*
Operative time, min, mean (SD) Estimated blood loss, mL, mean (SD) 1-yr mortality 1-yr leak 1-yr pulmonary embolus 1-yr any adverse event
78.1 (37.4) 32.2 (58.3) 21 (.11) 50 (.27) 26 (.14) 244 (1.3)
46.5 (8.4) 23.7 (28.4) 4 (.06) 8 (.13) 5 (.08) 55 (.89)
70.1 (7.2) 31.3 (44.3) 6 (.10) 19 (.32) 13 (.22) 89 (1.50)
117.4 (37.7) 41.5 (85.0) 11 (.18) 23 (.37) 8 (.13) 99 (1.60)
, .0001 , .0001 .1635 .0245 .1239 .0009
OT 5 operative time; SD 5 standard deviation. Values are presented as frequencies (%) unless otherwise indicated. * P value is comparison for all 3 terciles.
longest OT terciles revealed that longer OT was associated with higher mean age (43.1 versus 43.3 versus 44.3 yr, respectively; P , .0001); higher proportion of male gender (19.2% versus 25.5% versus 29.8%, respectively; P , .0001); higher mean BMI (45.2 versus 46.7 versus 48.1 kg/m2, respectively; P , .0001); and higher rates of congestive heart failure (1.1% versus 1.6% versus 1.7%, respectively; P 5 .0149), diabetes (25.5% versus 26.6% versus 27.9%, respectively; P 5 .0116), hypertension (47.4% versus 49.8% versus 49.8%; P 5 .0108), ischemic heart disease (2.6% versus 4.2% versus 5.9%, respectively; P , .0001), sleep apnea (34.6% versus 39.4% versus 41.8%, respectively; P , .0001), and impaired functional status (i.e., being bedridden or needing an assistance device to mobilize; 1.1% versus 1.5% versus 3.0%, respectively; P , .0001) but lower rates of alcohol use (19.3% versus 18.8% versus 16%, respectively; P , .0001).
operative blood loss (36.4 versus 40.2 versus 51.7 mL, respectively; P , .0001) and higher 1-year mortality rate (.13% versus .23% versus .25%, respectively; P 5 .0061), leak rate (.28% versus .38% versus .65%, respectively; P , .0001), and any-adverse-event rate (2.24% versus 2.73% versus 3.37%, respectively; P , .0001). Table 4 lists the operative characteristics and outcomes for LSG. The overall mean OT for LSG was 78.1 (SD 37.4) minutes. The mean OT of the shortest, middle, and longest terciles was 46.5 (8.4) minutes, 70.1 (SD 7.2) minutes, and 117.4 (SD 37.7) minutes, respectively. Univariate comparisons among the shortest, middle, and longest operative-time terciles revealed that longer OT was associated with higher mean operative blood loss (23.7 versus 31.3 versus 41.5 mL, respectively; P , .0001) and higher 1-year leak rate (.13% versus .32% versus .37%, respectively; P 5 .0245) and any-adverse-event rate (.89% versus 1.50% versus 1.60%, respectively; P 5 .0009).
Operative characteristics and outcomes Table 3 lists the operative characteristics and outcomes for LRYGB. The overall mean OT for LRYGB was 104.4 (standard deviation [SD] 46.6) minutes. The mean OT of the shortest, middle, and longest terciles was 61.3 (SD 10.8) minutes, 95.9 (SD 10.5) minutes, and 155.8 (SD 41.3) minutes, respectively. Univariate comparisons among the shortest, middle, and longest operative-time terciles revealed that longer OT was associated with higher mean
Multivariate logistic regression analysis Multivariate logistic regression analysis was used to determine the association between prolonged OT and 1-year postoperative outcomes after adjustment for patient demographic characteristics and comorbidities. The results for multivariate logistic regression for association between prolonged operative time and outcome are listed in Table 5.
Table 5 Multivariate logistic regression analysis for association between every 10 minutes of prolonged operative time and outcomes after laparoscopic bariatric surgery Outcome
LRYGB
LSG
AOR (95% CI)
P Value*
AOR (95% CI)
P Value*
Mortality Leak Pulmonary embolism Any adverse event
1.04 (1.01, 1.07) 1.07 (1.06, 1.09) 1.01 (.98, 1.05) 1.03 (1.02, 1.04)
.0195 ,.0001 .5528 ,.0001
1.06 (1.00, 1.12) 1.07 (1.03, 1.10) 1.05 (.99, 1.11) 1.03 (1.00, 1.05)
.0828 .0002 .1291 .0652
AOR 5 adjusted odds ratio; CI 5 confidence interval; LRYGB 5 laparoscopic Roux-en-Y gastric bypass; LSG 5 laparoscopic sleeve gastrectomy. Model is adjusted for age, BMI, blood loss, gender, race, ASA, and all preoperative comorbidities and characteristics. * P value is adjusted for false discovery rate.
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121
1117
Figure 1. Unadjusted outcome as a function of operative time for A) laparoscopic Roux-en-Y gastric bypass and B) laparoscopic sleeve gastrectomy.
1118
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121
For LRYGB, every additional 10 minutes of OT was associated with a statistically significant increase in the odds of 1-year mortality (AOR 1.04; P 5 .0195), leak (AOR 1.07; P , .0001), and any adverse event (AOR 1.03; P , .0001). The relationships between OT and unadjusted outcomes after LRYGB are demonstrated in Fig. 1A. For LSG, every additional 10 minutes of OT was associated with a statistically significant increase in the odds of 1-year leak (AOR 1.07; P 5 .0002). There was also a trend toward an increase in the odds of 1-year mortality (AOR 1.06; P 5 .0828) and any adverse event (AOR 1.03; P 5 .0652), but neither of these were statistically significant at the 2-sided P , .05 level. The relationships between OT and unadjusted outcomes after LSG are demonstrated in Fig. 1B. Discussion Using a national database to evaluate the effect of prolonged OT on 1-year outcomes after laparoscopic bariatric surgery, we found an association between prolonged OT and higher odds of 1-year mortality, leak, and any adverse event after LRYGB and higher odds of 1-year leak after LSG. These findings suggest that OT may be used as a surrogate measurement for quality in bariatric surgery and might be useful as part of the designation criteria for Bariatric Centers of Excellence. Several studies have reported an association between prolonged OT in laparoscopic bariatric surgery and increased complications, particularly increased risk of 30-day complications [5–8], 90-day venous thromboembolism [14], and postoperative sepsis [9]. Our study also found an association between prolonged OT and higher odds of mortality after LRYGB, as well as higher odds of leak after both LRYGB and LSG. Our findings likely reflect the large number of patients required to adequately power an analysis that can detect the effect of OT for events as rare as leak or mortality, with 1-year leak rates ,.45% and 1-year mortality rates ,.20% in this study. Our study did not find a significant effect of prolonged OT on rates of PE. This is contrary to the findings of a study by Chan et al. [14] that examined 500 laparoscopic bariatric procedures and found an increased relative risk for PE of .0002 per additional minute of OT (P 5 .009). In another study evaluating PE and deep vein thrombosis as a composite measure after 16,344 LRYGB cases, the authors found the rates to be significantly higher in cases performed by surgeons who operated in the longest versus shortest OT tercile (.32% versus .22%, respectively; P , .001) [5]. There is limited literature specifically evaluating the effect of OT on LSG outcomes. In a recently published institutional review of 233 patients who underwent LSG, the authors found that longer OT was significantly associated with increased risk of overall complications (AOR 1.01 per additional minute; P 5 .029) [8]. While we found that
prolonged OT was associated with an increase in odds of leak, there was a nonsignificant trend toward an increase in odds of overall complications (AOR 1.03; P 5 .0652). Given our findings that prolonged OT is correlated with increased patient morbidity and mortality, OT may be used as a marker for quality of care in bariatric surgery. Not only is OT correlated with patient outcomes, it is also correlated with perioperative factors that may themselves affect patient outcomes in bariatric surgery, such as the surgeon’s technical skill, case complexity, resident involvement, and dynamics of the surgical team [5,6]. In a study in which bariatric surgeons ranked their peers’ technical skill during LRYGB cases, surgeons whose skill was ranked in the lowest versus highest quartile had longer OT (137 versus 98 minutes, respectively; P , .001), higher overall complication rates (14.5% versus 5.2%, respectively; P , .001), and higher patient mortality (.26% versus .05%, respectively; P 5 .01), suggesting that better technical prowess is correlated with both decreased OT and improved outcomes [10]. This corroborates findings from other studies demonstrating decreased OT and lower rates of complications after overcoming the learning curve for LRYGB [15–17]. Another factor correlated with longer OT is case complexity, such as operating on patients with higher BMI, which is associated with both longer OT and increased complications [14,18]. To reduce bias from unmeasured patient characteristics that may contribute to case complexity, one study compared outcomes based on the median OT of specific bariatric surgeons performing LRYGB, instead of evaluating specific cases [5]. The authors still found that patients of surgeons who operated in the longest versus shortest OT tercile had higher adjusted complication rates (12.9% versus 8.1%, respectively P 5 .022). Resident involvement in bariatric surgery has also been associated with longer OT, but results are mixed regarding whether resident involvement in bariatric surgery is also associated with increased postoperative complications [19–24]. However, even after adjusting for the presence of resident involvement, surgeons with longer OT may have higher rates of overall complications [5,25]. Finally, using the same operative team throughout the day has been demonstrated to decrease OT for bariatric surgery, as well as improve teamwork and the safety climate [26]. Therefore, OT may function as a surrogate marker of various preoperative and intraoperative factors that may impact patient outcomes. As such, OT might be useful as a marker of surgical quality in laparoscopic bariatric surgery, similar to institutional accreditation status and higher case volume, which have both been shown to be associated with better bariatric surgery outcomes [15,27–30]. There are several limitations to this study. First, this was a retrospective database study, and data may be subject to selection bias, missing data, and coding errors. Second,
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121
data on patients lost to follow-up, as well as other long-term outcomes such as incidence of deep vein thrombosis, are not readily available in the database, meaning true complication rates may be underestimated. Finally, the database did not include hospital-level or surgeon-level data nor any data on trainee involvement in cases; thus, we were unable to account for the effect of these factors in our analysis. Despite these limitations, ours is the first study to examine the longterm effect of OT on morbidity and mortality after laparoscopic bariatric surgery. Conclusion Prolonged OT is associated with a significant increase in odds of mortality, leak, and any adverse event after LRYGB and an increase in odds of leak after LSG. Given these associations, OT may be a useful surrogate marker for quality of care in laparoscopic bariatric surgery, similar to institutional accreditation status or case volume. Future studies should also consider adjusting for OT when examining outcomes after bariatric surgery. Acknowledgments This study was partially supported by grant UL1 TR001414 from the National Center for Advancing Translational Sciences, National Institutes of Health (NIH), through the Biostatistics, Epidemiology and Research Design Unit at the University of California Irvine. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Disclosures The American Society for Metabolic and Bariatric Surgery and the surgeons and the hospitals participating in the Centers of Excellence program from 2007 to 2012 are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors. Inaba, Koh, Sujatha-Bhaskar, Gallagher, and Nguyen have no conflicts of interest to disclose. Chen works with a statistical team that receives funding from the National Institutes of Health. References [1] Procter LD, Davenport DL, Bernard AC, Zwischenberger JB. General surgical operative duration is associated with increased risk-adjusted infectious complication rates and length of hospital stay. J Am Coll Surg 2010;210(1):60–5. [2] Daley BJ, Cecil W, Clarke PC, Cofer JB, Guillamondegui OD. How slow is too slow? Correlation of operative time to complications: an analysis from the Tennessee Surgical Quality Collaborative. J Am Coll Surg 2015;220(4):550–8. [3] Scheer A, Martel G, Moloo H, et al. Laparoscopic colon surgery: does operative time matter? Dis Colon Rectum 2009;52(10): 1746–52.
1119
[4] Bailey MB, Davenport DL, Vargas HD, Evers BM, Mckenzie SP. Longer operative time: deterioration of clinical outcomes of laparoscopic colectomy versus open colectomy. Dis Colon Rectum 2014;57(5):616–22. [5] Reames BN, Bacal D, Krell RW, Birkmeyer JD, Birkmeyer NJ, Finks JF. Influence of median surgeon operative duration on adverse outcomes in bariatric surgery. Surg Obes Relat Dis 2015;11(1): 207–13. [6] Jackson TD, Wannares JJ, Lancaster RT, Rattner DW, Hutter MM. Does speed matter? The impact of operative time on outcome in laparoscopic surgery. Surg Endosc 2011;25(7):2288–95. [7] Chen SY, Stem M, Schweitzer MA, Magnuson TH, Lidor AO. Assessment of postdischarge complications after bariatric surgery: A National Surgical Quality Improvement Program analysis. Surgery 2015;158(3):777–86. [8] Major P, Wysocki M, Pedziwiatr M, et al. Risk factors for complications of laparoscopic sleeve gastrectomy and laparoscopic Rouxen-Y gastric bypass. Int J Surg 2017;37:71–8. [9] Blair LJ, Huntington CR, Cox TC, et al. Risk factors for postoperative sepsis in laparoscopic gastric bypass. Surg Endosc 2016;30(4): 1287–93. [10] Birkmeyer JD, Finks JF, O’reilly A, et al. Surgical skill and complication rates after bariatric surgery. N Engl J Med 2013;369(15): 1434–42. [11] Celik F, Bounif F, Fliers JM, et al. The impact of surgical complications as a main risk factor for venous thromboembolism: a multicenter study. Obes Surg 2014;24(10):1603–9. [12] Steele KE, Schweitzer MA, Prokopowicz G, et al. The long-term risk of venous thromboembolism following bariatric surgery. Obes Surg 2011;21(9):1371–6. [13] Omalu BI, Ives DG, Buhari AM, et al. Death rates and causes of death after bariatric surgery for Pennsylvania residents, 1995 to 2004. Arch Surg 2007;142(10):923–8. [14] Chan MM, Hamza N, Ammori BJ. Duration of surgery independently influences risk of venous thromboembolism after laparoscopic bariatric surgery. Surg Obes Relat Dis 2013;9(1): 88–93. [15] Doumouras AG, Saleh F, Anvari S, Gmora S, Anvari M, Hong D. Mastery in bariatric surgery: the long-term surgeon learning curve of Roux-en-Y gastric bypass. Ann Surg 2018;(3):489–94. [16] De Beaux AC. Simulation-based training and learning curves in laparoscopic Roux-en-Y gastric bypass. Br J Surg 2012;99(7): 887–95. [17] Ben David M, Maler I, Kashtan H, Keidar A. Learning curve in laparoscopic Roux-en-Y gastric bypass for the treatment of morbid obesity. Harefuah 2015;154(4):254–8. [18] Schwartz ML, Drew RL, Chazin-Caldie M. Laparoscopic Roux-en-Y gastric bypass: preoperative determinants of prolonged operative times, conversion to open gastric bypasses, and postoperative complications. Obes Surg 2003;13(5):734–8. [19] Inaba CS, Koh CY, Sujatha-Bhaskar S, Lee Y, Pejcinovska M, Nguyen NT. The effect of hospital teaching status on outcomes in bariatric surgery. Surg Obes Relat Dis 2017;13(10):1723–7. [20] Davis SS Jr, Husain FA, Lin E, Nandipati KC, Perez S, Sweeney JF. Resident participation in index laparoscopic general surgical cases: impact of the learning environment on surgical outcomes. J Am Coll Surg 2013;216(1):96–104. [21] Fanous M, Carlin A. Surgical resident participation in laparoscopic Roux-en-Y bypass: is it safe? Surgery 2012;152(1):21–5. [22] Krell RW, Birkmeyer NJ, Reames BN, et al. Effects of resident involvement on complication rates after laparoscopic gastric bypass. J Am Coll Surg 2014;218(2):253–60. [23] Tseng WH, Jin L, Canter RJ, et al. Surgical resident involvement is safe for common elective general surgery procedures. J Am Coll Surg 2011;213(1):19–26.
1120
Colette S. Inaba et al. / Surgery for Obesity and Related Diseases 15 (2019) 1113–1121
[24] Doyon L, Moreno-Koehler A, Ricciardi R, Nepomnayshy D. Resident participation in laparoscopic Roux-en-Y gastric bypass: a comparison of outcomes from the ACS-NSQIP database. Surg Endosc 2016;30(8):3216–24. [25] Major P, Wysocki M, Dworak J, Pedziwiatr M, Malczak P, Budzynski A. Are bariatric operations performed by residents safe and efficient? Surg Obes Relat Dis 2017;13(4):614–21. [26] Stepaniak PS, Heij C, Buise MP, Mannaerts GH, Smulders JF, Nienhuijs SW. Bariatric surgery with operating room teams that stayed fixed during the day: a multicenter study analyzing the effects on patient outcomes, teamwork and safety climate, and procedure duration. Anesth Analg 2012;115(6):1384–92.
[27] Nguyen NT, Nguyen B, Nguyen VQ, Ziogas A, Hohmann S, Stamos MJ. Outcomes of bariatric surgery performed at accredited vs nonaccredited centers. J Am Coll Surg 2012;215(4):467–74. [28] Young MT, Jafari MD, Gebhart A, Phelan MJ, Nguyen NT. A decade analysis of trends and outcomes of bariatric surgery in Medicare beneficiaries. J Am Coll Surg 2014;219(3): 480–8. [29] Morton JM, Garg T, Nguyen N. Does hospital accreditation impact bariatric surgery safety? Ann Surg 2014;260(3):504–8. [30] Jafari MD, Jafari F, Young MT, Smith BR, Phalen MJ, Nguyen NT. Volume and outcome relationship in bariatric surgery in the laparoscopic era. Surg Endosc 2013;27(12):4539–46.
Editorial comment
Comment on: Operative time as a marker of quality in bariatric surgery Does operative time (OT) matter and should it serve as a quality metric? Does shorter OT indicate superior and efficient surgical skills, or does it contribute to less attention to detail and more mistakes? Does longer OT suggest a safe and considered approach, or inefficiency, poor skills, and inexperience? OT results from the amalgamation of surgical skill, patient factors, case complexity, surgical trainee involvement, surgical team, equipment, and infrastructure. It is difficult to determine the exact magnitude of contribution of each factor to OT, but it is intuitive that a surgeon’s technical skill has great impact on OT and outcome. Inaba et al. [1] have used the Bariatric Outcomes Longitudinal Database to show an association between prolonged OT and increased risk of complications, such as leak and mortality, within 1 year after laparoscopic Roux-en-Y gastric bypass (LRYGB) and laparoscopic sleeve gastrectomy [1]. Their findings add to a number of studies that have consistently shown an association between increased OT and greater odds of early morbidity and mortality after laparoscopic bariatric and general surgery. The use of retrospective Bariatric Outcomes Longitudinal Database data, however, has limitations, such as possible selection bias, coding error, and loss to follow-up. Furthermore, it does not include surgeon- and hospital-level data, surgical trainee involvement, specific cause of death, and case complexity, such as intraoperative bleeding or leak. Ideally, a comparison of fast and slow surgeons may yield more useful information about the impact of surgeonspecific characteristics on OT than does the comparison of fast and slow cases. Regardless, it is safe to assume that poor surgical skill contributes to prolonged OT and increases complications. When blinded peer surgeons rated individual surgeons’ technical skill for LRYGB, those with low technical skill had longer OT and higher complication, reoperation, readmission, and mortality rates [2]. Several studies have shown that surgical resident and fellow participation in laparoscopic bariatric surgery
increases OT, but the impact on adverse outcome has been mixed. A study on the impact of resident involvement in bariatric surgery at 1 institution revealed longer OT, without an increase in the risk of reoperation, intraoperative adverse events, or surgical complications [3]. By contrast, Aminian et al. [4] found that fellow involvement increased OT and the odds of worse early postoperative outcomes after LRYGB but not after laparoscopic sleeve gastrectomy. Finally, a study by Reames et al. [5] showed that slow surgeons had higher rates of complication after LRYGB even when adjusting for trainee involvement. OT should be used as a quality metric because multiple studies have shown a relationship between prolonged OT and increased risk of morbidity and mortality. To improve OT, the following steps must be considered. Hospitals should provide consistent operating room teams and necessary equipment and infrastructure for bariatric surgery. We must continue to emphasize maximum technical proficiency for surgical trainees. The popularity of laparoscopic sleeve gastrectomy has deprived trainees of the LRYGB skill set. Simulation training serves a role in training, but actual operative experience is irreplaceable. We should involve fellows in as many advanced minimally invasive and bariatric cases as possible, with gradual participation tailored to their skill level. Young attending surgeons should have strong mentorship, and all of us should be receptive to blinded peer review to identify and improve deficiencies in our technical skill to improve outcome and quality. Saber Ghiassi, M.P.H., M.D., F.A.C.S., F.A.S.M.B.S Department of Surgery, Yale School of Medicine New Haven, Connecticut References [1] Inaba CS, Koh CY, Sujatha-Bhaskar S, et al. Operative time as a marker of quality in bariatric surgery. Surg Obes Relat Dis 2019;15(7):1113–21.