- Email: [email protected]
Association of operative time with outcomes in minimally invasive and abdominal myomectomy
ORIGINAL ARTICLE: REPRODUCTIVE SURGERY
The association of operative time with outcomes in minimally invasive and abdominal myomectomy Maria V. Vargas, M.D., M.S.,a Kathryn Denny Larson, MD,b Andrew Sparks, M.S.,c Samantha L. Margulies, M.D.,d Cherie Q. Marfori, M.D.,a Gaby Moawad, M.D.,a and Richard L. Amdur, Ph.D.c a
Department of Obstetrics and Gynecology, George Washington University School of Medicine and Health Sciences; School of Medicine and Health Sciences, George Washington University; and c Department of Surgery, George Washington University School of Medicine and Health Sciences, Washington, DC; and d Department of Obstetrics, Gynecology and Reproductive Health Sciences, Yale Medicine, New Haven, Connecticut b
Objective(s): To determine the association of operative time (ORT) with perioperative morbidity and whether there is an ORT at which minimally invasive myomectomy becomes inferior to laparotomy. Design: Retrospective cohort study. Setting: Not applicable. Patient(s): Myomectomy cases identified by CPT code from 2005 to 2016. Intervention(s): Cases were stratified and analyzed by surgical approach and 90-minute intervals. Main Outcome Measure(s): Thirty-day postoperative morbidity. Result(s): A total of 11,709 myomectomies were identified; 4,673 (39.9%) were minimally invasive, 6,997 (59.8%) were abdominal, and 39 (0.3%) were conversions. The incidence of complications significantly increased with ORT. After adjusting for confounders, mean ORT in minutes (95% confidence interval) was 113 (111–115) for abdominal, 156 (153–159) for minimally invasive, and 172 (148–200) for conversions. Despite shorter ORT, morbidity was greater in abdominal cases (16% vs. 5.7%), with the highest rates in converted cases (20.5%). The minimally invasive approach in general had lower odds of complications (odds ratio, 0.23; 95% confidence interval, 0.19–0.26). However, when minimally invasive surgery ORT reached R 270 minutes, the odds of a composite complication variable increased compared with abdominal cases <90 minutes (odds ratio, 2.30; 95% confidence interval, 1.69–3.13). Of minimally invasive cases, 88% were completed in <270 minutes. Conclusion(s): ORT was predictive of complications for both minimally invasive and abdominal myomectomies. Despite longer ORTs, minimally invasive procedures generally had superior 30-day outcomes up to 270 minutes. Careful patient counseling and preparation to increase surgical efficiency should be prioritized for either approach. (Fertil SterilÒ 2019;-:-–-. Ó2019 by American Society for Reproductive Medicine.) Key Words: Myomectomy, minimally invasive gynecologic surgery, laparoscopy, laparotomy, outcomes Discuss: You can discuss this article with its authors and other readers at https://www.fertstertdialog.com/users/16110-fertilityand-sterility/posts/43850-27172
U
terine fibroids are very common in reproductive-age women and have a significant societal impact, including negatively affecting quality of life and work productivity (1). In the United States, the estimated
annual direct costs related to medical care for fibroids, including surgery, hospital admissions, outpatient visits, and medications, is $4.1–9.4 billion (2). Although medical options exist, the management of large or symptom-
Received October 23, 2018; revised and accepted February 19, 2019. M.V.V. has nothing to disclose. K.D.L. has nothing to disclose. A.S. has nothing to disclose. S.L.M. has nothing to disclose. C.Q.M. has nothing to disclose. G.M. is a speaker for Intuitive Surgical. R.L.A. has nothing to disclose. The results of this study were presented at AAGL Global Congress on Minimally Invasive Gynecology, Orlando, Florida, November 2016; and the Annual Meeting of the Society for Reproductive Investigation, Orlando, Florida, March 2017. Reprint requests: Maria Victoria Vargas, M.D., 2150 Pennsylvania Avenue, NW, Washington, DC, 20037 (E-mail: [email protected]). Fertility and Sterility® Vol. -, No. -, - 2019 0015-0282/$36.00 Copyright ©2019 Published by Elsevier Inc. on behalf of the American Society for Reproductive Medicine https://doi.org/10.1016/j.fertnstert.2019.02.020 VOL. - NO. - / - 2019
atic fibroids has traditionally been surgical, and this remains the mainstay of treatment. In the United States, almost half of all hysterectomies are performed as the result of uterine fibroids (3). Myomectomy is the only uterus-sparing surgical management option and is commonly performed in women who wish to preserve reproductive potential (4). Myomectomies are increasingly being performed via minimally invasive surgery (MIS), including laparoscopy and robotic-assisted laparoscopy. MIS has proven clinical benefits over traditional laparotomy for myomectomy, including a faster return to 1
ORIGINAL ARTICLE: REPRODUCTIVE SURGERY normal activities, fewer febrile episodes, shorter hospital stay, lower intraoperative blood loss, fewer wound complications, and less postoperative pain (5, 6). Despite the established benefits of minimally invasive myomectomy, longer operative time is common with the use of laparoscopic and robotic-assisted laparoscopy compared with laparotomy (5). Longer operative times in minimally invasive approaches and laparotomy have been associated with adverse outcomes, such as surgical site infection (7) and venous thromboembolisms (8). An analysis of laparoscopic hysterectomy showed that at 4 hours, perioperative morbidity rises significantly (9). However, such an operative time has not been defined for myomectomy. Limited evidence exists to guide surgeons in both identifying patients at risk of excessive operating time and determining whether the at-risk patient may then benefit from an alternative surgical approach that may be more expeditious. The American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) was established to measure and improve outcomes across surgical specialties. The program captures over 150 patient variables from participating institutions across the United States. These variables include preoperative, intraoperative, and 30-day postoperative outcomes for patients undergoing major surgical procedures (10). The objective of this study is to use data from ACS NSQIP database to identify predictors of prolonged operative time for myomectomy, to compare perioperative outcomes between abdominal and MIS myomectomy by operative time, and to determine whether there is an operative time at which outcomes for MIS myomectomy become inferior to abdominal myomectomy.
MATERIALS AND METHODS Using the ACS NSQIP database, we conducted a retrospective cohort analysis of women undergoing myomectomy for benign indications from 2005 to 2016. The study was exempt from review by the George Washington University Institutional Review Board due to the use of a deidentified database. The methods for this study are reported according to ‘‘The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement’’ (11).
Source of Data The method of data collection through the ACS NSQIP database has been previously described and validated (10,12– 15). Surgical clinical reviewers at each site collect data from surgical and medical records. The number of cases per site is determined by site surgical volume. Exclusion criteria include minor cases, patients <18 years of age, and return to the operating room (OR) related to a previous procedure. The database contains only deidentified data and uses a sampling procedure intended to prevent bias in cases chosen. Selected cases have outcomes recorded for 30 days after surgery. The ACS NSQIP audits participating sites regularly for procedural compliance. 2
Case Selection We selected all cases from the ACS NSQIP database years 2005–2016 that had any of the following principal CPT codes: 58140, 58146 (abdominal), 58545, 58546 (MIS). Myomectomies with specimen weight >250 g and/or more than five intramural fibroids removed were indicated by codes 58146 or 58546. Emergent cases (n ¼ 108), those with disseminated cancer (n ¼ 6), and those with operative time <20 minutes (n ¼ 65) were excluded.
Variables Outcomes included superficial surgical site infection, deep wound infection, organ space infection, dehiscence (wound complications), pneumonia, reintubation, prolonged intubation (pulmonary complications), deep vein thrombosis or pulmonary embolism (clotting), myocardial infarction or cardiac arrest (cardiac), acute kidney injury or dialysis (renal), sepsis or septic shock (sepsis events), bleeding, urinary tract infection (UTI), return to OR, and hospital length of stay (LOS) >3 days. A composite complication outcome was defined as wound, pulmonary, clotting, sepsis, UTI, bleeding, return to OR, cardiac, or renal events. Predictors analyzed for significance included age, race, body mass index (BMI), hematocrit, diabetes mellitus (DM), hypertension (HTN), and specimen weight >250 g and/or five or more intramural fibroids removed.
Data Analysis Variable distributions were examined for outliers and skewed distributions. These were either log-transformed or recoded into quartiles. To examine associations between treatment variables (procedure: MIS/abdominal/conversion; operative time: <90 minutes, 90 to <180 minutes, 180 to <270 minutes, or R270 minutes) and categorical variables, we used c2 or Fisher's exact test. Associations with continuous variables were examined using t-tests or analysis of variance. First, we identified presurgery variables that were associated with operative time and with procedure. Second, to determine which variables had independent associations with operative time, we used a general linear model to predict natural log of operative time, including as predictors any pretreatment variables that could function as confounds, having at least a trend level of association (P< .10) either with procedure or with operative time. Third, to determine whether there was a single cut point at which increasing operative time was associated with worse outcomes within procedures, we first investigated whether the odds of having any complication as a function of operative time differed for MIS versus abdominal surgery using a logistic regression model. This model was also adjusted for other presurgery variables that could function as confounds. We then tested logistic regression models adjusted for the same set of covariates listed above, for longer MIS versus shorter abdominal procedures. SAS version 9.4 (SAS Institute) was used for all data analysis, and P< .05 was considered statistically significant. VOL. - NO. - / - 2019
Fertility and Sterility®
RESULTS There were 11,709 cases identified that met the inclusion criteria (Supplemental Fig. 1). Mean (SD) age was 36.9 6.5, mean (SD) BMI was 28.8 6.9, 4,626 patients (39.5%) were African American, and 3,436 (29.3%) were Caucasian. MIS procedures were used in 4,673 cases (39.9%) and abdominal myomectomy in 6,997 cases (59.8%); 39 (0.3%) had CPT codes for both, indicating conversion. Mean (SD) operative time was 145.9 77.5 minutes, but the distribution was positively skewed. Median operative time was 131 minutes (interquartile range, 90–184 minutes). Presurgery patient variables significantly associated with longer operative time included younger age, African American race, higher BMI, lower preoperative hematocrit, and the presences of >250 g of fibroids and/or more than five intramural fibroids (Supplemental Table 1). Procedure type was also significantly associated with age, race, BMI, hematocrit, HTN, and >250 g of fibroids and/or more than five intramural fibroids removed (Supplemental Table 2). Compared with patients who underwent abdominal surgery, those who underwent MIS myomectomy were older, were more likely to be Caucasian, had lower BMI, had higher hematocrit, and were less likely to have HTN or >250 g of fibroids and/or more than five intramural fibroids removed. In general, operative time was longer for MIS procedures compared with abdominal procedures (Table 1). There was a positive association between operative time and morbidity outcomes including wound, pulmonary, clotting, bleeding, and renal complications as well as return to the OR. There was also a positive association with operative time and hospital LOS >3 days (Table 2). The composite complication outcome had a linear association with operative time (P< .0001). Prolonged operative time was significantly more common in MIS procedures, whether this was defined as >3 hours (38.5% vs. 18%; P< .0001) or >4 hours (18.1% vs. 6%; P< .0001). However, in general, abdominal procedures had higher rates of complications than MIS procedures (Table 3). This was true for pulmonary complications, sepsis, UTI, and bleeding. Hospital LOS > 3 days was more frequent for abdominal procedures. The composite morbidity outcome was more likely in abdominal than in MIS procedures (16% vs. 5.7%) but was most likely in conversion cases (20.5%; P< .0001). The general linear model predicting the natural log of operative time was significant (F ¼ 226.44, P< .0001) with R2 ¼ 0.15, indicating a moderate association between the pre-
dictors and outcome. Significant independent predictors of operative time included procedure type, >250 g of fibroids and/or more than five intramural fibroids removed, race, age, hematocrit, and BMI (Supplemental Table 3). To compare outcomes for MIS versus abdominal procedures at various operative times, it was necessary to control for the effects of race, >250 g of fibroids and/or more than five intramural fibroids removed, age, BMI, presurgery hematocrit, and HTN. After adjusting for these covariates, the mean operative times were 113 minutes (95% confidence interval [CI], 111–115) for abdominal, 156 minutes (95% CI, 153– 159) for MIS, and 172 minutes (95% CI, 148–200) for conversions (Supplemental Table 3). The logistic regression model predicting any complication was significant (likelihood ratio c2 ¼ 1,035.08 P< .0001) with pseudo R2 ¼ .09 and with c ¼ .75 indicating moderately good discrimination. The overall operative time effect was significant (P< .0001). Compared with the shortest operative times (<90 minutes), the odds of reaching the composite morbidity endpoint were over two times higher, over four times higher, and over 11 times higher in surgeries lasting 90 to <180, 180 to <270, and R270 minutes, respectively. Compared with abdominal myomectomy, the adjusted odds ratio for the composite complications variable for MIS myomectomy was 0.23 (95% CI, 0.19–0.26; P< .0001) but was not significantly different for conversion cases compared with abdominal surgery (Supplemental Table 4). In comparing longer MIS versus shorter abdominal cases, we found several comparisons in which MIS procedures continued to have lower odds of complications (MIS 90 to <180 vs. abdominal <90; MIS 180 to <270 minutes vs. abdominal 90 to <180 minutes; MIS R270 minutes vs. abdominal 180 to <270 minutes), and one situation (MIS R 270 minutes vs. abdominal <90 minutes) where abdominal surgery had a significantly lower odds of complications (Table 4). All operation time combinations where the MIS procedure was shorter yielded significantly lower odds of complications for MIS myomectomy.
DISCUSSION This retrospective cohort study provides comparative surgical results for MIS versus abdominal myomectomy procedures stratified by operative time. Consistent with previous studies, we found improved outcomes for MIS approaches to myomectomy when compared with abdominal myomectomy
TABLE 1 Association of procedure with operative time (in minutes). Operative time, min (%) Procedure
n (column %)
<90 (24.4)
90 to <180 (48.8)
180 to <270 (19.9)
‡270 (6.8)
P value
MIS Abdominal Conversion
4,673 (39.9) 6,997 (59.8) 39 (0.3)
756 (26.4) 2,100 (73.4) 4 (0.1)
2,090 (36.5) 3,616 (63.2) 13 (0.2)
1,270 (54.4) 1,050 (45.0) 13 (0.6)
557 (69.9) 231 (30.0) 9 (1.1)
< .0001
Note: Data are shown as n (row %), except in the n column, where column % is shown. Vargas. Impact of operative on outcomes for myomectomy . Fertil Steril 2019.
VOL. - NO. - / - 2019
3
ORIGINAL ARTICLE: REPRODUCTIVE SURGERY
TABLE 2 Outcomes by operative time (in minutes). Operative time, min (%) Variable
<90 (n [ 2,860)
90 to <180 (n [ 5,719)
180 to <270 (n [ 2,333)
‡270 (n [ 797)
P value
149 (5.2) 29 (1.0) 4 (0.1) 2 (0.1) 7 (0.2) 30 (1.1) 92 (3.2) 15 (0.5) 0 0 106 (3.7)
628 (11.0) 77 (1.4) 11 (0.2) 12 (0.2) 26 (0.5) 48 (0.8) 484 (8.5) 47 (0.8) 0 0 328 (5.7)
407 (17.5) 39 (1.7) 11 (0.5) 16 (0.7) 8 (0.3) 29 (1.2) 342 (14.7) 20 (0.9) 1 (0.1) 3 (0.1) 186 (8.0)
211 (26.5) 18 (2.3) 6 (0.8) 7 (0.9) 5 (0.6) 10 (1.3) 185 (23.2) 14 (1.8) 0 1 (0.1) 82 (10.3)
< .0001 .03 .004 < .0001 .33 .32 < .0001 .0084 .27 .008 < .0001
Any complication Wound complications Pulmonary Clotting Sepsis UTI Bleeding Return to OR Cardiac Renal Hospital LOS > 3 d
Note: Wound complications included superficial surgical site infection, deep wound infection, organ space infection, and dehiscence. Pulmonary complications included pneumonia, reintubation, and prolonged intubation. Clotting complications included deep vein thrombosis or pulmonary embolism. Cardiac complications included myocardial infarction or cardiac arrest. Renal complications included acute kidney injury or dialysis. Any complication included wound, pulmonary, clotting, sepsis, UTI, bleeding, return to OR, cardiac, or renal events. Vargas. Impact of operative on outcomes for myomectomy . Fertil Steril 2019.
(5, 6, 16–21) and worsened outcomes with increasing operative time for both MIS and abdominal approaches (22– 24). MIS myomectomy was associated with lower incidence of the composite complications variable and a 77% lower odds of having any complication compared with abdominal myomectomy (95% CI, 74%–81% lower). Specifically, MIS myomectomy had lower rates of pulmonary complications, sepsis, UTI, and bleeding. Our study also supports findings from prior work showing longer operative times for MIS cases (5, 6, 17, 24), which were about 40 minutes longer on average than abdominal cases in our cohort.
TABLE 3 Outcomes by procedure. Variable
MIS
Abdominal Conversion
Any complication 265 (5.7) 1,122 (16.0) Wound 51 (1.1) 112 (1.6) complications Pulmonary 5 (0.1) 27 (0.4) Clotting 10 (0.2) 27 (0.4) Sepsis 8 (0.2) 38 (0.5) UTI 30 (0.6) 85 (1.2) Bleeding 165 (3.5) 932 (13.3) Return to OR 28 (0.6) 68 (1.0) Cardiac 0 1 (0.01) Renal 0 (0) 4 (0.1) Hospital LOS 44 (0.9) 657 (9.4) >3 d Surgery time, h >3 1,799 (38.5) 1,257 (18.0) >4 844 (18.1) 416 (6.0)
P value
8 (20.5) 0 (0)
< .0001 .054
0 (0) 0 (0) 0 (0) 2 (5.1) 6 (15.4) 0 (0) 0 0 (0) 1 (2.6)
.017 .25 .007 .0003 < .0001 .08 .99 .17 < .0001
21 (53.9) 11 (28.2)
< .0001 < .0001
Note: Data in parentheses are percentages. Wound complications included superficial surgical site infection, deep wound infection, organ space infection, and dehiscence. Pulmonary complications included pneumonia, reintubation, and prolonged intubation. Clotting complications included deep vein thrombosis or pulmonary embolism. Cardiac complications included myocardial infarction or cardiac arrest. Renal complications included acute kidney injury or dialysis. Any complication included wound, pulmonary, clotting, sepsis, UTI, bleeding, return to OR, cardiac, or renal events. Vargas. Impact of operative on outcomes for myomectomy . Fertil Steril 2019.
4
When stratified by duration, longer MIS cases had lower odds of complications when compared with shorter abdominal cases in several scenarios (MIS 90 to <180 vs. abdominal <90; MIS 180 to <270 minutes vs. abdominal 90 to <180 minutes; MIS R270 minutes vs. abdominal 180 to <270 minutes). However, abdominal cases completed in <90 minutes had lower odds of complications compared with MIS cases >270 minutes. Most MIS cases (88%) were completed in <270 minutes, while about 30% of abdominal myomectomies were completed in <90 minutes. In a meta-analysis comparing robotic myomectomy with the abdominal approach, Iavazzo and colleagues (6) showed a decrease in total complications and LOS despite an increased operative time of 85 minutes. This finding is also reflected in Griffin and colleagues’ (24) comparison demonstrating a longer operative time with roboticassisted MIS myomectomy (261.1 vs. 124.8 minutes; P< .001) but quicker recovery, shorter LOS, and faster return to work than with an abdominal approach. Barakat et al. also showed that both robotic and traditional laparoscopy had longer surgical times (181 and 155 minutes, respectively) compared with abdominal myomectomy (126 minutes) but decreased LOS and blood loss (16). In our cohort, younger age, African American race, higher BMI, an MIS approach, and specimen weight >250 g and/or more than five intramural fibroids removed were the strongest predictors of operative time. Notably, patients who had an abdominal myomectomy were significantly more likely to be African American, were younger, and had a higher mean BMI. These characteristics could signify a higher surgical complexity, prompting the surgeon to select an abdominal approach, and/or a difference in access to surgeons trained to complete MIS myomectomies for patients with the above characteristics. MIS myomectomy is a technically challenging surgery requiring advanced MIS skills such as suturing and extraction (21, 22). Currently, the Accreditation Council on Graduate Medical Education does not require a minimum number of MIS myomectomies for graduation VOL. - NO. - / - 2019
Fertility and Sterility®
TABLE 4 Association of MIS versus abdominal procedures with the composite morbidity outcome. MIS time (min) 90 to <180 180 to <270 R270 180 to <270 R270 R270 <90 <90 <90 90 to <180 90 to <180 180 to <270
Abdominal time (min)
No. of cases
Adjusted odds ratio for MIS vs. abdominal (95% CI)
P value
<90 <90 <90 90 to <180 90 to <180 180 to <270 90 to <180 180 to <270 R270 180 to <270 R270 R270
4,207 3,387 2,670 4,912 4,195 1,629 4,389 1,823 1,000 3,166 2,343 1,523
0.70 (0.53–0.93) 1.03 (0.76–1.38) 2.30 (1.69–3.13) 0.44 (0.35–0.57) 0.98 (0.75–1.27) 0.42 (0.32–0.56) 0.16 (0.09–0.27) 0.08 (0.04–0.13) 0.03 (0.01–0.05) 0.14 (0.11–0.18) 0.05 (0.03–0.07) 0.06 (0.04–0.09)
.0143a .8685 < .0001b < .0001a .8652 < .0001a < .0001 < .0001 < .0001 < .0001 < .0001 < .0001
Note: Adjusted for race, specimen weight > 250 g and/or more than five intramural fibroids removed, HTN, age, BMI, and hematocrit. a Lower odds of composite morbidity outcome for longer minimally invasive myomectomy compared with shorter abdominal myomectomy. b Higher odds of composite morbidity outcome for longer minimally invasive compared with shorter abdominal myomectomy. Vargas. Impact of operative on outcomes for myomectomy . Fertil Steril 2019.
from residency in obstetrics and gynecology (25). It is possible that many surgeons performing MIS myomectomy in the United States are subspecialty trained and that access is limited to these procedures. Patients who underwent a conversion were more likely to have these characteristics as well, which would seem to support that these factors (African American, weight, higher BMI, and specimen weight >250 g and/or more than five intramural fibroids removed) are associated with surgical complexity. These factors were adjusted for in the multivariate logistic regression analysis comparing outcomes in MIS and abdominal cases by operative times. Other reports have similarly shown specimen weight as a predictor of operative time (23) and longer operative time, number and/or weight of fibroid, and intraoperative hemorrhage as factors associated with conversion (26). Of importance is that operative time was linearly associated with surgical morbidity and that this did not differ by approach. This suggests that although MIS myomectomy had a lower morbidity overall, surgical efficiency is still of paramount importance. Surgical efficiency may be of particular importance for converted cases, which had the highest risk of operating time >270 minutes (23.1% for conversions vs. 3.3% for abdominal and 11.9% for MIS myomectomy). Morbidity was highest for conversions in our cohort, and it is possible that an earlier decision to convert the case could have decreased operative time and minimized morbidity. Modifiable factors that have been shown to be associated with surgical efficiency in general include surgeon and hospital volume (27–31). The surgical factors shown to be associated with conversion were previously mentioned (operative time, specimen weight, and intraoperative hemorrhage). More data are needed to guide surgeons about the appropriate time for conversion in myomectomy. Based on the results in our cohort, there is no clear time point at which the MIS approach to myomectomy becomes more morbid than the abdominal approach. Even when MIS cases were very long (>270 minutes) there was a protective effect against morbidity compared with shorter abdominal myomectomies (180–270 minutes). In the case when MIS VOL. - NO. - / - 2019
myomectomy had higher morbidity than abdominal myomectomy (>270 vs. <90 minutes), it is possible that these cases did not represent similar complexity or that surgeon experience was a factor. Unfortunately, surgical complexity could not be evaluated beyond specimen weight >250 g and/or greater than five intramural fibroids removed in our cohort. Neither complexity scores nor surgeon experience is a variable in the ACS NSQIP database. Our results for improved outcomes for minimally invasive myomectomy should therefore be cautiously interpreted with these limitations in mind. The authors recognize several limitations of this work due to the retrospective study design, such as the possibility of selection bias and misclassification. We used a rigorously managed database and adjusted for potential confounders identified in univariate analysis to minimize these effects. However, as previously mentioned, the ACS NSQIP database does not include variables for surgical complexity such as final specimen weight, prior surgeries, or a surgical complexity score. A cutoff of 250 g or more than five fibroids removed may not be detailed enough to truly represent surgical complexity in myomectomy cases. The outcomes between conventional MIS and robotic procedures could not be differentiated given that they have the same CPT codes. Surgeon volume or expertise is not accounted for within the database nor is unintended visceral injury at the time of surgery. These factors could all significantly impact the duration of surgery. It is therefore likely that variation in complexity and intraoperative complications were not completely accounted for in our analyses. Indication for surgery was not included for analysis in this study, and this factor could have potentially impacted the route and duration of surgery. Additionally, long-term surgical outcomes such as fibroid recurrence, rates of conception, and uterine rupture could not be evaluated due to the limited follow-up of 30 days within the ACS NSQIP database. The study is strengthened by the data obtained from a blinded, risk-adjusted, case-mix-adjusted database that reports outcomes from multiple institutions across the country. Furthermore, the sample size is large and from a 5
ORIGINAL ARTICLE: REPRODUCTIVE SURGERY nationwide cohort representing a number of hospitals and surgeons.
11.
Conclusions
12.
In conclusion, our results showed that operative time was predictive of complications for both minimally invasive and abdominal myomectomies. Despite a longer mean operative time of approximately 40 minutes, MIS myomectomies had lower odds of complications overall. When stratified by operative time, longer MIS myomectomies generally had lower odds of morbidity when compared with more expeditiously completed abdominal myomectomies. Surgeons should be encouraged to consider both the clinical implications of operative time and surgical approach on morbidity. An MIS approach can be offered for many patients undergoing myomectomy, and preparation to increase surgical efficiency should be prioritized for either approach. Future study should focus on interventions to minimize surgeon-dependent factors that increase operative time, identifying patients at risk for prolonged operative time, enhancing access to surgeons offering advanced surgical techniques, and optimizing training in the surgical approaches that maximize clinical benefit. Acknowledgments: 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 the statistical validity of the data analysis or the conclusions derived by the authors.
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
6
Downes E, Sikirica V, Gilabert-Estelles J, Bolge SC, Dodd SL, Maroulis C, et al. The burden of uterine fibroids in five European countries. Eur J Obstet Gynecol Reprod Biol 2010;152:96–102. Cardozo ER, Clark AD, Banks NK, Henne MB, Stegmann BJ, Segars JH. The estimated annual cost of uterine leiomyomata in the United States. Am J Obstet Gynecol 2012;206:211.e1–9. Cohen SL, Vitonis AF, Einarsson JI. Updated hysterectomy surveillance and factors associated with minimally invasive hysterectomy. JSLS 2014;18, e2014.00096. Wechter ME, Stewart EA, Myers ER, Kho RM, Wu JM. Leiomyoma-related hospitalization and surgery: prevalence and predicted growth based on population trends. Am J Obstet Gynecol 2011;205:492.e1–5. Bhave Chittawar P, Franik S, Pouwer AW, Farquhar C. Minimally invasive surgical techniques versus open myomectomy for uterine fibroids. Cochrane Database Syst Rev 2014;10:CD004638. Iavazzo C, Mamais I, Gkegkes ID. Robotic assisted vs laparoscopic and/or open myomectomy: systematic review and meta-analysis of the clinical evidence. Arch Gynecol Obstet 2016;294:5–17. Lake AG, McPencow AM, Dick-Biascoechea MA, Martin DK, Erekson EA. Surgical site infection after hysterectomy. Am J Obstet Gynecol 2013;209: 490.e1–9. Barber EL, Neubauer NL, Gossett DR. Risk of venous thromboembolism in abdominal versus minimally invasive hysterectomy for benign conditions. Am J Obstet Gynecol 2015;212:609.e1–7. Catanzarite T, Saha S, Pilecki MA, Kim JY, Milad M. The effect of operative time on perioperative morbidity after laparoscopic hysterectomy. Obstet Gynecol 2014;123(Suppl 1):123S. American College of Surgeons. American College of Surgeons National Surgical Quality Improvement Program user guide. Chicago: American College of Surgeons; 2016.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23. 24.
25.
26.
27.
28. 29.
30.
31.
von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. Int J Surg 2014;12:1495–9. Ko CY, Hall BL, Hart AJ, Cohen ME, Hoyt DB. The American College of Surgeons National Surgical Quality Improvement Program: achieving better and safer surgery. Jt Comm J Qual Patient Saf 2015;41:199–204. Henderson WG, Daley J. Design and statistical methodology of the National Surgical Quality Improvement Program: why is it what it is? Am J Surg 2009; 198:S19–27. Cohen ME, Ko CY, Bilimoria KY, Zhou L, Huffman K, Wang X, et al. Optimizing ACS NSQIP modeling for evaluation of surgical quality and risk: patient risk adjustment, procedure mix adjustment, shrinkage adjustment, and surgical focus. J Am Coll Surg 2013;217:336–46.e1. Ingraham AM, Richards KE, Hall BL, Ko CY. Quality improvement in surgery: the American College of Surgeons National Surgical Quality Improvement Program approach. Adv Surg 2010;44:251–67. Barakat EE, Bedaiwy MA, Zimberg S, Nutter B, Nosseir M, Falcone T. Roboticassisted, laparoscopic, and abdominal myomectomy: a comparison of surgical outcomes. Obstet Gynecol 2011;117:256–65. Buckley VA, Nesbitt-Hawes EM, Atkinson P, Won HR, Deans R, Burton A, et al. Laparoscopic myomectomy: clinical outcomes and comparative evidence. J Minim Invasive Gynecol 2015;22:11–25. Gobern JM, Rosemeyer CJ, Barter JF, Steren AJ. Comparison of robotic, laparoscopic, and abdominal myomectomy in a community hospital. JSLS 2013; 17:116–20. Kalogiannidis I, Prapas N, Xiromeritis P, Prapas Y. Laparoscopically assisted myomectomy versus abdominal myomectomy in short-term outcomes: a prospective study. Arch Gynecol Obstet 2010;281:865–70. Nash K, Feinglass J, Zei C, Lu G, Mengesha B, Lewicky-Gaupp C, et al. Robotic-assisted laparoscopic myomectomy versus abdominal myomectomy: a comparative analysis of surgical outcomes and costs. Arch Gynecol Obstet 2012;285:435–40. Sangha R, Eisenstein DI, George A, Munkarah A, Wegienka G. Surgical outcomes for robotic-assisted laparoscopic myomectomy compared to abdominal myomectomy. J Robotic Surg 2010;4:229–33. Vargas MV, Moawad GN, Sievers C, Opoku-Anane J, Marfori CQ, Tyan P, et al. Feasibility, safety, and prediction of complications for minimally invasive myomectomy in women with large and numerous myomata. J Minim Invasive Gynecol 2017;24:315–22. Sinha R, Hegde A, Warty N, Patil N. Laparoscopic excision of very large myomas. J Am Assoc Gynecol Laparosc 2003;10:461–8. Griffin L, Feinglass J, Garrett A, Henson A, Cohen L, Chaudhari A, et al. Postoperative outcomes after robotic versus abdominal myomectomy. JSLS 2013;17:407–13. Accreditation Council of Graduate Medical Education. Obstetrics and gynecology case logs: National data report 2016–2017. https://www.acgme.org/acgmeweb/tabid/274/DataCollectionSystems/ResidentCaseLogSystem/CaseLogsStatisticalReports.aspx. Sandberg EM, Cohen SL, Jansen FW, Einarsson JI. Analysis of risk factors for intraoperative conversion of laparoscopic myomectomy. J Minim Invasive Gynecol 2016;23:352–7. 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;25:2288–95. Vree FEM, Cohen SL, Chavan N, Einarsson JI. The impact of surgeon volume on perioperative outcomes in hysterectomy. JSLS 2014;18:174–81. Perri JL, Nolan BW, Goodney PP, DeMartino RR, Brooke BS, Arya S, et al. Factors affecting operative time and outcome of carotid endarterectomy in the Vascular Quality Initiative. J Vasc Surg 2017;66:1100–8. Maruthappu M, Duclos A, Lipsitz SR, Orgill D, Carty MJ. Surgical learning curves and operative efficiency: a cross-specialty observational study. BMJ Open 2015;5, e006679,2014-006679. Amato L, Fusco D, Acampora A, Bontempi K, Rosa AC, Colais P, et al. Volume and health outcomes: evidence from systematic reviews and from evaluation of Italian hospital data. Epidemiol Prev 2017;41:1– 128.
VOL. - NO. - / - 2019
Fertility and Sterility®
SUPPLEMENTAL FIGURE 1
Outline of cases analyzed. Vargas. Impact of operative on outcomes for myomectomy . Fertil Steril 2019.
VOL. - NO. - / - 2019
6.e1
The association of operative time with outcomes in minimally invasive and abdominal myomectomy Maria V. Vargas, M.D., M.S.,a Kathryn Denny Larson, MD,b Andrew Sparks, M.S.,c Samantha L. Margulies, M.D.,d Cherie Q. Marfori, M.D.,a Gaby Moawad, M.D.,a and Richard L. Amdur, Ph.D.c a
Department of Obstetrics and Gynecology, George Washington University School of Medicine and Health Sciences; School of Medicine and Health Sciences, George Washington University; and c Department of Surgery, George Washington University School of Medicine and Health Sciences, Washington, DC; and d Department of Obstetrics, Gynecology and Reproductive Health Sciences, Yale Medicine, New Haven, Connecticut b
Objective(s): To determine the association of operative time (ORT) with perioperative morbidity and whether there is an ORT at which minimally invasive myomectomy becomes inferior to laparotomy. Design: Retrospective cohort study. Setting: Not applicable. Patient(s): Myomectomy cases identified by CPT code from 2005 to 2016. Intervention(s): Cases were stratified and analyzed by surgical approach and 90-minute intervals. Main Outcome Measure(s): Thirty-day postoperative morbidity. Result(s): A total of 11,709 myomectomies were identified; 4,673 (39.9%) were minimally invasive, 6,997 (59.8%) were abdominal, and 39 (0.3%) were conversions. The incidence of complications significantly increased with ORT. After adjusting for confounders, mean ORT in minutes (95% confidence interval) was 113 (111–115) for abdominal, 156 (153–159) for minimally invasive, and 172 (148–200) for conversions. Despite shorter ORT, morbidity was greater in abdominal cases (16% vs. 5.7%), with the highest rates in converted cases (20.5%). The minimally invasive approach in general had lower odds of complications (odds ratio, 0.23; 95% confidence interval, 0.19–0.26). However, when minimally invasive surgery ORT reached R 270 minutes, the odds of a composite complication variable increased compared with abdominal cases <90 minutes (odds ratio, 2.30; 95% confidence interval, 1.69–3.13). Of minimally invasive cases, 88% were completed in <270 minutes. Conclusion(s): ORT was predictive of complications for both minimally invasive and abdominal myomectomies. Despite longer ORTs, minimally invasive procedures generally had superior 30-day outcomes up to 270 minutes. Careful patient counseling and preparation to increase surgical efficiency should be prioritized for either approach. (Fertil SterilÒ 2019;-:-–-. Ó2019 by American Society for Reproductive Medicine.) Key Words: Myomectomy, minimally invasive gynecologic surgery, laparoscopy, laparotomy, outcomes Discuss: You can discuss this article with its authors and other readers at https://www.fertstertdialog.com/users/16110-fertilityand-sterility/posts/43850-27172
U
terine fibroids are very common in reproductive-age women and have a significant societal impact, including negatively affecting quality of life and work productivity (1). In the United States, the estimated
annual direct costs related to medical care for fibroids, including surgery, hospital admissions, outpatient visits, and medications, is $4.1–9.4 billion (2). Although medical options exist, the management of large or symptom-
Received October 23, 2018; revised and accepted February 19, 2019. M.V.V. has nothing to disclose. K.D.L. has nothing to disclose. A.S. has nothing to disclose. S.L.M. has nothing to disclose. C.Q.M. has nothing to disclose. G.M. is a speaker for Intuitive Surgical. R.L.A. has nothing to disclose. The results of this study were presented at AAGL Global Congress on Minimally Invasive Gynecology, Orlando, Florida, November 2016; and the Annual Meeting of the Society for Reproductive Investigation, Orlando, Florida, March 2017. Reprint requests: Maria Victoria Vargas, M.D., 2150 Pennsylvania Avenue, NW, Washington, DC, 20037 (E-mail: [email protected]). Fertility and Sterility® Vol. -, No. -, - 2019 0015-0282/$36.00 Copyright ©2019 Published by Elsevier Inc. on behalf of the American Society for Reproductive Medicine https://doi.org/10.1016/j.fertnstert.2019.02.020 VOL. - NO. - / - 2019
atic fibroids has traditionally been surgical, and this remains the mainstay of treatment. In the United States, almost half of all hysterectomies are performed as the result of uterine fibroids (3). Myomectomy is the only uterus-sparing surgical management option and is commonly performed in women who wish to preserve reproductive potential (4). Myomectomies are increasingly being performed via minimally invasive surgery (MIS), including laparoscopy and robotic-assisted laparoscopy. MIS has proven clinical benefits over traditional laparotomy for myomectomy, including a faster return to 1
ORIGINAL ARTICLE: REPRODUCTIVE SURGERY normal activities, fewer febrile episodes, shorter hospital stay, lower intraoperative blood loss, fewer wound complications, and less postoperative pain (5, 6). Despite the established benefits of minimally invasive myomectomy, longer operative time is common with the use of laparoscopic and robotic-assisted laparoscopy compared with laparotomy (5). Longer operative times in minimally invasive approaches and laparotomy have been associated with adverse outcomes, such as surgical site infection (7) and venous thromboembolisms (8). An analysis of laparoscopic hysterectomy showed that at 4 hours, perioperative morbidity rises significantly (9). However, such an operative time has not been defined for myomectomy. Limited evidence exists to guide surgeons in both identifying patients at risk of excessive operating time and determining whether the at-risk patient may then benefit from an alternative surgical approach that may be more expeditious. The American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) was established to measure and improve outcomes across surgical specialties. The program captures over 150 patient variables from participating institutions across the United States. These variables include preoperative, intraoperative, and 30-day postoperative outcomes for patients undergoing major surgical procedures (10). The objective of this study is to use data from ACS NSQIP database to identify predictors of prolonged operative time for myomectomy, to compare perioperative outcomes between abdominal and MIS myomectomy by operative time, and to determine whether there is an operative time at which outcomes for MIS myomectomy become inferior to abdominal myomectomy.
MATERIALS AND METHODS Using the ACS NSQIP database, we conducted a retrospective cohort analysis of women undergoing myomectomy for benign indications from 2005 to 2016. The study was exempt from review by the George Washington University Institutional Review Board due to the use of a deidentified database. The methods for this study are reported according to ‘‘The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement’’ (11).
Source of Data The method of data collection through the ACS NSQIP database has been previously described and validated (10,12– 15). Surgical clinical reviewers at each site collect data from surgical and medical records. The number of cases per site is determined by site surgical volume. Exclusion criteria include minor cases, patients <18 years of age, and return to the operating room (OR) related to a previous procedure. The database contains only deidentified data and uses a sampling procedure intended to prevent bias in cases chosen. Selected cases have outcomes recorded for 30 days after surgery. The ACS NSQIP audits participating sites regularly for procedural compliance. 2
Case Selection We selected all cases from the ACS NSQIP database years 2005–2016 that had any of the following principal CPT codes: 58140, 58146 (abdominal), 58545, 58546 (MIS). Myomectomies with specimen weight >250 g and/or more than five intramural fibroids removed were indicated by codes 58146 or 58546. Emergent cases (n ¼ 108), those with disseminated cancer (n ¼ 6), and those with operative time <20 minutes (n ¼ 65) were excluded.
Variables Outcomes included superficial surgical site infection, deep wound infection, organ space infection, dehiscence (wound complications), pneumonia, reintubation, prolonged intubation (pulmonary complications), deep vein thrombosis or pulmonary embolism (clotting), myocardial infarction or cardiac arrest (cardiac), acute kidney injury or dialysis (renal), sepsis or septic shock (sepsis events), bleeding, urinary tract infection (UTI), return to OR, and hospital length of stay (LOS) >3 days. A composite complication outcome was defined as wound, pulmonary, clotting, sepsis, UTI, bleeding, return to OR, cardiac, or renal events. Predictors analyzed for significance included age, race, body mass index (BMI), hematocrit, diabetes mellitus (DM), hypertension (HTN), and specimen weight >250 g and/or five or more intramural fibroids removed.
Data Analysis Variable distributions were examined for outliers and skewed distributions. These were either log-transformed or recoded into quartiles. To examine associations between treatment variables (procedure: MIS/abdominal/conversion; operative time: <90 minutes, 90 to <180 minutes, 180 to <270 minutes, or R270 minutes) and categorical variables, we used c2 or Fisher's exact test. Associations with continuous variables were examined using t-tests or analysis of variance. First, we identified presurgery variables that were associated with operative time and with procedure. Second, to determine which variables had independent associations with operative time, we used a general linear model to predict natural log of operative time, including as predictors any pretreatment variables that could function as confounds, having at least a trend level of association (P< .10) either with procedure or with operative time. Third, to determine whether there was a single cut point at which increasing operative time was associated with worse outcomes within procedures, we first investigated whether the odds of having any complication as a function of operative time differed for MIS versus abdominal surgery using a logistic regression model. This model was also adjusted for other presurgery variables that could function as confounds. We then tested logistic regression models adjusted for the same set of covariates listed above, for longer MIS versus shorter abdominal procedures. SAS version 9.4 (SAS Institute) was used for all data analysis, and P< .05 was considered statistically significant. VOL. - NO. - / - 2019
Fertility and Sterility®
RESULTS There were 11,709 cases identified that met the inclusion criteria (Supplemental Fig. 1). Mean (SD) age was 36.9 6.5, mean (SD) BMI was 28.8 6.9, 4,626 patients (39.5%) were African American, and 3,436 (29.3%) were Caucasian. MIS procedures were used in 4,673 cases (39.9%) and abdominal myomectomy in 6,997 cases (59.8%); 39 (0.3%) had CPT codes for both, indicating conversion. Mean (SD) operative time was 145.9 77.5 minutes, but the distribution was positively skewed. Median operative time was 131 minutes (interquartile range, 90–184 minutes). Presurgery patient variables significantly associated with longer operative time included younger age, African American race, higher BMI, lower preoperative hematocrit, and the presences of >250 g of fibroids and/or more than five intramural fibroids (Supplemental Table 1). Procedure type was also significantly associated with age, race, BMI, hematocrit, HTN, and >250 g of fibroids and/or more than five intramural fibroids removed (Supplemental Table 2). Compared with patients who underwent abdominal surgery, those who underwent MIS myomectomy were older, were more likely to be Caucasian, had lower BMI, had higher hematocrit, and were less likely to have HTN or >250 g of fibroids and/or more than five intramural fibroids removed. In general, operative time was longer for MIS procedures compared with abdominal procedures (Table 1). There was a positive association between operative time and morbidity outcomes including wound, pulmonary, clotting, bleeding, and renal complications as well as return to the OR. There was also a positive association with operative time and hospital LOS >3 days (Table 2). The composite complication outcome had a linear association with operative time (P< .0001). Prolonged operative time was significantly more common in MIS procedures, whether this was defined as >3 hours (38.5% vs. 18%; P< .0001) or >4 hours (18.1% vs. 6%; P< .0001). However, in general, abdominal procedures had higher rates of complications than MIS procedures (Table 3). This was true for pulmonary complications, sepsis, UTI, and bleeding. Hospital LOS > 3 days was more frequent for abdominal procedures. The composite morbidity outcome was more likely in abdominal than in MIS procedures (16% vs. 5.7%) but was most likely in conversion cases (20.5%; P< .0001). The general linear model predicting the natural log of operative time was significant (F ¼ 226.44, P< .0001) with R2 ¼ 0.15, indicating a moderate association between the pre-
dictors and outcome. Significant independent predictors of operative time included procedure type, >250 g of fibroids and/or more than five intramural fibroids removed, race, age, hematocrit, and BMI (Supplemental Table 3). To compare outcomes for MIS versus abdominal procedures at various operative times, it was necessary to control for the effects of race, >250 g of fibroids and/or more than five intramural fibroids removed, age, BMI, presurgery hematocrit, and HTN. After adjusting for these covariates, the mean operative times were 113 minutes (95% confidence interval [CI], 111–115) for abdominal, 156 minutes (95% CI, 153– 159) for MIS, and 172 minutes (95% CI, 148–200) for conversions (Supplemental Table 3). The logistic regression model predicting any complication was significant (likelihood ratio c2 ¼ 1,035.08 P< .0001) with pseudo R2 ¼ .09 and with c ¼ .75 indicating moderately good discrimination. The overall operative time effect was significant (P< .0001). Compared with the shortest operative times (<90 minutes), the odds of reaching the composite morbidity endpoint were over two times higher, over four times higher, and over 11 times higher in surgeries lasting 90 to <180, 180 to <270, and R270 minutes, respectively. Compared with abdominal myomectomy, the adjusted odds ratio for the composite complications variable for MIS myomectomy was 0.23 (95% CI, 0.19–0.26; P< .0001) but was not significantly different for conversion cases compared with abdominal surgery (Supplemental Table 4). In comparing longer MIS versus shorter abdominal cases, we found several comparisons in which MIS procedures continued to have lower odds of complications (MIS 90 to <180 vs. abdominal <90; MIS 180 to <270 minutes vs. abdominal 90 to <180 minutes; MIS R270 minutes vs. abdominal 180 to <270 minutes), and one situation (MIS R 270 minutes vs. abdominal <90 minutes) where abdominal surgery had a significantly lower odds of complications (Table 4). All operation time combinations where the MIS procedure was shorter yielded significantly lower odds of complications for MIS myomectomy.
DISCUSSION This retrospective cohort study provides comparative surgical results for MIS versus abdominal myomectomy procedures stratified by operative time. Consistent with previous studies, we found improved outcomes for MIS approaches to myomectomy when compared with abdominal myomectomy
TABLE 1 Association of procedure with operative time (in minutes). Operative time, min (%) Procedure
n (column %)
<90 (24.4)
90 to <180 (48.8)
180 to <270 (19.9)
‡270 (6.8)
P value
MIS Abdominal Conversion
4,673 (39.9) 6,997 (59.8) 39 (0.3)
756 (26.4) 2,100 (73.4) 4 (0.1)
2,090 (36.5) 3,616 (63.2) 13 (0.2)
1,270 (54.4) 1,050 (45.0) 13 (0.6)
557 (69.9) 231 (30.0) 9 (1.1)
< .0001
Note: Data are shown as n (row %), except in the n column, where column % is shown. Vargas. Impact of operative on outcomes for myomectomy . Fertil Steril 2019.
VOL. - NO. - / - 2019
3
ORIGINAL ARTICLE: REPRODUCTIVE SURGERY
TABLE 2 Outcomes by operative time (in minutes). Operative time, min (%) Variable
<90 (n [ 2,860)
90 to <180 (n [ 5,719)
180 to <270 (n [ 2,333)
‡270 (n [ 797)
P value
149 (5.2) 29 (1.0) 4 (0.1) 2 (0.1) 7 (0.2) 30 (1.1) 92 (3.2) 15 (0.5) 0 0 106 (3.7)
628 (11.0) 77 (1.4) 11 (0.2) 12 (0.2) 26 (0.5) 48 (0.8) 484 (8.5) 47 (0.8) 0 0 328 (5.7)
407 (17.5) 39 (1.7) 11 (0.5) 16 (0.7) 8 (0.3) 29 (1.2) 342 (14.7) 20 (0.9) 1 (0.1) 3 (0.1) 186 (8.0)
211 (26.5) 18 (2.3) 6 (0.8) 7 (0.9) 5 (0.6) 10 (1.3) 185 (23.2) 14 (1.8) 0 1 (0.1) 82 (10.3)
< .0001 .03 .004 < .0001 .33 .32 < .0001 .0084 .27 .008 < .0001
Any complication Wound complications Pulmonary Clotting Sepsis UTI Bleeding Return to OR Cardiac Renal Hospital LOS > 3 d
Note: Wound complications included superficial surgical site infection, deep wound infection, organ space infection, and dehiscence. Pulmonary complications included pneumonia, reintubation, and prolonged intubation. Clotting complications included deep vein thrombosis or pulmonary embolism. Cardiac complications included myocardial infarction or cardiac arrest. Renal complications included acute kidney injury or dialysis. Any complication included wound, pulmonary, clotting, sepsis, UTI, bleeding, return to OR, cardiac, or renal events. Vargas. Impact of operative on outcomes for myomectomy . Fertil Steril 2019.
(5, 6, 16–21) and worsened outcomes with increasing operative time for both MIS and abdominal approaches (22– 24). MIS myomectomy was associated with lower incidence of the composite complications variable and a 77% lower odds of having any complication compared with abdominal myomectomy (95% CI, 74%–81% lower). Specifically, MIS myomectomy had lower rates of pulmonary complications, sepsis, UTI, and bleeding. Our study also supports findings from prior work showing longer operative times for MIS cases (5, 6, 17, 24), which were about 40 minutes longer on average than abdominal cases in our cohort.
TABLE 3 Outcomes by procedure. Variable
MIS
Abdominal Conversion
Any complication 265 (5.7) 1,122 (16.0) Wound 51 (1.1) 112 (1.6) complications Pulmonary 5 (0.1) 27 (0.4) Clotting 10 (0.2) 27 (0.4) Sepsis 8 (0.2) 38 (0.5) UTI 30 (0.6) 85 (1.2) Bleeding 165 (3.5) 932 (13.3) Return to OR 28 (0.6) 68 (1.0) Cardiac 0 1 (0.01) Renal 0 (0) 4 (0.1) Hospital LOS 44 (0.9) 657 (9.4) >3 d Surgery time, h >3 1,799 (38.5) 1,257 (18.0) >4 844 (18.1) 416 (6.0)
P value
8 (20.5) 0 (0)
< .0001 .054
0 (0) 0 (0) 0 (0) 2 (5.1) 6 (15.4) 0 (0) 0 0 (0) 1 (2.6)
.017 .25 .007 .0003 < .0001 .08 .99 .17 < .0001
21 (53.9) 11 (28.2)
< .0001 < .0001
Note: Data in parentheses are percentages. Wound complications included superficial surgical site infection, deep wound infection, organ space infection, and dehiscence. Pulmonary complications included pneumonia, reintubation, and prolonged intubation. Clotting complications included deep vein thrombosis or pulmonary embolism. Cardiac complications included myocardial infarction or cardiac arrest. Renal complications included acute kidney injury or dialysis. Any complication included wound, pulmonary, clotting, sepsis, UTI, bleeding, return to OR, cardiac, or renal events. Vargas. Impact of operative on outcomes for myomectomy . Fertil Steril 2019.
4
When stratified by duration, longer MIS cases had lower odds of complications when compared with shorter abdominal cases in several scenarios (MIS 90 to <180 vs. abdominal <90; MIS 180 to <270 minutes vs. abdominal 90 to <180 minutes; MIS R270 minutes vs. abdominal 180 to <270 minutes). However, abdominal cases completed in <90 minutes had lower odds of complications compared with MIS cases >270 minutes. Most MIS cases (88%) were completed in <270 minutes, while about 30% of abdominal myomectomies were completed in <90 minutes. In a meta-analysis comparing robotic myomectomy with the abdominal approach, Iavazzo and colleagues (6) showed a decrease in total complications and LOS despite an increased operative time of 85 minutes. This finding is also reflected in Griffin and colleagues’ (24) comparison demonstrating a longer operative time with roboticassisted MIS myomectomy (261.1 vs. 124.8 minutes; P< .001) but quicker recovery, shorter LOS, and faster return to work than with an abdominal approach. Barakat et al. also showed that both robotic and traditional laparoscopy had longer surgical times (181 and 155 minutes, respectively) compared with abdominal myomectomy (126 minutes) but decreased LOS and blood loss (16). In our cohort, younger age, African American race, higher BMI, an MIS approach, and specimen weight >250 g and/or more than five intramural fibroids removed were the strongest predictors of operative time. Notably, patients who had an abdominal myomectomy were significantly more likely to be African American, were younger, and had a higher mean BMI. These characteristics could signify a higher surgical complexity, prompting the surgeon to select an abdominal approach, and/or a difference in access to surgeons trained to complete MIS myomectomies for patients with the above characteristics. MIS myomectomy is a technically challenging surgery requiring advanced MIS skills such as suturing and extraction (21, 22). Currently, the Accreditation Council on Graduate Medical Education does not require a minimum number of MIS myomectomies for graduation VOL. - NO. - / - 2019
Fertility and Sterility®
TABLE 4 Association of MIS versus abdominal procedures with the composite morbidity outcome. MIS time (min) 90 to <180 180 to <270 R270 180 to <270 R270 R270 <90 <90 <90 90 to <180 90 to <180 180 to <270
Abdominal time (min)
No. of cases
Adjusted odds ratio for MIS vs. abdominal (95% CI)
P value
<90 <90 <90 90 to <180 90 to <180 180 to <270 90 to <180 180 to <270 R270 180 to <270 R270 R270
4,207 3,387 2,670 4,912 4,195 1,629 4,389 1,823 1,000 3,166 2,343 1,523
0.70 (0.53–0.93) 1.03 (0.76–1.38) 2.30 (1.69–3.13) 0.44 (0.35–0.57) 0.98 (0.75–1.27) 0.42 (0.32–0.56) 0.16 (0.09–0.27) 0.08 (0.04–0.13) 0.03 (0.01–0.05) 0.14 (0.11–0.18) 0.05 (0.03–0.07) 0.06 (0.04–0.09)
.0143a .8685 < .0001b < .0001a .8652 < .0001a < .0001 < .0001 < .0001 < .0001 < .0001 < .0001
Note: Adjusted for race, specimen weight > 250 g and/or more than five intramural fibroids removed, HTN, age, BMI, and hematocrit. a Lower odds of composite morbidity outcome for longer minimally invasive myomectomy compared with shorter abdominal myomectomy. b Higher odds of composite morbidity outcome for longer minimally invasive compared with shorter abdominal myomectomy. Vargas. Impact of operative on outcomes for myomectomy . Fertil Steril 2019.
from residency in obstetrics and gynecology (25). It is possible that many surgeons performing MIS myomectomy in the United States are subspecialty trained and that access is limited to these procedures. Patients who underwent a conversion were more likely to have these characteristics as well, which would seem to support that these factors (African American, weight, higher BMI, and specimen weight >250 g and/or more than five intramural fibroids removed) are associated with surgical complexity. These factors were adjusted for in the multivariate logistic regression analysis comparing outcomes in MIS and abdominal cases by operative times. Other reports have similarly shown specimen weight as a predictor of operative time (23) and longer operative time, number and/or weight of fibroid, and intraoperative hemorrhage as factors associated with conversion (26). Of importance is that operative time was linearly associated with surgical morbidity and that this did not differ by approach. This suggests that although MIS myomectomy had a lower morbidity overall, surgical efficiency is still of paramount importance. Surgical efficiency may be of particular importance for converted cases, which had the highest risk of operating time >270 minutes (23.1% for conversions vs. 3.3% for abdominal and 11.9% for MIS myomectomy). Morbidity was highest for conversions in our cohort, and it is possible that an earlier decision to convert the case could have decreased operative time and minimized morbidity. Modifiable factors that have been shown to be associated with surgical efficiency in general include surgeon and hospital volume (27–31). The surgical factors shown to be associated with conversion were previously mentioned (operative time, specimen weight, and intraoperative hemorrhage). More data are needed to guide surgeons about the appropriate time for conversion in myomectomy. Based on the results in our cohort, there is no clear time point at which the MIS approach to myomectomy becomes more morbid than the abdominal approach. Even when MIS cases were very long (>270 minutes) there was a protective effect against morbidity compared with shorter abdominal myomectomies (180–270 minutes). In the case when MIS VOL. - NO. - / - 2019
myomectomy had higher morbidity than abdominal myomectomy (>270 vs. <90 minutes), it is possible that these cases did not represent similar complexity or that surgeon experience was a factor. Unfortunately, surgical complexity could not be evaluated beyond specimen weight >250 g and/or greater than five intramural fibroids removed in our cohort. Neither complexity scores nor surgeon experience is a variable in the ACS NSQIP database. Our results for improved outcomes for minimally invasive myomectomy should therefore be cautiously interpreted with these limitations in mind. The authors recognize several limitations of this work due to the retrospective study design, such as the possibility of selection bias and misclassification. We used a rigorously managed database and adjusted for potential confounders identified in univariate analysis to minimize these effects. However, as previously mentioned, the ACS NSQIP database does not include variables for surgical complexity such as final specimen weight, prior surgeries, or a surgical complexity score. A cutoff of 250 g or more than five fibroids removed may not be detailed enough to truly represent surgical complexity in myomectomy cases. The outcomes between conventional MIS and robotic procedures could not be differentiated given that they have the same CPT codes. Surgeon volume or expertise is not accounted for within the database nor is unintended visceral injury at the time of surgery. These factors could all significantly impact the duration of surgery. It is therefore likely that variation in complexity and intraoperative complications were not completely accounted for in our analyses. Indication for surgery was not included for analysis in this study, and this factor could have potentially impacted the route and duration of surgery. Additionally, long-term surgical outcomes such as fibroid recurrence, rates of conception, and uterine rupture could not be evaluated due to the limited follow-up of 30 days within the ACS NSQIP database. The study is strengthened by the data obtained from a blinded, risk-adjusted, case-mix-adjusted database that reports outcomes from multiple institutions across the country. Furthermore, the sample size is large and from a 5
ORIGINAL ARTICLE: REPRODUCTIVE SURGERY nationwide cohort representing a number of hospitals and surgeons.
11.
Conclusions
12.
In conclusion, our results showed that operative time was predictive of complications for both minimally invasive and abdominal myomectomies. Despite a longer mean operative time of approximately 40 minutes, MIS myomectomies had lower odds of complications overall. When stratified by operative time, longer MIS myomectomies generally had lower odds of morbidity when compared with more expeditiously completed abdominal myomectomies. Surgeons should be encouraged to consider both the clinical implications of operative time and surgical approach on morbidity. An MIS approach can be offered for many patients undergoing myomectomy, and preparation to increase surgical efficiency should be prioritized for either approach. Future study should focus on interventions to minimize surgeon-dependent factors that increase operative time, identifying patients at risk for prolonged operative time, enhancing access to surgeons offering advanced surgical techniques, and optimizing training in the surgical approaches that maximize clinical benefit. Acknowledgments: 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 the statistical validity of the data analysis or the conclusions derived by the authors.
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
6
Downes E, Sikirica V, Gilabert-Estelles J, Bolge SC, Dodd SL, Maroulis C, et al. The burden of uterine fibroids in five European countries. Eur J Obstet Gynecol Reprod Biol 2010;152:96–102. Cardozo ER, Clark AD, Banks NK, Henne MB, Stegmann BJ, Segars JH. The estimated annual cost of uterine leiomyomata in the United States. Am J Obstet Gynecol 2012;206:211.e1–9. Cohen SL, Vitonis AF, Einarsson JI. Updated hysterectomy surveillance and factors associated with minimally invasive hysterectomy. JSLS 2014;18, e2014.00096. Wechter ME, Stewart EA, Myers ER, Kho RM, Wu JM. Leiomyoma-related hospitalization and surgery: prevalence and predicted growth based on population trends. Am J Obstet Gynecol 2011;205:492.e1–5. Bhave Chittawar P, Franik S, Pouwer AW, Farquhar C. Minimally invasive surgical techniques versus open myomectomy for uterine fibroids. Cochrane Database Syst Rev 2014;10:CD004638. Iavazzo C, Mamais I, Gkegkes ID. Robotic assisted vs laparoscopic and/or open myomectomy: systematic review and meta-analysis of the clinical evidence. Arch Gynecol Obstet 2016;294:5–17. Lake AG, McPencow AM, Dick-Biascoechea MA, Martin DK, Erekson EA. Surgical site infection after hysterectomy. Am J Obstet Gynecol 2013;209: 490.e1–9. Barber EL, Neubauer NL, Gossett DR. Risk of venous thromboembolism in abdominal versus minimally invasive hysterectomy for benign conditions. Am J Obstet Gynecol 2015;212:609.e1–7. Catanzarite T, Saha S, Pilecki MA, Kim JY, Milad M. The effect of operative time on perioperative morbidity after laparoscopic hysterectomy. Obstet Gynecol 2014;123(Suppl 1):123S. American College of Surgeons. American College of Surgeons National Surgical Quality Improvement Program user guide. Chicago: American College of Surgeons; 2016.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23. 24.
25.
26.
27.
28. 29.
30.
31.
von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. Int J Surg 2014;12:1495–9. Ko CY, Hall BL, Hart AJ, Cohen ME, Hoyt DB. The American College of Surgeons National Surgical Quality Improvement Program: achieving better and safer surgery. Jt Comm J Qual Patient Saf 2015;41:199–204. Henderson WG, Daley J. Design and statistical methodology of the National Surgical Quality Improvement Program: why is it what it is? Am J Surg 2009; 198:S19–27. Cohen ME, Ko CY, Bilimoria KY, Zhou L, Huffman K, Wang X, et al. Optimizing ACS NSQIP modeling for evaluation of surgical quality and risk: patient risk adjustment, procedure mix adjustment, shrinkage adjustment, and surgical focus. J Am Coll Surg 2013;217:336–46.e1. Ingraham AM, Richards KE, Hall BL, Ko CY. Quality improvement in surgery: the American College of Surgeons National Surgical Quality Improvement Program approach. Adv Surg 2010;44:251–67. Barakat EE, Bedaiwy MA, Zimberg S, Nutter B, Nosseir M, Falcone T. Roboticassisted, laparoscopic, and abdominal myomectomy: a comparison of surgical outcomes. Obstet Gynecol 2011;117:256–65. Buckley VA, Nesbitt-Hawes EM, Atkinson P, Won HR, Deans R, Burton A, et al. Laparoscopic myomectomy: clinical outcomes and comparative evidence. J Minim Invasive Gynecol 2015;22:11–25. Gobern JM, Rosemeyer CJ, Barter JF, Steren AJ. Comparison of robotic, laparoscopic, and abdominal myomectomy in a community hospital. JSLS 2013; 17:116–20. Kalogiannidis I, Prapas N, Xiromeritis P, Prapas Y. Laparoscopically assisted myomectomy versus abdominal myomectomy in short-term outcomes: a prospective study. Arch Gynecol Obstet 2010;281:865–70. Nash K, Feinglass J, Zei C, Lu G, Mengesha B, Lewicky-Gaupp C, et al. Robotic-assisted laparoscopic myomectomy versus abdominal myomectomy: a comparative analysis of surgical outcomes and costs. Arch Gynecol Obstet 2012;285:435–40. Sangha R, Eisenstein DI, George A, Munkarah A, Wegienka G. Surgical outcomes for robotic-assisted laparoscopic myomectomy compared to abdominal myomectomy. J Robotic Surg 2010;4:229–33. Vargas MV, Moawad GN, Sievers C, Opoku-Anane J, Marfori CQ, Tyan P, et al. Feasibility, safety, and prediction of complications for minimally invasive myomectomy in women with large and numerous myomata. J Minim Invasive Gynecol 2017;24:315–22. Sinha R, Hegde A, Warty N, Patil N. Laparoscopic excision of very large myomas. J Am Assoc Gynecol Laparosc 2003;10:461–8. Griffin L, Feinglass J, Garrett A, Henson A, Cohen L, Chaudhari A, et al. Postoperative outcomes after robotic versus abdominal myomectomy. JSLS 2013;17:407–13. Accreditation Council of Graduate Medical Education. Obstetrics and gynecology case logs: National data report 2016–2017. https://www.acgme.org/acgmeweb/tabid/274/DataCollectionSystems/ResidentCaseLogSystem/CaseLogsStatisticalReports.aspx. Sandberg EM, Cohen SL, Jansen FW, Einarsson JI. Analysis of risk factors for intraoperative conversion of laparoscopic myomectomy. J Minim Invasive Gynecol 2016;23:352–7. 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;25:2288–95. Vree FEM, Cohen SL, Chavan N, Einarsson JI. The impact of surgeon volume on perioperative outcomes in hysterectomy. JSLS 2014;18:174–81. Perri JL, Nolan BW, Goodney PP, DeMartino RR, Brooke BS, Arya S, et al. Factors affecting operative time and outcome of carotid endarterectomy in the Vascular Quality Initiative. J Vasc Surg 2017;66:1100–8. Maruthappu M, Duclos A, Lipsitz SR, Orgill D, Carty MJ. Surgical learning curves and operative efficiency: a cross-specialty observational study. BMJ Open 2015;5, e006679,2014-006679. Amato L, Fusco D, Acampora A, Bontempi K, Rosa AC, Colais P, et al. Volume and health outcomes: evidence from systematic reviews and from evaluation of Italian hospital data. Epidemiol Prev 2017;41:1– 128.
VOL. - NO. - / - 2019
Fertility and Sterility®
SUPPLEMENTAL FIGURE 1
Outline of cases analyzed. Vargas. Impact of operative on outcomes for myomectomy . Fertil Steril 2019.
VOL. - NO. - / - 2019
6.e1