Adoption of enhanced recovery after laparotomy in gynecologic oncology

Abstract

Introduction

Enhanced recovery after surgery (ERAS) pathways combine a comprehensive set of perioperative practices that have been demonstrated to hasten patient post-operative recovery. We aimed to evaluate the adoption of ERAS components and assess attitudes towards ERAS among gynecologic oncologists.

Methods

We developed and administered a cross-sectional survey of attending, fellow and resident physicians who were members of the Society of Gynecologic Oncology in January 2018. The Chi-square test was used to compare adherence to individual components of ERAS.

Results

There was a 23% survey response rate and we analyzed 289 responses. 76% were attending physicians, 59% were from academic institutions, and 64% were from institutions with an established ERAS pathway. Respondents from ERAS institutions were significantly more likely to adhere to recommendations regarding pre-operative fasting for liquids (ERAS: 51%, non-ERAS: 28%; p<0.001), carbohydrate loading (63% vs. 16%; p<0.001), intra-operative fluid management (78% vs. 32%; p<0.001), and extended duration of deep vein thrombosis prophylaxis for malignancy (69% vs.55%, p= 0.003). We found no difference in the use of mechanical bowel preparation, use of peritoneal drainage, or use of nasogastric tubes between ERAS and non-ERAS institutions. Nearly all respondents (92%) felt that ERAS pathways were safe.

Discussion

Practicing at an institution with an ERAS pathway increased adoption of many ERAS elements, however, adherence to certain guidelines remains highly variable. Use of bowel preparation, nasogastric tubes, and peritoneal drainage catheters remain common. Future work should identify barriers to implementation of ERAS and its components.

INTRODUCTION

Enhanced recovery pathways are a set of perioperative practices that seek to hasten patient recovery after surgery. An increasing body of evidence demonstrates improved surgical outcomes and cost savings among patients managed with enhanced recovery pathways across multiple surgical subspecialties, including colorectal surgery, general surgery, and urology^1,2,3. A 2014 systematic review of gynecologic-oncologists found that perioperative practices were highly variable, demonstrating a need for a formalized enhanced recovery pathways for patients undergoing gynecologic oncology surgery⁴. Observational studies of many individual ERAS components and nascent ERAS pathways within gynecologic surgery had demonstrated decreased complications, length of stay and cost^5-11. The Enhanced Recovery After Surgery Society (ERAS Society) undertook a systematic review of this literature that led to the publication of ERAS guidelines in gynecologic oncology in 2016^12,13. In March 2019, these were further revised and updated ERAS guidelines were published for gynecologic oncology¹⁴.

ERAS for gynecologic oncology includes specific recommendations for preoperative, intraoperative, and postoperative care. Preoperative ERAS recommendations include permitting oral intake of clear fluids until two hours prior to surgery, use of carbohydrate loading, and avoidance of preoperative bowel preparation. Intraoperative ERAS recommendations include deep vein thrombosis and antimicrobial prophylaxis, goal-directed intravenous fluid therapy, and maintenance of normothermia. Postoperative ERAS recommendations include deep vein thrombosis prophylaxis, initiation of regular diet within 24 hours, a multimodal approach to analgesia, avoidance of peritoneal drainage and nasogastric tubes, removal of urinary catheters within 24 hours, and early mobilization^12,13.

Since the release of the initial guidelines in 2016, adoption of ERAS practices have not been systematically assessed. This study aimed to compare adoption of ERAS perioperative management practices and attitudes towards ERAS pathways among practicing gynecologic oncologists in the United States from both ERAS and non-ERAS institutions.

METHODS

We developed a cross-sectional survey to assess perioperative management practices and to evaluate adoption of and attitudes towards ERAS pathways (Supplemental 1). The survey posed questions relating to pre-, intra-, and post-operative practices included in the 2016 ERAS guidelines for gynecologic oncology. The survey asked respondents whether their institution had an established ERAS pathway for laparotomy and asked about practice and demographic characteristics, as well as attitudes toward ERAS. Many questions were based on a five-point Likert scale (never/rarely/sometimes/often/always or strongly disagree/disagree/undecided/ agree/strongly agree). The survey was piloted and further refined based on feedback. The survey was emailed to members of the Society of Gynecologic Oncology in January 2018. Study data were collected and managed using REDCap¹⁵, an electronic data capture tool hosted at Harvard University.

We excluded respondents who did not practice in the United States and those who were unsure if their institution had an ERAS pathway for laparotomy. We further restricted the analysis to obstetrics and gynecology resident physicians, gynecologic oncology fellows and gynecologic oncology attending physicians. Because of their close involvement in the perioperative management of patients, inclusion of responses from resident and fellow trainees was deemed acceptable.

Data are presented as frequency with percent and median with interquartile range. For most analyses, the five-point Likert scales were collapsed into fewer categories. We compared data from respondents at institutions that have and have not implemented an ERAS pathway for laparotomy using the Chi-square or Fisher’s exact test. We analyzed data with Stata version 14 (StataCorp, College Station, TX); p values <0.05 were considered statistically significant. The institutional review board at Beth Israel Deaconess Medical Center approved this study.

RESULTS

We sent the survey to 1,612 Society of Gynecologic Oncology members and received 367 responses, yielding a 23% response rate. After restricting to respondents practicing in the United States and those who were obstetrics and gynecology resident physicians, gynecologic oncology fellows and gynecologic oncology attending physicians, 321 respondents remained. Of those, 12 were not sure if their institution had implemented an ERAS pathway for laparotomy and 20 did not answer the question, leaving 289 respondents in the final analysis. Of those, 186 (64%) were at institutions that had adopted an ERAS pathway for laparotomy and 103 (36%) were at institutions that had not. In both groups, the majority of respondents were gynecologic oncology attending physicians from academic institutions. The groups were similar with regard to geographic region of practice, whether they worked with residents and years in practice. Respondents from institutions with an ERAS pathway for laparotomy were more likely to work with fellows and had more physicians in their practice than those from institutions without an ERAS pathway (Table 1).

Table 1.

Respondent Characteristics

	ERAS Institution n=186	Non-ERAS Institution n=103
Levels of training
Obstetrics and gynecology resident	7 (4)	6 (6)
Gynecologic oncology fellow	38 (20)	10 (10)
Gynecologic oncology attending	141 (76)	87 (84)
U.S. Census Bureau Region of Practice
Northeast	44 (24)	32 (31)
South	65 (35)	28 (27)
Midwest	42 (23)	17 (17)
West	35 (19)	26 (25)
Type of institution
Academic	110 (59)	54 (52)
Private with academic affiliation	45 (24)	25 (24)
Other	31 (17)	24 (23)
Work with residents	157 (85)	88 (85)
Work with fellows	95 (51)	25 (24)
Years in practice	14 ±11	12 ±10
Number or physicians in practice	6 ±4	3 ±2

Respondent characteristics stratified by ERAS and non-ERAS institutions. Data presented as n (%) or mean ± standard deviation.

Adherence to ERAS bowel preparation guidelines was similar between respondents from ERAS and non-ERAS institutions for planned laparotomy, planned ovarian cancer debulking, and when there was concern for bowel surgery (all p≥0.23; Figure 1). For both groups, the frequency of adherence was highest for routine planned laparotomies and lowest when there was concern for bowel surgery. While use of mechanical and enema bowel preparations was similar, the ERAS group was more likely to use antibiotic bowel preparation (46%) than non-ERAS (28%; p=0.003). Respondents from ERAS institutions also were more likely to adhere to recommendations for preoperative fasting for liquids and carbohydrate loading (both p<0.001), but not for preoperative fasting for solids (p=0.14). Preoperative prophylaxis for deep vein thrombosis did not differ between the groups and exceeded 90%. Preoperative components of ERAS guidelines are shown in Table 2.

Figure 1.

Association of frequency of use of bowel preparation and use of peritoneal drainage for laparotomy with institution type.

Table 2.

Preoperative and Intraoperative Components of ERAS Guidelines for Laparotomy

	ERAS Institution n=186	Non-ERAS Institution n=103	p
Type of bowel preparation *
Mechanical	135 (73)	82 (80)	0.20
Antibiotic	86 (46)	29 (28)	0.003
Enema	39 (16)	15 (15)	0.87
None	22 (12)	14 (14)	0.66
Initiation of preoperative fasting, solids
≤6 hours	46 (25)	17 (17)	0.14
6-8 hours	69 (37)	37 (26)
≥8 hours	71 (38)	48 (47)
No response	0 (0)	1 (1)
Initiation of preoperative fasting, liquids
≤2 hours	97 (52)	29 (28)	<0.001
2-6 hours	57 (31)	24 (23)
>6 hours	29 (16)	49 (48)
No response	3 (2)	1 (1)
Use of carbohydrate loading
Yes	117 (63)	16 (16)	<0.001
No	65 (35)	85 (83)
Not sure/no response	4 (2)	2 (2)
Type of carbohydrate loading *
Gatorade®	62 (33)	8 (8)	<0.001
Clear fast	38 (20)	2 (2)	<0.001
GU/marathon carbohydrate	12 (6)	3 (3)	0.27
Other	19 (10)	4 (4)	0.07
Use of deep vein thrombosis prophylaxis
Yes	170 (91)	97 (94)	0.48
No	15 (8)	5 (5)
Not sure/no response	1 (1)	1 (1)
Mode of deep vein thrombosis prophylaxis
Chemical	29 (16)	20 (19)	0.71
Mechanical	140 (75)	77 (75)
Other	1 (1)	0 (0)
No response	16 (9)	6 (6)
Use of epidural
Rarely/never	104 (56)	61 (59)	0.004
Sometimes	28 (15)	25 (24)
Often/always	52 (28)	13 (13)
No response	2 (1)	4 (4)
Use of nerve block
Rarely/never	70 (38)	25 (24)	0.75
Sometimes	53 (28)	25 (24)
Often/always	60 (32)	32 (31)
No response	3 (2)	1 (1)
Intraoperative fluid management
By anesthesia	58 (31)	28 (27)	<0.001
Goal-directed, non-invasive	81 (44)	5 (5)
Goal-directed, invasive	6 (3)	0 (0)
None	21 (11)	54 (52)
Not sure/no response	20 (11)	16 (16)
Core body temperature measured continuously
Yes	147 (79)	80 (78)	0.93
No	6 (3)	3 (3)
Not sure/no response	33 (18)	20 (19)

Association of preoperative and intraoperative components of ERAS guidelines for laparotomy with institution type.

^*Percentages may sum to more than 100%, as participants could choose more than one option

Routine use of thoracic epidural or nerve block is not strongly recommended in the ERAS guidelines; however, a difference was noted in that respondents from ERAS institutions were more likely to report often or always using an epidural than those from non-ERAS institutions (28% vs. 13%, p=0.004), but there was no difference in the use of nerve blocks (p=0.75). Compared to the non-ERAS group, the ERAS group was more likely to adhere to guidelines for goal-directed invasive or non-invasive fluid management (47% vs. 5%, p<0.001); the groups were similar with respect to continuous monitoring of core body temperature (p=0.93).Table 2 shows intraoperative components of ERAS guidelines.

Despite guidelines recommending against postoperative nasogastric intubation, many respondents reported its use, particularly for small bowel resection and ligation of the short gastric vessels during omentectomy. This was similar between respondents from ERAS and non-ERAS institutions. There also were no differences in use of peritoneal drainage after lymphadenectomy, bowel resection, liver resection, splenectomy, or urologic procedures (Figure 1). Nearly all respondents in both groups reported adhering to guidelines for the duration of postoperative deep vein thrombosis prophylaxis for benign pathology. In contrast, adherence to guidelines for extended duration (1 month) of deep vein thrombosis prophylaxis for malignant pathology was higher in the ERAS group than non-ERAS group (p=0.003). Timing of termination of intravenous fluids was not uniform, but the ERAS group was more likely to adhere to ERAS guidelines (p=0.001). Respondents in the two groups reported similarly low adherence with other postoperative recommendations, such as initiation of regular diet within 24 hours after surgery, urinary catheter discontinuation, mobilization and use of a bowel regimen to hasten return of function. Interestingly, 46% of respondents from ERAS institutions and 36% from non-ERAS institutions reported use of chewing gum as a bowel regimen (p=0.14). Post-operative components of ERAS guidelines are shown in Table 3.

Table 3.

Postoperative Components of ERAS Guidelines for Laparotomy

	ERAS Institution n=186	Non-ERAS Institution n=103	p
Nasogastric intubation postoperatively
Small bowel resection	57 (31)	37 (36)	0.36
Large bowel resection	19 (10)	15 (15)	0.34
Splenectomy	10 (5)	8 (8)	0.45
Ligation of short gastric vessels	61 (33)	46 (45)	0.06
Never used	68 (37)	32 (31)	0.37
Postoperative deep vein thrombosis prophylaxis
Yes	182 (98)	99 (96)	0.25
No	4 (2)	2 (2)
No response	0 (0)	2 (2)
Postoperative deep vein thrombosis prophylaxis for benign pathology
Only during hospital stay	171 (92)	93 (90)	0.511
<1 month	7 (4)	3 (3)
=1 month	1 (1)	3 (3)
>1 month	1 (1)	0 (0)
No sure/no response	6 (3)	4 (4)
Postoperative deep vein thrombosis prophylaxis for cancerous pathology
Only during hospital stay	14 (8)	23 (22)	0.003
<1 month	34 (18)	13 (13)
=1 month	128 (69)	57 (55)
>1 month	6 (3)	6 (6)
Not sure/no response	4 (2)	4 (4)
Termination of intravenous fluids
<12 hours after surgery	27 (14)	6 (6)	0.001
12-24 hours	71 (38)	29 (28)
>24 hours	8 (4)	15 (15)
When tolerating liquid	80 (43)	52 (50)
No response	0 (0)	1 (1)
Initiation of regular diet
<24 hours	134 (72)	60 (58)	0.08
24-48 hours	44 (24)	34 (33)
>48 hours	6 (3)	7 (7)
No response	2 (1)	2 (2)
Urinary catheter discontinuation
<24 hours	143 (77)	73 (71)	0.43
>24 hours	24 (13)	19 (18)
No response	19 (10)	11 (11)
Mobilization
Day of surgery	125 (67)	59 (57)	0.18
Postoperative day 1 or after	60 (32)	43 (42)
No response	1 (1)	1 (1)
Use of a bowel regimen to hasten return of bowel function
Yes	124 (67)	59 (57)	0.23
No	61 (33)	43 (42)
No response	1 (1)	1 (1)

Association of postoperative components of ERAS guidelines for laparotomy with institution type.

A majority of respondents from ERAS and non-ERAS institutions agreed or strongly agreed that ERAS practices improve patient outcomes and are safe (Figure 2). Those from ERAS institutions were more likely to agree or strongly agree that ERAS practices improve patient satisfaction (77%) than those from non-ERAS institutions (66%). Approximately half of respondents from both groups were undecided as to whether ERAS practices increase complication rates or decrease unscheduled visits. Attitudes toward ERAS guidelines are shown in Figure 2.

Figure 2.

Association of attitudes towards ERAS and its components with institution type.

DISCUSSION

Perioperative practices are both highly individual and deeply held personal habits among surgeons. They trace back to a surgeon’s residency or fellowship training, carrying the weight of learned and lived experience. In the age of ERAS, where practitioners are asked to abandon key tenets of their training and disregard clinical pearls that were once the backbone of morning rounds, it is not surprising that practice habits are slow to change. While we found that respondents from ERAS institutions were more likely to report adherence with ERAS guidelines, for several perioperative practices the overall adherence to ERAS Society recommendations was low.

We identified several areas where practice is clearly aligned with ERAS guidelines, among respondents from both ERAS and non-ERAS institutions. These include intraoperative deep vein thrombosis prophylaxis and continuous monitoring of core body temperature, along with post-operative early removal of urinary catheters, mobilization, initiation of regular diet and regular use of bowel regimens. Unsurprisingly, those areas where the ERAS and non-ERAS institutions diverge with respect to guideline adherence represent practices that are active changes from routine care. These include allowing oral liquids until 2 hours prior to surgery, carbohydrate loading, goal directed intra-operative fluid management, and prolonged postoperative deep vein thrombosis prophylaxis for malignancy.

The establishment of an ERAS pathway may be the first step toward change of these perioperative practices and improvement in patient outcomes. Practitioners within an institution that have established an ERAS pathway report improved adherence to preoperative fasting and carbohydrate treatment. They are also far more likely to utilize goal directed fluid management and restrictive fluid strategies, which have been shown to decrease length of stay and complication rates^16,17. ERAS institutions are also more likely to administer guideline-compliant extended post-operative deep vein thrombosis prophylaxis, a practice that carries high clinical significance towards the reduction of vein thromboembolism^18-21.

Despite these ERAS successes among the subset of SGO members who responded, adherence is far from universal. Even though 69% of ERAS programs offer postoperative anticoagulation as recommended, 8% of respondents from ERAS institutions only use anticoagulation for malignancy while patients are in the hospital. While most ERAS pathways employ a strategy of carbohydrate loading (63%), more than one-third still do not. Similarly, rates of pre-operative fasting and perioperative fluid management have room for improvement. The updated guidelines published in March 2019 highlight the need for an “ERAS Audit and Reporting” as it has been demonstrated that increased compliance with ERAS guidelines is directly related to decreased complications, length of stay, and cost^22-25.

Our survey was an informal audit of adoption of ERAS components, and notable for three perioperative practice where compliance with guidelines is moderate at best: bowel preparation, use of nasogastric tubes, and peritoneal drains. First, use of bowel preparation remains particularly controversial. It is a sea change to ask surgeons to one day prep the bowel, and the next day abandon the practice completely. The updated 2019 ERAS guidelines specifically state bowel preparation is “discouraged before open laparotomy in gynecologic surgery/gynecologic oncology, especially within an established ERAS pathway….high quality data from the colorectal literature have shown that mechanical bowel preparation alone does not decrease post-operative morbidity and should thus be abandoned”¹⁵. These new guidelines acknowledge the possible benefits of oral antibiotics and recent trends towards reintroduction of mechanical bowel prep with antibiotics and state that if a surgeon feels bowel preparation is necessary, they limit to surgeries with planned colon resections. Our survey demonstrates in spite of the 2016 guidelines, bowel preparation remains a common practice at both ERAS and non-ERAS institutions with approximately 30% of respondents using bowel prep for laparotomy, nearly 50% prior to ovarian cancer debulking surgeries and 68% of providers at ERAS institutions and 61% at non-ERAS institutions using bowel preparation when there is a concern for bowel surgery. The 2016 ERAS Society guidelines base their strong recommendation to avoid bowel prep on a moderate level of quality literature that demonstrates a lack of clear benefit to patients.

Avoidance of peritoneal drains and nasogastric tubes post-operatively is strongly recommended. Interestingly, both ERAS and non-ERAS institutions continue to employ these practices at similar rates: peritoneal drains are left in place in approximately 1/3 of procedures involving a bowel resection and 50% of the time after splenectomy. Nasogastric tubes are reportedly left in place after small bowel resections and ligation of the short gastric vessels by more than one-third of surgeons.

ERAS is well-established and proven to be a safe, effective bundle of peri-operative practices that improves the quality of perioperative care delivered. In our survey of gynecologic oncologists, ERAS met with overall positive reviews. Most agree that it is safe and improves patient outcomes and satisfaction. Yet, 63% of providers who have not yet implemented an ERAS pathway believe it is difficult to implement. Furthermore, a wide range of practice patterns and significant deviations from the guidelines that were laid out in 2016 and updated in 2019 remain, suggesting more work needs to be done in auditing those programs already in existence. Future work should be directed at better understanding barriers to ERAS implementation, developing interventions and tools to overcome these barriers. Focused investigation around the specific elements which were not widely adopted, such as bowel preparation, may also be helpful. As gynecologic oncology-specific ERAS pathways continue to evolve and more programs are developed, implementation of program-level audits will be critical to improve compliance in order to maximize the clinical and cost benefits.