Impact of enhanced recovery after surgery programs on pancreatic surgery: A meta-analysis.

AIM: To evaluate the impact of enhanced recovery after surgery (ERAS) programs on postoperative complications of pancreatic surgery.
METHODS: Computer searches were performed in databases (including PubMed, Cochrane Library and Embase) for randomized controlled trials or case-control studies describing ERAS programs in patients undergoing pancreatic surgery published between January 1995 and August 2017. Two researchers independently evaluated the quality of the studies' extracted data that met the inclusion criteria and performed a meta-analysis using RevMan5.3.5 software. Forest plots, demonstrating the outcomes of the ERAS group vs the control group after pancreatic surgery, and funnel plots were used to evaluate potential publication bias.
RESULTS: Twenty case-control studies including 3694 patients, published between January 1995 and August 2017, were selected for the meta-analysis. This study included the ERAS group (n = 1886) and the control group (n = 1808), which adopted the traditional perioperative management. Compared to the control group, the ERAS group had lower delayed gastric emptying rates [odds ratio (OR) = 0.58, 95% confidence interval (CI): 0.48-0.72, P < 0.00001], lower postoperative complication rates (OR = 0.57, 95%CI: 0.45-0.72, P < 0.00001), particularly for the mild postoperative complications (Clavien-Dindo I-II) (OR = 0.71, 95%CI: 0.58-0.88, P = 0.002), lower abdominal infection rates (OR = 0.70, 95%CI: 0.54-0.90, P = 0.006), and shorter postoperative length of hospital stay (PLOS) (WMD = -4.45, 95%CI: -5.99 to -2.91, P < 0.00001). However, there were no significant differences in complications, such as, postoperative pancreatic fistulas, moderate to severe complications (Clavien-Dindo III- V), mortality, readmission and unintended reoperation, in both groups.
CONCLUSION: The perioperative implementation of ERAS programs in pancreatic surgery is safe and effective, can decrease postoperative complication rates, and can promote recovery for patients.

Core tip: Enhanced recovery after surgery (ERAS) programs have been launched in a variety of surgical fields, including colorectal, orthopedics, urology, esophageal and gynecology, demonstrating favorable outcomes. Pancreatic surgery is considered a high-risk abdominal surgery, due to increased surgical trauma and high incidence of postoperative complications. In this meta-analysis we aimed to evaluate the impact of ERAS on complications of pancreatic surgery. The present study demonstrates that ERAS could reduce complication rates, especially of mild complications, delayed gastric emptying, abdominal infection and postoperative length of hospital stay, while not affecting the rates of postoperative pancreatic fistulas, reoperation, readmission and mortality during the perioperative period.

INTRODUCTION

Enhanced recovery after surgery (ERAS; also called ‘fast-track surgery’) was first introduced by Kehlet H, a Danish surgeon, in 1997[1]. ERAS is a multidisciplinary and evidence-based framework developed to decrease perioperative surgical stress, accelerate postoperative recovery and significantly reduce the postoperative length of hospital stay (PLOS). ERAS programs were initially implemented in colorectal surgery and have been shown to be effective for reducing PLOS and complications[2]. Subsequently, ERAS programs have been published in numerous areas of surgery, such as orthopedics, urology, esophageal, gynecology, breast and hepatobiliary[3-8].

An array of studies has shown that the perioperative implementation of ERAS programs can reduce PLOS without increasing complications or mortality. However, pancreatic surgery is still considered a high-risk abdominal surgery, due to the anatomical location of the pancreas and high rate of complications (30%-60%). Postoperative complications, such as postoperative pancreatic fistula (POPF), delayed gastric emptying (DGE), abdominal infection, and so on, are the main reasons for delayed recovery and the frequent need for additional interventions, without which the complications are potentially life threatening. For these reasons, the implementation of ERAS programs has lagged for pancreatic surgeries.

There had been an increasing number of ERAS programs implemented in pancreatic surgery when the ERAS group published evidence-based consensus recommendations for pancreatic surgery in 2012[9]. The benefit of implementing ERAS programs on postoperative complications in pancreatic surgery has not reached consensus. For this reason, we performed a meta-analysis of the available studies on ERAS programs compared with traditional perioperative management in patients undergoing pancreatic surgery.

MATERIALS AND METHODS

Search strategy

A search was performed by two researchers (Ji HB and Wang XX) in August 2017 of the PubMed, Cochrane Library and Embase database, spanning the period from January 1995 to August 2017. The search language was restricted to English, using the search terms “enhanced recovery after surgery”, “fast track surgery”, “ERAS”, “clinical pathways”, “pancreatectomy”, “pancreatoduodenectomy” and “duodenopancreatectomy”, and using the Boolean operators “AND” and “OR”. Synonyms of all these terms were used in this search. The PubMed search strategy for the meta-analysis is shown in Table 1.

Table 1

The search strategy for the PubMed database1

Search number	Description	Number of publications
1	Enhanced recovery after surgery [Title/Abstract] OR ERAS [Title/Abstract] OR fast track surgery [Title/Abstract]	3333
2	Clinical pathways [MeSH Terms]	5848
3	1 OR 2	9130
4	Pancreatectomy [MeSH Terms] OR Pancreatectomy* [Title/Abstract] OR Pancreatoduodenectomy [MeSH Terms] OR Pancreatoduodenectom* [Title/Abstract] OR duodenopancreatectomy [MeSH Terms] OR duodenopancreatectom* [Title/Abstract]	21497
5	3 AND 4 NOT (animals[mh] NOT humans[mh])	69
6	5 limited to English	68

1Date of search: August 1, 2017.

Inclusion/exclusion criteria

Studies meeting all of the following selection criteria were eligible for inclusion: (1) studies concerning patients undergoing pancreatic surgery; (2) the ERAS group implemented ERAS programs management, and the control group adopted traditional perioperative management; (3) measures in perioperative management were described in both groups; and (4) studies reported at least the following outcome measures, POPF, DGE, abdominal infection, mortality and PLOS, and explained their diagnostic criteria for postoperative complications.

Exclusion criteria were (1) sample size of less than 10; (2) comments, guidelines, reviews, case reports, abstracts, letters and non-comparative studies; (3) repeated publication of the same study population; and (4) incomplete clinical data.

Outcomes of interest

The outcomes of interest were POPF, DGE, PLOS, abdominal infection, mortality, readmission, unintended reoperation and occurrence of any complication within a postoperative period of 30 d. POPF was defined using the International Study Group of Pancreatic Fistula (ISGPF) guidelines describing a drain output of any measurable volume of fluid on or after postoperative day (POD) 3, with an amylase content greater than three times the serum amylase activity or as defined by the study’s authors[10]. DGE was defined according to the International Study Group of Pancreatic Surgery’s (ISGPS) recommendation that patients needing maintenance of a nasogastric tube (NGT) for > 3 d, needing to reinsert the NGT for persistent vomiting after POD 3, or unable to tolerate a solid diet by POD 7, should be considered DGE. In addition, there are another two widely used definitions for DGE after pancreatic resection (1) Yeo defined DGE as an NGT left in place for ≥ 10 d plus one of the following, or for < 10 d plus two of the following (a) repeated emesis after removal of the NGT, (b) need for prokinetic agents after POD 10, (c) need for reinsertion of the NGT, or (d) failure to progress with the diet. (2) Van Berge Henegouwen et al[11] defined DGE as gastric stasis requiring NGT for ≥ 10 d or the inability to tolerate a regular diet after POD 14. PLOS was defined as the span from the day of surgery to the day of actual discharge from the hospital. Abdominal infection was defined by the study’s authors. Mortality was defined as the range from the day of hospitalization to the first 30 d after actual discharge. Readmission was defined as the patient needing medical attention again within 30 d after discharge. Overall postoperative complications included any complication from the time of surgery to discharge, or within 30 d, with severity grading and classification relying on the Clavien-Dindo system[12]. Unintended reoperation was defined as patients with complications or other reasons that required reoperation within 30 d after discharge.

Data extraction

Data were extracted from each study by two authors (Ji HB and Wei Q) independently. The main parameters included common information (time of study publication, country, study type, and authors), characteristics of the study population (sex and age), elements of ERAS programs, and postoperative outcomes (overall complications, POPF, DGE, abdominal infection, PLOS, mortality, readmission, and unintended reoperation). All continuous outcome variables were described using the means and standard deviations for this meta-analysis. We needed to estimate means and standard deviations via the methodologies reported by Hozo et al[13] if the original data were expressed as medians or ranges.

Quality assessment

The quality assessment of each study was done by two authors (Zhu WT and Ji HB) independently via the Methodological Index for Non-Randomized Studies (MINORS) checklist. It was then summarized by a French surgeon, and if there was a disagreement, the third researcher was involved in the negotiation or adjudication, until a consensus was achieved. The MINORS checklist includes eight methodological items for non-comparative studies and an additional four items for comparative studies. The items are scored 0 (not reported), 1 (reported but inadequate), or 2 (reported and adequate). The overall ideal scores were 24 for comparative studies.

Statistical analysis

The meta-analysis was performed using RevMan5.3.5 software (Ji HB and Wang HB). Continuous and categorical variables were calculated as weighted mean differences (WMDs) or odds ratios (ORs) with their corresponding 95% confidence interval (CI), respectively. Heterogeneity was assessed using a chi-square test, where P > 0.05 was considered non-significant. I2 values were used for the evaluation of statistical heterogeneity, and the I2 value of 50% or more indicated the presence of heterogeneity. The fixed-effects model was used for studies of homogeneity (I2 < 50%), and the random-effects model was applied when studies indicated heterogeneity (I2 ≥ 50%). In addition, funnel plots were used to evaluate potential publication bias based on the incidence of POPF and mortality.

Eligible studies

The search strategy initially identified 159 relevant studies. No randomized control trials were identified. Figure 1 shows the process of selecting the studies for meta-analysis. After removing duplicates, the titles and abstracts of 118 studies were reviewed. Of these, 68 studies were not related to ERAS in pancreatic surgery, 12 studies did not have a control group, 6 studies did not have the outcomes of interest reported, 10 studies only had an abstract or we were unable to get the full text, 1 study did not have enough data, and 1 study was published in a language other than English. A total of 20 studies met the inclusion criteria for the meta-analysis.

Figure 1

The diagram of selected studies for this meta-analysis.

Study characteristics and quality assessment

The characteristics and quality assessments of the included studies are shown in Table 2[14-33]. All studies clearly described an ERAS program. The major components are summarized in Table 2. All of the studies used a retrospective case-control model, and of those, there were 16 studies that had sample sizes greater than 100. A total of 3694 patients were included, of which there were 1886 patients and 1808 patients included in the ERAS group and control group, respectively. In addition, there were 17 studies with MINORS scores > 12.

Table 2

Study Characteristics and Quality Assessment.

Study	Year	Country	Study design	Sample size		ERAS programs1	MINORS Score
				ERAS group	Control group
Kennedy et al[14]	2007	United States	Case-control	91	44	e, f, g, h	16/24
Vanounou et al[15]	2007	United States	Case-control	145	64	c, d, g, h	15/24
Balzano et al[16]	2008	Italy	Case-control	252	252	d, e, f, g, h	13/24
Kennedy et al[17]	2009	United States	Case-control	71	40	d, e, f, g, h	11/24
Abu Hilal et al[18]	2013	Britain	Case-control	20	24	b, e, f, g, h	15/24
Braga et al[19]	2014	Italy	Case-control	115	115	a, b, c, d, e, f, g, h	17/24
Pillai et al[20]	2014	India	Case-control	20	20	c, d, e, f, g, h	17/24
Coolsen et al[21]	2014	Holland	Case-control	86	97	b, c, d, e, f, g, h	12/24
Nussbaum et al[22]	2014	United States	Case-control	100	142	c, e, f, g, h	11/24
Yui et al[23]	2014	Japan	Case-control	57	52	e, g, h	13/24
Nussbaum et al[24]	2014	United States	Case-control	50	100	c, e, f, g, h	16/24
Kobayashi et al[25]	2014	Japan	Case-control	100	90	a, e, g, h	13/24
Shao et al[26]	2015	China	Case-control	325	310	d, e, f, g, h	15/24
Joliat et al[27]	2015	Switzerland	Case-control	74	87	a, b, c, d, e, f, g, h	15/24
Partelli et al[28]	2015	Italy	Case-control	22	66	a, c, d, e, f, g, h	13/24
Williamsson et al[29]	2015	Sweden	Case-control	50	50	c, d, e, f, g, h	17/24
Morales Soriano et al[30]	2015	Spain	Case-control	41	44	a, b, c, d, e, f, g, h	17/24
Bai et al[31]	2016	China	Case-control	124	63	a, d, e, f, g, h	15/24
Zouros et al[32]	2016	Greece	Case-control	75	50	a, b, c, d, e, f, g, h	16/24
Dai et al[33]	2017	China	Case-control	68	98	a, b, c, e, f, g, h	15/24

ERAS programs: a: No bowel preparation in the preoperative period; b: Clear fluids until 2-3 h before surgery; c: Restrictive policy of intravenous fluids in the intra-operative period; d: Multimodal analgesia of the postoperative period; e: Clear fluids or food intakes in the early period; f: Enhanced mobilization in the early period; g: Removal of the drainage tube; h: Others. ERAS: enhanced recovery after surgery; MINORS score: Methodological Index for Non-Randomized Studies checklist.

RESULTS

Pancreatic fistula

Eighteen studies reported the rates of POPF. The overall results (OR = 0.87, 95%CI: 0.74-1.03, P = 0.10; Figure 2), or only those using the ISGPF definition (OR = 0.90, 95%CI: 0.76-1.07, P = 0.24), showed that there were no significant differences present in either group. Furthermore, there was no significant difference in A (OR = 1.05, 95%CI: 0.81-1.36, P = 0.71), B (OR = 1.13, 95%CI: 0.85-1.51, P = 0.40), and C (OR = 0.90, 95%CI: 0.60-1.33, P = 0.59) grade of POPF between the ERAS group and control group.

Figure 2

Forest plots demonstrating the outcomes of postoperative pancreatic fistula.

DGE

Eighteen studies reported the rates of DGE. Compared to the control group, the ERAS group had a lower incidence of DGE (OR = 0.58, 95%CI: 0.48-0.72, P < 0.00001; Figure 3). The difference persisted when including only studies that adopted the ISGPS definition (OR = 0.50, 95%CI: 0.39-0.65, P < 0.00001).

Figure 3

Forest plots demonstrating the outcomes of delayed gastric emptying.

Postoperative complications

The rate of overall postoperative complications was lower in the ERAS group (OR = 0.57, 95%CI: 0.45-0.72, P < 0.00001; Figure 4). Additionally, the incidence of mild postoperative complications (Clavien-Dindo I-II), which relies on the Clavien-Dindo definition of severity and classification, was lower in the ERAS group (OR = 0.71, 95%CI: 0.58-0.88, P = 0.002; Figure 5). There were no statistical differences in the moderate to severe complication rates (Clavien-Dindo III-V) between the ERAS group and control group (OR = 0.90, 95%CI: 0.73-1.11, P = 0.32).

Figure 4

Forest plots demonstrating the outcomes of overall complications.

Figure 5

Forest plots demonstrating the outcomes of mild complications.

Abdominal infection

A total of 12 studies reported the rates of abdominal infection. The incidence of abdominal infection was lower (OR = 0.70, 95%CI: 0.54-0.90, P = 0.006; Figure 6) in the ERAS group.

Figure 6

Forest plots demonstrating the outcomes of abdominal infection.

PLOS

A total of 13 studies reported the PLOS, and they showed that the ERAS group had shorter PLOS (WMD = -4.45, 95%CI: -5.99 to -2.91, P < 0.00001; Figure 7) than the control group.

Figure 7

Forest plots demonstrating the outcomes of postoperative length of hospital stay.

In addition, there were no significant differences in rates of mortality (OR = 0.85, 95%CI: 0.54-1.36, P = 0.51; Figure 8), readmission (OR = 1.04, 95%CI: 0.83-1.30, P = 0.75; Figure 9), and unintended reoperation (OR = 0.87, 95%CI: 0.63-1.20, P = 0.40; Figure 10).

Figure 8

Forest plots demonstrating the outcomes of mortality.

Figure 9

Forest plots demonstrating the outcomes of readmission.

Figure 10

Forest plots demonstrating the outcomes of unintended reoperation.

Subgroup analysis

The subgroup analysis, which included only larger size studies (n ≥ 100) generated similar results in postoperative outcomes (Table 3). Furthermore, the analysis of only high-quality studies (MINORS score > 12) also yielded parallel results in postoperative outcomes (Table 3). However, the heterogeneity for overall complications and PLOS still exists in larger studies and high-quality studies.

Table 3

Results of Subgroup Analysis

Outcomes of interest	Studies	Patients	OR/WMD	95%CI	P-value	Heterogeneity P-value	I2, %
Studies with cases ≥ 100
POPF	14	3067	0.87	0.73-1.03	0.11	0.02	48
DGE	14	3117	0.58	0.47-0.71	< 0.00001	0.07	39
Overall complications	14	3045	0.57	0.45-0.72	< 0.00001	0.006	55
Mild complications	9	1470	0.74	0.59-0.93	0.009	0.06	46
Abdominal infection	10	2087	0.67	0.51-0.87	0.003	0.07	42
PLOS	10	2394	-4.64	-6.37 to -2.91	< 0.00001	0.009	65
Mortality	15	2802	0.83	0.51-1.37	0.47	1	0
Readmission	12	2877	1.05	0.83-1.33	0.68	0.23	22
Unintended reoperation	9	1955	0.85	0.60-1.21	0.38	0.96	0
MINORS score > 12
POPF	15	2784	0.84	0.70-1.00	0.05	0.13	30
DGE	16	2949	0.52	0.42-0.64	< 0.00001	0.12	31
Overall complications	16	3008	0.56	0.44-0.71	< 0.00001	0.01	51
Mild complications	11	1504	0.67	0.54-0.83	0.0003	0.24	21
Abdominal infection	10	1791	0.63	0.46-0.85	0.002	0.05	46
PLOS	11	2272	-4.35	-5.97 to -2.72	< 0.00001	0.003	66
Mortality	16	2523	0.96	0.56-1.65	0.89	0.99	0
Readmission	13	2598	1.09	0.84-1.43	0.52	0.82	0
Unintended reoperation	11	1787	0.96	0.65-1.41	0.83	0.94	0

CI: Confidence interval; DGE: Delayed gastric emptying; MINORS score: Methodological Index for Non-Randomized Studies checklist; OR: Odds ratio; PLOS: Postoperative length of hospital stay; POPF: Postoperative pancreatic fistula; WMD: Weighted mean difference.

Sensitivity analysis

We aimed to investigate the influence of a single study on the overall results by omitting one study in each turn. This analysis revealed that no single study generated an especially strong influence on the results, with estimates ranging from an OR of 0.54 to 0.62 (Table 4).

Table 4

Results of sensitivity analysis by omitting one study in each turn

Studies	OR	95%CI	P-value
Omitting Vanounou et al[15]	0.56	0.44-0.72	< 0.00001
Omitting Kennedy et al[14]	0.56	0.44-0.71	< 0.00001
Omitting Balzano et al[16]	0.56	0.43-0.73	< 0.0001
Omitting Kennedy et al[17]	0.58	0.46-0.74	< 0.00001
Omitting Abu Hilal et al[18]	0.58	0.45-0.73	< 0.00001
Omitting Yui et al[23]	0.56	0.44-0.72	< 0.00001
Omitting Kobayashi et al[25]	0.58	0.45-0.74	< 0.00001
Omitting Coolsen et al[21]	0.54	0.43-0.69	< 0.00001
Omitting Braga et al[19]	0.55	0.43-0.71	< 0.00001
Omitting Pillai et al[20]	0.57	0.45-0.73	< 0.00001
Omitting Joliat et al[27]	0.57	0.45-0.73	< 0.0001
Omitting Partelli et al[28]	0.55	0.44-0.68	< 0.00001
Omitting Williamsson et al[29]	0.56	0.44-0.71	< 0.00001
Omitting Morales Soriano et al[30]	0.58	0.46-0.74	< 0.00001
Omitting Shao et al[26]	0.57	0.44-0.74	< 0.0001
Omitting Zouros et al[32]	0.57	0.44-0.73	< 0.00001
Omitting Bai et al[31]	0.56	0.44-0.71	< 0.00001
Omitting Dai et al[33]	0.62	0.52-0.71	< 0.00001
Overall effect	0.57	0.45-0.72	< 0.00001

CI: Confidence interval; OR: Odds ratio.

Publication bias

Funnel plots based on the incidence of POPF and mortality were used to evaluate potential publication bias in this study (Figure 11). There was no evidence of publication bias of POPF, mortality or other outcomes of this study (other figures not shown).

Figure 11

The funnel plots were used to evaluate potential publication bias. A: POPF; B: Mortality.

DISCUSSION

ERAS requires surgical, nursing, anesthesia, nutritionist and other specialties to work together and uses a series of optimal and evidence-based management measures to lessen perioperative surgical stress while promoting the recovery of organ function in the early postoperative period[34,35]. ERAS programs were initially implemented in colorectal surgery, with recommendations for each step to achieve optimal perioperative care[36]. Subsequently, ERAS programs had been launched in numerous fields of surgery, such as orthopedics, urology, esophageal and gynecology.

The literature from these disciplines has suggested that standardizing ERAS measures could reduce the incidence of complications, accelerate recovery for patients, reduce hospitalization costs and save medical resources in perioperative care[3,4,7,8]. Pancreatic surgery is an effective treatment of pancreatic tumors, periampullary tumors, duodenal tumors and distal bile duct tumors. Currently, despite surgical techniques, anesthesia, and preoperative imaging assessment making great progress and the mortality of the procedure dropping to approximately 2% in high-volume medical centers, it is still considered a complicated and high-risk abdominal surgery[37].

Coolsen et al[38] analyzed 8 studies, which related to pancreatic surgery, and suggested that the ERAS group had shorter PLOS and lower postoperative complication rates; however, there were no significant differences in rates of DGE, POPF, readmission, and mortality. Kagedan et al[39] analyzed 10 studies suggesting that the ERAS group had only shorter PLOS and no differences in other complications. As mentioned above, we may reasonably conclude that the influence of ERAS programs on the postoperative complications of pancreatic surgery is controversial. Hence, the application of ERAS programs in the perioperative period of pancreatic surgery is still being explored in our practices.

The main measures of the ERAS programs include no bowel preparation and clear fluids until 2-3 h before surgery, multimodal analgesia of postoperative, clear fluids or food intakes, enhanced mobilization and removal of the drainage tube in early period. The ERAS group has reduced time of fasting in the preoperative period, which can decrease the insulin resistance in the postoperative period. We adopted multimodal analgesia in the postoperative period, which was able to reduce the stress caused by pain. The programs, such as, no bowel preparation before surgery, clear fluids or food intakes, enhanced mobilization in the early postoperative period which may promote rehabilitation of gastrointestinal function[40].

The ERAS programs aimed to reduce the incidence of complications and accelerate recovery for patients. Among them, gastrointestinal function rehabilitation is an important part of the rapid recovery in abdominal surgery. In addition, the early postoperative oral feeding, which may play an important role in the gastrointestinal function rehabilitation in the postoperative period. This is because early postoperative oral feeding is more in line with human physiology of the digestive tract, and which may have a beneficial effect on immunological, inflammatory and nutritional status. In addition, early postoperative oral feeding can promote the recovery of gastrointestinal motility, protect the gastrointestinal mucosal barrier, shorten time to gas and stools passage, and reduce the incidence of complications.

A total of 20 studies and 3694 patients were included in our meta-analysis. Compared with the control group, the ERAS group had lower rates of DGE, lower postoperative complication rates, particularly lower mild postoperative complication rates, lower abdominal infection rates, and shorter PLOS. However, no significant differences existed in POPF, moderate to severe complications, mortality, readmission or unintended reoperation in both groups.

Many factors, such as age, nutritional status, and serious comorbidity, can influence patients’ postoperative complication rates and the process of postoperative recovery[41,42]. The patients’ demographic data in the included studies was basically identical, so these influences may be eliminated for the outcomes in this study. In addition, all of the included studies described the diagnostic criteria for postoperative complications.

Despite our careful work on this meta-analysis of currently available evidence, some limitations should be acknowledged. First, the diagnostic criteria of some postoperative complications were not uniformly defined, though all the included studies gave a description of the diagnostic criteria. Therefore, to a certain extent, information bias was possible, because some complications did not have national criteria. Second, only retrospective case control studies were included in this analysis. Therefore, to a certain extent, the outcomes of this study may be influenced by the selection bias. Third, the degree of implementation of ERAS programs and the compliance of patients may be different between studies. Finally, there was no evidence to indicate that major publication bias existed in these studies, and potential publication bias is impossible to completely rule out in small studies. Hence, these factors had some influence on our results.

In summary, the results from our present study demonstrate that the implementation of ERAS programs could reduce overall complication rates, especially of mild complications, DGE, rates of abdominal infection, and PLOS, while not affecting the rates of POPF, reoperation, readmission, and mortality during the perioperative period for pancreatic surgery. The perioperative period for pancreatic surgery is safe and effective to implement ERAS programs that can decrease postoperative complication rates and promote recovery. However, in the future, we need to include more high-quality and strict prospective studies to assess the contributions of individual program components.

ARTICLE HIGHLIGHTS

Research background

Enhanced recovery after surgery (ERAS) is a multidisciplinary and evidence-based framework, developed to decrease perioperative surgical stress, accelerate postoperative recovery and significantly reduce the postoperative length of hospital stay (PLOS). ERAS programs have been launched in a variety of other fields of surgery, such as colorectal, orthopedics, urology, esophageal, and gynecology, and have demonstrated favorable outcomes. The implementation of ERAS programs has lagged surrounding pancreatic surgeries because of the anatomical location of the pancreas and the high rate of postoperative complications (30%-60%). It is very important to promote the postoperative recovery for this high-risk abdominal surgery via implementing ERAS programs during the perioperational period.

Research motivation

ERAS requires surgical, nursing, anesthesia and other specialties to work together and uses a series of optimal or evidence-based management measures to lessen perioperative surgical stress while promoting the recovery of organ function in the early postoperative period. The implementation of ERAS programs may play a very important role in the perioperational period for pancreatic surgery.

Research objectives

This study evaluated the impact of ERAS programs on postoperative complications and PLOS of pancreatic surgery.

Research methods

Computer searches were performed in databases (including PubMed, Cochrane Library, and Embase) for randomized controlled trials or case-control studies describing ERAS programs in patients undergoing pancreatic surgery published between January 1995 and August 2017. Two researchers independently evaluated the quality of the studies’ extracted data that met inclusion criteria and performed a meta-analysis using RevMan5.3.5 software. Forest plots, demonstrating the outcomes of the ERAS group versus the control group after pancreatic surgery, and funnel plots were used to evaluate potential publication bias.

Research results

Twenty case-control studies, published between January 1995 and August 2017, including 3694 patients, were selected for the meta-analysis. They included the ERAS group (n = 1886) and control group (n = 1808), which adopted the traditional perioperative management. Compared to the control group, the ERAS group had lower delayed gastric emptying (DGE) rates (odds ratio (OR) = 0.58, 95% confidence interval (CI): 0.48-0.72, P < 0.00001), lower postoperative complication rates (OR = 0.57, 95%CI: 0.45-0.72, P < 0.00001), particularly for mild postoperative complications (Clavien-Dindo I- II) (OR = 0.71, 95%CI: 0.58-0.88, P = 0.002), lower abdominal infection rates (OR = 0.70, 95%CI: 0.54-0.90, P = 0.006) and shorter PLOS (weighted mean difference (WMD) = -4.45, 95%CI: -5.99 to -2.91, P < 0.00001). However, there were no significant differences in postoperative pancreatic fistulas (POPF), moderate to severe complications (Clavien-Dindo III- IV), mortality, readmission and unintended reoperation in both groups.

Research conclusions

The results from our present study demonstrate that the implementation of ERAS programs could reduce overall complication rates, especially of mild complications, DGE, rate of abdominal infection and PLOS, while not affecting the rates of POPF, reoperation, readmission and mortality during the perioperative period for pancreatic surgery. The perioperative period for pancreatic surgery is safe and effective to implement ERAS programs that can decrease postoperative complication rates and promote recovery

Research perspectives

We need to include more high-quality and strict prospective studies to assess the contributions of individual program components, such as clear fluids or food intakes in the early period, and removal of the drainage tube.