J Gastrointest Surg DOI 10.1007/s11605-015-2879-z
ORIGINAL ARTICLE
Fast-Track Programs for Liver Surgery: A Meta-Analysis Si-Jia Wu 1 & Xian-Ze Xiong 1 & Jiong Lu 1 & Yao Cheng 1 & Yi-Xin Lin 1 & Rong-Xing Zhou 1 & Nan-Sheng Cheng 1
Received: 19 January 2015 / Accepted: 22 June 2015 # 2015 The Society for Surgery of the Alimentary Tract
Abstract Background and Objectives Plentiful publications have inspected the feasibility of fast-track surgery programs during hepatic surgery, but the potency of these studies has not been discussed profoundly so far. Our goal was to assess the effects of fast-track programs on surgical outcomes compared with traditional surgical plans for liver surgery. Methods The following databases were searched: PubMed, Cochrane library, Embase, Science Citation Index Expanded, etc. Studies meeting our inclusion criteria were included. All interrelated data and the methodological quality of included studies were extracted and assessed. We applied risk ratio and weighted mean difference as the estimated effect measures. Sensitivity analysis was performed to perceive the reliability of our findings. Results Altogether, 14 studies with 1400 patients were analyzed. Meta-analysis of randomized controlled trials demonstrated that implementation of fast-track surgery programs could observably decrease the total length of hospital stay, complication rate, postoperative first flatus time, and hospitalization expense, and did not compromise mortality and readmission rate. The above findings were also in line with the results of case-control studies. Conclusions Fast-track surgery programs are feasible and effective for liver surgery. Future studies should optimize fast-track surgery programs catering to liver surgery. Keywords Fast track . Enhanced recovery . Liver surgery . Hepatectomy . Meta-analysis
Introduction Liver surgery has become an extremely effective treatment means for numerous primary and secondary liver diseases but it will bring some adverse effects to human body. Postoperative hemorrhage, serious infection, biliary leakage, * Nan-Sheng Cheng
[email protected] Si-Jia Wu
[email protected] 1
Department of Bile Duct Surgery, West China Hospital, Sichuan University, No. 37 Guo Xue Xiang, Chengdu 610041, Sichuan Province, China
as well as postoperative liver dysfunction are the most common postoperative complications after liver surgery.1 Plenty of effective strategies emphasizing on improving skills or enhancing perioperative care have been put into use and the present mortality after liver surgery has declined below 5 %.2,3 Be that as it may, complication rate after liver surgery remains above 30 %,2,3 thus other necessary measures must be adopted to reduce harmful outcomes regarding liver surgery. First described by Kehlet et al., Fast-Track (FT) or Enhanced Recovery after Surgery (ERAS) programs containing multidisciplinary care strategies were applied successfully in colonic surgery.4–6 The primary target of FT programs was to ease the metabolic and inflammatory reaction, relieve surgical stress, and maintain the physiological functions of vital organs.5–7 The concrete contents of FT programs include many aspects such as preoperative counseling, patient education, less or short-acting anesthetic agents, less use or early removal of peritoneal drainages and nasogastric tubes, restoring gastrointestinal function, oral nutrition, fluid management, reasonable pain control, and early postoperative ambulation.8–10 In addition to
J Gastrointest Surg
colonic surgery, FT programs have also been utilized successfully in vascular surgery, musculoskeletal surgery, orthopedic surgery, gynecologic surgery, urological surgery, breast surgery, and other abdominal surgeries.11–21 These previous studies certified that FT programs could decrease postoperative complications, postoperative length of stay (LoS), hospitalization costs, and improve patient satisfaction.4–6,11–21 A good deal of publications has inspected the feasibility of FT programs for hepatic surgery, but the potency of these studies has not been discussed profoundly. As a consequence, we performed this meta-analysis for the sake of assessing the effects of implementation of FT programs on surgical outcomes compared with the traditional treatment plans following liver surgery.
(4) The studies investigated the effects of implementation of FT or ERAS programs compared with the traditional treatment plans in liver surgery. (5) The studies reported pivotal surgical outcomes such as total length of hospital stay (TLoS), complication rate, mortality, readmission rate, and so on. Exclusion criteria are as follows: (1) The studies did not adopt FT or ERAS programs. (2) The studies were centered on a single FT program in perioperative care. (3) The studies were case report, protocol, letter, review, meta-analysis, meeting minutes, and research summaries. (4) The studies adopted FT or ERAS programs but compared other interventions in both arms. (5) The studies did not report key surgical outcomes, such as TLoS, complication rate, mortality, readmission rates, etc. (6) The studies employed other types of surgery, liver surgery combined with a second concomitant procedure (e.g., bowel or bile duct repair) or emergency surgery.
Methods and Materials Selection of Outcomes We conducted this study according to the PRISMA guidelines for meta-analysis.22 Literature Search and Selection of Studies The list of relevant studies was obtained until November 2014 by searching the following electronic databases: Medline, Cochrane Central Register of Controlled Trials, PubMed, Embase, Science Citation Index Expanded, China National Knowledge Infrastructure Whole Article Database, Chinese Biomedical Database, and the Google Scholar. We also did a systematic search of reference lists. The following keywords were used: fast track, FT, enhanced recovery, ERAS, early discharge, clinical pathway, liver, liver surgery, hepatic, and hepatectomy. There was no restriction in language and publication date. The selection of studies included the following several steps: the primary author (Wu SJ) looked through the titles and abstracts to exclude obviously irrelevant studies; then, the preliminary results were reviewed by two authors (Wu SJ and Cheng Y) independently via the titles, abstracts, keywords, and MeSH in order to identify all possibly related articles; finally, all full texts of appropriate articles were obtained for review. During this process, any discrepancy was solved by discussion. Inclusion and Exclusion Criteria Inclusion criteria are as follows: (1) The studies discussed patients who underwent open or laparoscopic liver surgery. (2) The studies described Fast-Track (FT) or Enhanced Recovery after Surgery (ERAS) programs, which contained at least nine perioperative items. (3) The studies were randomized controlled trials (RCTs) or case-control studies (CCSs).
Primary outcomes: Total length of hospital stay (TLoS), defined as the number of nights spent in hospital including nights after readmission within 30 days after surgery. Complication rate was calculated as the percentage of patients with at least one complication after surgery. Mortality was defined as death in hospital or within 30 days after surgery. The last primary outcome was readmission. Secondary outcomes: Secondary outcomes were hospitalization expense and first flatus time after operation. Data Extraction and Assessment of Quality Two authors (Wu SJ, Cheng TY) independently extracted all interrelated data onto a spreadsheet including studies and population characteristics, information on the outcomes, the quality of literature, and other relevant information. Any discrepancy was solved in consultation with each other. To chase down if there were missing data or inaccurate information, we also contacted the authors of included studies. Four authors (Wu SJ, Xiong XZ, Lu J and Lin YX) independently assessed the methodological quality of included studies. The GRADE guidelines were used to evaluate the methodological quality of all included studies.23 Ramified views were also solved by dint of discussion during this process. Statistical Methods and Data Analysis So that the included studies could be divided into two categories (randomized controlled trials and case-control studies), we conducted meta-analysis of them, respectively, according to the study design. For dichotomous data, we applied risk
J Gastrointest Surg
ratio (RR) as the estimated effect measures and weighted mean difference (WMD) for continuous data, both were reported with 95 % confidence intervals (CI).24 The chi-square (χ2) test was used to detect the heterogeneity and heterogeneity was regarded as statistical significance at a P value of <0.10, and the heterogeneity was acceptable on the occasion of I2 ≤50 %.24 The fixed-effects model was used to compound data when heterogeneity was receivable (I2 ≤50 %); if not, the random effects model was used.24 In case remarkable heterogeneity existed, we performed sensitivity analysis to perceive the reliability of our findings. Publication bias was evaluated by means of funnel plot.25 The appearance of the funnel plot presented an asymmetric form might imply the existence of potential bias and vice versa.25 Relying on contacting the authors of the included studies and the method reported by Hozo et al., 26 we processed the expression form of some data and estimated them in order to form a unified data presentation and analyze them conveniently. For some missing or unclear data, we also attempted to get detailed information by contacting the authors. We performed this meta-analysis via the software package Review Manager (version 5.1). Four authors (Wu SJ, Cheng NS, Lu J, Lin YX) separately entered all correlative data into software and analyzed them. Any divergence during this process was solved through mutual discussion. Fig. 1 Flowchart of literature screening process
Results Study Identification After systematically searching, we identified 381 potential references. Of these, 335 records were identified ineligible based on the titles and abstracts for the following reasons: duplicated studies (n=61), review articles (n=56), and irrelevant records (n=218). The full text references of the remaining 46 records were reviewed for further appraisal. Then, 32 records were excluded; of these, 11 references did not meet our inclusion criteria, 8 references were review articles, and 13 records were duplicates of the same trials. Finally, altogether, 14 studies (including 9 randomized controlled trials and 5 case-control studies) were considered as eligible for our meta-analysis.27–40 The PRISMA flowchart of literature screening process was shown in Fig. 1. Study Characteristics A total of 14 studies27–40 with 1400 patients (675 patients in the fast track surgery (FTS) group and 725 patients in the conventional surgery (CS) group) were analyzed. Nine documents27–35 with 933 patients (459 patients in the FTS group and 474 patients in the CS group) were randomized controlled trials (RCTs) and five literatures36–40 with 467
FTS vs. CS programs FTS vs. CS programs FTS vs. CS programs FTS vs. CS programs FTS vs. CS programs FTS vs. CS programs FTS vs. CS programs FTS vs. CS programs FTS vs. CS programs FTS vs. CS programs FTS vs. CS programs FTS vs. CS programs FTS vs. CS programs FTS vs. CS programs Chinese Journal of General Surgery Chinese Journal of Clinical Oncology Journal of Modern Clinical Medicine Parenteral & Enteral Nutrition China Journal of Endoscopy Eur J Surg Oncol British Journal of Surgery Chin J Dig Surg Hapetogastroenterology British Journal of Surgery HPB Cell Biochem Biophys World J Gastrointest Surg Military Medical Journal of Southeast China 20 52 30 35 30 80 45 20 162 100 13 61 17 60 20 63 30 35 30 80 46 20 135 61 13 56 26 60 40 115 60 70 60 160 91 40 297 161 26 117 43 120
FTS fast-track surgery, CS conventional surgery
Randomized controlled trial Randomized controlled trial Randomized controlled trial Randomized controlled trial Randomized controlled trial Randomized controlled trial Randomized controlled trial Randomized controlled trial Randomized controlled trial Case-control study Case-control study Case-control study Case-control study Case-control study
Low quality (two plus: ++) Low quality (two plus: ++) Low quality (two plus: ++) Low quality (two plus: ++) Low quality (two plus: ++) High quality (four plus: ++++) High quality (four plus: ++++) Moderate quality (three plus: +++) Moderate quality (three plus: +++) Moderate quality (three plus: +++) Moderate quality (three plus: +++) Moderate quality (three plus: +++) Moderate quality (three plus: +++) Moderate quality (three plus: +++)
Journal CS (n) FTS (n) Sample size (n) Literature quality (GRADE guidelines)
Fan et al. (2011) Chi et al. (2012) Pu et al. (2012) Wang et al. (2013) Huang et al. (2013) Ni et al. (2013) Jones et al. (2013) Shou et al. (2014) Lu et al. (2014) van Dam et al. (2008) Stoot et al. (2009) Lin et al. (2011) Sánchez-Pérez et al. (2012) Du et al. (2013)
FTS vs. CS—Total Length of Hospital Stay (TLoS) Data on TLoS could be obtained from all nine RCTs27–35 with a total of 933 patients (459 in the FTS group and 474 in the CS group). Meta-analysis indicated that TLoS was significantly shorter in the FTS group than the CS group and this difference was statistically significant (WMD=−2.25, 95 % CI=−3.10 to
Study design
Primary Outcome Data of Randomized Controlled Trials
Author (year)
The primary and secondary outcome data of CCSs comparing FTS with CS were shown in Table 3. Analogously, we also extracted the relevant data of RCTs and presented them in Table 4.
Characteristics of included studies comparing FTS with CS
Meta-Analysis and Outcome Measures
Table 1
patients (216 patients in the FTS group and 251 patients in the CS group) were case-control studies (CCSs). All studies were published between 2008 and 2014 with the sample sizes ranging from 26 to 297 patients (Table 1). The results of quality assessment-conducted according the GRADE guidelines23 showed that two RCTs32,33 were identified as high quality (four plus: ++++), five RCTs27–31 were deemed to be low quality (two plus: ++), and the rest of the seven studies including two RCTs34,35 and all five CCSs36–40 were ranked as moderate quality (three plus: +++) (Table 1). The baseline characteristics of patients and operative details between the FTS and CS groups were also similar (Table 2). Laparoscopic hepatectomy was adopted in 4 studies,30,31,37,39 of which 2 were RCTs30,31 and the other 2 were CCSs, 37,39 respectively, while the other 10 included studies27–29,32–36,38,40 employed an open hepatectomy technique (Table 2). The overall surgical procedures between the FTS and CS groups were also parallel (Table 2). Detailed baseline data of patients and operative information were presented in Table 2. Although the key FTS items used in the included studies were generally similar, the detailed ingredients of FT programs were still varied among different studies due to the fact that no unified international standards and guidelines concerning perioperative FT programs for liver surgery have been published so far.27–40 All included studies precisely described the detailed ingredients of FT programs.27–40 The key items most frequently applied were as follows: preoperative counseling, without bowel preparation, carbohydrate drink before surgery, avoiding anesthetic premedication, nasogastric tube avoidance, perioperative fluids control, less urinary catheter and routine drains, intraoperative temperature control, postoperative pain management, early postoperative feeding, and positive postoperative mobilization.27–40 All studies described FT programs containing at least 10 key items and the number of elements utilized in FT programs ranged from 10 to 19 with an average of 14 elements (Table 2).
Comparison
J Gastrointest Surg
44.1:48.7
46.5:45.8
53.3:55.6
NS NS 48.4:50.1
64:67
51:48
54.0:52.6
Fan et al. (2011)
Chi et al. (2012)
Pu et al. (2012)
Wang et al. (2013) Huang et al. (2013) Ni et al. (2013)
Jones et al. (2013)
Shou et al. (2014)
Lu et al. (2014)
NM
0:43:3:0/2:38:5:0
NM NM 76:4:0:0/78:2:0:0
NM
NM
NM
ASA class (I:II:III:IV n) FTS/CS
3:10/2:11 31:25/34:27 15:11/10:7 48:12/47:13
55:45 57:55
58.3:52.5
58.6:58.8
Stoot et al. (2009) Lin et al. (2011)
Sánchez-Pérez et al. (2012) Du et al. (2013) 46:14:0:0/49:11:0:0
3:9:1:0/6:6:1:0 43 (I–II):11:2/50 (I–II):10:1 0:13:13:0/0:8:9:0 60:0/60:0
12:14/3:14
5:8/2:11 NM
11:42:8:0/14:64:22:0 57:4/86:14
Operative blood loss (mean ml) FTS/CS Volume transfused (mean ml) FTS/CS
NM
NM NM
NM
143:146
NM
NM
50:250 760:850
750:800
NM
NS 550:600
NM
NM
200:240
NM
5:1
NS 21:25
NM
24:32
NM
7/3
NM NM 14/12
NM
NM
NS
major resection (≥3 segments) 40:40; minor resection 14:14; unregular hepatectomy 6:6 NS
Major resection (≥3 segments) 4:8; minor resection 16:12 NS Regular hepatectomy; unregular hepatectomy NS Major resection (≥ 3 segments) 30:30 NS Laparoscopic hepatectomy 35:35 NS Laparoscopic hepatectomy 30:30 NS Major resection (≥3 segments) 73:69; minor resection 7:11 NS Major resection (≥3 segments) 21:12; Minor resection 25:33 NS Major resection (≥ 3 segments) 17:15; minor resection 3:5 NS Major resection (≥ 3 segments) 16:26; minor resection 119:136 NS Major resection (≥3 segments) 51:79; minor resection 10:21 NS Laparoscopic hepatectomy 13:13 NS Major resection (≥3 segments) 19:21; Minor resection 37:40 NS Laparoscopic hepatectomy 26:17 NS
Number of Type of resection (n) patients FTS/CS transfused (n) FTS/CS
121.5:183.3 453.8:481.0 482.8:496.8 NM
180:177
118:180 110:125
220:270
440:480
366:331
350/340
NS NM NM NS NM NM 141.2/132.1 313.4/358.2 280/240
NS
NS
130.0/150.0 153.0/200.0 NS
Duration of surgery (mean min) FTS/CS
135:0/162:0 110:150
20:0/20:0
45:1/36:9
NM NS 80:0/80:0
0:30/0:30
63:0/52:0
20:0/20:0
Pathology (malign/ benign n) FTS/CS
FTS fast track surgery, CS conventional surgery, ASA American Society of Anesthesiologists, NM not mentioned, NS no statistically significant difference
35:26/51:49
van Dam et al. (2008) 62:60
111:24/133:29 71:64:0:0/88:74:0:0
18:2/18:2
31:15/23:22
NS NS 66:14/59:21
12:18/14:16
46:20/34:15
16:4/13:7
Age (mean Sex year) FTS/ (male/ CS female) FTS/CS
Characteristics of patients and operative details of included studies comparing FTS with CS
Author (year)
Table 2
13
12
14 15
17
16
15
19
13 13 17
13
10
16
Number of key FTS items (n) FTS group
J Gastrointest Surg
J Gastrointest Surg Table 3
Outcomes of case-control studies comparing FTS with CS
Author (year)
TLoS (days) FTS/CS
Complication rate (n/total) FTS/CS
Mortality (n/total) FTS/CS
Readmission (n/total) FTS/CS
First flatus time after operation (h) FTS/CS
Hospitalization expense (10,000 yuan) FTS/CS
van Dam et al. (2008) Stoot et al. (2009) Lin et al. (2011) Sánchez-Pérez et al. (2012) Du et al. (2013)
6 (3–82)/8 (4–68)a 5 (3–10)/7 (3–12)a 7 (3–26)/11 (4–37)a 2.5 (1–39)/3 (1–22)a 5.1±1.4/11.9±3.1b
25:61/31:100 2:13/2:13 26:56/27:61 3:26/2:17 NM
0:61/2:100 0:13/0:13 1:56/1:61 0:26/0:17 0:60/0:60
8:61/10:100 0:13/0:13 4:56/2:61 1:26/1:17 0:60/0:60
NM NM NM NM 20.3±1.7/50.1±1.2b
NM NM NM NM 2.5±0.2/3.8±0.3b
NM not mentioned, TLoS total length of hospital stay, FTS fast track surgery, CS conventional surgery a
Median (range)
b
Mean±standard deviation
−1.40, P<0.00001, I2 =92 %) (Fig. 2). We used random-effect model to evaluate this outcome. Clinical heterogeneity analysis found that four studies30,31,37,39 adopted laparoscopic hepatectomy while open liver resection was adopted in the other 10 studies.27–29,32–36,38,40 FTS vs. CS—Complication Rate Information on complication rate could be acquired from all nine RCTs27–35 with a total of 933 patients (459 in the FTS group and 474 in the CS group). The merged complication rate in the FTS group was 19 % (85 of 459) and that in the CS group was 27 % (129 of 474). Meta-analysis demonstrated that the complication rate was statistically significantly lower in the FTS group than the CS group (RR=0.65, 95 % CI=0.52 to 0.81, P=0.0001, I2 =0 %)
Table 4
(Fig. 3). Fixed-effect model was used. Funnel plot showed a by and large symmetry, which indicated little publication bias existed (Fig. 4). FTS vs. CS—Mortality All trials27–35 with 933 patients (459 in the FTS group and 474 in the CS group) reported mortality. Jones et al.33 reported one patient died in the FTS group and it was the same situation in the CS group (RR = 0.98, 95 % CI = 0.06 to15.17, P = 0.987), the difference between the FTS and CS groups was not statistically significant. However, eight other studies27–32,34,35 reported that no patient died in either the FTS or CS group. Meta-analysis showed that there was no statistically significant difference in perioperative mortality between the FTS and CS groups.
Outcomes of randomized controlled trials comparing FTS with CS
Author (year)
TLoS (days) FTS/CS
Complication Mortality (n/ Readmission First flatus time after rate (n/total) total) FTS/CS (n/total) FTS/CS operation (h) FTS/CS FTS/CS
Fan et al. (2011) Chi et al. (2012) Pu et al. (2012) Wang et al. (2013)
6.5±1.0/7.5±1.5b 10.2±1.9/15.5±2.7b 7.40±2.00/8.39±2.98b 10.3±5.7/13.9±6.4b
3:20/2:20 15:63/16:52 4:30/6:30 21:35/26:35
0:20/0:20 0:63/0:52 0:30/0:30 0:35/0:35
NM NM NM NM
1:30/3:30 24:80/37:80
0:30/0:30 0:80/0:80
NM NM
Huang et al. (2013) 6.5±1.7/8.5±2.1b Ni et al. (2013) 6.9±2.8/8.0±3.7b Jones et al. (2013) Shou et al. (2014) Lu et al. (2014)
4 (3–5)/7 (6–8)a 8:46/14:45 1:46/1:45 7.0±0.8/8.5±0.9b 2:20/13:20 0:20/0:20 11.56±3.10/13.87±5.08b 7:135/12:162 0:135/0:162
2:46/0:45 0:20/0:20 0:135/0:162
46.35±5.76/68.70±6.45b 18.6±1.7/53.4±2.9b 45.25±4.66/67.60±5.35b 53.76±36.72/ 89.52±40.08b 14.3±3.2/30.8±7.6b 55.2±36/76.8±52.8b
1.7±0.45/2.1±0.50b 5.4±0.3/6.2±0.2b NM 2.31±0.97/2.38±1.25b
NM 60±12/103.2±16.8b 61.83±20.78/ 67.90±24.22b
NM 3.6±0.3/4.1±0.3b 3.84±0.19/3.95±0.36b
NM not mentioned, TLoS total length of hospital stay, FTS fast track surgery, CS conventional surgery a
Median (range)
b
Mean±standard deviation
Hospitalization expense (10,000 yuan) FTS/CS
1.82±0.21/2.53±0.35b NM
J Gastrointest Surg
Fig. 2 Meta-analysis of TLoS in RCTs comparing the FTS group with the CS group
FTS vs. CS—Readmission Three RCTs33–35 with 428 patients (201 in the FTS group and 227 in the CS group) provided data on this outcome. Jones et al.33 reported that two patients in the FTS group were readmitted compared with none in the CS group (RR = 4.89, 95 % CI =0.24 to 99.18, P =0.153), and this difference was not statistically significant. But two other trials34,35 reported that no patient was readmitted in each group. Hence, the entire difference with respect to the outcome-readmission was not statistically significant between two groups.
Secondary Outcome Data of Randomized Controlled Trials FTS vs. CS—First Flatus Time After Operation Data on first flatus time after operation could be found from eight studies27–32,34,35 with 842 patients (413 in the FTS group and 429 in the CS group). Meta-analysis demonstrated that first flatus time after operation in FTS group was statistically significantly shorter than CS group (WMD=−24.69, 95 % CI = −32.90 to −16.48, P < 0.00001, I 2 = 98 %) (Fig. 5). Random-effect model was utilized as I2 >50 %.
FTS vs. CS—Hospitalization Expense We could gain information on hospitalization expense from six RCTs27,28,30,31,34,35 with 622 patients (303 in the FTS group and 319 in the CS group). Meta-analysis manifested that hospitalization expense was less in the FTS group than the CS group, and this difference was statistically significant (WMD = −0.45, 95 % CI = −0.78 to −0.12, P = 0.007, I2 =97 %) (Fig. 6). In this case, random-effect model was used.
Primary Outcome Data of Case-Control Studies FTS vs. CS—Total Length of Hospital Stay (TLoS) All five case-control studies (CCSs)36–40 with 467 patients (216 in the FTS group and 251 in the CS group) provided data on TLoS. Meta-analysis also proved that TLoS was still shorter in the FTS group than the CS group, and this difference was statistically significant (WMD=−3.21, 95 % CI=−6.18 to −0.25, P=−0.03, I2 =89 %) (Fig. 7). Random-effect model was utilized. The results of CCSs about this outcome were consistent with aforementioned results of RCTs. Funnel plot for this outcome also showed a little asymmetry, and we considered some degree of publication bias might be existed (Fig. 8).
Fig. 3 Meta-analysis of complication rate in RCTs comparing the FTS group with the CS group
J Gastrointest Surg Fig. 4 Funnel plot of outcome: complication rate in RCTs
FTS vs. CS—Complication Rate Detailed data concerning this outcome could be gained from four CCSs36–39 with 347 patients (156 in the FTS group and 191 in the CS group). The pooled complication rate in the FTS group was 36 % (56 of 156) and that in the CS group was 32 % (62 of 191). Metaanalysis indicated no statistically significant difference in complication rate between two groups (RR = 1.16, 95 % CI=0.88 to 1.54, P=0.29, I2 =0 %) (Fig. 9). As I2 =0 %, fixed-effect model was used. FTS vs. CS—Mortality We could get data on mortality from all five CCSs36–40 with 467 patients (216 in the FTS group and 251 in the CS group). Lin et al.38 indicated that there was one death in each group, van Dam et al.36 reported that there were no deaths in the FTS group and two patients died in the CS group (P=0.526). The remaining three CCSs37,39,40 described no deaths in each group. The merged mortality in the FTS group was 0.4 % (1 of 216) and that in the CS group was 1.2 % (3 of 251). Meta-analysis demonstrated that perioperative mortality was lower in the FTS group than the CS group, but the difference between two groups was not statistically
significant (RR = 0.58, 95 % CI = 0.08 to 4.13, P = 0.59, I2 =0 %) (Fig. 10). Fixed-effect model was also used. FTS vs. CS—Readmission Information on readmission could be obtained from all five CCSs36–40 with 467 patients (216 in the FTS group and 251 in the CS group). Stoot et al.37 and Du et al.40 reported that no patient was readmitted in either the FTS or CS group. The pooled readmission rate in the FTS group was 6 % (13 of 216) and that in the CS group was 5 % (13 of 251). Meta-analysis of the remaining three CCSs36,38,39 showed that the difference between two groups was not statistically significant with respect to this outcome (RR=1.39, 95 % CI=0.67 to 2.91, P=0.38, I2 =0 %) (Fig. 11). Fixedeffect model was also utilized. Secondary Outcome Data of Case-Control Studies FTS vs. CS—First Flatus Time After Operation and Hospitalization Expense Only from one CCS40 could we obtain the specific data on first flatus time after operation and hospitalization expense. Thus, our meta-analysis was limited for
Fig. 5 Meta-analysis of first flatus time after operation in RCTs comparing the FTS group with the CS group
J Gastrointest Surg
Fig. 6 Meta-analysis of hospitalization expense in RCTs comparing the FTS group with the CS group
these two outcomes owing to be short of relevant data. Our analysis demonstrated that first flatus time after operation was statistically significantly shorter in the FTS group than the CS group (WMD=−29.8, 95 % CI=−30.33 to −29.27, P<0.05) and hospitalization expense was also statistically significantly less in the FTS group than the CS group (WMD=−1.30, 95 % CI=−1.39 to −1.21, P<0.05). Sensitivity Analysis After excluding some low-quality studies, reanalyzing the data, and comparing the later results with the above-mentioned findings, we found that the posterior results and our abovementioned findings were in line with each other. Our results still did not change if we converted the statistic model between fixed-effect model and random effects model. We also estimated some missing or variable data reasonably and reanalyzed them, and the above findings still did not change. As a consequence, the sensitivity of our meta-analysis was low and we believed that the significant potential bias might not exist. In other words, our above findings were somewhat more credible.
Discussion A certain amount of clinical trials including randomized controlled trials, case-control studies, cohort studies, and even systematic review and meta-analyses have been published with the purpose of investigating the superiority of FT programs so far; however, the majority of them have been designed and
performed mainly in colorectal surgery.41 Relative to being used successfully and maturely in colorectal surgery, FT programs are performed infrequently in other types of surgery such as hepatic resection. Our present meta-analysis investigated the effects of perioperative FT programs compared with CS programs on surgical outcomes in liver resection. Four previous systematic reviews42–45 have explored this theme by now; however, these previous reviews included no RCTs (Coolsen et al.45 included two RCTs that did not purely compared FT programs with CS programs—both control and experimental groups in these two RCTs were all treated with FT programs) and analyzed some case-control studies or retrospective case series studies that solely discussed the effects of FTS programs without setting a control group, this may reduce the potency of systematic review. A recent systematic review46 contained two RCTs and published in 2014, also investigated this topic and showed more cogency than previous studies. Even so, previous reviews were still limited due to the finite number of included RCTs. The authors of these systematic reviews performed observational and descriptive systematic review rather than an essential statistical meta-analysis. Compared with previous systematic reviews, our present research included altogether nine RCTs and five CCSs, all of which specially compared FT programs with CS programs (control group) in liver surgery. The total sample size of our present research was large enough to come to a relatively convinced conclusion. For these reasons, our meta-analysis represents the current best available evidence. Our meta-analysis demonstrated that perioperative FT programs resulted in decreased TLoS, complication rate,
Fig. 7 Meta-analysis of TLoS in CCSs comparing the FTS group with the CS group
J Gastrointest Surg Fig. 8 Funnel plot of outcome: TLoS in CCSs
postoperative first flatus time, and hospitalization expense, but this decrease was not repeated in perioperative mortality and readmission rate. The above-mentioned findings were in accordance with the results of meta-analysis of CCSs. Death and readmissions were rare events in all included studies. Just because the data from included studies appeared to be an approximately equivalent mortality and readmission rates between the FTS and the CS groups, we did not find any decrease in perioperative mortality and readmission rate between the two groups. The results of RCTs and CCSs aligned fairly well means that our findings have a relative high credibility. Although most of our findings were compatible with previous systematic reviews,42–46 it should be pointed out that previous reviews found that the complication rate did not reduce after applying FT programs, but our research discovered a notable reduction in complication rate in the FTS groups. Taking different statistical analysis methods, our findings are still in line with the most results of previous reviews;9,42–46 this made us convince our results and it appears that FT programs for liver surgery are feasible and effective. The varied components of FT programs among studies were also found. Roughly similar key FTS items were utilized
in studies, but there were still some individual differences regarding the detailed ingredients of FT programs existing. For instance, Lu et al.35 presented that oral feeding 6 h after surgery but liquid and slag free diet till postoperative day 5, while Jones et al.33 presented normally feeding on postoperative day 0. In the item of early mobilization, Ni et al.32 recommended patients mobilized out of bed at least four times per day on postoperative day 1, Chi et al.28 recommended patients mobilized on postoperative day 2, Lu et al.35 suggested that out of bed ambulation on postoperative day 1 (less than 2 h) and gradually increase from postoperative day 2, Fan et al.27 indicated in bed ambulation on postoperative day 0 and out of bed ambulation on postoperative day 1 (less than 2 h), whereas Jones et al.33 recommended patients mobilized out of bed twice per day on postoperative day 1. Similar differences in other FTS items among included studies were also revealed. The number of FTS items used in studies was also varied. Therefore, we could not know what favorable effects a specific FTS item could give and recommend the first-rank and thorough FT programs due to the fact that no unified perioperative FT programs for liver surgery have been published. Although our research revealed the benefits of application of FT programs during liver surgery, some particular items
Fig. 9 Meta-analysis of complication rate in CCSs comparing the FTS group with the CS group
J Gastrointest Surg
Fig. 10 Meta-analysis of mortality in CCSs comparing the FTS group with the CS group
successfully used in colorectal surgery are still controversial when applied in hepatic resection. For example, a recommended thoracic epidural in FT programs during colorectal surgery might not be for liver surgery because epidurals might affect postoperative recovery.46–50 Paracetamol were utilized as a usual analgesic in some FT programs, but it might induce liver damage and increase postoperative complications especially in major liver resection, thus we should discreetly use some anesthetic that may cause liver damage.46,51 Compared with other operations, liver surgery has its own particularity. The liver is the human body’s metabolism center with complex anatomical structures. Ischemia reperfusion injury, intraoperative blood loss, and perioperative use of drugs can lead to postoperative liver function instability. Therefore, ideal FT programs for liver surgery should be refined according to its own characteristics other than simply duplicate versions of other types of surgery. Future studies should optimize some FTS items and combine pre-, intra-, and postoperative programs perfectly according to the characteristics of liver surgery. One other thing to be noted is that adherence to FT programs was rarely mentioned in either included studies27–40 or other studies.52 Adherence is a very important indicator that may have an influence on the effects of FT programs and affect some outcomes such as TLoS, complication rate, mortality, readmission, and hospitalization expense. Under some circumstances, adherence rate even contributes to the heterogeneity of research. Thus, future studies should value it and report it in detail.
Our meta-analysis examined and confirmed the beneficial effects produced by FT programs in liver surgery, but some limitations also existed. Firstly, not all included studies employed open hepatectomy and this could generate clinical heterogeneity. From another perspective, similar FT programs could also provide benefits when applied to different types of surgery. Next, the majority of included trials were single center studies which might limit the potency and feasibility of studies. Thus, more high quality, multicenter RCTs concerning this theme are urgently needed to optimize FT programs. Finally, discharge criterion, which can largely affect TLoS, was rarely mentioned. Although some studies explicitly put forward discharge criterions, they were varied from each other. This limitation would also result in clinical heterogeneity. Future studies should optimize some FTS items that are not fit for liver surgery, contain detailed information on adherence and discharge criterions, and combine pre-, intra-, and postoperative programs perfectly according to the characteristics of liver surgery. We also look forward to unified international FT programs for liver surgery will come out soon.
Fig. 11 Meta-analysis of readmission in CCSs comparing FTS group with CS group. FT fast track, FTS fast track surgery, CS conventional surgery, TLoS total length of hospital stay, RR risk ratio, WMD weighted
mean difference, CI confidence intervals, RCTs randomized controlled trials, CCSs case-control studies
Conclusions In summary, our present meta-analysis demonstrated that perioperative implementation of FT programs in liver surgery could observably decrease TLoS, complication rate, postoperative first flatus time, and hospitalization expense, and did not
J Gastrointest Surg
compromise mortality and readmission rate. Thus, it appears that FT programs in liver surgery are feasible and effective. Future studies should report detailed information on adherence and discharge criterions, and optimize FT programs catering to liver surgery. Conflicts of interest The authors declare that they have no conflict of interest. Author contributions Wu SJ and Cheng NS designed the study, Wu SJ and Cheng Y performed the literature research and collected the data; Wu SJ, Xiong XZ, Lu J and Lin YX assessed literature quality; Cheng NS, Wu SJ, Lu J, Lin YX and Zhou RX analyzed the data; Wu SJ wrote the paper.
References 1.
2.
3.
4.
5. 6. 7.
8.
9. 10.
11.
12.
13.
14.
15.
van den Broek MA, van Dam RM, van Breukelen GJ, et al.: Development of a composite endpoint for randomized controlled trials in liver surgery. Br J Surg. 2011; 98:1138–1145. Dimick JB, Wainess RM, Cowan JA, et al.: National trends in the use and outcomes of hepatic resection. J Am Coll Surg. 2004; 199: 31–38. Jarnagin WR, Gonen M, Fong Y, et al.: Improvement in perioperative outcome after hepatic resection: analysis of 1803 consecutive cases over the past decade. Ann Surg. 2002; 236: 397–407. Basse L, Hjort Jakobsen D, Billesbolle P, et al.: A clinical pathway to accelerate recovery after colonic resection. Ann Surg. 2000; 232: 51–57. Kehlet H. Multimodal approach to control postoperative pathophysiology and rehabilitation.: Br J Anaesth. 1997; 78: 606–617. Kehlet H, Wilmore DW.: Multimodal strategies to improve surgical outcome. Am J Surg. 2002; 183: 630–641. Holte K, Kehlet H.: Epidural anaesthesia and analgesia-effects on surgical stress response and implications for postoperative nutrition. Clin Nutr. 2002; 21: 99–206. Kehlet H, Dahl JB.: The value of Bmultimodal^ or Bbalanced analgesia^ in postoperative pain treatment. Anesth Analg. 1993; 77(5): 1048–1056. Kehlet H, Wilmore DW.: Evidence-based surgical care and the evolution of fast-track surgery. Ann Surg. 2008; 248(2): 189–198. Basse L, Raskov HH, Hjort Jakobsen D, et al.: Accelerated postoperative recovery programme after colonic resection improves physical performance, pulmonary function and body composition. Br J Surg. 2002; 89(4): 446–453. Brustia P, Renghi A, Gramaglia L, et al.: Mininvasive abdominal aortic surgery. Early recovery and reduced hospitalization after multidisciplinary approach. J Cardiovasc Surg. 2003; 44: 629–635. Podore PC, Throop EB.: Infrarenal aortic surgery with a 3-day hospital stay: a report on success with a clinical pathway. J Vasc Surg. 1999; 29: 787–792. Barbieri A, Vanhaecht K, Van Herck P, et al.: Effects of clinical pathways in the joint replacement: a meta-analysis. BMC Med. 2009; 7: 32. Arsalani-Zadeh R, ElFadl D, Yassin N, et al.: Evidence-based review of enhancing postoperative recovery after breast surgery. Br J Surg. 2011; 98: 181–196. Rawlinson A, Kang P, Evans J, et al.: A systematic review of enhanced recovery protocols in colorectal surgery. Ann R Coll Surg Engl. 2011; 93: 583–588.
16.
McCarty TM, Arnold DT, Lamont JP, et al.: Optimizing outcomes in bariatric surgery: outpatient laparoscopic gastric bypass. Ann Surg. 2005; 242: 494–501. 17. Kirsh EJ, Worwag EM, Sinner M, et al.: Using outcome data and patient satisfaction surveys to develop policies regarding minimum length of hospitalization after radical prostatectomy. Urology. 2000; 56: 101–107. 18. Scott N, McDonald D, Campbell J, et al.: The use of enhanced recovery after surgery (ERAS) principles in Scottish orthopaedic units – an implementation and follow-up at 1 year, 2010–2011: a report from the Muculoskeletal Audit, Scotland. Arch Orthop Trauma Surg. 2013; 133: 117–124. 19. Lv D, Wang X, Shi G.: Perioperative enhanced recovery programmes for gynaecological cancer patients. Cochrane Database Syst Rev. 2010; (6): CD008239 20. Koch MO, Smith JA.: Influence of patient age and co-morbidity on outcome of a collaborative care pathway after radical prostatectomy and cystoprostatectomy. J Urol. 1996; 155: 1681–1684. 21. Leibman BD, Dillioglugil O, Abbas F, et al.: Impact of a clinical pathway for radical retropubic prostatectomy. Urology. 1998; 52: 94–99. 22. Moher D, Liberati A, Tetzlaff J, et al.: Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009; 6: e1000097. 23. Balshem H, Helfand M, Schunemann HJ, et al.: GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol. 2011; 64: 401–406. 24. Deeks JJ, Higgins JPT, Altman DG (editors): Chapter 9: Analyzing data and undertaking meta-analyses. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration 2011. Available at: www.cochrane-handbook.org. Accessed August 1, 2011. 25. Sterne JAC, Egger M, Moher D (editors): Chapter 10: Addressing reporting biases. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration 2011. Available at: www.cochrane-handbook.org. Accessed August 1, 2011. 26. Hozo SP, Djulbegovic B, Hozo I.: Estimating the mean and variance from the median, range, and the size of a sample. BMC Medical Research Methodology. 2005; 5: 13. 27. Fan XJ, Tan HY, Xiao YM, et al.: Application of fast track surgery in perioperative management of patients with primary hepatic cancer. Chinese Journal of General Surgery. 2011; 20 (7): 683–686. 28. Chi MH, Zeng YY, Liu JF.: Fast-track surgical treatment of liver cancer patients through liver resection within the perioperative period. Chinese Journal of Clinical Oncology. 2012; 39 (23): 1939– 1942. 29. Pu CR, Ji Q, Zhang JM, et al.: Application of nursing of fast recovering in perioperative period of hepatectomy. Journal of Modern Clinical Medicine. 2012; 38 (5): 375–377. 30. Wang XQ, Lv SD, Wang YB.: Application of nursing of fast recovering model during perioperative period of laparoscopic hepatectomy. Parenteral and Enteral Nutrition. 2013; 20 (2): 126–128. 31. Huang H, Zhang HC, Mo SF, et al.: The application of fast track surgery in laparoscopic precise liver resection. China Journal of Endoscopy. 2013; 19 (6): 603–606. 32. Ni CY, Yang Y, Chang YQ, et al.: Fast-track surgery improves postoperative recovery in patients undergoing partial hepatectomy for primary liver cancer: A prospective randomized controlled trial. Eur J Surg Oncol. 2013; 39: 542–547. 33. Jones C, Kelliher L, Dickinson M, et al.: Randomized clinical trial on enhanced recovery versus standard care following open liver resection. Br J Surg. 2013; 100: 1015–1024.
J Gastrointest Surg 34.
Shou ZX, Zheng DW, Luo YX, et al.: Value of fast track surgery principles in the perioperative management of liver cancer patients after hepatectomy. Chin J Dig Surg. 2014; 13 (6): 456–460. 35. Lu H, Fan Y, Zhang F, et al.: Fast-track surgery improves postoperative outcomes after hepatectomy. Hepatogastroenterology. 2014; 61(129): 168–172. 36. van Dam R, Hendry P, Coolsen M, et al.: Initial experience with a multimodal enhanced recovery programme in patients undergoing liver resection. Br J Surg. 2008; 95: 969–975. 37. Stoot J, van Dam R, Busch O, et al.: The effect of multimodal fasttrack programme on outcomes in laparoscopic liver surgery: a multicentre pilot study. HPB. 2009; 11: 140–144. 38. Lin DX, Li X, Ye QW, et al.: Implementation of a fast-track clinical pathway decreases postoperative length of stay and hospital charges for liver resection. Cell Biochem Biophys. 2011; 61: 413–419. 39. Sànchez-Pérez B, Aranda-Narvaez J, Suarez-Munoz M, et al.: Fasttrack programme in laparoscopic liver surgery: theory of fact. World J Gastrointest Surg. 2012; 4: 246–250. 40. Du ZS, Zhao LT, He Q, et al.: Application of fast track surgery in liver cancer hepatectomy. Military Medical Journal of Southeast China. 2013; 15(6): 588–591. 41. Lemmens L, van Zelm R, Borel Rinkes I, et al.: Clinical and organizational content of clinical pathways for digestive surgery: a systematic review. Dig Surg. 2009; 26: 91–99. 42. Olsén MF, Wennberg E.: Fast-Track Concepts in Major Open Upper Abdominal and Thoracoabdominal Surgery: A Review. World J Surg. 2011; 35: 2586–2593. 43. Spelt L, Ansari D, Sturesson C, et al.: Fast-track programmes for hepatopancreatic resections: where do we stand? HPB. 2011; 13: 833–838.
44.
Hall TC, Dennison AR, Bilku DK, et al.: Enhanced recovery programmes in hepatobiliary and pancreatic surgery: a systematic review. Ann R Coll Surg Engl. 2012; 94: 318–326. 45. Coolsen MME, Wong-Lun-Hing EM, van Dam RM, et al.: A systematic review of outcomes in patients undergoing liver surgery in an enhanced recovery after surgery pathways. HPB. 2013; 15: 245– 251. 46. Hughes MJ, McNally S, Wigmore SJ.: Enhanced recovery following liver surgery: a systematic review and meta-analysis. HPB. 2014; 16: 699–706. 47. Lassen K, Soop M, Nygren J, et al.: Consensus review of optimal perioperative care in colorectal surgery. Arch Surg. 2009; 144: 961– 969. 48. Koea J, Young Y, Gunn K.: Fast track liver resection: the effect of a comprehensive care package and analgesia with single dose intrathecal morphine with gabapentin or continuous epidural analgesia. HPB Surg. 2009: 271986. 49. Revie E, McKeown D, Wilson J, et al.: Randomized clinical trial of local infiltration plus patient-controlled opiate analgesia vs. epidural analgesia following liver resection surgery. HPB. 2012; 14: 611– 618. 50. Tzimas P, Prout J, Papadopolous G, et al.: Epidural anaesthesia and analgesia for liver resection. Anaesthesia. 2013; 68: 628–635. 51. Galinski M, Delhotal-Landes B, Lockey D, et al.: Reduction of paracetamol metabolism after hepatic resection. Pharmacology. 2006; 77: 161–165. 52. Ahmed J, Khan S, Lim M, et al.: Enhanced recovery after surgery protocols – compliance and variations in practice during routine colorectal surgery. Col Dis. 2011; 14: 1045–1051.