Available online at www.sciencedirect.com
ScienceDirect EJSO 40 (2014) 1763e1770
www.ejso.com
Systematic review of nasogastric or nasojejunal decompression after gastrectomy for gastric cancer Z.-W. Wei a,b,e, J.-L. Li c,e, Z.-S. Li d, Y.-T. Hao c, Y.-L. He a,b,*, W. Chen a,b, C.-H. Zhang a,b,* a
Department of Gastrointestinopancreatic Surgery of the First Affiliated Hospital of Sun Yat-sen University, 58 Zhongshan 2nd Road, Guangzhou, Guangdong 510080, China b Gastric Cancer Center of Sun Yat-sen University, 58 Zhongshan 2nd Road, Guangzhou, Guangdong 510080, China c Department of Medical Statistics and Epidemiology, School of Public Health, 74 Zhongshan 2nd Road, Guangzhou, Guangdong 510080, China d Department of General Surgery of People’s Hospital of Chizhou, 3 Baiya Road, Chizhou, Anhui 247000, China Accepted 26 May 2014 Available online 3 July 2014
Abstract Background: The aim of this meta-analysis was to evaluate the necessity of nasogastric or nasojejunal decompression after gastrectomy for gastric cancer. Methods: Medline, Embase and the Cochrane Library were searched. Only prospective randomized controlled trials (RCTs) that compared subjects with and without nasogastric or nasojejunal decompression after gastrectomy were eligible in this meta-analysis. Time to flatus, time to first oral intake, length of hospital stay, reinsertion rate, anastomotic leakage, pulmonary complications, morbidity and mortality were evaluated. Results: Eight studies finally fulfilled the inclusion criteria. This meta-analysis enrolled 1141 patients, 570 randomized to routine decompression and 571 randomized to no decompression. Time to first oral intake was significantly shorter in the non-decompression group (WMD ¼ 0.53, 95% CI: 0.28 to 0.77; p < 0.001). Additionally, subjects with nasogastric or nasojejunal decompression experienced a longer hospital stay ( p ¼ 0.001). Time to flatus, anastomotic leakage, reinsertion rates, pulmonary complications, morbidity and mortality rates were similar between the two groups. Conclusion: Nasogastric or nasojejunal decompression does not facilitate the recovery of bowel function or reduce the risk of postoperative complications. Therefore, routine nasogastric or nasojejunal decompression is unnecessary after gastrectomy for gastric cancer. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Gastric cancer; Gastrectomy; Nasogastric decompression; Nasojejunal decompression
Introduction Nasogastric or nasojejunal decompression has been introduced into the treatment of acute intestinal obstruction and postoperative ileus since the 1930s.1 It was considered that nasogastric or nasojejunal decompression could be either * Corresponding authors. Department of Gastrointestinopancreatic Surgery, The First Affiliated Hospital of Sun Yat-sen University, 58 Zhongshan 2nd Road, Guangzhou, Guangdong 510080, China. Tel./fax: þ86 20 87331059. E-mail addresses: [email protected] (Y.-L. He), zhchangh@ mail.sysu.edu.cn (C.-H. Zhang). e Z.-W. Wei and J.-L. Li contributed equal to this work. http://dx.doi.org/10.1016/j.ejso.2014.05.013 0748-7983/Ó 2014 Elsevier Ltd. All rights reserved.
therapeutic or diagnostic approaches in major abdominal operations. Nowadays, such decompressions are considered to be unnecessary after colorectal surgery.2,3 However, until relatively recent such decompressions have still been used in operations for gastric cancer as a prophylactic measure and described as “the standard of care” and “common practice”.4 Nasogastric or nasojejunal decompression was once considered to be necessary to decrease postoperative ileus (nausea, vomiting, and gastric distension) and to reduce anastomotic leakage after gastrectomy. However, the necessity of nasogastric decompression following elective gastrectomy among gastric cancer patients has been increasingly debated over the last several years. Several randomized and non-
1764
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
randomized trials were performed to evaluate the efficacy of nasogastric or nasojejunal decompression after gastrectomy.5e14 A meta-analysis included 5 small randomized trials of only 717 patients indicated that routine nasogastric or nasojejunal decompression was not necessary after gastrectomy for gastric cancer.15 However, this metaanalysis had some limitations. First, the sample size of this systematic analysis was limited. It was published in 2008 and excluded some continuous outcomes reported in the form of median, which led to a smaller sample size. In addition, several randomized clinical trials10e12 have been published after 2008. Therefore, it is necessary to carry out another metaanalysis with a larger sample size to evaluate the necessity of nasogastric or nasojejunal decompression after gastrectomy for gastric cancer. There was still another meta-analysis collected 8 trials regarding the necessity of indwelling gastrointestinal decompressions after gastrectomy.16 However, it included some studies incorporating gastrectomy for benign diseases. Compared to benign diseases, gastrectomy for gastric cancer is often combined with lymph nodes dissection (especially D2 and D3), which may increase surgical morbidity and gut dysfunction. Thus, it is necessary to conduct a meta-analysis to evaluate the necessity of nasogastric or nasojejunal decompression after gastrectomy for gastric cancer.
Methods and materials Literature search To identify all published clinical studies evaluating nasogastric or nasojejunal decompression after
gastrectomy, a comprehensive search of medical databases (Medline, Embase and the Cochrane Library) were performed using the terms nasogastric decompression, nasojejunal decompression, gastrointestinal decompression, nasogastric tube insertion, gastrectomy and gastric cancer surgery (Fig. 1). Inclusion criteria and exclusion criteria The inclusion criteria were (1) studies comparing nasogastric or nasojejunal decompression with no nasogastric or nasojejunal decompression after gastrectomy; (2) prospective RCTs; (3) clinical studies contained patients underwent gastrectomy for gastric cancer between 1921 and 2013; (4) English language publications. The following articles were excluded: (1) emergency surgery; (2) non-randomized trials; (3) articles not reporting outcomes of interest; (4) duplicate publication. Data extraction Data were extracted independently by two authors (ZW Wei and JL Li) and cross-checked to reach consensus. The following characteristics were extracted for each study: author, publication date, journal, number of patients, research design, geographical region, type of operation, time to first flatus, time to first oral intake, reinsertion of tubes, anastomotic leakage, pulmonary complications, and length of hospital stay, morbidity and mortality. This study was approved by the Ethical Review Committee of Sun Yat-sen University.
Figure 1. Systematic search and selection strategy.
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
Assessment of study quality Study quality was assessed independently by two investigators (W Chen and ZS Li) according to the Jadad composite scale.17 The maximum possible score was 5 and high quality trials scored more than 2.18 Statistical analysis
1765
50 patients in each group, respectively. Statistical analyses were performed using the software Review Manager Version 5.2 (Nordic Cochrane Centre, the Cochrane Collaboration). Results Description of studies
Continuous variables were assessed using the weighted mean difference (WMD) when both means and standard deviations were reported. But two studies5,6 just reported medians and ranges for continuous variables and we estimated the means and standard deviations from medians, ranges and sample sizes according to the formula described by Hozo.19 Dichotomous variables were analyzed using risk ratio (RR) since all the studies included were randomized controlled trials.20 The equibalance for WMD between groups is set at 0, while for RR is set at 1. When an overall summary effect with a 95% confidence interval that straddles a WMD of 0, or straddles a RR of 1, it corresponds to having no statistically significant difference between groups. Fixed effect model was used to estimate the effect magnitude when P > 0.05. Otherwise, random effect model was used. Sensitivity analysis was performed according to high-quality (3 points) studies and trials with more than
Eight RCTs of 1141 patients after gastrectomy for gastric cancer were enrolled in this meta-analysis, which contained 570 patients in the decompression group and 571 patients in the non-decompression group (Table 1). We collected the data of tube and no-tube group from the study performed by Li and his colleagues, which allocated subjects into three groups: tube group, intra-operative group and no-tube group.10 In addition, two studies were excluded because they included some patients underwent gastrectomy for benign diseases.13,14 Quality assessment of studies Jadad composite scale was used to assess the methodological quality of included studies. The mean of Jadad scores of the included studies was 3 (Table 2).
Table 1 Main characteristic of included randomized controlled trials. Author
Year 5
Wu et al. DB NDB Yoo et al.6 DB NDB Lee et al.7 DB NDB Doglietto et al.8 DB NDB Hsu et al. 9 DB NDB Li et al.10 DB NDB Tavassoli et al.11 DB NDB Pacelli et al.12 DB NDB
1994
2002
2002
2004
2007
2011
2011
2013
Country
Chinese Taipei
Korea
Korea
Italy
Chinese Taipei
China
Iran
Italy
No. of patients
74 37 37 136 69 67 119 63 56 237 116 121 151 76 75 104 50 54 50 25 25 270 134 136
DB ¼ decompression group; NDB ¼ no-decompression group. NR, not reported. BⅠ ¼ Billroth Ⅰ, BⅡ ¼ Billroth Ⅱ, RY ¼ Roux-en-Y. a Median age.
Mean age(years)
Gender
Gastrectomy
Reconstruction
M
F
Subtotal
Total
BⅠ
BⅡ
RY
62a 60a
NR. NR
NR NR
37 37
0 0
0 0
37 37
0 0
58a 55a
52 48
17 19
52 49
17 18
7 12
42 31
20 24
57 59
51 41
12 15
37 34
26 22
11 10
26 24
26 22
63 63
66 72 83 41 42 66 27 39 33 17 16 143 62 81
50 49 68 35 33 38 23 15 17 8 9 127 72 55
0 0
116 121
NR NR
NR NR
NR NR
53 50
23 25
0 0
0 0
76 75
? ?
? ?
36 40
1 1
13 13
0 0
25 25
0 0
0 0
25 25
134 136
0 0
0 0
71 63
63 73
53 58 59.2 57.5 58.4 56.12 68.9 67.7
1766
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
Table 2 Jadad scores of included studies.
Time to first oral intake
Author
Randomization
Double-blind
Withdrawals
Score
Wu et al. Yoo et al. Lee et al. Doglietto et al. Hsu et al. Li et al. Tavassoli et al. Pacelli et al.
2 2 2 2 2 2 1 2
0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1
3 3 3 3 3 3 2 3
Time to flatus Seven studies reported the time to flatus after gastrectomy. Yoo et al.6 found that the median time in the decompression group was longer than that in the nondecompression group. However, a multicenter prospective RCT12 revealed that the time to flatus was significantly shorter in the decompression group after subtotal gastrectomy. The random effect model was used to analyze the data owing to the significant heterogeneity. As shown in Fig. 2A, the time to flatus was shorter in the nondecompression group. However, there were no significant differences between them (WMD ¼ 0.14, 95% CI: 0.16 to 0.44; p ¼ 0.37). Moreover, the difference between the two groups was not significant in time to flatus stratified by total gastrectomy (WMD ¼ 0.14, 95% CI: 0.19 to 0.46; p ¼ 0.41) (Fig. 2B). Only Pacelli et al.12 reported time to flatus after subtotal gastrectomy and revealed that time to flatus were significantly shorter in the nondecompression group after subtotal gastrectomy.
All the studies enrolled in this meta-analysis investigated the effect of nasogastric or nasojejunal tubes on interval time from surgery to first oral intake. Two studies5,6 reported the median time to first oral diet and we estimated the mean time with the formulas described by Hozo.19 Other studies in this meta-analysis reported mean time to first oral intake. Three trials revealed that patients with nasogastric or nasojejunal decompression endured longer intervals to first oral intake.7,9,12 This meta-analysis showed that time to first oral intake was significantly shorter among patients without decompression compared to those with decompression (WMD ¼ 0.53, 95% CI: 0.28e0.77; p < 0.001) (Fig. 3A). When by subtotal gastrectomy, time to first oral intake was still shorter in the non-decompression group (WMD ¼ 0.47, 95% CI: 0.08e0.86; p ¼ 0.02) (Fig. 3B). However, this difference in time to first oral intake was not significant by stratified analysis of total gastrectomy (WMD ¼ 0.20, 95% CI: 0.20 to 0.59; p ¼ 0.33), as shown in Fig. 3B. Length of hospital stay Two studies5,6 reported median hospital stay and mean time was estimated.19 This meta-analysis revealed that the length of hospital stay was significantly shorter among patients without nasogastric or nasojejunal decompression compared to those with decompression (WMD ¼ 0.79, 95% CI: 0.32e1.27; p ¼ 0.001), without significant heterogeneity (Fig. S1A). However, no significant difference was found between the two groups when stratified by subtotal
Figure 2. A. Effects of nasogastric or nasojejunal decompression on time to flatus. B. Effects of nasogastric or nasojejunal decompression on time to flatus after total gastrectomy.
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
1767
Figure 3. A Effects of nasogastric or nasojejunal decompression on time to first oral intake. B. Effects of nasogastric or nasojejunal decompression on time to first oral intake for subgroup analysis.
(WMD ¼ 0.30, 95% CI: 1.33 to 1.92; p ¼ 0.72) or total gastrectomy (WMD ¼ 0.29, 95% CI: 0.58 to 1.16; p ¼ 0.51) (Fig. S1B). Reinsertion of tubes Six studies reported the postoperative reinsertion rate in the decompression group and insertion rate in the nondecompression group. As shown in Fig. S2A, our results revealed that there was no significant difference between the postoperative reinsertion rate in the decompression group and insertion rate in the non-decompression group (RR ¼ 0.82, 95% CI: 0.48e1.42; p ¼ 0.49). Two studies reported the postoperative reinsertion rate in the decompression group and insertion rate in the nondecompression group after subtotal gastrectomy. When stratified by subtotal gastrectomy, no significant difference was found between the postoperative reinsertion rate in the decompression group and insertion rate in the nondecompression group (RR ¼ 0.71, 95% CI: 0.36e1.40; p ¼ 0.32) (Fig. S2B). No studies reported postoperative reinsertion rate after total gastrectomy. Anastomotic leakage All the trials enrolled in this study reported the incidence rate of anastomotic leakage and one study10 did not report
the exact anastomotic leakage incidence in subgroup of subtotal and total gastrectomy. As shown in Fig. S3A, no difference was detected in this outcome (RR ¼ 1.26, 95% CI: 0.67e2.37; p ¼ 0.47), when combining subtotal and total gastrectomy as a whole. Among them, 5 studies reported anastomotic leakage for subtotal gastrectomy and metaanalysis showed that the incidence of leakage after subtotal gastrectomy was similar between the two groups (RR ¼ 1.01, 95% CI: 0.42e2.41; p ¼ 0.99) without heterogeneity. There was no difference for total gastrectomy (RR ¼ 1.39, 95% CI: 0.60e3.22; p ¼ 0.44), either (Fig. S3B). Pulmonary complications The incidence of pulmonary complications was reported in all the eight studies. Pulmonary complications in this meta-analysis mainly included pneumonia, atelectasis and pulmonary embolus. Our results revealed that more pulmonary complications occurred in the nasogastric or nasojejunal group, though the difference was not statistically significant (RR ¼ 1.31, 95% CI: 0.92e1.89; p ¼ 0.14) (Fig. S4A). When stratified by total gastrectomy, pulmonary complications were still more common in the decompression group, though the difference was not statistically significant (RR ¼ 1.39, 95% CI: 0.84e2.31; p ¼ 0.20) (Fig. S4B). However, no difference was found stratified
1768
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
by subtotal gastrectomy (RR ¼ 1.22, 95% CI: 0.63e2.33; p ¼ 0.56). Major and minor complications All the eight studies enrolled in this meta-analysis reported postoperative complications. Major complications in this meta-analysis included pneumonia, abdominal abscess, fasciitis, septic shock, bacteremia, septic coagulopathy, anastomotic leakage, wound dehiscence, digestive tract complications, bleeding, perforation, obstruction, ischemia and pancreatitis. As shown in Fig. S5A, metaanalysis showed no significant difference in major complications between the two groups (RR ¼ 1.27, 95% CI: 0.92e1.76; p ¼ 0.15). There was no difference in major complications between the two groups when stratified by subtotal gastrectomy (RR ¼ 0.98, 95% CI: 0.58e1.65; p ¼ 0.94) (Fig. S5B). Only one study8 reported major complications after total gastrectomy. In our study, minor complications were defined as urinary tract infection, wound infections, fever, atelectasis, pleural effusion, hepatic dysfunction, nausea, regurgitation, vomiting and abdominal distension. Our results showed that minor complications were still similar between the two groups after gastrectomy (RR ¼ 0.90, 95% CI: 0.63e1.29; p ¼ 0.57) (Fig. S6A). As shown in Fig. S6B, no significant difference was found between the two groups when stratified by subtotal (RR ¼ 1.01, 95% CI: 0.32e3.20; p ¼ 0.98) and total (RR ¼ 0.86, 95% CI: 0.53e1.40; p ¼ 0.55) gastrectomy. Morbidity Morbidity rates were reported in 7 studies. In the pooled data, no significant difference in morbidity rates was found between the two groups (RR ¼ 1.10, 95% CI: 0.87e1.40; p ¼ 0.41), without heterogeneity (Fig. S7A). As shown in Fig. S7B, the morbidity rate between the two groups was similar when stratified by subtotal (RR ¼ 0.99, 95% CI: 0.69e1.41; p ¼ 0.94) and total gastrectomy (RR ¼ 1.22, 95% CI: 0.82e1.83; p ¼ 0.33). Mortality Patient deaths occurred in 3 studies.8,9,12 The heterogeneity was not obvious for mortality rates in these studies. In the pooled data, no significant difference was found between the two groups with respect to mortality rates (RR ¼ 1.26, 95% CI: 0.34e4.67; p ¼ 0.72) (Fig. S8). Sensitivity analysis Sensitivity analysis was subsequently performed based on the quality (Jadad scores 3) and the sample size of studies (each group with more than 50 patients) (Table S1). Results of sensitivity analyses based on either the quality or the sample size in this study demonstrated that the
differences of length of hospital stay and time to first oral intake were still statistically significant between the two groups. Moreover, the risk ratio, WMD and significance of differences of the time to flatus, anastomotic leakage, reinsertion of tube, pulmonary complications, morbidity rates and mortality rates were not substantially affected. Publication bias The funnel plot of pulmonary complications (Fig. S9) showed that there was no evidence of obvious publications bias. Discussion Gastrointestinal decompression after abdominal operations was advocated in the past century and commonly used in clinic currently. However, gastrointestinal decompression (GI) after small bowel resection and colorectal surgery was considered to be unnecessary in recent years.2,3 Some studies found that GI decompression was not necessary for patients with gastrectomy.5,21 Because the oesophagojejunal, gastrojejunal or gastroduodenal anastomoses are high risk factors for early postoperative fistula, GI decompression is still routinely used. Gastric cancer is the fourth most prevalent cancer and the second common cause of cancer related death throughout the world.22e24 Gastrectomy with lymph nodes dissection is widely performed for gastric cancer.22,25 Morbidity and mortality rates associated with gastrectomy remain about 13.6% and 0.6%, respectively.26 Thus, GI decompression was still widely used after gastrectomy for gastric cancer to prevent anastomotic leakage and duodenal stump leak.27 However, the necessity of nasogastric or nasojejunal decompression after gastrectomy for gastric cancer remains controversial. It is traditionally considered that gut motility of gastric cancer patients after gastrectomy attenuates, which delays the recovery of bowel function, and GI decompression may facilitate its recovery. The present meta-analysis showed that the time to flatus in the non-decompression group was even shorter, though this difference was not statistically significant. Moreover, the time to first oral intake of subjects with decompression was significantly longer than that of those without. Our results demonstrated that gastrointestinal decompression might delay postoperative recovery. Recent studies about fast-track surgery have demonstrated that early oral intake could accelerate postoperative recovery combined with other various techniques used in the care for patients undergoing elective surgeries.28 Moreover, some studies29,30 reported that nasogastric or nasojejunal decompression delayed the recovery of gut function after elective abdominal surgeries, which was similar to our findings after gastrectomy for gastric cancer in meta-analysis.
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
Reinsertion of nasogastric or nasojejunal tubes is required when postoperative complications occur such as anastomotic leakage, bleeding and ileus. Our metaanalysis demonstrated that the postoperative reinsertion rate in the decompression group was similar to the insertion rate in the non-decompression group. Routine decompression is more commonly used in clinic in patients with subtotal gastrectomy compared to total gastrectomy. In this study, there was no significant difference in reinsertion rate between the two groups regarding to stratified analysis of subtotal gastrectomy for gastric cancer. Thus, gastrointestinal decompression did not decrease reinsertion rate of nasogastric or nasojejunal tube after gastrectomy for gastric cancer. To reduce postoperative complications, especially anastomotic leakage, was an important reason for indwelling nasogastric or nasojejunal tubes after gastrectomy. Our study demonstrated that the incidence rate of anastomotic leakage was similar between the two groups no matter after subtotal gastrectomy or total gastrectomy. Although the placement of nasogastric or nasojejunal decompression is an option for the treatment of postoperative anastomotic leakage,31 outcomes of the present meta-analysis demonstrated that such decompression did not decrease the risk of anastomotic leakage after gastrectomy for gastric cancer. Some studies showed that risk factors for developing anastomotic leakage were ischemia, neoadjuvant therapy, comorbid conditions, splenectomy and pancreatectomy.32,33 It is also well established that inadequate vascularization and tension are major issues for leading to anastomotic leakage.34 Pulmonary complications are common after gastrectomy. Many surgeons considered that nasogastric or nasojejunal decompression could reduce the risk of aspiration and decrease postoperative pulmonary complications. Our meta-analysis demonstrated that the incidence of pulmonary complications was lower in the non-decompression group, but this difference was not statistically significant. Ephgrave et al. suggested that postoperative decompression would restrict cough, expectoration and breathing exercises, which might contribute to pulmonary complications.35 Moreover, one study showed that nasogastric or nasojejunal decompression was an independent risk factor for pulmonary complications after elective non-thoracic surgery.36 In addition, our meta-analysis showed that there was no significant difference between the decompression and non-decompression group, with respect to major or minor postoperative complications. Morbidity rates between the two groups were similar in the present meta-analysis. Furthermore, nasogastric or nasojejunal tubes may cause discomfort such as sore throat, nasal discomfort and vomiting. Carrireand and his colleagues reported in 2007 that complains related to the nasogastric or nasojejunal tubes were moderate to severe in 72% of patients after a total gastrectomy.13 A total of 11 patients died in all the eight studies included in our meta-analysis. The major causes of death
1769
were anastomotic leakage (n ¼ 6), duodenal stump leak (n ¼ 3), acute heart failure (n ¼ 1) and pulmonary embolism (n ¼ 1). Our meta-analysis showed that the mortality rate was similar between the two groups. In conclusion, there is no evidence that nasogastric or nasojejunal decompression after gastrectomy can reduce the risk of postoperative complications, such as anastomotic leakage. Conversely, the time to first oral intake and length of hospital stay in the decompression group were even longer. Although the mean Jadad score of eligible trials was 3, the methodological quality of the studies was still a limitation of this meta-analysis. Our results showed that routine nasogastric or nasojejunal decompression was not necessary after elective gastrectomy for gastric cancer. Appendix A. Supplementary data Supplementary data related to this chapter can be found at http://dx.doi.org/10.1016/j.ejso.2014.05.013. Grants This study was supported by National Natural Science Foundation of China (30700805, 81272637 and 81272643). Conflicts of interest We declare that we have no conflicts of interest.
References 1. Wangensteen OH, Paine JR. Treatment of acute intestinal obstruction by suction with the duodenal tube. J Am Med Assoc 1933;101(20): 1532–9. 2. Wolff BG, Pembeton JH, van Heerden JA, et al. Elective colon and rectal surgery without nasogastric decompression. A prospective, randomized trial. Ann Surg 1989;209(6):670–3. 3. Otchy DP, Wolff BG, van Heerden JA, et al. Does the avoidance of nasogastric decompression following elective abdominal colorectal surgery affect the incidence of incisional hernia? Results of a prospective, randomized trial. Dis Colon Rectum 1995;38(6):604–8. 4. Soybel DI. Nasogastric decompression following laparotomy. Gastroenterology 1996;110(2):645–6. 5. Wu CC, Hwang CR, Liu TJ. There is no need for nasogastric decompression after partial gastrectomy with extensive lymphadenectomy. Eur J Surg 1994;160(6e7):369–73. 6. Yoo CH, Son BH, Han WK, et al. Nasogastric decompression is not necessary in operations for gastric cancer: prospective randomised trial. Eur J Surg 2002;168(7):379–83. 7. Lee JH, Hyung WJ, Noh SH. Comparison of gastric cancer surgery with versus without nasogastric decompression. Yonsei Med J 2002; 43(4):451–6. 8. Doglietto GB, Papa V, Tortorelli AP, et al. Nasojejunal tube placement after total gastrectomy: a multicenter prospective randomized trial. Arch Surg 2004;139(12):1309–13. 9. Hsu S-D, Yu J-C, Chen T-W, et al. Role of nasogastric tube insertion after gastrectomy. Viszeralmedizin 2007;23(3):303–6.
1770
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
10. Li C, Mei JW, Yan M, et al. Nasogastric decompression for radical gastrectomy for gastric cancer: a prospective randomized controlled study. Dig Surg 2011;28(3):167–72. 11. Tavassoli A, Rajabi MT, Abdollahi A, et al. Efficacy and necessity of nasojejunal tube after gastrectomy. Int J Surg 2011;9(3):233–6. 12. Pacelli F, Rosa F, Marrelli D, et al. Naso-gastric or naso-jejunal decompression after partial distal gastrectomy for gastric cancer. Final results of a multicenter prospective randomized trial. Gastric Cancer 2013. 13. Carrere N, Seulin P, Julio CH, et al. Is nasogastric or nasojejunal decompression necessary after gastrectomy? A prospective randomized trial. World J Surg 2007;31(1):122–7. 14. Rossetti G, Fei L, Docimo L, et al. Is nasogastric decompression useful in prevention of leaks after laparoscopic sleeve gastrectomy? A randomized trial. J Invest Surg 2014;27(4):234–9. 15. Yang Z, Zheng Q, Wang Z. Meta-analysis of the need for nasogastric or nasojejunal decompression after gastrectomy for gastric cancer. Br J Surg 2008;95(7):809–16. 16. Ding J, Liao G, Xia Y, et al. The necessity of indwelling gastrointestinal decompression after gastrectomy: a meta-analysis. J Surg Res 2013;179(1):e71–81. 17. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17(1):1–12. 18. Moher D, Pham B, Jones A, et al. Does quality of reports of randomised trials affect estimates of intervention efficacy reported in meta-analyses? Lancet 1998;352(9128):609–13. 19. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol 2005;5:13. 20. Deeks JJ. Issues in the selection of a summary statistic for metaanalysis of clinical trials with binary outcomes. Stat Med 2002; 21(11):1575–600. 21. Akbaba S, Kayaalp C, Savkilioglu M. Nasogastric decompression after total gastrectomy. Hepatogastroenterology 2004;51(60):1881–5. 22. Zhang CH, Zhan WH, He YL, et al. Spleen preservation in radical surgery for gastric cardia cancer. Ann Surg Oncol 2007;14(4):1312–9. 23. Wei ZW, Li JL, Wu Y, et al. Impact of pre-existing type-2 diabetes on patient outcomes after radical resection for gastric cancer: a retrospective cohort study. Dig Dis Sci 2014;59(5):1017–24.
24. Jemal A, Bray F, Center MM, et al. Global cancer statistics. CA Cancer J Clin 2011;61(2):69–90. 25. Zhang C, He Y, Schwarz RE, et al. Evaluation of para-aortic nodal dissection for locoregionally advanced gastric cancer with 1e3 involved para-aortic nodes. Chin Med J (Engl) 2014;127(3):435–41. 26. Degiuli M, Sasako M, Calgaro M, et al. Morbidity and mortality after d1 and d2 gastrectomy for cancer: interim analysis of the Italian Gastric Cancer Study Group (IGCSG) randomised surgical trial. Eur J Surg Oncol 2004;30(3):303–8. 27. Robinson JR. Duodenal stump leakage: causes and prevention. Am J Surg 1950;79(4):549–56. 28. Wind J, Polle SW, Fung Kon Jin PH, et al. Systematic review of enhanced recovery programmes in colonic surgery. Br J Surg 2006; 93(7):800–9. 29. Nelson R, Edwards S, Tse B. Prophylactic nasogastric decompression after abdominal surgery. Cochrane Database Syst Rev 2007;(3): CD004929. 30. Nelson R, Tse B, Edwards S. Systematic review of prophylactic nasogastric decompression after abdominal operations. Br J Surg 2005; 92(6):673–80. 31. Lang H, Piso P, Stukenborg C, et al. Management and results of proximal anastomotic leaks in a series of 1114 total gastrectomies for gastric carcinoma. Eur J Surg Oncol 2000;26(2):168–71. 32. Sierzega M, Kolodziejczyk P, Kulig J. Impact of anastomotic leakage on long-term survival after total gastrectomy for carcinoma of the stomach. Br J Surg 2010;97(7):1035–42. 33. Briel JW, Tamhankar AP, Hagen JA, et al. Prevalence and risk factors for ischemia, leak, and stricture of esophageal anastomosis: gastric pull-up versus colon interposition. J Am Coll Surg 2004;198(4):536– 41. [discussion 541e532]. 34. Holscher AH, Vallbohmer D, Brabender J. The prevention and management of perioperative complications. Best Pract Res Clin Gastroenterol 2006;20(5):907–23. 35. Ephgrave KS, Kleiman-Wexler R, Pfaller M, et al. Postoperative pneumonia: a prospective study of risk factors and morbidity. Surgery 1993;114(4):815–9. [discussion 819e821]. 36. McAlister FA, Bertsch K, Man J, et al. Incidence of and risk factors for pulmonary complications after nonthoracic surgery. Am J Respir Crit Care Med 2005;171(5):514–7.
ScienceDirect EJSO 40 (2014) 1763e1770
www.ejso.com
Systematic review of nasogastric or nasojejunal decompression after gastrectomy for gastric cancer Z.-W. Wei a,b,e, J.-L. Li c,e, Z.-S. Li d, Y.-T. Hao c, Y.-L. He a,b,*, W. Chen a,b, C.-H. Zhang a,b,* a
Department of Gastrointestinopancreatic Surgery of the First Affiliated Hospital of Sun Yat-sen University, 58 Zhongshan 2nd Road, Guangzhou, Guangdong 510080, China b Gastric Cancer Center of Sun Yat-sen University, 58 Zhongshan 2nd Road, Guangzhou, Guangdong 510080, China c Department of Medical Statistics and Epidemiology, School of Public Health, 74 Zhongshan 2nd Road, Guangzhou, Guangdong 510080, China d Department of General Surgery of People’s Hospital of Chizhou, 3 Baiya Road, Chizhou, Anhui 247000, China Accepted 26 May 2014 Available online 3 July 2014
Abstract Background: The aim of this meta-analysis was to evaluate the necessity of nasogastric or nasojejunal decompression after gastrectomy for gastric cancer. Methods: Medline, Embase and the Cochrane Library were searched. Only prospective randomized controlled trials (RCTs) that compared subjects with and without nasogastric or nasojejunal decompression after gastrectomy were eligible in this meta-analysis. Time to flatus, time to first oral intake, length of hospital stay, reinsertion rate, anastomotic leakage, pulmonary complications, morbidity and mortality were evaluated. Results: Eight studies finally fulfilled the inclusion criteria. This meta-analysis enrolled 1141 patients, 570 randomized to routine decompression and 571 randomized to no decompression. Time to first oral intake was significantly shorter in the non-decompression group (WMD ¼ 0.53, 95% CI: 0.28 to 0.77; p < 0.001). Additionally, subjects with nasogastric or nasojejunal decompression experienced a longer hospital stay ( p ¼ 0.001). Time to flatus, anastomotic leakage, reinsertion rates, pulmonary complications, morbidity and mortality rates were similar between the two groups. Conclusion: Nasogastric or nasojejunal decompression does not facilitate the recovery of bowel function or reduce the risk of postoperative complications. Therefore, routine nasogastric or nasojejunal decompression is unnecessary after gastrectomy for gastric cancer. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Gastric cancer; Gastrectomy; Nasogastric decompression; Nasojejunal decompression
Introduction Nasogastric or nasojejunal decompression has been introduced into the treatment of acute intestinal obstruction and postoperative ileus since the 1930s.1 It was considered that nasogastric or nasojejunal decompression could be either * Corresponding authors. Department of Gastrointestinopancreatic Surgery, The First Affiliated Hospital of Sun Yat-sen University, 58 Zhongshan 2nd Road, Guangzhou, Guangdong 510080, China. Tel./fax: þ86 20 87331059. E-mail addresses: [email protected] (Y.-L. He), zhchangh@ mail.sysu.edu.cn (C.-H. Zhang). e Z.-W. Wei and J.-L. Li contributed equal to this work. http://dx.doi.org/10.1016/j.ejso.2014.05.013 0748-7983/Ó 2014 Elsevier Ltd. All rights reserved.
therapeutic or diagnostic approaches in major abdominal operations. Nowadays, such decompressions are considered to be unnecessary after colorectal surgery.2,3 However, until relatively recent such decompressions have still been used in operations for gastric cancer as a prophylactic measure and described as “the standard of care” and “common practice”.4 Nasogastric or nasojejunal decompression was once considered to be necessary to decrease postoperative ileus (nausea, vomiting, and gastric distension) and to reduce anastomotic leakage after gastrectomy. However, the necessity of nasogastric decompression following elective gastrectomy among gastric cancer patients has been increasingly debated over the last several years. Several randomized and non-
1764
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
randomized trials were performed to evaluate the efficacy of nasogastric or nasojejunal decompression after gastrectomy.5e14 A meta-analysis included 5 small randomized trials of only 717 patients indicated that routine nasogastric or nasojejunal decompression was not necessary after gastrectomy for gastric cancer.15 However, this metaanalysis had some limitations. First, the sample size of this systematic analysis was limited. It was published in 2008 and excluded some continuous outcomes reported in the form of median, which led to a smaller sample size. In addition, several randomized clinical trials10e12 have been published after 2008. Therefore, it is necessary to carry out another metaanalysis with a larger sample size to evaluate the necessity of nasogastric or nasojejunal decompression after gastrectomy for gastric cancer. There was still another meta-analysis collected 8 trials regarding the necessity of indwelling gastrointestinal decompressions after gastrectomy.16 However, it included some studies incorporating gastrectomy for benign diseases. Compared to benign diseases, gastrectomy for gastric cancer is often combined with lymph nodes dissection (especially D2 and D3), which may increase surgical morbidity and gut dysfunction. Thus, it is necessary to conduct a meta-analysis to evaluate the necessity of nasogastric or nasojejunal decompression after gastrectomy for gastric cancer.
Methods and materials Literature search To identify all published clinical studies evaluating nasogastric or nasojejunal decompression after
gastrectomy, a comprehensive search of medical databases (Medline, Embase and the Cochrane Library) were performed using the terms nasogastric decompression, nasojejunal decompression, gastrointestinal decompression, nasogastric tube insertion, gastrectomy and gastric cancer surgery (Fig. 1). Inclusion criteria and exclusion criteria The inclusion criteria were (1) studies comparing nasogastric or nasojejunal decompression with no nasogastric or nasojejunal decompression after gastrectomy; (2) prospective RCTs; (3) clinical studies contained patients underwent gastrectomy for gastric cancer between 1921 and 2013; (4) English language publications. The following articles were excluded: (1) emergency surgery; (2) non-randomized trials; (3) articles not reporting outcomes of interest; (4) duplicate publication. Data extraction Data were extracted independently by two authors (ZW Wei and JL Li) and cross-checked to reach consensus. The following characteristics were extracted for each study: author, publication date, journal, number of patients, research design, geographical region, type of operation, time to first flatus, time to first oral intake, reinsertion of tubes, anastomotic leakage, pulmonary complications, and length of hospital stay, morbidity and mortality. This study was approved by the Ethical Review Committee of Sun Yat-sen University.
Figure 1. Systematic search and selection strategy.
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
Assessment of study quality Study quality was assessed independently by two investigators (W Chen and ZS Li) according to the Jadad composite scale.17 The maximum possible score was 5 and high quality trials scored more than 2.18 Statistical analysis
1765
50 patients in each group, respectively. Statistical analyses were performed using the software Review Manager Version 5.2 (Nordic Cochrane Centre, the Cochrane Collaboration). Results Description of studies
Continuous variables were assessed using the weighted mean difference (WMD) when both means and standard deviations were reported. But two studies5,6 just reported medians and ranges for continuous variables and we estimated the means and standard deviations from medians, ranges and sample sizes according to the formula described by Hozo.19 Dichotomous variables were analyzed using risk ratio (RR) since all the studies included were randomized controlled trials.20 The equibalance for WMD between groups is set at 0, while for RR is set at 1. When an overall summary effect with a 95% confidence interval that straddles a WMD of 0, or straddles a RR of 1, it corresponds to having no statistically significant difference between groups. Fixed effect model was used to estimate the effect magnitude when P > 0.05. Otherwise, random effect model was used. Sensitivity analysis was performed according to high-quality (3 points) studies and trials with more than
Eight RCTs of 1141 patients after gastrectomy for gastric cancer were enrolled in this meta-analysis, which contained 570 patients in the decompression group and 571 patients in the non-decompression group (Table 1). We collected the data of tube and no-tube group from the study performed by Li and his colleagues, which allocated subjects into three groups: tube group, intra-operative group and no-tube group.10 In addition, two studies were excluded because they included some patients underwent gastrectomy for benign diseases.13,14 Quality assessment of studies Jadad composite scale was used to assess the methodological quality of included studies. The mean of Jadad scores of the included studies was 3 (Table 2).
Table 1 Main characteristic of included randomized controlled trials. Author
Year 5
Wu et al. DB NDB Yoo et al.6 DB NDB Lee et al.7 DB NDB Doglietto et al.8 DB NDB Hsu et al. 9 DB NDB Li et al.10 DB NDB Tavassoli et al.11 DB NDB Pacelli et al.12 DB NDB
1994
2002
2002
2004
2007
2011
2011
2013
Country
Chinese Taipei
Korea
Korea
Italy
Chinese Taipei
China
Iran
Italy
No. of patients
74 37 37 136 69 67 119 63 56 237 116 121 151 76 75 104 50 54 50 25 25 270 134 136
DB ¼ decompression group; NDB ¼ no-decompression group. NR, not reported. BⅠ ¼ Billroth Ⅰ, BⅡ ¼ Billroth Ⅱ, RY ¼ Roux-en-Y. a Median age.
Mean age(years)
Gender
Gastrectomy
Reconstruction
M
F
Subtotal
Total
BⅠ
BⅡ
RY
62a 60a
NR. NR
NR NR
37 37
0 0
0 0
37 37
0 0
58a 55a
52 48
17 19
52 49
17 18
7 12
42 31
20 24
57 59
51 41
12 15
37 34
26 22
11 10
26 24
26 22
63 63
66 72 83 41 42 66 27 39 33 17 16 143 62 81
50 49 68 35 33 38 23 15 17 8 9 127 72 55
0 0
116 121
NR NR
NR NR
NR NR
53 50
23 25
0 0
0 0
76 75
? ?
? ?
36 40
1 1
13 13
0 0
25 25
0 0
0 0
25 25
134 136
0 0
0 0
71 63
63 73
53 58 59.2 57.5 58.4 56.12 68.9 67.7
1766
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
Table 2 Jadad scores of included studies.
Time to first oral intake
Author
Randomization
Double-blind
Withdrawals
Score
Wu et al. Yoo et al. Lee et al. Doglietto et al. Hsu et al. Li et al. Tavassoli et al. Pacelli et al.
2 2 2 2 2 2 1 2
0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1
3 3 3 3 3 3 2 3
Time to flatus Seven studies reported the time to flatus after gastrectomy. Yoo et al.6 found that the median time in the decompression group was longer than that in the nondecompression group. However, a multicenter prospective RCT12 revealed that the time to flatus was significantly shorter in the decompression group after subtotal gastrectomy. The random effect model was used to analyze the data owing to the significant heterogeneity. As shown in Fig. 2A, the time to flatus was shorter in the nondecompression group. However, there were no significant differences between them (WMD ¼ 0.14, 95% CI: 0.16 to 0.44; p ¼ 0.37). Moreover, the difference between the two groups was not significant in time to flatus stratified by total gastrectomy (WMD ¼ 0.14, 95% CI: 0.19 to 0.46; p ¼ 0.41) (Fig. 2B). Only Pacelli et al.12 reported time to flatus after subtotal gastrectomy and revealed that time to flatus were significantly shorter in the nondecompression group after subtotal gastrectomy.
All the studies enrolled in this meta-analysis investigated the effect of nasogastric or nasojejunal tubes on interval time from surgery to first oral intake. Two studies5,6 reported the median time to first oral diet and we estimated the mean time with the formulas described by Hozo.19 Other studies in this meta-analysis reported mean time to first oral intake. Three trials revealed that patients with nasogastric or nasojejunal decompression endured longer intervals to first oral intake.7,9,12 This meta-analysis showed that time to first oral intake was significantly shorter among patients without decompression compared to those with decompression (WMD ¼ 0.53, 95% CI: 0.28e0.77; p < 0.001) (Fig. 3A). When by subtotal gastrectomy, time to first oral intake was still shorter in the non-decompression group (WMD ¼ 0.47, 95% CI: 0.08e0.86; p ¼ 0.02) (Fig. 3B). However, this difference in time to first oral intake was not significant by stratified analysis of total gastrectomy (WMD ¼ 0.20, 95% CI: 0.20 to 0.59; p ¼ 0.33), as shown in Fig. 3B. Length of hospital stay Two studies5,6 reported median hospital stay and mean time was estimated.19 This meta-analysis revealed that the length of hospital stay was significantly shorter among patients without nasogastric or nasojejunal decompression compared to those with decompression (WMD ¼ 0.79, 95% CI: 0.32e1.27; p ¼ 0.001), without significant heterogeneity (Fig. S1A). However, no significant difference was found between the two groups when stratified by subtotal
Figure 2. A. Effects of nasogastric or nasojejunal decompression on time to flatus. B. Effects of nasogastric or nasojejunal decompression on time to flatus after total gastrectomy.
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
1767
Figure 3. A Effects of nasogastric or nasojejunal decompression on time to first oral intake. B. Effects of nasogastric or nasojejunal decompression on time to first oral intake for subgroup analysis.
(WMD ¼ 0.30, 95% CI: 1.33 to 1.92; p ¼ 0.72) or total gastrectomy (WMD ¼ 0.29, 95% CI: 0.58 to 1.16; p ¼ 0.51) (Fig. S1B). Reinsertion of tubes Six studies reported the postoperative reinsertion rate in the decompression group and insertion rate in the nondecompression group. As shown in Fig. S2A, our results revealed that there was no significant difference between the postoperative reinsertion rate in the decompression group and insertion rate in the non-decompression group (RR ¼ 0.82, 95% CI: 0.48e1.42; p ¼ 0.49). Two studies reported the postoperative reinsertion rate in the decompression group and insertion rate in the nondecompression group after subtotal gastrectomy. When stratified by subtotal gastrectomy, no significant difference was found between the postoperative reinsertion rate in the decompression group and insertion rate in the nondecompression group (RR ¼ 0.71, 95% CI: 0.36e1.40; p ¼ 0.32) (Fig. S2B). No studies reported postoperative reinsertion rate after total gastrectomy. Anastomotic leakage All the trials enrolled in this study reported the incidence rate of anastomotic leakage and one study10 did not report
the exact anastomotic leakage incidence in subgroup of subtotal and total gastrectomy. As shown in Fig. S3A, no difference was detected in this outcome (RR ¼ 1.26, 95% CI: 0.67e2.37; p ¼ 0.47), when combining subtotal and total gastrectomy as a whole. Among them, 5 studies reported anastomotic leakage for subtotal gastrectomy and metaanalysis showed that the incidence of leakage after subtotal gastrectomy was similar between the two groups (RR ¼ 1.01, 95% CI: 0.42e2.41; p ¼ 0.99) without heterogeneity. There was no difference for total gastrectomy (RR ¼ 1.39, 95% CI: 0.60e3.22; p ¼ 0.44), either (Fig. S3B). Pulmonary complications The incidence of pulmonary complications was reported in all the eight studies. Pulmonary complications in this meta-analysis mainly included pneumonia, atelectasis and pulmonary embolus. Our results revealed that more pulmonary complications occurred in the nasogastric or nasojejunal group, though the difference was not statistically significant (RR ¼ 1.31, 95% CI: 0.92e1.89; p ¼ 0.14) (Fig. S4A). When stratified by total gastrectomy, pulmonary complications were still more common in the decompression group, though the difference was not statistically significant (RR ¼ 1.39, 95% CI: 0.84e2.31; p ¼ 0.20) (Fig. S4B). However, no difference was found stratified
1768
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
by subtotal gastrectomy (RR ¼ 1.22, 95% CI: 0.63e2.33; p ¼ 0.56). Major and minor complications All the eight studies enrolled in this meta-analysis reported postoperative complications. Major complications in this meta-analysis included pneumonia, abdominal abscess, fasciitis, septic shock, bacteremia, septic coagulopathy, anastomotic leakage, wound dehiscence, digestive tract complications, bleeding, perforation, obstruction, ischemia and pancreatitis. As shown in Fig. S5A, metaanalysis showed no significant difference in major complications between the two groups (RR ¼ 1.27, 95% CI: 0.92e1.76; p ¼ 0.15). There was no difference in major complications between the two groups when stratified by subtotal gastrectomy (RR ¼ 0.98, 95% CI: 0.58e1.65; p ¼ 0.94) (Fig. S5B). Only one study8 reported major complications after total gastrectomy. In our study, minor complications were defined as urinary tract infection, wound infections, fever, atelectasis, pleural effusion, hepatic dysfunction, nausea, regurgitation, vomiting and abdominal distension. Our results showed that minor complications were still similar between the two groups after gastrectomy (RR ¼ 0.90, 95% CI: 0.63e1.29; p ¼ 0.57) (Fig. S6A). As shown in Fig. S6B, no significant difference was found between the two groups when stratified by subtotal (RR ¼ 1.01, 95% CI: 0.32e3.20; p ¼ 0.98) and total (RR ¼ 0.86, 95% CI: 0.53e1.40; p ¼ 0.55) gastrectomy. Morbidity Morbidity rates were reported in 7 studies. In the pooled data, no significant difference in morbidity rates was found between the two groups (RR ¼ 1.10, 95% CI: 0.87e1.40; p ¼ 0.41), without heterogeneity (Fig. S7A). As shown in Fig. S7B, the morbidity rate between the two groups was similar when stratified by subtotal (RR ¼ 0.99, 95% CI: 0.69e1.41; p ¼ 0.94) and total gastrectomy (RR ¼ 1.22, 95% CI: 0.82e1.83; p ¼ 0.33). Mortality Patient deaths occurred in 3 studies.8,9,12 The heterogeneity was not obvious for mortality rates in these studies. In the pooled data, no significant difference was found between the two groups with respect to mortality rates (RR ¼ 1.26, 95% CI: 0.34e4.67; p ¼ 0.72) (Fig. S8). Sensitivity analysis Sensitivity analysis was subsequently performed based on the quality (Jadad scores 3) and the sample size of studies (each group with more than 50 patients) (Table S1). Results of sensitivity analyses based on either the quality or the sample size in this study demonstrated that the
differences of length of hospital stay and time to first oral intake were still statistically significant between the two groups. Moreover, the risk ratio, WMD and significance of differences of the time to flatus, anastomotic leakage, reinsertion of tube, pulmonary complications, morbidity rates and mortality rates were not substantially affected. Publication bias The funnel plot of pulmonary complications (Fig. S9) showed that there was no evidence of obvious publications bias. Discussion Gastrointestinal decompression after abdominal operations was advocated in the past century and commonly used in clinic currently. However, gastrointestinal decompression (GI) after small bowel resection and colorectal surgery was considered to be unnecessary in recent years.2,3 Some studies found that GI decompression was not necessary for patients with gastrectomy.5,21 Because the oesophagojejunal, gastrojejunal or gastroduodenal anastomoses are high risk factors for early postoperative fistula, GI decompression is still routinely used. Gastric cancer is the fourth most prevalent cancer and the second common cause of cancer related death throughout the world.22e24 Gastrectomy with lymph nodes dissection is widely performed for gastric cancer.22,25 Morbidity and mortality rates associated with gastrectomy remain about 13.6% and 0.6%, respectively.26 Thus, GI decompression was still widely used after gastrectomy for gastric cancer to prevent anastomotic leakage and duodenal stump leak.27 However, the necessity of nasogastric or nasojejunal decompression after gastrectomy for gastric cancer remains controversial. It is traditionally considered that gut motility of gastric cancer patients after gastrectomy attenuates, which delays the recovery of bowel function, and GI decompression may facilitate its recovery. The present meta-analysis showed that the time to flatus in the non-decompression group was even shorter, though this difference was not statistically significant. Moreover, the time to first oral intake of subjects with decompression was significantly longer than that of those without. Our results demonstrated that gastrointestinal decompression might delay postoperative recovery. Recent studies about fast-track surgery have demonstrated that early oral intake could accelerate postoperative recovery combined with other various techniques used in the care for patients undergoing elective surgeries.28 Moreover, some studies29,30 reported that nasogastric or nasojejunal decompression delayed the recovery of gut function after elective abdominal surgeries, which was similar to our findings after gastrectomy for gastric cancer in meta-analysis.
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
Reinsertion of nasogastric or nasojejunal tubes is required when postoperative complications occur such as anastomotic leakage, bleeding and ileus. Our metaanalysis demonstrated that the postoperative reinsertion rate in the decompression group was similar to the insertion rate in the non-decompression group. Routine decompression is more commonly used in clinic in patients with subtotal gastrectomy compared to total gastrectomy. In this study, there was no significant difference in reinsertion rate between the two groups regarding to stratified analysis of subtotal gastrectomy for gastric cancer. Thus, gastrointestinal decompression did not decrease reinsertion rate of nasogastric or nasojejunal tube after gastrectomy for gastric cancer. To reduce postoperative complications, especially anastomotic leakage, was an important reason for indwelling nasogastric or nasojejunal tubes after gastrectomy. Our study demonstrated that the incidence rate of anastomotic leakage was similar between the two groups no matter after subtotal gastrectomy or total gastrectomy. Although the placement of nasogastric or nasojejunal decompression is an option for the treatment of postoperative anastomotic leakage,31 outcomes of the present meta-analysis demonstrated that such decompression did not decrease the risk of anastomotic leakage after gastrectomy for gastric cancer. Some studies showed that risk factors for developing anastomotic leakage were ischemia, neoadjuvant therapy, comorbid conditions, splenectomy and pancreatectomy.32,33 It is also well established that inadequate vascularization and tension are major issues for leading to anastomotic leakage.34 Pulmonary complications are common after gastrectomy. Many surgeons considered that nasogastric or nasojejunal decompression could reduce the risk of aspiration and decrease postoperative pulmonary complications. Our meta-analysis demonstrated that the incidence of pulmonary complications was lower in the non-decompression group, but this difference was not statistically significant. Ephgrave et al. suggested that postoperative decompression would restrict cough, expectoration and breathing exercises, which might contribute to pulmonary complications.35 Moreover, one study showed that nasogastric or nasojejunal decompression was an independent risk factor for pulmonary complications after elective non-thoracic surgery.36 In addition, our meta-analysis showed that there was no significant difference between the decompression and non-decompression group, with respect to major or minor postoperative complications. Morbidity rates between the two groups were similar in the present meta-analysis. Furthermore, nasogastric or nasojejunal tubes may cause discomfort such as sore throat, nasal discomfort and vomiting. Carrireand and his colleagues reported in 2007 that complains related to the nasogastric or nasojejunal tubes were moderate to severe in 72% of patients after a total gastrectomy.13 A total of 11 patients died in all the eight studies included in our meta-analysis. The major causes of death
1769
were anastomotic leakage (n ¼ 6), duodenal stump leak (n ¼ 3), acute heart failure (n ¼ 1) and pulmonary embolism (n ¼ 1). Our meta-analysis showed that the mortality rate was similar between the two groups. In conclusion, there is no evidence that nasogastric or nasojejunal decompression after gastrectomy can reduce the risk of postoperative complications, such as anastomotic leakage. Conversely, the time to first oral intake and length of hospital stay in the decompression group were even longer. Although the mean Jadad score of eligible trials was 3, the methodological quality of the studies was still a limitation of this meta-analysis. Our results showed that routine nasogastric or nasojejunal decompression was not necessary after elective gastrectomy for gastric cancer. Appendix A. Supplementary data Supplementary data related to this chapter can be found at http://dx.doi.org/10.1016/j.ejso.2014.05.013. Grants This study was supported by National Natural Science Foundation of China (30700805, 81272637 and 81272643). Conflicts of interest We declare that we have no conflicts of interest.
References 1. Wangensteen OH, Paine JR. Treatment of acute intestinal obstruction by suction with the duodenal tube. J Am Med Assoc 1933;101(20): 1532–9. 2. Wolff BG, Pembeton JH, van Heerden JA, et al. Elective colon and rectal surgery without nasogastric decompression. A prospective, randomized trial. Ann Surg 1989;209(6):670–3. 3. Otchy DP, Wolff BG, van Heerden JA, et al. Does the avoidance of nasogastric decompression following elective abdominal colorectal surgery affect the incidence of incisional hernia? Results of a prospective, randomized trial. Dis Colon Rectum 1995;38(6):604–8. 4. Soybel DI. Nasogastric decompression following laparotomy. Gastroenterology 1996;110(2):645–6. 5. Wu CC, Hwang CR, Liu TJ. There is no need for nasogastric decompression after partial gastrectomy with extensive lymphadenectomy. Eur J Surg 1994;160(6e7):369–73. 6. Yoo CH, Son BH, Han WK, et al. Nasogastric decompression is not necessary in operations for gastric cancer: prospective randomised trial. Eur J Surg 2002;168(7):379–83. 7. Lee JH, Hyung WJ, Noh SH. Comparison of gastric cancer surgery with versus without nasogastric decompression. Yonsei Med J 2002; 43(4):451–6. 8. Doglietto GB, Papa V, Tortorelli AP, et al. Nasojejunal tube placement after total gastrectomy: a multicenter prospective randomized trial. Arch Surg 2004;139(12):1309–13. 9. Hsu S-D, Yu J-C, Chen T-W, et al. Role of nasogastric tube insertion after gastrectomy. Viszeralmedizin 2007;23(3):303–6.
1770
Z.-W. Wei et al. / EJSO 40 (2014) 1763e1770
10. Li C, Mei JW, Yan M, et al. Nasogastric decompression for radical gastrectomy for gastric cancer: a prospective randomized controlled study. Dig Surg 2011;28(3):167–72. 11. Tavassoli A, Rajabi MT, Abdollahi A, et al. Efficacy and necessity of nasojejunal tube after gastrectomy. Int J Surg 2011;9(3):233–6. 12. Pacelli F, Rosa F, Marrelli D, et al. Naso-gastric or naso-jejunal decompression after partial distal gastrectomy for gastric cancer. Final results of a multicenter prospective randomized trial. Gastric Cancer 2013. 13. Carrere N, Seulin P, Julio CH, et al. Is nasogastric or nasojejunal decompression necessary after gastrectomy? A prospective randomized trial. World J Surg 2007;31(1):122–7. 14. Rossetti G, Fei L, Docimo L, et al. Is nasogastric decompression useful in prevention of leaks after laparoscopic sleeve gastrectomy? A randomized trial. J Invest Surg 2014;27(4):234–9. 15. Yang Z, Zheng Q, Wang Z. Meta-analysis of the need for nasogastric or nasojejunal decompression after gastrectomy for gastric cancer. Br J Surg 2008;95(7):809–16. 16. Ding J, Liao G, Xia Y, et al. The necessity of indwelling gastrointestinal decompression after gastrectomy: a meta-analysis. J Surg Res 2013;179(1):e71–81. 17. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17(1):1–12. 18. Moher D, Pham B, Jones A, et al. Does quality of reports of randomised trials affect estimates of intervention efficacy reported in meta-analyses? Lancet 1998;352(9128):609–13. 19. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol 2005;5:13. 20. Deeks JJ. Issues in the selection of a summary statistic for metaanalysis of clinical trials with binary outcomes. Stat Med 2002; 21(11):1575–600. 21. Akbaba S, Kayaalp C, Savkilioglu M. Nasogastric decompression after total gastrectomy. Hepatogastroenterology 2004;51(60):1881–5. 22. Zhang CH, Zhan WH, He YL, et al. Spleen preservation in radical surgery for gastric cardia cancer. Ann Surg Oncol 2007;14(4):1312–9. 23. Wei ZW, Li JL, Wu Y, et al. Impact of pre-existing type-2 diabetes on patient outcomes after radical resection for gastric cancer: a retrospective cohort study. Dig Dis Sci 2014;59(5):1017–24.
24. Jemal A, Bray F, Center MM, et al. Global cancer statistics. CA Cancer J Clin 2011;61(2):69–90. 25. Zhang C, He Y, Schwarz RE, et al. Evaluation of para-aortic nodal dissection for locoregionally advanced gastric cancer with 1e3 involved para-aortic nodes. Chin Med J (Engl) 2014;127(3):435–41. 26. Degiuli M, Sasako M, Calgaro M, et al. Morbidity and mortality after d1 and d2 gastrectomy for cancer: interim analysis of the Italian Gastric Cancer Study Group (IGCSG) randomised surgical trial. Eur J Surg Oncol 2004;30(3):303–8. 27. Robinson JR. Duodenal stump leakage: causes and prevention. Am J Surg 1950;79(4):549–56. 28. Wind J, Polle SW, Fung Kon Jin PH, et al. Systematic review of enhanced recovery programmes in colonic surgery. Br J Surg 2006; 93(7):800–9. 29. Nelson R, Edwards S, Tse B. Prophylactic nasogastric decompression after abdominal surgery. Cochrane Database Syst Rev 2007;(3): CD004929. 30. Nelson R, Tse B, Edwards S. Systematic review of prophylactic nasogastric decompression after abdominal operations. Br J Surg 2005; 92(6):673–80. 31. Lang H, Piso P, Stukenborg C, et al. Management and results of proximal anastomotic leaks in a series of 1114 total gastrectomies for gastric carcinoma. Eur J Surg Oncol 2000;26(2):168–71. 32. Sierzega M, Kolodziejczyk P, Kulig J. Impact of anastomotic leakage on long-term survival after total gastrectomy for carcinoma of the stomach. Br J Surg 2010;97(7):1035–42. 33. Briel JW, Tamhankar AP, Hagen JA, et al. Prevalence and risk factors for ischemia, leak, and stricture of esophageal anastomosis: gastric pull-up versus colon interposition. J Am Coll Surg 2004;198(4):536– 41. [discussion 541e532]. 34. Holscher AH, Vallbohmer D, Brabender J. The prevention and management of perioperative complications. Best Pract Res Clin Gastroenterol 2006;20(5):907–23. 35. Ephgrave KS, Kleiman-Wexler R, Pfaller M, et al. Postoperative pneumonia: a prospective study of risk factors and morbidity. Surgery 1993;114(4):815–9. [discussion 819e821]. 36. McAlister FA, Bertsch K, Man J, et al. Incidence of and risk factors for pulmonary complications after nonthoracic surgery. Am J Respir Crit Care Med 2005;171(5):514–7.
