| Background and objectivePancreaticoduodenectomy resection is one of the most complex surgery ingeneral surgery. The case-fatality rate can be controlled below5%with theaccumulation of experience and technology improved in the large Pancreas CenterOver in the past decade. However, because PD involved more organs, wide scope andlong time, its incidence of postoperative complications still reach20%~40%,andpancreatic leakage and fistula (PF) has always been the most common and seriouscomplications. According to reports in the literature, the pancreatic leakage and fistularate of pancreaticoduodenectomy is range from5%to13%, and the mortality of thepancreatic leakage and fistula can be as high as20%~50%. So only take effectivemeasures to prevent the occurrence of PF is the key to successful operation andreduce the operation mortality. How to effectively prevent PF and reduce theincidence of complications has been the problem which surgeons continuously try tosolve. The effective of prophylacticly use of somatostatin in the clinical study ofpancreaticoduodenectomy has been controversial, now there has many randomizedcontrolled trials to assess its effect. But subjecting to the insufficient sample size of case study makes the individual results poor stability. So we apply prospective,randomized, single center clinical research to estimate the effect of somatostatin inpancreatic juice secretion and PF after pancreaticoduodenectomy. Futhermore, usingMeta analysis to make comprehensive analysis of results in a previous study. Withevidence-based medical evidence for quantitative evaluation of the prophylactic useof somatostatin in the clinical effect after pancreaticoduodenectomy. In order toreduce the incidence of PF and mortality after pancreaticoduodenectomy. Through theimprovement methord of end to end pancreatic anastomosis, we are to explore itsimpact on the incidence of PF after pancreaticoduodenectomy.MethodsIn order to observe the effective of Outreotide on inhibiting pancreatic juicesecretion after PD and the action of prevention and treatment of PF, we study90casesof PD of patients with prospective, randomized and single center research betweenSep2009to July2012. Pubmed, EMBASE database,China journal full-text database,VIP database, and combining the method of literature trace, online query(www.baidu.com;www.google.com) are taken to collecte the researches ofprophylactic use of somatostatin in the clinical curative effect of PD. According to themethod recommended by the Cochrane collaboration to meet eligibility criteria of the11Meta analysis study, a total of1041patients. Publication bias funnel chart is usedto assess. With evidence of evidence-based medicine evaluating the effect ofpreventive use of somatostatin after PD. Finally through the retrospective analysis ofour hospital of the clinical data of396cases of PD from January2001to January2011,inorder to compare different ways jejunum anastomosis with the incidence of PF,anddiscuss improvement of pancreatic anastomosis end to end set into type’s influence onthe incidence of PF after PD.Statistical analysis was performed by SPSS software package version17.0(Chicago, IL).Categorical data was tested by Chi-squareor Fisher’s exact test,continuous variables were compared by the Student’s test.P≤0.05was consideredsignificant.Data analysis was performed by using Review Manager4.2software (The Cochrane Collaboration, Oxford, United Kingdom). Each study was weighted bymeans of sample size; statistical heterogeneity was assessed as the proportion ofvariation because of heterogeneity with the I2statistic.To account for clinicalheterogeneity, based on different definitions of outcome parameters, variability ofinterventions, and perioperative management, overall estimates were calculated byusing the random effects models according to DerSimonian and Laird. For this reason,the results of the MA (odds ratio (OR) and corresponding95%confidence interval(95%CI) are presented as more conservative estimates compared with an analysiswith a fixed-effect model in absence of clinical heterogeneity. Potential publicationbias was determined by conducting informal visual inspection of funnel plots basedon the major complication PF.Results1.90patients with PD for prospective, randomized, single center clinical observationshows that three groups of patients had no perioperative death, A total of4cases ofpancreatic fistula (all are grade A and all recovered after conservative treatment),there was no statistically significant difference incidence of pancreatic fistula in threegroups(P＜0.05）.2. Meta analysis shows that the11RCTs are included in the incidence of PF, the fixedeffects model analysis (heterogeneity test p=0.14). There was no statisticallysignificant difference in the incidence of PF after PD between intervention group andcontrol group (OR0.76CI0.5395%~1.08, P=0.13). With the incidence ofpostoperative pancreatic fistula results funnel chart analysis,11RCTs included inbasic within95%CI and basic symmetric distribution, graphics are basically invertedfunnel shape of figure, which suggest that the possibility of publication bias is small,and the conclusion is more reliable. What’s more, these11RCTs are included in theperioperative mortality, by using the fixed effects model analysis (heterogeneity test p=0.67), there was no statistically significant difference in perioperative mortality afterPD between intervention group and control group (OR1.12CI0.6395%~2.01, p=0.69). This study included11RCTS,7research provides two groups of postoperativecomplication rate of total situation, by using the fixed effects model analysis (heterogeneity test p=0.06), the difference complications after PD betweenintervention group and control group is statistically significant (OR0.73CI0.5495%~0.98, p=0.04),which illustrate that the rate of complication of preventively usingsomatostatin in the intervention group is lower than control group. This articleincluded11RCTS, the6research provides two groups of patients with postoperativehospitalization days, using random effects model analysis (heterogeneity test p=0.006), there was no statistically significant difference in hospital number of daysbetween intervention group and control group after PD(OR1.12,95%CI4.54~2.31,p=0.52).3. Through retrospective analysis from January2001to January2011between theenforcement of PD,396cases of clinical data show that there are no statisticaldifference in operative death, operation time, intraoperative blood loss and theaverage length of hospital stay between these two groups（P＞0.05）. The totalincidence of PF is7.6%(30/396), there is no postoperative pancreatic fistula inimproved group, but there are30cases of postoperative pancreatic fistula in classicgroup (26cases of simple pancreatic fistula,4cases of intestinal anastomotic fistulaof pancreas), improved group had a significantly lower than the incidence ofpancreatic fistula of classic group (P <0.01). Classic group occurred in patients withpancreatic fistula by keeping the drainage unobstructed, use of somatostatin andparenteral nutrition, achieves the pancreatic fistula recovered after conservativetreatment.Conclusion1. While somatostatin for the secretion of pancreatic juice has obvious inhibitoryeffect, it does not effectively reduce the incidence of postoperative PF after PD.Prophylactic use of somatostatin, to some extent, can prevent or reduce theoccurrence of postoperative complications. But the side effects can’t reduce theperioperative mortality, and shorten the hospitalization days as a result.2. From the view of evidence-based medicine, through the Meta analysis of thisarticle11RCTs, the authors think that although somatostatin for the secretion of pancreatic juice has obvious inhibitory effect, it does not effectively reduce theoccurrence of postoperative pancreatic fistula. Prevention of complications after PD inaddition to the choice of the reasonable operation, careful operation, intravenousnutrition, proper drainage way, preventive use of somatostatin, can prevent or reducethe occurrence of postoperative complications to some extent. But the side effectscan’t reduce the perioperative mortality, and shorten the hospitalization days as aresult. In addition, this article included in the document failed to show the treatmentcost, so it can’t estimate economics evaluation.3. Improved end side set into the pancreatic anastomosis type method is simple andconvenient operation, stitching, pancreatic between sealing degree high, applicationscope widly, safe and effective, and it has great clinical application value. But if it canas conventional additional surgery pf PD remains to be seen in the multicenterprospective randomized controlled study confirmed. |
PDF Full Text Request