Lower Rectal Cancer

Gastrointestinal OncologyMultimediaComparison of Functional and Surgical Outcomes of Laparoscopic-Assisted Colonic J-Pouch Versus Straight Reconstruction After Total Mesorectal Excision for LowerRectal CancerJin-Tung Liang,MD,PhD,1Hong-Shiee Lai,MD,PhD,1Po-Huang Lee,MD,PhD,1andKuo-Chin Huang,MD,PhD 21Department of Surgery,National Taiwan University Hospital and College of Medicine,No.7,Chung-Shan South Road,Taipei,Taiwan,Republic of China2Department of Family Medicine,National Taiwan University Hospital and College of Medicine,No.7,Chung-Shan South Road,Taipei,Taiwan,Republic of ChinaBackground:To compare the functional and surgical outcomes of colonic J-pouch and straight anastomosis in the context that both reconstruction procedures were performed laparoscopically.Methods:The present study was a randomized prospective clinical trial.Patients with lower rectal cancer requiring laparoscopic total mesorectal excision were equally randomized to either laparoscopic-assisted colonic J-pouch reconstruction or laparoscopic straight end-to-end anastomosis.The techniques of the laparoscopic-assisted colonic J-pouch reconstruction are shown in the attached video.The primary end point was the comparison of functional results in both reconstruction methods.The secondary end points included the safety (surgical morbidity and mortality),surgical efficiency,and postoperative recovery.Results:A total of 48patients were recruited within 2-year periods,in consideration of sta-tistical power of 90%for comparison.There was no marked difference between patient groups undergoing colonic J-pouch surgery (n =24)and straight anastomosis (n =24)in various demographic and clinicopathogic parameters.The anorectal function of patients by colonic J-pouch were better than those by straight anastomosis in 3months after operation,as evaluated by stool frequency (mean ±standard deviation:4.0±2.0vs.7.0±2.4times/day,P <.001);use of antidiarrheal agents (29.2%[n =7]vs.75.0%[n =18],P =.004);and perineal irritation (45.8%[n =11]vs.79.2%[n =19],P =.037).Because of the relatively better bowel function in immediate postoperative period,patients by colonic J-pouch reconstruction were less disabled after surgery and had quicker return to partial activity (P =.039),full activity (P <.001),and work (P <.001).Both reconstruction methods were performed with similar amounts of blood loss,complication rates,and postoperative recovery.However,the operation time was signifi-cantly longer in the colonic J-pouch group (274.4±34.0vs.202.0±28.0minutes,P <.001).Received June 2,2006;accepted December 26,2006;published online:April 13,2007.Electronic supplementary material:The online version of this article (doi:10.1245/s10434-007-9355-2)contains supplementary material,which is available to authorized users.Address correspondence and reprint requests to:Jin-Tung Liang,MD,PhD;E-mail:jintung@.twPublished by Springer Science +Business Media,LLC Ó2007The Society of Surgical Oncology,Inc.Annals of Surgical Oncology 14(7):1972–1979DOI:10.1245/s10434-007-9355-21972Conclusions:Because laparoscopic-assisted creation of a colonic J-pouch achieved better short-term functional results of the anorectum and did not increase surgical morbidity,as compared with laparoscopic straight anastomosis,this reconstruction procedure could be recommended to patients with lower rectal cancer requiring laparoscopic total mesorectal excision.Key Words:Laparoscopic surgery—Colonic J-pouch—Rectal cancer—Total mesorectal exci-sion.Surgical treatment of rectal cancer seeks tofind a balance between radical excision of the tumor and preservation of defecation and genitourinary function. Total mesorectal excision,which involves the precise excision of the entire rectum and pararectal lymph nodes en bloc,can achieve adequate oncologic clear-ance with low local recurrence rate of3%–6%.1On the other hand,with the introduction of circular stapling devices,restoration of bowel continuity is possible most of the time,without compromise of oncologic clearance.However,traditional end-to-end(straight) anastomosis at the level of the anorectal junction re-sults in compromised bowel function.After a total mesorectal excision,the compliant rectum that has been removed is reconstructed by a less compliant segment of colon,which is physiologically less suit-able for storing and regulating feces.The clinical manifestations for such patients include excessive stool frequency,urgency,and varying degrees of fecal incontinence.To overcome such anorectal functional disorders,Lazorthes et al.2first advocated the colonic J-pouch procedures.Although research has shown that the colonic J-pouch reconstruction method was superior to straight anastomosis in postoperative function of anorectum,the conclusions were drawn on the basis of both reconstruction methods were per-formed by traditional open surgery.3–14Yet more and more procedures of total mesorectal excision for the radical extirpation of middle and lower rectal cancers are being performed by a laparoscopic approach.15,16 Therefore,we conducted this study to evaluate whe-ther the prevailing functional benefits of the colonic J-pouch procedure over traditional end-to-end straight reconstruction,as shown in traditional open surgery,can be reproduced when both reconstruction procedures were performed laparoscopically.MATERIALS AND METHODSPatient Selection and RandomizationWe conducted this prospective randomized trial to compare the functional and surgical outcomes of colonic J-pouch and straight end-to-end anastomosis for the reconstruction of patients with middle or low rectal cancer after total mesorectal excision.The study population comprised patients who were will-ing to undergo both reconstruction methods laparo-scopically.Patients were well informed regarding the details of both reconstruction methods,the potential advantages and disadvantages,and the possible rmed consent was obtained from all patients.This study was approved by the Institu-tional Ethics Committee of National Taiwan Uni-versity Hospital.The primary end point was the anorectal functional outcomes in both reconstruction methods.The secondary end points were safety, surgical efficiency,and complications.We hypothe-sized that the colonic J-pouch procedure was better in postoperative anorectal function than straight anas-tomosis,even when both methods were performed laparoscopically.Before entry onto the study,patients had to meet certain inclusion and exclusion criteria.The inclusion criteria were:(1)curative and elective surgery;(2) rectal adenocarcinoma below the peritoneal reflection and sphincter preservation was possible;and(3) American Society of Anesthesiology(ASA)function class I–III.The exclusion criteria were:(1)emergency or urgent surgery;(2)cancer located at upper rectum (above the peritoneal reflection),or very low-lying rectal cancer requiring an abdominal-perineal resec-tion;(3)evidence of invasion of adjacent organs or distant metastasis;(4)previous major abdominal or pelvic surgery;(5)anal incontinence before surgery;(6)body mass index(BMI)‡35kg/m2;and(7)pre-vious chemoradiotherapy.The estimation of sample size was based on our preliminary data that the3-month bowel frequency (mean±standard deviation)in J-pouch and straight methods was4.0±2.0and6.0±2.0respectively, when both procedures were performed by traditional open surgery.A sample size of23in each group will have90%power to detect a difference in means of2.0 (the difference between a group1mean of4.0and a group2mean of6.0),assuming that the commonLAPAROSCOPIC J-POUCH VERSUS STRAIGHT RECONSTRUCTION1973Ann.Surg.Oncol.Vol.14,No.7,2007standard deviation is2.0by a two-group t-test with a.050two-sided significance level.Patients were assigned to either the colonic J-pouch or the straight anastomosis group by means of sealed opaque enve-lopes containing computer-generated random num-bers.To prevent selection bias,random numbers were generated by an investigator who was not in-volved in enrollment of patients.Patients were ran-domized in the operating room right before surgery. Operative TechniquesThe surgical techniques of laparoscopic-assisted end-to-end straight anastomosis have been described in our previous multimedia article.16The details of the laparoscopic-assisted colonic J-pouch recon-struction are shown in the video attached to the present article.Briefly,complete mobilization of left-sided colon is performed to the level of the middle colic vessels by laparoscopic medial-to-lateral dis-section technique,as we showed in our previous article.17After ligation of the inferior mesenteric vessel proximal to the left colic vessels,the dissection was shifted to the mobilization of rectum.After total mesorectal excision,the rectum was clamped and transected at least2cm distal to the distal edge of the tumor,the bowel was then exteriorized,and the proximal colon was divided with a linear cutting stapler.The level of this transection must allow a well-vascularized segment of descending(preferably) or sigmoid(rarely)colon for anastomosis.18An8-cm colonic J-pouch was made by folding the colon and creating a side-to-side anastomosis with two linear cutting staplers(45· 3.5mm,Tyco)introduced through the apex of the st,a stapled colo-anal anastomosis was performed.Functional AssessmentThe anorectal function was evaluated by question-naire-based interview of patients preoperatively and then3and6months after surgery.The preoperative functional questionnaire was based on the recollection of function before the development of symptoms from the rectal cancer.Those with a defunctioning colos-tomy were evaluated3and6months after stoma reversal.Functional questionnaires regarding fecal continence and bowel function were completed by an assistant who was not aware of the randomization status of the patient.Continence was recorded as grade 1(perfect continence),grade2(incontinence offlatus), grade3(occasional minor soiling),grade4(frequent major soiling),and grade5(total incontinence).19Urgency was recorded in patients who did not have the ability to defer defecation for more than15minutes. Furthermore,episode and frequency of fecal leakage and the need of pad were recorded.Questions con-cerning bowel function included daily stool frequency, use of antidiarrheal agent or laxatives,presence of incomplete defecation and/or fragmentation,and need of enema or digital evacuation of stool.Fragmentation of stools was defined as the inability to defecate and empty the reservoir in one attempt.Fragmented stools were counted as multiple bowel movements. Anorectal Manometry and Volumetric StudyThe data of anorectal manometry and volumetric study were validated by the comparison with those of a control group,which were available in our colo-rectal physiology laboratory.The assessment was performed twice in3and6months,respectively,after one-stage tumor excision orfinal colostomy reversal surgery.Anal resting and squeeze pressures were measured by an open,water-perfused catheter con-nected by a transducer to a recorder(Albyn Medical, Griffon,United Kingdom).Pressure profile was measured by manual pull-through of the catheter at 1-cm intervals.A catheter with a latex balloon was inserted through the anal canal and above the anas-tomosis,and the balloon wasfilled with water in increments of20mL.Threshold volume was defined as the infused volume that made patientsfirst aware of the presence of the balloon.Maximum tolerable volume was defined as the volume at which the pa-tient could not tolerate further infusion.20Rectal compliance(D V/D P)was measured by changes in volume(mL)per unit of pressure(cm H2O). Evaluation of Surgical OutcomesWe assessed the surgical efficiency by using parameters that included length of operation time (counted from the beginning of skin incision to the final skin closure),blood loss(measured by the amount of blood in suction bottle and the number of blood-soaked gauzes),conversion rate,intraoperative and postoperative complications,and wound size. The wound size was measured by summation of the length of one major wound(generally5cm in size) for tumor retrieval and four working ports(5–12mm in diameter).The operative complications,if present, were individually listed.The functional recovery was compared by length of postoperative restoration offlatus passage and hospitalization,degree of postoperative pain,andJ.-T.LIANG ET AL. 1974Ann.Surg.Oncol.Vol.14,No.7,2007disability.The visual analog scale was used in assessing postoperative pain on thefirst postoperative day.A standardized questionnaire was given to pa-tients to assess disability,which included the number of days until return to partial activity,full activity, and work;responses were subjective.21All parame-ters of functional recovery were evaluated by research assistants who were blinded to the study groups. StatisticsData were analyzed on the basis of the intention-to-treat principle.The two-tailed Fisher exact test or v2test was used to analyze the categorical data.The continuous data were compared by the Student t-test. The significance level of all tests was set at P<.05.RESULTSBetween May2004and April2006,a total of48 patients were recruited and equally allocated to the two study groups.All patients were followed up until October2006(median,18months;range,6–30 months).The colonic J-pouch and straight anasto-mosis group of patients were well matched(P>.05) for age,sex,BMI,ASA functional class,tumor dis-tance above anal verge,distal resection margin,cir-cumferential resection margin,number of collected lymph nodes,and tumor,node,metastasis system stage(Table1).Table2demonstrated manometric and volumetric data of the patients.In the early postoperative period (3months),there was no marked difference in recto-anal inhibitory reflex and pressure profile among the two study groups and the control group.This implied that the sphincteric function remained intact after the two surgical procedures.On the other hand,although the threshold volume,maximal tolerable volume,and rectal compliance were considerably reduced after both colonic J-pouch reconstruction and straight anastomosis,they were significantly lower(P<.001) in patients with straight anastomosis than in those with J-pouch.The manometric and volumetric data of both group improved very little in the repeated eval-uation performed6months after surgery. Functional results are listed in Table3.There was no significant difference(P>.05)between colonic J-pouch and straight anastomosis groups of patients in continence grading and feces/flatus differentiation. Likewise,there was no statistically significant differ-ence between the groups in urgency.With regard to bowel function,we found that a considerable per-centage of patients in both groups experienced in-creased stool frequency,incomplete defecation,and fragmentation postoperatively,but only the degree of the bowel frequency was significantly lower(P< .001)in patients with colonic J-pouch reconstruction than those with straight anastomosis.The derange-ment of bowel movement frequently resulted in per-ineal irritation and the necessity of antidiarrheal medication,as demonstrated by the degree of peri-neal irritation(P=.037)and antidiarrheal medica-TABLE1.Demographics and clinicopathologic data between groups aParameter J-pouch(n=24)Straight anastomosis(n=24)P value Age(y)(mean±SD)64.4±10.462.4±9.8.496 Sex(n).771 Male1314Female1110BMI(weight/height2,kg/m2)(mean±SD)24.8±2.825.2±3.0.635 ASA functional class(n).840I1210II1113III11Tumor location(cm above anal verge)(mean±SD) 6.4±1.2 6.8±1.6.332 Distal resection margin(cm)(mean±SD) 2.8±.6 2.6±.4.181 Circumferential resection margin(mm)(mean±SD)8.4±2.48.2±2.6.783 No.Collected lymph nodes(mean±SD)17.0±5.416.0±4.0.470 TNM stage(pathologic)(n).992T3N0M01514T3N1M078T3N2M011T3N3M011BMI,body mass index;ASA,American Society of Anesthesiology;TNM,tumor,node,metastasis system.a Continuous data are presented as mean±SD.LAPAROSCOPIC J-POUCH VERSUS STRAIGHT RECONSTRUCTION1975Ann.Surg.Oncol.Vol.14,No.7,2007tion (P =.004),which were significantly lower in J-pouch group than in straight anastomosis group.Although the rate of incomplete defecation and fragmentation stool was lower in J-pouch than in straight anastomosis group,it did not reach statistical significance (P >.05).Additionally,it was noted that the construction of a relatively large pouch (up to 8cm in size)did not increase the incidence of incom-plete stool evacuation,as the use of laxative and digital evacuation of stool were very infrequently seen in both groups of patients.Again,we found that the bowel derangement seen at 3-month time point improved very little at a 6-month reevaluation.The comparison of various parameters of surgical outcomes between J-pouch and straight anastomosis groups of patients is shown in Table 4.Construction of the colonic J-pouch tended to take more time than the straight anastomosis (P <.001).There was no significant difference (P >.05)between the two groups in blood loss,wound size,restoration offlatus passage,hospitalization,postoperative pain,and postoperative complications.There was no anastomotic leakage in both groups.However,4patients (16.7%)in the J-pouch group and 3patients (12.5%)in the straight anastomosis group were pro-tected by a diverting ileostomy.Because of the better postoperative bowel function,the patients in the J-pouch group had less disability,as shown by the quicker return to partial activity (P =.039),full activity (P <.001),and work (P <.001).Addi-tionally,it was found that colonic J-pouch facilitated the use of a bigger size of CEEA stapler (Premium Plus CEEA,Autosuture,Tyco).DISCUSSIONThe present randomized prospective controlled clinical trial demonstrated that colonic J-pouch reconstruction provided better short-term functionalTABLE 2.Anorectal manometry and volumetric test between groups aParameterControl (n =24)J-pouch (n =24)Straight anastomosis(n =24)P value 3mo 6mo 3mo 6mo Resting pressure (mm Hg)75±2074±1976±2172±2475±28.750Maximal squeeze pressure (mm Hg)160±40144±34148±37134±40142±46.356Rectoanal inhibitory reflex present (%)100100100100100–Minimal volume (mL)14±815±714±613±815±6.362Threshold volume (mL)75±2470±1872±1649±644±10<.001Maximal tolerable volume (mL)180±90148±34152±3078±2074±22<.001Rectal compliance (mL/cm H 2O)12.4±4.59.8±2.812.0±4.45.4±1.45.8±1.6<.001aData are presented as mean ±SD unless otherwise indicated.TABLE 3.Functional assessment between groupsParameterJ-pouch (n =24)Straight anastomosis (n =24)P value a Before surgery At 3mo At 6mo Before surgery At 3mo At 6mo Continence,n (%).942Normal24(0)20(83.3)21(87.5)24(0)18(75)20(83.3)Incontinent to gas 0(0)2(8.3)1(4.2)0(0)2(8.3)1(4.2)Occasional minor leak 0(0)2(8.3)2(8.3)0(0)4(16.7)3(12.5)Frequent major soiling 0(0)0(0)0(0)0(0)0(0)0(0)Total incontinence0(0)0(0)0(0)0(0)0(0)0(0)Absence of feces/flatus differentiation ,n (%)0(0)0(0)0(0)0(0)1(4.2)1(4.2)–Urgency,n (%)0(0)6(25)5(2.8)0(0)5(20.8)4(16.7).142Stool frequency (mean ±SD) 1.5±.4 4.0±2.0 4.0±2.9 1.4±0.67.0±2.4 6.5±1.9<.001Perineal irritation,n (%)0(0)11(45.8)9(37.5)0(0)19(79.2)14(58.3).037Incomplete defecation,n (%)1(4.2)14(58.3)14(58.3)1(4.2)20(83.3)18(75).112Fragmentation,n (%)0(0)13(54.2)13(54.2)0(0)20(83.3)18(75).062Antidiarrheal medication,n (%)1(4.2)7(29.2)7(29.2)1(4.2)18(75)15(62.5).004Use of laxative,n (%)2(8.3)1(4.2)1(4.2)1(4.2)1(4.2)1(4.2).471Digital evacuation,n (%)0(0)0(0)0(0)0(0)0(0)0(0)–aP value was calculated on the basis of the comparison of anorectal function between groups of J-pouch and straight anastomosis in the 3months after the operation.J.-T.LIANG ET AL.1976Ann.Surg.Oncol.Vol.14,No.7,2007results than the conventional straight anastomosis without adding surgical complications;both proce-dures were performed laparoscopically.It has been generally accepted that a surgical trial is quite dif-ferent from a medical trial in essence.The techniques and pitfalls for the conduction of a randomized prospective clinical trial regarding the laparoscopic resection of colorectal cancer have been addressed in our previous articles.22Compared with the previous randomized trials in the literature,21,22the present study has the following strengths.First,because both J-pouch and straight reconstruction methods were widely used by the colorectal surgeons in traditional open surgery and the patients usually did not have any idea regarding the details of surgical procedures (i.e.,the adoption of a reconstruction method was at the discretion of surgeons and included in the in-formed consent of our daily routine practice for lower rectal cancer),the ethical issues commonly involved in the randomized clinical trials could be circum-vented,and therefore,it was relatively easy to recruit enough patients for study.Second,the evaluation of functional outcomes,including clinical symptoms and the manometric and volumetric data of patients, would be vulnerable to a subjective placebo effect if the clinical trial lacked blinding between patients and assessors.In the present study,the abdominal wounds were the same between the two study groups, and therefore,the functional parameters were evalu-ated in a double-blind manner,because the patients and assessors were both ignorant of the surgical procedure that had been performed.In this study,we showed that a colonic J-pouch procedure is more time-consuming than straight anastomosis.This could be explained by the simple fact that shaping an additional colonic J-pouch is more complex than doing traditional straight end-to-end anastomosis.However,we thought that colonic J-pouch reconstruction was also technically more difficult than the end-to-end anastomosis,especially when both procedures were performed laparoscopi-cally.First,the colonic J-pouch reconstruction needs more extensive dissection to completely take down the colonic splenicflexure,thus facilitating the con-struction of a J-pouch at descending(preferably)or sigmoid colon and ensuring the followed tension-free J-pouch–anal anastomosis.18Second,in comparison with straight anastomosis,the colonic J-pouch was relatively bulky;this might hinder a precise J-pouch–anal stapling anastomosis,particularly in the male narrow pelvis or when the patient was fatty.Third, during the fashioning of a colonic J-pouch,the pa-tient underwent an additional staple to open the septum of the folded colon.This additional8-cm stapling line might theoretically increase the rate of anastomotic leakage.There was no anastomotic leak in the present study population.This may be because of our efforts to fully mobilize the left-sided colon, thus ensuring a tension-free pouch-anal anastomosis, and because,most importantly,the J-pouch–anal anastomosis was protected by a defunctioning ileos-tomy in a subset of patients(14.6%,n=7)who had comorbid factors,poor nutritional status,edematous bowel,and/or technical insecurity of stapling.23The better anorectal function of a colonic J-pouch was generally considered to be due to the relatively large capacity and reversed peristalsis in the neorectal reservoir.3–14The volumetric data in the presentTABLE4.Parameters related to the surgical outcomes between groupsParameter J-pouch(n=24)Straight anastomosis(n=24)P value Operating time(min)(mean±SD)274.4±34.0202.8±28.0<.001 Blood loss(mL)(mean±SD)94.0±24.088.0±30.0.448 Flatus passage(h)(mean±SD)50.0±10.548±12.0.542 Hospitalization(d)(mean±SD)8.5±1.59.0±2.0.332 Postoperative pain(visual analog scale)(mean±SD) 4.2±1.0 4.8±1.2.066 Wound size(cm)(mean±SD)10.8±1.69.8±1.0.144 Postoperative complications,n(%).150 Total4(16.7%)3(12.5%)Wound infection21Urinary tract infection10Urinary retention11Anastomotic leakage00Deep vein thrombosis01Protective ileostomy,n(%)4(16.7%)3(12.5%).150 CEEA stapler(31/28/25mm)24/0/08/10/6<.001 Disability,mean±SD(%patients responding to questionnaires)Return to partial activity(wk) 2.4±.6(100%) 2.8±.7(100%).039 Return to full activity(wk) 4.0±1.0(100%) 6.0±.8(100%)<.001 Return to work(wk) 4.6±1.2(62.5%)7.4±.9(66.7%)<.001 LAPAROSCOPIC J-POUCH VERSUS STRAIGHT RECONSTRUCTION1977Ann.Surg.Oncol.Vol.14,No.7,2007study supported the concept of larger capacity and compliance in J-pouch neorectum.Some researchers stress the importance of the antiperistaltic character of a colonic J-pouch and advocate that a small J-pouch(5cm in size)can achieve similar bowel function and simultaneously prevent the potential incomplete evacuation of stool from a large J-pouch (8cm).24–28In the present study,we did notfind any increased rate of incomplete evacuation of stool in the J-pouch group of patients.Remarkably,the comparison of functional outcomes between the small and large colonic J-pouches was inconsistent in the literature.3–14,24–28Furthermore,the present study showed that the bowel continuity of patients in colonic J-pouch group tends to be reconstructed by the bigger size of CEEA stapler,as compared with that of patients in the straight anastomosis group. Although there was no anastomotic stricture in the present colonic J-pouch and straight anastomosis groups of patients,it was conceivable that the bigger coloanal anastomotic ring would have a positive effect on the postoperative anorectal function.To further explore the underlying mechanisms for the better anorectal function of the patients in the colonic J-pouch group,continuous long-term follow-up of the present study population will be necessary.ACKNOWLEDGMENTSupported by a grant from National Taiwan University Hospital(96-S557).REFERENCES1.Heald RJ,Moran BJ,Ryall RD,Sexton R,MacFarlane JK.Rectal cancer:the Basingstoke experience of total mesorectal excision,1978–1997.Arch Surg1998;133:894–9.zorthes F,Fages P,Chiotasso P,Lemozy J,Bloom E.Resection of the rectum with construction of a colonic reser-voir and colo-anal anastomosis for carcinoma of the rectum.Br J Surg1986;73:136–8.3.Jiang JK,Yang SH,Lin JK.Transabdominal anastomosisafter low anterior resection:A prospective,randomized,con-trolled trial comparing long-term results between side-to-end anastomosis and colonic J-pouch.Dis Colon Rectum2005;48:2100–8.4.Heriot AG,Tekkis PP,Constantinides V,et al.Meta-analysisof colonic reservoirs versus straight coloanal anastomosis after anterior resection.Br J Surg2006;93:19–32.5.Machado M,Nygren J,Goldman S,Ljungqvist O.Functionaland physiologic assessment of the colonic reservoir or side-to-end anastomosis after low anterior resection for rectal cancer:a two-year follow-up.Dis Colon Rectum2005;48:29–36.6.Machado M,Nygren J,Goldman S,Ljungqvist O.Similaroutcome after colonic pouch and side-to-end anastomosis in low anterior resection for rectal cancer:a prospective ran-domized trial.Ann Surg2003;238:214–20.7.Sailer M,Fuchs KH,Fein M,Thiede A.Randomized clinicaltrial comparing quality of life after straight and pouch coloanal reconstruction.Br J Surg2002;89:1108–17.8.Oya M,Komatsu J,Takase Y,Nakamura T,Ishikawa H.Comparison of defecatory function after colonic J-pouch anastomosis and straight anastomosis for stapled low anterior resection:results of a prospective randomized trial.Surg Today 2002;32:104–10.9.Mantyh CR,Hull TL,Fazio VW.Coloplasty in low colorectalanastomosis:manometric and functional comparison with straight and colonic J-pouch anastomosis.Dis Colon Rectum 2001;44:37–42.10.Ho YH,Seow-Choen F,Tan M.Colonic J-pouch function atsix months versus straight coloanal anastomosis at two years: randomized controlled trial.World J Surg2001;25:876–81. 11.Ho YH,Tan M,Leong AF,Seow-Choen F.Ambulatorymanometry in patients with colonic J-pouch and straight col-oanal anastomoses:randomized,controlled trial.Dis Colon Rectum2000;43:793–9.12.Barrier A,Martel P,Gallot D,Dugue L,Sezeur A,MalafosseM.Long-term functional results of colonic J pouch versus straight coloanal anastomosis.Br J Surg1999;86:1176–9. 13.Dehni N,Tiret E,Singland JD,et al.Long-term functionaloutcome after low anterior resection:comparison of low colorectal anastomosis and colonic J-pouch–anal anastomosis.Dis Colon Rectum1998;41:817–22.14.Joo JS,Latulippe JF,Alabaz O,Weiss EG,Nogueras JJ,WexnerSD.Long-term functional evaluation of straight coloanal anastomosis and colonic J-pouch:is the functional superiority of colonic J-pouch sustained?Dis Colon Rectum1998;41:740–6.15.Chung CC,Ha JP,Tsang WW,Li paroscopic-assistedtotal mesorectal excision and colonic J pouch reconstruction in the treatment of rectal cancer.Surg Endosc2001;15:1098–101.16.Liang JT,Lai HS,Lee paroscopic total mesorectalexcision for rectal cancers.Dis Colon Rectum2006;49:517–8.17.Liang JT,Lai HS,Lee paroscopic medial-to-lateralapproach for the curative left hemicolectomy.Dis Colon Rec-tum2005;49:2142–3.18.Wexner SD,Alabaz O.Anastomotic integrity and function:role of the colonic J-pouch.Semin Surg Oncol1998;15:91–100.19.Kirwan WO,Turnbull RB Jr,Fazio VW,Weakley FL.Pull-through operation with delayed anastomosis for rectal cancer.Br J Surg1978;65:695–8.20.Wang JY,You YT,Chen HH,Chiang JM,Yeh CY,Tang R.Stapled colonic J-pouch–anal anastomosis without a diverting colostomy for rectal carcinoma.Dis Colon Rectum1997;40:30–4.21.Liang JT,Lai HS,Huang KC,et parison of medial-to-lateral versus traditional lateral-to-medial laparoscopic dissection sequences for resection of rectosigmoid cancers: randomized controlled clinical trial.World J Surg2003;27:190–6.22.Liang JT,Huang KC,Lai HS,Lee PH,Jeng YM.Oncologicresults of laparoscopic versus conventional open surgery for stage II or III left-sided colon cancers:a randomized controlled trial.Ann Surg Oncol2007;14:109–17.23.Dehni N,Schlegel RD,Cunningham C,Guiguet M,Tiret E,Parc R.Influence of a defunctioning stoma on leakage rates after low colorectal anastomosis and colonic J pouch–anal anastomosis.Br J Surg1998;85:1114–7.24.Amin AI,Hallbook O,Lee AJ,Sexton R,Moran BJ,HealdRJ.A5-cm colonic J pouch colo-anal reconstruction following anterior resection for low rectal cancer results in acceptable evacuation and continence in the long term.Colorectal Dis 2003;5:33–7.25.Furst A,Burghofer K,Hutzel L,Jauch KW.Neorectal reser-voir is not the functional principle of the colonic J-pouch: the volume of a short colonic J-pouch does not differ froma straight coloanal anastomosis.Dis Colon Rectum2002;45:660–7.J.-T.LIANG ET AL. 1978Ann.Surg.Oncol.Vol.14,No.7,2007。