World J Surg (2011) 35:1925–1932 DOI 10.1007/s00268-011-1135-2

Compression Anastomosis Revisited: Prospective Audit of Short- and Medium-term Outcomes in 62 Rectal Anastomoses Bernhard Dauser • Thomas Winkler Gerhard Loncsar • Friedrich Herbst

•

Published online: 4 May 2011 Ó Socie´te´ Internationale de Chirurgie 2011

Abstract Background The ColonRingTM represents a further advancement in compression anastomosis by using superelastic materials. In contrast to other compression devices, its field of application is identical to circular staplers. Yet, limited data are available and there are no reports on the use of this compression device for anastomoses in the lower rectum and in diverted cases. Methods Between November 2008 and March 2010 a nonrandomized prospective study of patients undergoing colorectal resection was done. Results Sixty-two patients (30 female) with a median age of 67 years (range = 23-87) underwent surgery for benign (35) and malignant (27) indications;the surgery was performed laparoscopically in 53 (85.5%) cases. There were no leaks or other major complications related to the compression anastomosis. One patient died from sepsis caused by a leak of a colonic J-pouch. Three of 12 (25%) patients after (ultra-)low anterior resection experienced pain caused by the presence of the ring. Of 53 undiverted patients, rings were passed spontaneously in 52 (98.1%). No spontaneous passage occurred in patients with a stoma. Their rings were removed transanally prior to stoma closure. After a median of 6 months, endoscopic follow-up was performed in 60 patients (96.8%). Two anastomotic stenoses were detected (3.3%).

The study was registered at ClinicalTrials.gov under NCT01056913. B. Dauser
T. Winkler
G. Loncsar
F. Herbst (&) Department of Surgery, St. John of God Hospital, Vienna, Johannes von Gott Platz 1, 1020 Vienna, Austria e-mail: [email protected] B. Dauser e-mail: [email protected]

Conclusion Construction of rectal anastomosis using this novel compression device is feasible and safe and appears to cause only few anastomotic strictures in the medium term. Discomfort or pain from the presence of the ring at anastomosis following (ultra-)low anterior resection can be observed in a relevant number of patients.

Introduction Morbidity and mortality in colorectal surgery is associated mainly with anastomotic complications. Leakage rate after anterior resection using stapling devices is still inadmissibly high,with clinical sequelae ranging from local infection to sepsis and death [1–9]. Compression anastomoses have demonstrated to be at least comparable to conventional stapling devices in terms of anastomotic leakage and stenosis. In addition, bleeding from the anastomosis seems to be no issue at all [10, 11]. Despite that, various compression anastomotic devices such as the AKA-2 or the Biofragmentable Anastomosis Ring (BAR) did not gain broader acceptance among colorectal surgeons [12]. A possible explanation could be that compared to staplers these older devices differed considerably with respect to handling, were impractical to use in laparoscopic surgery, and required purse-string sutures for both bowel ends [13]. The idea of connecting healthy intestine without the use of foreign materials such as staples, which may crush into the tissue and cause local inflammation, is comprehensible. The ColonRingTM (NiTi Surgical Solutions Ltd., Netanya, Israel) represents a further development in compression anastomosis, and it is handledin a manner quite similar to standard circular stapling devices. It may also be used on rectal stump after a stapled resection, making it suitable for laparoscopic procedures and distal rectal anastomosis.

123

1926

In this device, superelastic nitinol leaf springs allow a wellbalanced necrosis-healing process at the anastomotic site [14, 15]. We hereby present short- and medium-term results from using this compression device for construction of rectal anastomoses.

Patients and methods Patients With institutional review board approval, a prospective audit was performed on all ColonRing procedures between November 2008 and March 2010. This period included our learning curve with the device. Informed consent was obtained from all patients. The audit was terminated when the first patient was entered into the COMPRES study (COMpression Anastomosis Ring [CARTM 27/ColonRing] Post maRketing Evaluation Study; http://clinicaltrials.gov/ ct2/show/NCT01091155). There were 62 consecutive patients who underwent surgery during this period. The operation was planned for 61 patients, and one patient was operated on within 2 hafter perforation at the rectosigmoid junction occurred during colonoscopy.There was only minimal pelvic inflammation. As recommended by the manufacturer, patients requiring emergency surgery for acute inflammation, colonic perforation, or bowel obstruction or those having a known allergy against nickel were excluded. After successful enrolment of the initial three cases, all consecutive colorectal resections were done using the compression device and conventional circular staplers were used only in patients with one of the aforementioned exclusion criteria or who refused to participate in this audit. Altogether, there were 14 additional restorative rectal resections with stapled (n = 2) or hand-sewn transanal anastomosis (n = 12). All operations were performed or directly supervised by the senior author. Perioperative management Apart from construction of the anastomosis, no change in pre-, intra-, and postoperative management was made. On the day of surgery a saline enema (200 ml) was administered in patients who did not require a diverting stoma. Mechanical bowel preparation was done only in patients requiring stoma (n = 8) or in those undergoing colonoscopy the day before surgery (n = 8). A fast-track regimen was employed but epidural catheters were offered only in open cases. The inferior mesenteric artery (IMA) was preserved in case of benign indication or ligated at its origin when surgery was for malignant disease. Takedown of the splenic flexure was routinely performed. In laparoscopic cases a small Pfannenstielincision was used for specimen retrieval.

123

World J Surg (2011) 35:1925–1932

Description and use of the device The device itself resembles a current circular stapling instrument. However, in contrast to staplers, the base part of the compression ring has to be mounted to the tip of the instrument before use. The anvil is different in that it contains a hard plastic disc, part of which is integrated into the compression ring upon construction of the anastomosis. Preparation for anastomosis includes closure of the rectal stump by linear staplers (alternatively, a purse-string suture may be used) and insertion of the anvil into the proximal bowel. With the ring mounted, the device is inserted transanally and advanced to the end of the rectal stump. A central spike is advanced by turning a knob at the base of the instrument to perforate the rectal wall close to the transverse staple line. The anvil then is attached to the spike and the instrument is closed. A safety lever makes firing of the device impossible before complete closure. When the instrument is closed fully, barbed wire spikes attach the base part of the ring to the hard plastic of the anvil. Upon firing the instrument, a circular knife cuts out the central part of the anvil and the completed compression ring is moved from the instrument. Therefore, the device is distal to the anastomosis in its entirety as soon as the anastomosis is constructed and may be removed immediately. Anastomotic rings are checked for integrity and a leak test is performed.

Postoperative follow-up Postoperatively, stool softeners were given on demand. The first day that the patient tolerates solids, has a bowel movement, and has any complication or adverse event was recorded. As ring expulsion usually happened after discharge by subjects without stoma, patients were asked to report by telephone when passage of the ring was recognized. Before closure of stoma, a water-soluble contrast enema and flexible endoscopy were performed to check anastomotic integrity. Rings were removed using a Roth NetÒ (US Endoscopy, Mentor, OH, USA) (Fig. 1). During follow-up, all patients were contacted to record any (late) procedure-related complications or the presence of symptoms attributable to the anastomosis within 6 months. In addition, patients were invited for endoscopic evaluation of the anastomosis. Endoscopy was performed using a 12.3-mm flexible video endoscope (CF-H180AI, Olympus, Hamburg, Germany) after a median of 6.5 months (range = 0.9-17.7). Anastomotic stenosis was defined as the impossibility of passing the anastomosis with the endoscope, as suggested by most authors [16–19]. This study was registered at ClinicalTrials.gov under NCT01056913.

World J Surg (2011) 35:1925–1932

1927

Fig. 1 a Water-soluble contrast enema. b Endoscopic view of the compression anastomosis ring located at the efferent limb of a side-to-end jejunorectal anastomosis in a diverted patient 6 weeks postoperatively. c Ring removal (diameter = 27 mm) through anastomosis performed using the Roth NetÒ

Results Sixty-two (30 female) patients who were to undergo colorectal surgery for benign (n = 35) and malignant (n = 27) indications were enrolled. Median age was 67 years (range = 23-87) and median body mass index (BMI) was 24.8 kg/m2 (range = 16.8-35.3). ASA score was I, II, and III in 40 (64.5%), 19 (30.7%), and 3 (4.8%) patients, respectively. A laparoscopic approach was chosen in 53 (85.5%) cases and an open in 9 cases. Anastomosis was located in the upper rectum (distance from the anal verge [ 12 cm), in the midrectum ([6 cm), and the lower rectum (0-6 cm) in 33, 17, and 12 cases, respectively (Tables 1 and 2). The median operation time was 210 min (range = 117-355). Four of 12 patients who underwent low anterior resection for rectal cancer underwent preoperative chemoradiation and all were diverted during the initial procedure. Altogether, eight loop ileostomies were raised—six times after low anterior resection, twice after difficult Hartmann reconstruction, with the anastomosis located in the midrectum. One female patient underwent correction of a previous accidentally constructed colovaginal anastomosis and was already diverted. Therefore, total diversion rate was 14.5% (9/62). Median time to tolerating solids was 3 days (range = 1-14). Stool softeners were given to 29 of 53 (54.7%) undiverted patients if bowel movement was not present on the first day of tolerating solids. First bowel movement was on POD 3

(median; range = 1-6). Median length of hospital stay was 8 days (range = 4-183). Complications Intraoperative No device-related problems could be observed. Doughnuts were complete in all cases. A positive air leak test was recognized once (1/62, 1.6%) during laparoscopic sigmoidectomy. Additional sutures were used at the anterior aspect of the anastomosis and postoperative course was uneventful. Postoperative No early or late leaks or bleeding from the compression anastomosis was observed. Relaparotomy because of bleeding from the pelvic sidewall was required on the day of operation in one patient after low anterior resection. Postoperative ileus necessitating insertion of a nasogastric tube was observed in three patients (4.8%). There were no wound infections. One patient was readmitted on POD 18 with signs of pelvic infection. CT scan with rectal contrast showed free fluid in the pelvic cavity without circumscribed abscess formation and a healed anastomosis. Conservative treatment using intravenous antibiotics was successful. In one patient, leakage from the efferent limb of a colonic J-pouch was seen after closure of the diverting stoma. Preoperative radiologic investigation had not shown pathological findings but diffuse fecal peritonitis developed on POD 3. During relaparotomy staple line failure was detected at the pouch while showing

123

1928

World J Surg (2011) 35:1925–1932

Table 1 Indications and operations in 62 consecutive compression device procedures Indication

N (%)

Sigmoidectomy Left hemicolectomy

Diverticular disease

21 (33.9) 9

Rectal cancer

18 (29)

11

Anterior resection

Low anterior resection

Intestinal reconstruction

1

–

–

–

–

6

12

–

Cancer of colon

8 (12.9) 5

2

1

–

–

Hartmann’s reconstructiona

4 (6.5)

–

–

–

–

4

Colon stenosisb

4 (6.5)

–

1

3

–

–

Adenoma

3 (4.8)

2

–

1

–

–

Iatrogenic perforation

1 (1.6)

1

–

–

–

–

Slow transit constipation

1 (1.6)

–

1

–

–

–

Rectal prolapse

1 (1.6)

1

–

–

–

–

Colovaginal fistula

1 (1.6)

–

–

–

–

1

a

Crohn’s disease (n = 2), anastomotic leakage (n = 2)

b

Crohn’s disease (n = 1), endometriosis (n = 1), diverticular disease (n = 2)

Table 2 Technical data: location of rectal anastomoses for benign and malignant indication, surgical approach, type of reconstruction, and fecal diversion Rectal third N (%)

Upper

Mid

Benign

35 (56.5)

27

8

–

Malignant

27 (43.5)

6

9

12

Lower

53 (85.5)

29

13

11

9 (14.5)

4

4

1

50 (80.7)

Indication

Access Laparoscopic Open Type of anastomosis End-to-end

33

14

3

Side-to-end

5 (8)

–

2

3

Pouch

7 (11.3)

–

1

6

9 (14.5)

–

3

6

Stoma

Distance of anastomosis from the anal verge: upper rectal third [ 12 cm; mid rectal third [ 6 cm; lower rectal third 0-6 cm

a healed pouch–rectal anastomosis. The patient died 2 months postoperatively due to sepsis despite multiple revisions. Fate of the ring: patients without stoma The ring was passed spontaneously by 52 (98.1%) of the 53 patients without stoma. The remaining patient underwent long-term ventilation because of respiratory failure and had liquid stools due to artificial enteral nutrition. The retained compression ring (mobile in the nonaltered rectum) was removed digitally on POD 141 in the ICU. In patients with spontaneous ring evacuation, 29% (15/52) sensed the

123

passage, whereas 71% (37/52) did not (Fig. 2). Among these 37 patients, 12 found the ring in the toilet or heard a metallic sound while defecating. In 27 (50.9%)of 53 undiverted patients, rings were passed after a median of 10 days postoperatively (range = 7-16). Fate of the ring: patients with diverting stoma No spontaneous ring passage occurred in the nine patients who had a diverting stoma. The ring was removed either endoscopically (n = 7) or digitally (n = 2) after 42 days (median; range = 28–99). Eight stomas were closed at a median of 55 days (range = 15–101) postoperatively. One patient is still awaiting closure of stoma while showing a well-healed anastomosis at the end of the study period. Three of 12 (25%) patients who had an ultralow anastomosis reported pain while sitting (n = 2) or taking a walk (n = 1) while the ring was in place. Symptoms subsided in all patients after ring removal at 6 weeks postoperatively before closure of stoma.

Follow-up and late complications Two anastomoses could not be passed with the endoscope (3.3%). They were located at 10 and 14 cm from the anal verge and had been constructed after resection for cancer and diverticular disease, respectively. Lumen diameter measured with an opened biopsy forceps was 9 mm in both patients. Anastomoses responded easily to endoscopic balloon dilatation with resolution of symptoms. Early and late complications according to the Clavien-Dindo Classification [20] and corresponding interventions are presented in Table 3.

World J Surg (2011) 35:1925–1932

1929

Fig. 2 Fate of the ring

Table 3 Early and late complications and corresponding interventions Complication

N (%)

Clavien-Dindo Relation Female Male Benign Malignant Upper Mid Lower Intervention classification to device rectal rectal rectal third third third

Early postoperative Bleeding (pelvic side wall)

1 (1.6) III

No

–

1

–

1

–

–

1

Operative revision

Postoperative ileus

3 (4.8) II

No

1

2

–

3

–

1

2

Nasogastric tube insertion

Symptoms related to ring

3 (4.8) I

Yes

1

2

–

3

–

–

3

Digital ring mobilization

Late Pelvic infection

1 (1.6) II

No

1

–

1

–

1

–

–

IV antibiotics

J-pouch leakage (efferent limb)

1 (1.6) IVb,V

No

–

1

–

1

–

–

1

Operative revision

2a (3.3) IIIa

Yes

–

2

1

1

1

1

–

1 (1.6) IVa

No

1

–

1

–

1

–

–

Endoscopic balloon dilatation Long-term ventilation

Anastomotic stenosis Respiratory failure a

Of 60 patients available for endoscopic follow-up

Discussion The double-stapling technique for restoration of intestinal continuity after anterior resection has greatly facilitated the construction of low rectal and also laparoscopically performed anastomoses [21–23]. Several studies were performed during the last two decades comparing stapled versus hand-sewn methods for colorectal anastomosis showing that both methods have similar rates of anastomotic leakage, but there is significantly more anastomotic stenosis in the stapled group [24]. There is no uniformly accepted definition and grading of anastomotic leakage. This leads to considerably varying rates reported and makes simple comparison impossible, as Rahbari et al. pointed out recently [25]. However, leak rate still is unacceptably high, especially when taking into account

that up to 50% of postoperative deaths after colorectal surgery are caused by sepsis due to leakage [26–28]. Meticulous technique, a tension-free anastomosis, and an appropriate blood supply [29] are basic requirements for undisturbed healing. Beyond that, the surgeon cannot influence the anastomotic healing process. There are a number of potential advantages to compression anastomosis when compared to hand-sutured or stapled techniques. An immediately and completely air- and watertight anastomosis is created. Because of the 100% circumferential compression, immediate and delayed hemorrhage may be abolished. As no foreign body materials are incorporated into the anastomosis, only minimal local inflammatory reaction is observed (Fig. 3). Because of its 27-mm external diameter, insertion of the device is greatly facilitated. In addition, final diameter of

123

1930

World J Surg (2011) 35:1925–1932

Fig. 3 a Pouch-rectal anastomosis 3 cm above the dentate line: endoscopic view immediately after ring removal 6 weeks postoperatively. b After 10 weeks no inflammation is seen at compression area (*), but granulation tissue is visible where the rectum was transected with a stapler (\) and in the colon pouch

anastomosis after removal of the ring is also 27 mm. As long as the ring stays in place, the internal luminal diameter available for stool passage is 20 mm, which compares well to the anastomotic diameters of frequently used circular staplers (external/anastomotic diameter: Covidien CEEA, 31/21.2 mm; Ethicon CDH, 29/20.4 mm). Compression anastomoses have been shown to be comparable to stapled anastomoses in terms of anastomotic leakage. The rate of symptomatic leakage (2.5%) in the AKA-2 trial involving 442 subjects is promising, even though a diverting stoma was raised in 25% [11]. Thiede et al. [10] reported a clinical leak rate in the lower gastrointestinal tract of 4.5% (20/448) while using the BAR device. In this series, we recorded one case of leakage from a colonic J-pouch but no clinical leaks from the compression anastomosis at all. Formerly used compression devices such as the BAR and the AKA-2 were not suitable for laparoscopic surgery and impractical for the construction of distal rectal anastomoses. The Colon Ring instrument represents the first compression anastomotic device that has the same field of application as conventional circular staplers. In addition, with three nitinol springs in the compression ring, a variable tissue thickness can be accommodated around the circumference. This technical feature represents an advancement compared to the AKA-2 device, as the force of compression does not vary depending on the thickness of entrapped tissue. The force of the springs has been adjusted in order to delay ring expulsion. Compared to the AKA-2 device where ring expulsion was observed after 4 and 6 days, the compression anastomosis ring is passed after a median of 10 days [11, 30]. Therefore, especially for (delayed) anastomotic healing, this device may outmatch the AKA-2 device, theoretically leading to better anastomotic leak rates. Technical features, differences, and pitfalls concerning the handling of this instrument have been discussed extensively in previous publications [15, 30–34]. A potential drawback of this compression anastomotic device could be that conservative management of leaks

123

using transanal drainage or endo-sponge-assisted treatment [35, 36] may be impossible as long as the ring is in place. On the other hand, early removal of the ring in case of a leak may result in complete separation of the anastomosis, which in term might increase the need for reoperation and Hartmann’s procedure [37]. Three of 12 patients (25%) who had a low anterior resection experienced pain while the ring was in place. Symptoms subsided after ring removal. It may be a point of discussion whether supra-anal anastomoses should be constructed using this device due to its potential for pain and because of limited local damage control. However, leak rates are still are inadmissibly high especially in stapled low rectal anastomoses. If leakage can be significantly decreased using the compression anastomosis ring for distal rectal anastomoses, temporary discomfort represents a minor problem and may be acceptable when explained to patients. No anastomotic strictures were seen in diverted patients, and we assume that spontaneous passage of rings with the fecal stream after closure of stoma can be expected in these cases. In case of stricture formation, ring removal may become challenging later on. Therefore, it appears prudent to remove rings in diverted patients before stoma closure. As with anastomotic leakage, there is no uniform definition of anastomotic stenosis [38], which explains the wide range of reported stenosis rates (0–30%). Furthermore, the lack of a relationship between the lumen diameter and symptoms makes it even more difficult to compare studies that refer to this problem. Publications investigating stenosis of stapled anastomoses using routine clinical and endoscopic follow-up are rare, especially for benign indications. Ambrosetti et al. [19] performed sequential rectoscopies in 68 patients who underwent elective laparoscopic sigmoidectomy for diverticular disease. Surprisingly, a symptomatic anastomotic stenosis was found in 12 (17.6%) patients, with a mean diameter of 7 mm and necessitating endoscopic dilatation. In our series, only 1 of 23 patients (4.3%) operated for diverticular disease developed anastomotic stenosis. In the AKA-2 trial, only 2

World J Surg (2011) 35:1925–1932

of 442 patients (0.45%) and none of the patients treated with the BAR device suffered from symptomatic anastomotic stenosis [10, 11]. There is some evidence that the absence of the fecal stream in diverted patients, possibly dilating the anastomosis, increases the risk of stenosis formation [39–41]. In our series, all diverted patients with anastomoses located in the lower (n = 6) and the mid (n = 3) rectal third underwent flexible rectoscopy at 6 weeks and 6 months postoperatively. No anastomotic stenosis was observed. Minimal local inflammation and the large anastomotic diameter of 27 mm may be especially advantageous in this patient population (Figs. 1 and 3). Finally, bleeding from the anastomosis—a rare but potentially life-threatening event during the early postoperative period [42]—seems to be no issue at all when using compression anastomotic devices [10, 11]. Our results support this observation, as we did not observe anastomotic hemorrhage in a single patient. Published clinical data are scarce regarding this novel device. No leaks were detected in three small case series that investigated short-term results in a total of 32 patients [30–32]. Among those, we reported our own initial clinical experience using the device in 12 patients with favorable results. Buchberg et al. [33] recently published a pilot study that included 23 patients who underwent left-sided colectomy with anastomoses located more than 10 cm from the anal verge. Anastomotic leakage was reported in one patient (4.3%) and anastomotic stenosis in two cases (8.7%). To our knowledge, our study currently represents the largest series of patients in which this novel device was used and the only one that includes patients who were diverted or had (ultra-)low anterior resections. In conclusion, rectal anastomosis using this compression device is safe and appears to result in only very few anastomotic strictures. Persistent anal sensation may be observed in a relevant number of patients with low colorectal/anal anastomoses as long as the ring is in place. Larger and comparative studies are required to evaluate its full potential. Disclosure There were no financial ties with the manufacturer (NiTi Surgical Solutions Ltd., Netanya, Israel) or the supplier (AFS Medical GmbH, Teesdorf, Austria) of the devices used in this study. There was no financial support for study personnel, free equipment, or writing services.

References 1. Heald RJ, Leicester RJ (1981) The low stapled anastomosis. Br J Surg 68:333–337 2. Docherty JG, McGregor JR, Akyol AM et al (1995) Comparison of manually constructed and stapled anastomoses in colorectal surgery. West of Scotland and Highland Anastomosis Study Group. Ann Surg 221:176–184

1931 3. Graf W, Gimelius B, Bergstro¨m R et al (1991) Complications after double and single stapling in rectal surgery. Eur J Surg 157:543–547 4. Rullier E, Laurent C, Garrelon JL et al (1998) Risk factors for anastomotic leakage after resection of rectal cancer. Br J Surg 85:355–358 5. Carlsen E, Schlichting E, Guldvog I et al (1998) Effect of the introduction of total mesorectal excision for the treatment of rectal cancer. Br J Surg 85:526–529 6. Matthiessen P, Hallbo¨o¨k O, Andersson M et al (2004) Risk factors for anastomotic leakage after anterior resection of the rectum. Colorectal Dis 6:462–469 7. Jech B, Felberbauer FX, Herbst F (2007) Complications of elective surgery for rectal cancer. Eur Surg 39:8–14 8. den Dulk M, Noter SL, Hendriks ER et al (2009) Improved diagnosis and treatment of anastomotic leakage after colorectal surgery. Eur J Surg Oncol 35:420–426 9. Taflampas P, Christodoulakis M, Tsiftsis DD (2009) Anastomotic leakage after low anterior resection for rectal cancer: facts, obscurity, and fiction. Surg Today 39:183–188 10. Thiede A, Geiger D, Dietz UA et al (1998) Overview on compression anastomoses: biofragmentable anastomosis ring multicenter prospective trial of 1666 anastomoses. World J Surg 22:78–86. doi:10.1007/s002689900353 11. Wullstein C, Gross E (2000) Compression anastomosis (AKA-2) in colorectal surgery: results in 442 consecutive patients. Br J Surg 87:1071–1075 12. Kaidar-Person O, Rosenthal R, Wexner SD et al (2008) Compression anastomosis: history and clinical considerations. Am J Surg 195:818–826 13. Chen TC, Ding KC, Yang MJ et al (1994) New device for biofragmentable anastomotic ring in low anterior resection. Dis Colon Rectum 37:834–836 14. Szold A (2008) New concepts for a compression anastomosis: Superelastic clips and rings. Minim Invasive Ther Allied Technol 17:168–171 15. Stewart D, Hunt S, Pierce R et al (2007) Validation of the NITI Endoluminal Compression Anastomosis Ring (EndoCAR) Device and Comparison to the Traditional Circular Stapled Colorectal Anastomosis in a Porcine Model. Surg Innov 14:252–260 16. Fasth S, Hedlund H, Svaninger G et al (1982) Autosuture of low colorectal anastomosis. Acta Chir Scand 148:535–539 17. Shimada S, Matsuda M, Uno K et al (1996) A new device for the treatment of coloproctostomic stricture after double stapling anastomoses. Ann Surg 224:603–608 18. Schlegel RD, Dehni N, Parc R et al (2001) Results of reoperations in colorectal anastomotic strictures. Dis Colon Rectum 44: 1464–1468 19. Ambrosetti P, Francis K, De Peyer R et al (2008) Colorectal anastomotic stenosis after elective laparoscopic sigmoidectomy for diverticular disease: a prospective evaluation of 68 patients. Dis Colon Rectum 51:1345–1349 20. Dindo D, Demartines N, Clavien PA (2004) Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 240:205–213 21. Ravitch MM, Steichen FM (1972) Technics of staple suturing in the gastrointestinal tract. Ann Surg 175:815–837 22. Knight CD, Griffen F (1980) An improved technique for low anterior resection of the rectum using the EEA stapler. Surgery 88:710–714 23. Griffen FD, Knight CD Sr, Whitaker JM et al (1990) The double stapling technique for low anterior resection. Results, modifications, and observations. Ann Surg 211:745–751 24. Lustosa SA, Matos D, Atallah AN et al (2002) Stapled versus handsewn methods for colorectal anastomosis surgery: a

123

1932

25.

26.

27.

28.

29.

30.

31.

32.

33.

systematic review of randomized controlled trials. Sao Paulo Med J 120:132–136 Rahbari NN, Weitz J, Hohenberger W et al (2010) Definition and grading of anastomotic leakage following anterior resection of the rectum: a proposal by the International Study Group of Rectal Cancer. Surgery 147:339–351 Hansen O, Schwenk W, Hucke HP et al (1996) Colorectal stapled anastomoses. Experiences and results. Dis Colon Rectum 39: 30–36 Eckmann C, Kujath P, Schiedeck TH et al (2004) Anastomotic leakage following low anterior resection: results of a standardized diagnostic and therapeutic approach. Int J Colorectal Dis 19: 128–133 Khan AA, Wheeler JM, Cunningham C et al (2008) The management and outcome of anastomotic leaks in colorectal surgery. Colorectal Dis 10:587–592 Hirano Y, Omura K, Tatsuzawa Y et al (2006) Tissue oxygen saturation during colorectal surgery measured by near-infrared spectroscopy: pilot study to predict anastomotic complications. World J Surg 30:457–461. doi:10.1007/s00268-005-0271-y D’Hoore A, Hompes D, Folkesson J et al (2008) Circular ‘superelastic’ compression anastomosis: from the animal lab to clinical practice. Minim Invasive Ther Allied Technol 17:172–175 Dauser B, Herbst F (2009) NITI Endoluminal Compression Anastomosis Ring (NITI CAR 27Ò): a breakthrough in compression anastomoses? Eur Surg 41:116–119 Tulchinsky H, Kashtan H, Rabau M et al (2010) Evaluation of the NiTi Shape Memory BioDynamix ColonRing in colorectal anastomosis: first in human multi-center study. Int J Colorectal Dis 25:1453–1458 Buchberg BS, Masoomi H, Bergman H et al (2011) The use of a compression device as an alternative to hand-sewn and stapled

123

World J Surg (2011) 35:1925–1932

34. 35.

36.

37.

38.

39.

40.

41.

42.

colorectal anastomoses: is three a crowd? J Gastrointest Surg 15(2):304–310 Ho YH, Ashour MA (2010) Techniques for colorectal anastomosis. World J Gastroenterol 16:1620–1621 Riss S, Stift A, Meier M et al (2010) Endo-sponge assisted treatment of anastomotic leakage following colorectal surgery. Colorectal Dis 12:104–108 van Koperen PJ, van Berge Henegouwen MI, Rosman C et al (2009) The Dutch multicenter experience of the endo-sponge treatment for anastomotic leakage after colorectal surgery. Surg Endosc 23:1379–1383 Maggiori L, Bretagnol F, Lefe`vre JH et al (2011) Conservative management is associated with decreased risk of definitive stoma after anastomotic leakage complicating sphincter-saving resection for rectal cancer. Colorectal Dis 13. doi:10.1111/j.14631318.2010.02252.x Bannura GC, Cumsille MA, Barrera AE et al (2004) Predictive factors of stenosis after stapled colorectal anastomosis: prospective analysis of 179 consecutive patients. World J Surg 28:921–925. doi:10.1007/s00268-004-7375-7 Suchan KL, Muldner A, Manegold BC (2003) Endoscopic treatment of postoperative colorectal anastomotic strictures. Surg Endosc 17:1110–1113 Virgilio C, Cosentino S, Favara C et al (1995) Endoscopic treatment of postoperative colonic strictures using an achalasia dilator: short-term and long-term results. Endoscopy 27:219–222 Waxman BP, Ramsay AH (1986) The effect of stapler diameter and proximal colostomy on narrowing at experimental circular stapled large bowel anastomosis. Aust N Z J Surg 56:797–801 Martı´nez-Serrano MA, Pare´s D, Pera M et al (2009) Management of lower gastrointestinal bleeding after colorectal resection and stapled anastomosis. Tech Coloproctol 13:49–53