COLORECTAL ANZJSurg.com
Laparoscopic versus open reversal of Hartmann’s procedure: a retrospective review Phillip F. Yang and Matthew J. Morgan Department of Surgery, Bankstown-Lidcombe Hospital, Sydney, New South Wales, Australia
Key words Hartmann’s procedure, Hartmann’s reversal. Correspondence Dr Matthew J. Morgan, Department of Surgery, Bankstown-Lidcombe Hospital, Eldridge Road, Bankstown, NSW 2200, Australia. Email: [email protected] P. F. Yang MBBS; M. J. Morgan MBBS, MS, FRACS. Accepted for publication 12 April 2014. doi: 10.1111/ans.12667
Abstract Background: Restoration of bowel continuity following Hartmann’s procedure may be performed using a laparoscopic or open technique. This study is the first of its kind comparing laparoscopic with open reversal of Hartmann’s procedure in Australasia. Methods: This is a retrospective review of 107 patients who underwent either a laparoscopic (n = 43) or open (n = 64) reversal of Hartmann’s procedure between 2001 and 2012. Outcome measures were perioperative clinical outcomes and post-operative complications. Results: Patients in the two groups were comparable in age, body mass index, American Society of Anesthesiologists score and number of previous operations. The most common indication for the original Hartmann’s operation in both groups was diverticular disease. Total theatre time was longer for the laparoscopic group (276.4 versus 242.0 min; P = 0.02). Three patients in the laparoscopic group required conversion to laparotomy (7%). Laparoscopic reversal of Hartmann’s procedure was associated with shorter time to passage of flatus (2.8 versus 4.0 days; P < 0.001) and faeces (4.2 versus 5.6 days; P = 0.002), and shorter overall length of hospital stay (6.7 versus 10.8 days; P < 0.001). There were fewer patients in the laparoscopic group who had post-operative complications (14% versus 31%; P = 0.04), including fewer cases of post-operative ileus (2% versus 17%; P = 0.02). There were no cases of anastomotic leak or in-hospital mortality in either group. Conclusion: Laparoscopic reversal of Hartmann’s procedure is a safe and feasible alternative to open Hartmann’s reversal and may be associated with significantly faster recovery time and fewer post-operative complications.
Introduction Reversal of Hartmann’s procedure is a difficult colorectal operation associated with a high rate of post-operative complications. It involves the closure of a left-sided colostomy and restoration of bowel continuity between proximal colon and distal rectal stump.1 Patients who have undergone a successful reversal of Hartmann’s procedure usually report improved quality of life because they no longer face the physical and psychological difficulties of living with a stoma.2 However, this operation is associated with a perioperative morbidity rate ranging from 3% to 50% (mean 16.3%) and a mortality rate of up to 7.1% (mean 1%).3 Consequently, up to 40% to 60% of patients never have their colostomies reversed.4,5 Conventionally, reversal of Hartmann’s procedure is performed using an open technique (i.e. laparotomy); however, laparoscopic approaches have been described.6 In 2010, Siddiqui et al.7 published © 2014 Royal Australasian College of Surgeons
a systematic review and meta-analysis of eight retrospective studies8–15 comparing laparoscopic with open reversal of Hartmann’s procedure. Since this review, five further retrospective studies16–20 and one large multicentre study using prospectively collected data extracted from the American College of Surgeons National Surgical Quality Improvement Program database21 have also been published. These studies have consistently supported the hypothesis that laparoscopic reversal of Hartmann’s procedure is associated with fewer post-operative complications, a lower mortality rate and a shorter length of hospital stay compared with open reversal of Hartmann’s procedure. No Australasian group has published a study comparing laparoscopic with open reversal of Hartmann’s procedure. The purpose of this study is to see whether the overseas experience with laparoscopic reversal of Hartmann’s procedure is replicated in an Australian healthcare setting and to provide further data on the safety of this procedure. ANZ J Surg 84 (2014) 965–969
966
Methods Patients and study design Patients who underwent a reversal of Hartmann’s procedure performed by the senior author (MJM) between January 2001 and September 2012, as well as all patients who underwent this operation at Bankstown Hospital and Fairfield Hospital during the same period, were included in the study. Eligible patients were identified by interrogating the medical records database of each hospital for admissions that were coded with Medicare Benefits Schedule item 32033 (Restoration of bowel continuity following Hartmann’s procedure). Patients who underwent an operation that was incorrectly coded as a reversal of Hartmann’s procedure were excluded.
Laparoscopic reversal of Hartmann’s procedure: surgical technique The patient is placed in a supine position with legs in Lloyd-Davies stirrups, thighs level with the abdominal wall and arms adjacent to the body. A urinary catheter and a nasogastric tube are inserted. Two monitors are required as the extent of small bowel adhesiolysis that may be required varies. The surgeon usually stands on the patient’s left for the initial mobilization of the colostomy. The stoma is mobilized to the level of the abdominal wall and then freed from the fascia. The colon is transected to remove the stoma and the anvil of a circular stapler is secured in the proximal colon. It is then returned into the abdominal cavity. Any adhesions in the abdominal cavity on direct view are freed to enable insertion of 10-mm camera ports under direct vision in the midline and left upper quadrant. This allows different camera positions to be used for dissection of adhesions. The colostomy site can be closed using an Alexis (Applied Medical, Rancho Santa Margarita, CA, USA) wound protector/retractor device with a laparoscopic cap so that it can function as an additional working port. A pneumoperitoneum is then established. Dissection of adhesions from the anterior abdominal wall allows placement of additional ports in the right upper quadrant and right iliac fossa. The small bowel is mobilized from the left iliac fossa and out of the pelvis. Care must be taken to identify the gonadal vessels and left ureter in the left paracolic gutter, at the pelvic brim and on the pelvic side wall. The proximal descending colon would have been mobilized to a varying extent at the initial Hartmann’s procedure and this will need to be redone, including the splenic flexure. A rectal probe or circular stapler sizer is used to identify the rectal stump. Sometimes the rectum may appear to have enough length, but the proximal rectum is scarred and narrowed and does not allow passage of the shaft of the circular stapler. If this is the case, in order to perform an end-to-end anastomosis, further mobilization and limited resection of the rectum with a laparoscopic stapler and division of the mesorectum is required. Alternatively, if mobilization is difficult and the anterior rectal wall can clearly be identified and adequate colon length is available to allow a tension-free anastomosis, an end-toside anastomosis can be performed. A circular stapler is introduced into the rectum to fashion the anastomosis. The donuts are checked and a leak test is done by filling the pelvis with saline and insufflating the rectum with air.
Yang and Morgan
Wash-out is performed and haemostasis confirmed. A 19 Fr Blake (Ethicon, Bridgewater, NJ, USA) drain is placed in the pelvis. Pneumoperitoneum is released and ports removed. The fascia is then closed.
Outcome measures Data relating to patient demographics, indication for previous Hartmann’s procedure, perioperative parameters and post-operative outcomes were collected by reviewing each patient’s medical record. Total theatre time was defined as from the moment the patient entered the operating room to when they departed. Postoperative outcomes studied include time to restoration of bowel function, post-operative complications and total length of hospital stay. Definitions of post-operative complications were predetermined and are the same as those used in the Australasian Laparoscopic Colon Cancer Study (ALCCaS).22 A delayed complication was defined as one that occurred within 28 days of discharge from hospital. Ethics approval was obtained from the ethics and research governance office of the Local Health District.
Statistical methods Continuous data are presented as mean ± standard deviation and compared using t-tests. Categorical data are summarized as percentages and compared using chi-square tests. Statistical analyses were performed using SPSS 19.0 statistical software (SPSS, Chicago, IL, USA). All statistical tests were two-tailed and statistical significance was set at the P < 0.05 level.
Results The intention-to-treat analysis included 107 patients (43 laparoscopic, 64 open) who underwent a reversal of Hartmann’s procedure (Table 1). Thirty-seven of the 43 laparoscopic cases (86%) were performed by the senior author (MJM). The patients in the two groups were comparable in age, gender, body mass index, American Society of Anesthesiologists score and number of previous abdominal operations. There were differences in the indications for the original Hartmann’s procedure between the laparoscopic and open groups, with 61% of patients in the former having had the operation for a complication of diverticular disease compared with 56% in the latter, and 12% of patients in the former having had the operation for colon cancer compared with 28% in the latter (χ2 = 11.4; df = 4; P = 0.02). Perioperative data are summarized in Table 2. The average total theatre time for the laparoscopic group was 34 min longer than for the open group (t105 = 2.33; P = 0.02; 95% CI, 5.1 to 63.8 min). Three (7%) patients in the laparoscopic group required conversion to open because of failure to progress because of extensive adhesions. Two of these patients had undergone some form of previous pelvic surgery (hysterectomy and bladder resection, respectively) in addition to their previous Hartmann’s procedure. Three (7%) patients in the open group required formation of a diverting ileostomy. One of these patients subsequently had her ileostomy reversed; the other two did not undergo reversal procedures because of multiple medical co-morbidities. There was no significant difference between the proportions of patients requiring post-operative admission to the intensive care unit. © 2014 Royal Australasian College of Surgeons
Laparoscopic versus open Hartmann’s reversal
967
Table 1 Baseline characteristics of patients undergoing reversal of Hartmann’s procedure Characteristic
Laparoscopic (n = 43)
Open (n = 64)
P value
60.9 ± 14.4 26/17 29.7 ± 6.5 — 5 (12%) 26 (61%) 1 2 17 4 2 3 (7%) 5 (12%) — 9 (27%) 19 (56%) 6 (18%) — 31 (72%) 8 (19%) 3 (7%) 1 (2%)
59.7 ± 17.7 34/30 27.6 ± 6.7 — 18 (28%) 36 (56%) 0 1 27 7 1 1 (2%) 9 (14%) — 13 (21%) 37 (59%) 13 (21%) — 45 (70%) 12 (19%) 5 (8%) 2 (3%)
0.71 0.45 0.22 0.02 — — — — — — — — — 0.79 — — — 0.99 — — — —
Laparoscopic (n = 43)
Open (n = 64)
P value
276.4 ± 70.0 3 (7%) 4/41 (10%) 2.8 ± 0.9 4.2 ± 1.7 6.7 ± 2.6 0
242.0 ± 78.3 — 9/64 (14%) 4.0 ± 1.5 5.6 ± 2.2 10.8 ± 6.4 0
0.02 — 0.51
Laparoscopic versus open reversal of Hartmann’s procedure: a retrospective review Phillip F. Yang and Matthew J. Morgan Department of Surgery, Bankstown-Lidcombe Hospital, Sydney, New South Wales, Australia
Key words Hartmann’s procedure, Hartmann’s reversal. Correspondence Dr Matthew J. Morgan, Department of Surgery, Bankstown-Lidcombe Hospital, Eldridge Road, Bankstown, NSW 2200, Australia. Email: [email protected] P. F. Yang MBBS; M. J. Morgan MBBS, MS, FRACS. Accepted for publication 12 April 2014. doi: 10.1111/ans.12667
Abstract Background: Restoration of bowel continuity following Hartmann’s procedure may be performed using a laparoscopic or open technique. This study is the first of its kind comparing laparoscopic with open reversal of Hartmann’s procedure in Australasia. Methods: This is a retrospective review of 107 patients who underwent either a laparoscopic (n = 43) or open (n = 64) reversal of Hartmann’s procedure between 2001 and 2012. Outcome measures were perioperative clinical outcomes and post-operative complications. Results: Patients in the two groups were comparable in age, body mass index, American Society of Anesthesiologists score and number of previous operations. The most common indication for the original Hartmann’s operation in both groups was diverticular disease. Total theatre time was longer for the laparoscopic group (276.4 versus 242.0 min; P = 0.02). Three patients in the laparoscopic group required conversion to laparotomy (7%). Laparoscopic reversal of Hartmann’s procedure was associated with shorter time to passage of flatus (2.8 versus 4.0 days; P < 0.001) and faeces (4.2 versus 5.6 days; P = 0.002), and shorter overall length of hospital stay (6.7 versus 10.8 days; P < 0.001). There were fewer patients in the laparoscopic group who had post-operative complications (14% versus 31%; P = 0.04), including fewer cases of post-operative ileus (2% versus 17%; P = 0.02). There were no cases of anastomotic leak or in-hospital mortality in either group. Conclusion: Laparoscopic reversal of Hartmann’s procedure is a safe and feasible alternative to open Hartmann’s reversal and may be associated with significantly faster recovery time and fewer post-operative complications.
Introduction Reversal of Hartmann’s procedure is a difficult colorectal operation associated with a high rate of post-operative complications. It involves the closure of a left-sided colostomy and restoration of bowel continuity between proximal colon and distal rectal stump.1 Patients who have undergone a successful reversal of Hartmann’s procedure usually report improved quality of life because they no longer face the physical and psychological difficulties of living with a stoma.2 However, this operation is associated with a perioperative morbidity rate ranging from 3% to 50% (mean 16.3%) and a mortality rate of up to 7.1% (mean 1%).3 Consequently, up to 40% to 60% of patients never have their colostomies reversed.4,5 Conventionally, reversal of Hartmann’s procedure is performed using an open technique (i.e. laparotomy); however, laparoscopic approaches have been described.6 In 2010, Siddiqui et al.7 published © 2014 Royal Australasian College of Surgeons
a systematic review and meta-analysis of eight retrospective studies8–15 comparing laparoscopic with open reversal of Hartmann’s procedure. Since this review, five further retrospective studies16–20 and one large multicentre study using prospectively collected data extracted from the American College of Surgeons National Surgical Quality Improvement Program database21 have also been published. These studies have consistently supported the hypothesis that laparoscopic reversal of Hartmann’s procedure is associated with fewer post-operative complications, a lower mortality rate and a shorter length of hospital stay compared with open reversal of Hartmann’s procedure. No Australasian group has published a study comparing laparoscopic with open reversal of Hartmann’s procedure. The purpose of this study is to see whether the overseas experience with laparoscopic reversal of Hartmann’s procedure is replicated in an Australian healthcare setting and to provide further data on the safety of this procedure. ANZ J Surg 84 (2014) 965–969
966
Methods Patients and study design Patients who underwent a reversal of Hartmann’s procedure performed by the senior author (MJM) between January 2001 and September 2012, as well as all patients who underwent this operation at Bankstown Hospital and Fairfield Hospital during the same period, were included in the study. Eligible patients were identified by interrogating the medical records database of each hospital for admissions that were coded with Medicare Benefits Schedule item 32033 (Restoration of bowel continuity following Hartmann’s procedure). Patients who underwent an operation that was incorrectly coded as a reversal of Hartmann’s procedure were excluded.
Laparoscopic reversal of Hartmann’s procedure: surgical technique The patient is placed in a supine position with legs in Lloyd-Davies stirrups, thighs level with the abdominal wall and arms adjacent to the body. A urinary catheter and a nasogastric tube are inserted. Two monitors are required as the extent of small bowel adhesiolysis that may be required varies. The surgeon usually stands on the patient’s left for the initial mobilization of the colostomy. The stoma is mobilized to the level of the abdominal wall and then freed from the fascia. The colon is transected to remove the stoma and the anvil of a circular stapler is secured in the proximal colon. It is then returned into the abdominal cavity. Any adhesions in the abdominal cavity on direct view are freed to enable insertion of 10-mm camera ports under direct vision in the midline and left upper quadrant. This allows different camera positions to be used for dissection of adhesions. The colostomy site can be closed using an Alexis (Applied Medical, Rancho Santa Margarita, CA, USA) wound protector/retractor device with a laparoscopic cap so that it can function as an additional working port. A pneumoperitoneum is then established. Dissection of adhesions from the anterior abdominal wall allows placement of additional ports in the right upper quadrant and right iliac fossa. The small bowel is mobilized from the left iliac fossa and out of the pelvis. Care must be taken to identify the gonadal vessels and left ureter in the left paracolic gutter, at the pelvic brim and on the pelvic side wall. The proximal descending colon would have been mobilized to a varying extent at the initial Hartmann’s procedure and this will need to be redone, including the splenic flexure. A rectal probe or circular stapler sizer is used to identify the rectal stump. Sometimes the rectum may appear to have enough length, but the proximal rectum is scarred and narrowed and does not allow passage of the shaft of the circular stapler. If this is the case, in order to perform an end-to-end anastomosis, further mobilization and limited resection of the rectum with a laparoscopic stapler and division of the mesorectum is required. Alternatively, if mobilization is difficult and the anterior rectal wall can clearly be identified and adequate colon length is available to allow a tension-free anastomosis, an end-toside anastomosis can be performed. A circular stapler is introduced into the rectum to fashion the anastomosis. The donuts are checked and a leak test is done by filling the pelvis with saline and insufflating the rectum with air.
Yang and Morgan
Wash-out is performed and haemostasis confirmed. A 19 Fr Blake (Ethicon, Bridgewater, NJ, USA) drain is placed in the pelvis. Pneumoperitoneum is released and ports removed. The fascia is then closed.
Outcome measures Data relating to patient demographics, indication for previous Hartmann’s procedure, perioperative parameters and post-operative outcomes were collected by reviewing each patient’s medical record. Total theatre time was defined as from the moment the patient entered the operating room to when they departed. Postoperative outcomes studied include time to restoration of bowel function, post-operative complications and total length of hospital stay. Definitions of post-operative complications were predetermined and are the same as those used in the Australasian Laparoscopic Colon Cancer Study (ALCCaS).22 A delayed complication was defined as one that occurred within 28 days of discharge from hospital. Ethics approval was obtained from the ethics and research governance office of the Local Health District.
Statistical methods Continuous data are presented as mean ± standard deviation and compared using t-tests. Categorical data are summarized as percentages and compared using chi-square tests. Statistical analyses were performed using SPSS 19.0 statistical software (SPSS, Chicago, IL, USA). All statistical tests were two-tailed and statistical significance was set at the P < 0.05 level.
Results The intention-to-treat analysis included 107 patients (43 laparoscopic, 64 open) who underwent a reversal of Hartmann’s procedure (Table 1). Thirty-seven of the 43 laparoscopic cases (86%) were performed by the senior author (MJM). The patients in the two groups were comparable in age, gender, body mass index, American Society of Anesthesiologists score and number of previous abdominal operations. There were differences in the indications for the original Hartmann’s procedure between the laparoscopic and open groups, with 61% of patients in the former having had the operation for a complication of diverticular disease compared with 56% in the latter, and 12% of patients in the former having had the operation for colon cancer compared with 28% in the latter (χ2 = 11.4; df = 4; P = 0.02). Perioperative data are summarized in Table 2. The average total theatre time for the laparoscopic group was 34 min longer than for the open group (t105 = 2.33; P = 0.02; 95% CI, 5.1 to 63.8 min). Three (7%) patients in the laparoscopic group required conversion to open because of failure to progress because of extensive adhesions. Two of these patients had undergone some form of previous pelvic surgery (hysterectomy and bladder resection, respectively) in addition to their previous Hartmann’s procedure. Three (7%) patients in the open group required formation of a diverting ileostomy. One of these patients subsequently had her ileostomy reversed; the other two did not undergo reversal procedures because of multiple medical co-morbidities. There was no significant difference between the proportions of patients requiring post-operative admission to the intensive care unit. © 2014 Royal Australasian College of Surgeons
Laparoscopic versus open Hartmann’s reversal
967
Table 1 Baseline characteristics of patients undergoing reversal of Hartmann’s procedure Characteristic
Laparoscopic (n = 43)
Open (n = 64)
P value
60.9 ± 14.4 26/17 29.7 ± 6.5 — 5 (12%) 26 (61%) 1 2 17 4 2 3 (7%) 5 (12%) — 9 (27%) 19 (56%) 6 (18%) — 31 (72%) 8 (19%) 3 (7%) 1 (2%)
59.7 ± 17.7 34/30 27.6 ± 6.7 — 18 (28%) 36 (56%) 0 1 27 7 1 1 (2%) 9 (14%) — 13 (21%) 37 (59%) 13 (21%) — 45 (70%) 12 (19%) 5 (8%) 2 (3%)
0.71 0.45 0.22 0.02 — — — — — — — — — 0.79 — — — 0.99 — — — —
Laparoscopic (n = 43)
Open (n = 64)
P value
276.4 ± 70.0 3 (7%) 4/41 (10%) 2.8 ± 0.9 4.2 ± 1.7 6.7 ± 2.6 0
242.0 ± 78.3 — 9/64 (14%) 4.0 ± 1.5 5.6 ± 2.2 10.8 ± 6.4 0
0.02 — 0.51
