Laparoscopic vs. Open Right Hemicolectomies: Short Term Outcomes within an Enhanced Recovery After Surgery Programme

Randomised controlled trials have demonstrated that the benefits of laparoscopic colorectal surgery can be improved if these are carried in an ERAS setting [15-18]. However these trials have included a heterogenous group of colorectal resections:right, extended right, left, sigmoid and rectal [16,18,19]. Right hemicolectomy differs in the type of incision, anatomy, operative technique and potential complications [20,21]. Studies advocating laparoscopic right hemicolectomies over open resections have not integrated the ERAS programme [9,22-25]. It remains unclear whether the benefits of laparoscopy combined with ERAS can be applied to right hemicolectomies.

p<0.05). The majority of laparoscopic anastomoses were carried out through a right transverse incision (n=22, 68.8%). The remainder were performed either through a midline (7,21.9%) or pfannenstiel (3, 9.4%) incision. Two of the four laparoscopic procedures were converted to open as the tumour was more locally advanced than determined pre-operatively. The other two were due to adhesions from previous surgery. Nine (24.3%) open procedures were performed through a transverse incision and 28 (75.7%) through midline laparotomy. Table 3 outlines the post-operative recovery. Of note length of postoperative hospital stay for patients in Group A was statistically lower than those in Group B (day 5 vs median day 7, p<0.05). Although not significantly difference, patients in Group A passed flatus and opened their bowels one day earlier than patents from group B. There was no difference in the type of analgesia used for either group, with an epidural PCA being the most common. The epidural when used was taken down at the same day in both groups (median day 3). The urinary catheter was also removed at a similar stage (median day 3).
Complication rates for both groups are set out in Table 4. Some patients may be represented twice within this table, as a patient may have had an intra-abdominal collection as well as a wound infection. There were two reoperations in group B (5.4%), both for an anastomotic leak. Although the hospital stay was prolonged, both patients were eventually discharged. There were also two re-operations in group A (6.3%) -one for an iatrogenic perforation of the jejunum and another for bleeding. Two deaths (2.9%) occurred in total. The single death in group A occurred due to multi-organ system failure following the discharged with a urinary catheter in situ if they failed a trial without catheter.

Perioperative care
Patients were counselled regarding ERAS in pre-assessment clinic. They received a carbohydrate rich drink up to 2 hours before the operation. Bowel preparation was not given. Nasogastric tubes were not inserted routinely. Oral fluid was started on the operative day once fully recovered from anaesthesia. An enteral diet and mobilisation was encouraged from the 1 st postoperative day. Once this was established, parenteral analgesia was switched to oral and the urinary catheter was removed. The choice of perioperative analgesia was determined by the anaesthetist, which included epidural, TAPP block, spinal or parenteral analgesia. If required, laxatives or suppositories were given postoperatively. Prophylactic enoxaparin and anti-thromboembolic stockings were prescribed if there were no contra-indications.

Procedure technique
All procedures were carried out by one of four consultants. The decision to perform a right hemicolectomy for malignant or dysplastic lesions was made at the colorectal multi-disciplinary meeting. However the decision whether a patient should undergo an open or laparoscopic procedure was determined by the surgical consultant in conjunction with the patient on a case-by-case basis. Previous abdominal surgery was not a contraindication to laparoscopy. Procedures involving ligation of the middle colic artery were excluded. Open procedures were carried out either through a midline vertical or transverse incision. Anastomosis in both groups were formed using a linear stapler device between the ileum and transverse/ascending colon. During a laparoscopic resection colonic mobilisation was carried out intracorpeally. The anastomosis was performed extracorpeally through a right transverse, peri-umbilical midline or pfannenstiel incision. Individual surgeon preference determined whether the right colic artery ligation was carried out either extra-or intra-corporeally. Similarly medial to lateral or lateral to medial mobilisation of the right colon was carried out according to surgeon's technique. Procedures were converted to open to aid mobilisation. All patients received broad spectrum antibiotics at induction. Drains were not used routinely. The laparoscopic and extraction sites were routinely infiltrated with long acting Marcaine adrenaline solution to a maximum dose of 4mg/kg.

Statistical analysis
The data was analysed according to intention to treat basis. Therefore the laparoscopic converted to open procedures were included in group A. Chi-squared test was applied to discreet data, and Mann Whitney U-test to continuous data. Statistical significance was taken at p<0.05.

Results
Sixty-nine right hemicolectomies fulfilled the inclusion criteria. Thirty-two patients (46.3%) underwent a laparoscopic procedure (Group A). Four of the laparoscopic procedures (12.5%) were converted to open. The remaining thirty-seven patients had an open procedure (Group B). The demographics of both groups are outlined below in Table 1. The two groups were similar in terms of age, sex, and ASA. The mean BMI of group A (27.9kg/m 2 ) was significantly higher than for Group B (24.8kg/m 2 ). A similar percentage of patients in each group had undergone a previous abdominal operation.

Discussion
Our study demonstrates patients who underwent a laparoscopic right hemicolectomy (group A) within an ERAS setting had a significantly reduced postoperative stay compared to those that had an open procedure (group B). Although not statistically significant, those in group A also opened their bowels and passed flatus a day earlier. There was no difference in the number of days taken to stop parenteral analgesia, or to remove the urinary catheter. Operating time for the laparoscopic group was significantly longer. Tumour staging, lymph node clearance and complications were comparable between the two groups.
Since the introduction of laparoscopic colorectal resections and ERAS, various groups have investigated the optimum combination of operative technique and peri-operative management [17,18,[26][27][28][29][30]. Studies have demonstrated laparoscopic colorectal resections together with ERAS reduces hospital stay by 2 -4 days compared to open procedures with ERAS [4,16]. When comparing laparoscopic procedures within an ERAS setting or traditional post-operative management, the former combination has shown to be beneficial in terms of length of stay and wound infections [28].
However these studies have amalgamated right sided with extended right, left, sigmoid and rectal cancer resections. Due to the operative anatomy, technique and therefore unique complication profile, we believe it is inaccurate to apply these conclusions to right-sided operations. To date a subgroup analysis for right hemicolectomies within an ERAS setting has not been published.
Tan et al. [31] compared short term outcomes between laparoscopic and open right hemicolectomies within a peri-operative management protocol described as a co-ordinated clinical pathway (CCP). Both techniques resulted in the same median post-operative length of stay -5 days. All patients in this study received bowel prep, which differs to our ERAS protocol. This can worsen dehydration and prolong post op ileus [32,33]. Our results are similar to Khan et al [34]. The median length of stay in their study following a laparoscopic procedure was 4 days compared to 8 days for an open procedure. The laparoscopic group also had fewer anastomotic leaks (0% vs. 4%) and a lower mortality rate (0% vs 5.3%). Although all the procedures were carried in a ERAS setting similar to ours, this study differed in that 25% of the open and 10% of the laparoscopic resections were for tumours involving the transverse and splenic flexure, requiring an extended right hemicolectomy.
The position and length of incision are important factors in postoperative opiate use and risk of ileus. Tanis et al. [35] demonstrated transverse incision and laparoscopic right hemicolectomies were associated with a shorter hospital stay compared to midline laparotomies. Loshiriwat et al. [36] have shown no difference in short term outcomes between laparoscopic and transverse open right hemicolectomies, although Veenhof demonstrated a shorter hospital stay following a laparoscopic procedure [37]. Meta-analysis by Grantcharav et al. [38] described less pain for transverse compared to midline incisions. In our study, a similar percentage of patient in both groups used an epidural, and in both groups it was stopped at a similar stage (median post-operative day 3). However we were unable to determine if the mean opiate usage in milligrammes per kg body weight was different between the groups. Larger studies to compare transverse, midline incisions and laparoscopic right hemicolectomies within an ERAS setting are warranted. A transverse incision right hemicolectomy compared to laparoscopy has the added advantage of a shorter operating time and lower equipment costs.
Similar to other publications, a lower percentage of patients in Group A had pneumonia and required a blood transfusion compared to Group B [5,4]. One patient in Group A had an iatrogenic bowel perforation. This patient underwent a re-operation, however the repair leaked. This patient (ASA III) subsequently died. One death occurred following pneumonia, again in an ASA III patient. The overall anastomotic leak rate (2.9%) and 30-day mortality (2.9%) is comparable to larger series [39,40,3,34,41,35]. Both patients who had an anastomotic leak were re-operated and were eventually discharged home. The mean operating time for a laparoscopic procedure for 153 minutes compares well to other studies [39][40][41][42] Post-operatively a consultant or registrar would ensure the daily ERAS targets were being met. The consultants, whose data was included for this analysis, all had laparoscopic colorectal experience. Therefore both ERAS and laparoscopic surgery were well established before the first procedure included in this study.
Our study is not without its limitations. Although both groups were similar in terms of tumour characteristics and patient demographics, there is a lack of randomisation and blinding. Patients' and health professionals' perception that recovery after an open procedure is expected to be longer than a laparoscopic operation, may have increased the hospital stay. A number of patients in both groups could have been discharged earlier if their social circumstances had been assessed at pre-operatively. Cost analysis to determine if the increased cost of the laparoscopic procedure is balanced by the reduced cost of a shorter hospital stay is outside the aims of this study.

Conclusions
Our study demonstrates that patients undergoing a laparoscopic procedure for right sided tumours within an ERAS setting had a reduced hospital stay by 2 days compared to open procedures without a significant increase in complications or mortality. The operating time for laparoscopic procedures was significantly higher but there was no difference in lymph node yield. A randomised controlled trial comparing the different extraction sites for open procedures and laparoscopy is warranted.