Laparoscopic Reversal Hartmann's Operation
Laparoscopic Reversal Hartmann’s Procedure
S Slawik, AR Dixon
Dept Colorectal Surgery, North Bristol NHS Trust, Frenchay Hospital, Bristol BS16 1LE. UK
Correspondence to Mr Dixon. E Mail; Anthony.Dixon@nbt.nhs.uk. www.bristolsurgery.com
Key words: Laparoscopic colorectal surgery, Reversal Hartmanns’ procedure
Purpose. The aim of this study was to analyse the outcome of laparoscopic restoration of bowel continuity post Hartmann’s procedure.
Methods. A prospectively electronic database of colorectal laparoscopic procedures between April 2001 and May 2006 has been used to identify surgical outcomes in 22 consecutive patients who have undergone laparoscopic reversal of Hartmann’s procedure (LRH).
Results. 22 patients (9 male), median age 65 (32-89), median BMI 25 have undergone an attempted LRH over a 5 year period. 14 (63%) had previously undergone an open Hartmanns’ procedure for complicated diverticulitis and six (27%) for cancer. Two patients had undergone an open procedure following an anastomotic dehiscence post laparoscopic high anterior resection. Adhesions were minimal/moderate in 77%. There was one late conversion/lap assisted (to release a small bowel loop from the pelvis). A small, fibrotic rectum split whilst attempting a stapled anastomosis – the procedure was abandoned. The median operation time of the 21 successful LRHs was 85 minutes (50 - 195). There were no deaths. The median time to normal diet was 24 hrs and median hospital stay was 3 days (1-7). There were three re-admissions, two with wound infections and one with abdominal pain.
Conclusions. Laparoscopic reversal Hartmann’s procedure is a feasible, safe and largely predictive operation that allows for early return of gastrointestinal function and early hospital discharge.
When intestinal continuity may be considered unwise, rectosigmoidectomy with end colostomy has stood the test of time1. Restoration of bowel continuity continues to be a major undertaking with a mortality of between 5-10% and an anastomotic leak rate of around 15%2,3,4. This perceived morbidity results in fewer than 60% of patients electing a reversal5. Timing of reversal is crucial with no mortality or anastomotic disruption when patients had their second stage performed after six months2. As reversal necessitates a long midline incision, variable peritoneal lysis and considerable bowel handling, laparoscopic Hartmann’s reversal is an attractive proposition. There have however, been very few reports since it was first described in 19936 and all have largely employed laparoscopic-assisted techniques. Conversion rates are as high as 25% and this is generally due to dense adhesions or an inability to identify the rectal stump7,8. Other authors have advocated hand-assisted techniques9. We describe our up-to date experience of laparoscopic Hartmann’s reversal.
Patients and methods
A prospectively collected electronic database of all colorectal laparoscopic procedures performed between April 2001 and May 2006 has been used to identify surgical outcomes of 22 consecutive patients who have undergone laparoscopic reversal of Hartmann’s procedure. All received preoperative mechanical bowel preparation.
Patients were placed in the dorso-lithotomy position using Allen stirrups, arms extended by their side and minimal hip flexion. All received single dose gentamicin (4mg/kg) and metronidazole. The surgeon and camera-operator stand on the patient’s right. Monitors are placed on each side. Port placement is dependent on the original abdominal incision. In general, the peritoneum was accessed via a 12mm port, sited midway in the right upper quadrant, just to the left of the right rectus sheath using an open cutdown technique. The peritoneal cavity was insuflated to 12mm Hg and a visual inspection performed. A 300 laparoscope was used. Two 5mm threaded disposable ports were placed in the right iliac fossa, approx. 10-12cms apart. Port placement was dependent on body habitus and length of abdominal cavity. It was occasionally necessary to use a 5mm port sited in an adhesion free area to allow dissection and safe placement of the right sided ports.
The patient was tilted towards the surgeon before the momentum and small bowel was mobilised from behind the incision. The adhesiolysis was by sharp scissor dissection using DeBakey-type forceps for bowel retraction and in the main confined to the left-hand side of the abdomen. The descending colon and splenic flexure was then fully mobilised using ultrasonic shears. The patients were then placed in steep Trendelenberg and the small bowel mobilised from the pelvis to allow identification of the rectal stump. Once visualised, the stump was dissected as needed (taking care to avoid the bladder) to enable a stapled anastamosis. The rectum was then serially dilated. The colostomy was then excised and mobilised down to the peritoneum, the colon withdrawn and the anvil of a circular stapler (29 CDH – Ethicon Endosurgery) secured within its lumen. The colon was then returned to the abdomen and the stoma site closed using a polydioxanone (1) suture and infiltrated with local anaesthetic. The anvil was finally secured to the stapling device introduced through the rectum watching once more for the bladder, vagina and any rotation of the bowel. A suction drain was placed in the pelvis via one of the 5mm ports. The wounds were finally closed and infiltrated with 0.25% Bupivicane.
Patients were allowed fluids as tolerated, mobilised the same evening and encouraged to take a light diet. Analgesia was provided by intra-operative fentanyl, diclofenac followed in the first 12 patients by Patient Controlled Analgesia (morphine). The last 10 patients only received intravenous parecetamol. We have never used epidural anaesthesia in our laparoscopic colorectal practice. In the last 8 patients, urinary catheters were removed at completion of the surgery.
Laparoscopic restoration of bowel continuity post Hartmann’s procedure (LRH) was attempted in 22 patients (9 males). All had left sided colostomies. Indications for Hartmann’s were complicated diverticulitis (14), cancer (6) and anastomotic dehiscence (2). Their median age was 65 years (range 32-89), median weight 65 kg and median BMI 25 (range 21-29). There was one late conversion (lap assisted) to allow safe mobilisation of a small bowel loop adherent in the pelvis. Poor optical visualisation and fear of perforation precluded continuation of the laparoscopic mobilisation.
LRH was abandoned after 4 hours in a further patient when the rectum split following insertion of the staple gun. This followed a full mobilisation of the transverse colon, hepatic flexure and division of the middle colic vessels to gain sufficient colon length and mobility to reach the short rectal stump. It was considered inappropriate to convert and undertake a proctectomy with coloanal anastamosis.
The median operative time of the 21 completed LRHs was 85 minutes (range 50-195 minutes). The median time to resumption of light diet was 24hrs (range 4-72hrs). There were no other intra-operative complications or post-operative deaths. The median postoperative stay was 3 days (range 1-7). Three patients were readmitted, two with wound infections at their colostomy sites and one with abdominal pain. All responded to conservative measures. A CT scan demonstrated no cause for the abdominal pain.
Laparoscopic colorectal surgery has, more than any other laparoscopic procedure, created extreme controversy and debate, particularly within the establishment. On reflection this is not surprising given that few specialist surgeons are trained to do it! For those that have been trained, experience and continuous technological innovation has encouraged surgeons to attempt more complex interventions. The objectives are reduced postoperative pain, early mobilisation, reduced rates of wound sepsis, rapid return of gastrointestinal function, early discharge from hospital, return to normal life, avoidance of incisional hernias and where there has been no prior incision superior cosmesis.
Restoration of bowel continuity post Hartmann’s procedure continues to be a major technical challenge with a likely hood of substantial morbidity and extended hospital stay consequent to a reported anastomotic leak rate of around 15%2,3,4a and mortality of between 5-10%. As a result, fewer than 60% of patients elect to undergo a reversal5. Laparoscopic Hartmann’s reversal6 is therefore attractive. It is however a relatively new technique with only a small number of series7,9-11 each having no more than 15-20 patients, most of which employed laparoscopicaly assisted techniques. Whilst all authors have confirmed its feasibility, excessive pelvic adhesions or an inability to identify the rectal stump has lead to conversion rates of 25%,7,8.
Rosen et al., recently reported11 a largely laparoscopic series (20/22); conversion 9%, mean operation time 158 mins, length of stay 4.2 days. Wound infections were reported in 14%. Their operative approach differed to ours in that they began by dissecting the stoma from the abdominal wall, continued some of the adhesiolysis under direct vision before obtaining the pneumoperitoneum using a port secured at the former colostomy site. We have preferred to use the pneumoperitoneum to aid adhesiolysis utilising gravity as well as the stoma “as a second assistant”. We have confirmed that the laparoscopic reversal of a Hartmann’s procedure is feasible, largely predictable and represents a true alternative to open restoration of bowel continuity. The procedure results in minimal morbidity and a short hospital stay. Whilst conversion is seldom necessary, LRH is like its open counterpart, one of the more difficult laparoscopic colorectal interventions. Adhesiolysis can be particularly challenging. If an enterotomy is to be avoided, sharp scissor dissection is required and with the points always visible. We believe that it is important to suture any serosal tears immediately as laparoscopy probably underestimates their extent. The laparoscopic approach does have additional advantages in allowing visualisation of the splenic flexure, spleen and if the former has not been mobilised at the original surgery a relatively straightforward dissection. We do not however support triumph for technology over sound common sense and recognise the need to convert or abandon the planned procedure.
1. Hartmann H. Nouveau prodede d’ablation des cancers de la partie terminale du colon pelvien. Trentieme Congress de Chirurgie. Strasborg 1923; 411-3.
2. Pearce NW, Scott SD, Karran SJ. Timing and method of reversal of Hartmann’s procedure. Br.J.Surg 1992; 79:389-41.
3. Wigmore SJ, Duthie GS, Young IE et al., Restoration of bowel continuity following Hartmann’s procedure: The Lothian experience 1987-1992. Br.J.Surg. 1992; 79: 839-41.
4. Roe AM, Prabhu S, Ali A Reversal of Hartmann’s procedure: Timing and operative technique. Br.J.Surg. 1991; 82: 27-30.
5. AR Dixon, JT Holmes. Hartmann’s procedure for carcinoma of rectum and distal sigmoid colon: 5-year audit. J.R.Coll.Surg.Edinb, 1990; 35 166-68.
6. Gorey TF, O.Connell PR, Waldron D, Cronin K, Kerin M, Fitzpatrick JM. Laparoscopically assisted colostomy closure after Hartmann’s procedure. Br.J.Surg. 1993; 80: 109.
7. Sosa Jl, Sleeman D, Puente I, Mckenney MG, Hartmann R. Laparoscopic-assisted colostomy closure after Hartmann’s procedure. Dis.Colon Rectum,. 1994; 37: 149-52.
8. Vernava AM 3rd, Liebscher G, Longo WE. Laparoscopic restoration of bowel continuity after Hartmann procedure. Surg.Laparosc Endosc 1995; 5: 129-32.
9. Lucarini L, Galleano R, Lombezzi R, Ippoliti M, Ajraldi G. Laparoscopic assisted Hartmann’s reversal with the Dexterity Pneumo Sleve. Dis.Colon Rectum, 2000; 43: 1164-7.
10. Regadas FS, Siebra JA, Rodrigues LV et al. Laparoscopically assisted anastomosis post Hartmann’s procedure. Surg.Laparosc.Endosc 1996; 6: 1-4
11. Rosen MJ, COBB ws, Kercher KW, Sing RF, Heniford BT. Laparoscopic restoration of intestinal continuity after Hartmann’s procedure. Am.J.Surg 2005; 189: 670-74.