Local Anaesthetic Day-care Haemorrhoidectomy Challenges Traditional Concepts - a Randomised Controlled Trial

Overview

Background: Local anaesthetic day-care open haemorrhoidectomy (LH) is feasible, cheap and may be the cost-effective surgical approach to third degree haemorrhoids. This prospective randomised controlled trial compares patient's evaluation of LH with general anaesthetic day-care Park's modified Milligan-Morgan haemorrhoidectomy (GH). Methods: 41 patients with third degree haemorrhoids were randomised to LH (19 cases) and GH (22 cases). Demographics were comparable. Independent assessment (by a research nurse) and clinical evaluation ran parallel for 6 months. Outcome measures were average and expected pain scores for 10 days; satisfaction scores at 10 days, 6 weeks and 6 months. Secondary outcomes were journey time and cost in day surgery.

INTRODUCTION: Open haemorrhoidectomy is a commonly performed painful operation. In 1998-99, 26514 haemorrhoidectomies were performed in the UK of which 14373 were day cases. A further 5342 could have been performed as day surgery1. Improvements in multimodal2 and pre-emptive analgesia3, introduction of stapled anopexy4 and improvements in patient counselling5 have led to increasing number of day surgical treatment of haemorrhoids. Technique of LH is described6-11 but with the paucity of randomised controlled trials this is rarely offered as a routine choice to patients with third degree haemorrhoids. AIM: This randomised controlled trial compares patient-evaluation of postoperative pain and expected pain scores for 10 days and satisfaction scores at 10 days, 6 weeks and 6 months between local and general anaesthetic day surgical open haemorrhoidectomy. Clinical outcomes (for 6 months), cost comparison and evaluation of total journey time in day surgical unit formed part of the study. PATIENTS AND METHODS: This randomised controlled trial was approved by the Research and Development department and the Local Research Ethics committee. We performed a pilot evaluation of ten-day average pain and expectation scores after LH in 7 patients. Based on the data from this pilot study and the average pain scores from the literature for GH, we calculated a requirement of 40 patients to give a 90% power to the study at the 5% significance level (student's t-test) to detect a 22% difference in 10 day average pain scores with a standard deviation of 1.712. Null hypothesis was that there was no difference in the pain scores between the two groups. Forty two patients with third degree haemorrhoids had full explanation of the trial (verbal and written with copies sent to their GP) and were randomised \[Figure 1\] on the day of surgery after consent. Patients unfit for day-surgery were excluded. Computer generated random numbers with opaque envelopes \[produced by the principal researcher (GNR)\] were opened by the research nurse (WH) after obtaining the research consent. Verbal and written information of the expected postoperative course following open haemorrhoidectomy were given to all patients. This leaflet detailed the measures to take in the event of constipation or difficulty in passing urine, if such a problem arose. They were given the research nurse's contact number, in the event of a problem and were advised to contact out-of-hours GP service or the research nurse. Research nurse \[WH\] assessed their pain and expectation scores in the post operative period on the day of surgery and spoke to all the patients on the day after surgery to record their comments and reinforce the importance of accurate scoring. Patients then recorded their pain and expectation scores at home for 10 days. This was the average of the pain that the patients experienced before, during and after defecation or the average of the pain during the course of the day. Two patients failed to return their 10 day pain scores \[7%\]. Pain scores were recorded on a ten-point Visual-analogue scale with scores from 1 to 10, with 1 being 'no pain' and 10 being 'as bad as it could be'. Ten-point patient expectation score was -5 to 5 with -5 being 'much better than expected' and 5 being 'much worse than expected'. Seven point satisfaction score was -3 to 3 with -3 being 'extremely dissatisfied' and 3 being 'extremely satisfied'. Clinical follow-up by the surgeon (NGR) at 10 days, 6 weeks and 6 months evaluated the surgical outcome and recorded all surgical complications. Patients sent their satisfaction scores directly to the research nurse at these time-points and were un-influenced by the surgical out-patient visit. All haemorrhoidectomies \[Park's modified Milligan-Morgan technique13\] were performed by a single surgeon \[NGR\]. Prone-Jacknife position (with buttocks strapped apart) was used for those randomised to local anaesthesia while Lithotomy-Trendelenburg position was used for those undergoing the procedure under general anaesthesia. Those with Type B and C buttocks10 needed more Trendelenburg position. Perianal block was used for all patients irrespective of their randomisation. GH were given the block prior to commencement of surgery. This was performed with 20 ml of 1% Lignocaine (with 1 in 10000 Adrenaline) injected at four sites \[midline anterior and posterior; left and right lateral\], fanning the needle in three directions at each of the four sites; care being taken to inject outside the external sphincter, to avoid pain. Once the sphincter muscle relaxed with the perianal block, a further 10 ml of the same anaesthetic was injected submucosally raising a bleb in each of sites corresponding to the sites of perianal block. This bleb was massaged inferiorly. No sedation was used with local anaesthetic haemorrhoidectomy. Following diathermy excision of haemorrhoids, a tube of lignocaine gel (Instillagel) was instilled into the anal canal followed by insertion of a gram of metronidazole and 50 mg of voltarol unless a history of asthma contraindicated the use of Voltarol in which case that was omitted. A soft sponge 'pack' \[curisponge\] was subsequently inserted. Patients were warned pre-operatively to expect the passage of the gelatinous curisponge in the post-operative period with their first defaecation. A package of medications was given to take home (TTA). TTA contained Co-codamol, Ibuprofen, Metronidazole, Lactulose, Dulcolax, 5% lignocaine ointment and 2% Diltiazem cream, with detailed instructions on its usage. Message in the post-operative instruction leaflet was reiterated prior to discharge. Journey time was calculated as the time elapsed between the start of anaesthesia and patient going home. Infection was defined as presence of purulent discharge with heightened pain, a visit to their general practitioner and use of antibiotics over and above the Metronidazole prescribed on discharge. Sum of the individual costs of the suture material, anaesthesia, cost of day surgery staff, recovery room costs and post operative medications were the basis of cost analysis. Statistical analysis: Data collected on Microsoft Excel was analysed on 'Analyse-it for Microsoft Excel' \[Analyse-it Software Ltd, Leeds, UK\]. Chi-square test and Mann-Whitney U test were used. Where the data was normally distributed as in the comparison of age, student's t-test was used. Serial measurements on linear analogue pain score were summarised by calculating the average pain over the 10-day time period for each patient as a summary measure14.