AUDIT SYMPOSIUM

Abstracts presented at the Symposium of Audit on Surgical Practice, 9th March 2001

J.R.Coll.Surg.Edinb., 46, December 2001, 358-371

NB Due to pressure of space some abstracts may have been edited.

Acute laparoscopic cholecystectomy for acute gallstone disease
C. REDDY, D.J. ADAM, M. BUCHANAN, J. CASEY and S.J. NIXON Western General Hospital, Edinburgh, Scotland,U.K.

Background: Acute laparoscopic cholecystectomy (LC) has been advocated for selected patients with acute gallstone disease. Patient selection and exclusion of those who do not undergo acute LC affect the outcome of published series. In the absence of contra-indications, it is the senior author’s policy to attempt acute LC in this group of patients. This audit examines the management of a consecutive series of patients with acute gallstone disease. Methods: Between 1 January 1997 and 30th June 2000, the Lothian Surgical Audit database identified 131 consecutive patients (35 men and 96 women of median age 54, range 19=93 years) who presented to the senior author as an emergency with acute biliary colic, cholecystitis or gallstone pancreatitis. The hospital case notes were reviewed. Results: Five patients were not offered surgery due to significant co-morbidity. In eight patients LC was deferred due to significant cardio-respiratory co-morbidity (n=4), jaundice (n=3) or severe pancreatitis (n=1). A further two patients declined to undergo acute LC. Surgery was performed during the initial admission in 116 (92%) patients. Acute open cholecystectomy (OC) was performed in nine patients due to previous gastrectomy (n=3), multiple previous laparotomies (n=2), gallbladder mass (n=2), cholecyst-cutaneous fistula (n=1) and large incisional hernia (n=1). Acute LC was attempted in the remaining 107 (92%) patients and was successful in 98 (92%) patients. One patient (0.8%) had a post-operative bile leak which was managed by ERCP and stent insertion. There was no mortality. Conclusions: These data demonstrate that acute cholecystectomy can be performed in over 90% of patients with acute gallstone disease. Of these patients, acute LC was attempted in over 90% and successful in over 90%. Even with an aggressive surgical approach, however, there remains a small proportion (13.5%) of patients in whom acute LC is contra-indicated. A reasonable standard for the management of acute gallstone disease is a) acute cholecystectomy should be performed in over 80% of patients, b) 80% of these should be attempted laparoscopically, and c) 80% of these should be completed laparoscopically.

Computerised colonography: the gold standard?

H.D.E. ATKINSON, M.R. BROADBENT, M.L. GREGORIADES,I.R. PARKER and D.N. ANDERSON St John’s Hospital, Livingston, Scotland,U.K.

Background: Computerised (CT) colonography (CTC) is the latest radiological technique for large bowel imaging, and combines 3-dimensional (3D) rendering of high-resolution axial spiral CT sections with the latest computer technology. Its diagnostic accuracy and clinical yield are currently under investigation, and are compared with barium enema and colonoscopy. We describe our experience of this novel diagnostic modality, with a view to its being used for routine imaging.
Methods: A symptomatic cohort of 103 patients with suspected colorectal cancer or colonic polyps underwent 3D CTC between July 1999 and December 2000. Ages ranged between 43 and 88 years mean 68 years. 47 were male and 56 were female. The examination required full bowel preparation and insufflation of the colon with gas, following paralysis with hyoscine butylbromide. A prone scan of the pelvis was performed with a subsequent supine scan, from perineum to diaphragm during the injection of an intravenous contrast bolus, which enhanced the bowel mucosa. The images were reconstructed, and then reviewed by a single consultant radiologist. This technique was not used in this hospital prior to July 1999. Results: Eighteen of the 103 patients had colonic carcinoma. Seventeen patients underwent surgery; 16 neoplasms were confirmed at operation and one patient had extrinsic compression of the colon by an ovarian cyst. One patient was too unwell to undergo operation and died of metastatic disease. This gave a true positive rate of 17/18, with a false positive rate of 1/18. CTC was unable to exclude colorectal carcinoma in three patients, who had some bowel wall thickening. All three had subsequent normal colonoscopies. We are not aware of any false negative results in our series. Twenty-one patients had suspected colonic polyps on CTC, ranging in size between 4 and 9mm. Subsequently 18 patients had either flexible sigmoidoscopy or colonoscopy. Nine of these patients had no polyp present, 9 had benign polyps removed, one had complete excision of an adenocarcinomatous polyp, one patient died of other pathology before further investigation could be carried out and one has had no follow-up, as yet. The data suggests that CTC gives a true positive rate of 10/19, with a false positive rate of 9/19 for polyp detection. CTC also gave valuable extracolonic information, which would otherwise have been missed, and revealed 8 patients to have serious occult neoplastic disease as well as 163 other incidental findings. Technical difficulties were experienced with 22 (21.4%) of patients, but only 2(1.94%) had an inadequate visualisation of the colon, necessitating subsequent barium enema. Eighty-one (78.6%) patients had no technical difficulties with their scans. Conclusion: CTC has good patient compliance and is a useful diagnostic modality in detecting colorectal neoplasms and enabled the evaluation of extracolonic structures and occult disease. The diagnosis of colonic polyps is still questionable, and possibly less accurate than with colonoscopy. It is possible that CTC will become the mainstay radiological investigation for colorectal cancer, despite technical issues including the length of the examination time, and training to interpret the images. It will be necessary to undertake a randomised prospective study comparing CTC with colonoscopy, to ultimately establish which is the superior method.

Ovarian cancer: the chicken or the egg?
C. BOURNE and D. ANDERSON St John’s Hospital, Livingston, Scotland, UK

Background and Objectives: The differential diagnosis between metastatic colonic and primary ovarian cancer is difficult to make, despite immunohistochemical tests. However it can have important implications on management and prognosis. The objectives of this study, therefore, was a) to observe the natural history of the disease process in women diagnosed with dual ovarian and colonic pathology, b) to assess the delay in identifying those who required colonic resection after their primary ovarian surgery, c) to highlight those features that would aid the diagnosis of possible dual pathology and d) to construct optimal diagnostic and treatment pathways. Patients and Methods: A cohort of 91 women in West Lothian, between 1992-1999, who where diagnosed and under-went ovarian cancer surgery. Women who required colonic resection, within a reasonable time-frame either after or before what was recorded as primary ovarian surgery, and the time lapse between the two. The pathological assessment of the ovary in those women who required colonic resection in conjunction with ovarian surgery, in comparison to those who had true primary ovarian lesions resected. Results: Eighty-three of 91 case notes were reviewed. 26 (31%) of the women had had colonic resections; of these 18 had colonic resections at the time of or after their ovarian surgery. They had an average age of 55 years and the most common pathological assessment was primary adenocarcinoma, metastatic adenocarcinoma and mucinous cancer. Fifty-seven women did not require colonic resections, their average age was 32, and the most common pathological entities were papillary and serous cystadenocarcinoma. Eight women had colonic resections prior to their ovarian surgery, the average age was 60 years and the majority of these ovaries depicted changes consistent with metastatic adenocarcinoma. Discussion: Nearly a third of all ovarian cancer patients in this study had at some time in the natural history of their disease, a colonic resection. The problems were highlighted and raised the following questions: a) should all women diagnosed with ovarian cancer have routine bowel visualisation preoperatively? b) should a combination of ovarian and colonic adjuvant chemo radiotherapy be used in conflict cases? and c) should we advocate the routine removal of both ovaries in postmenopausal women undergoing a colonic cancer resection. Further research, including genetic exploration, is required into this diagnostic and management problem.

An audit of outcomes of surgery for colon cancer in octagenerians
A.J. CLARK, C. WAKEFIELD, I. POPE, R.G. WILSON and M.G. DUNLOP Western General Hospital, Edinburgh, Scotland, U.K.

Introduction: The highest incidence of colorectal cancer is in the oldest age groups, but elderly patients are often denied radical curative resection due to preconceptions about poor outcomes. We carried out a detailed evaluation of key outcome parameters using prospective data capture from a cohort of patients aged >80 years undergoing resection for colorectal cancer. Methods: Prospectively captured operative and clinical data from Lothian Surgical Audit were augmented by case note review of all patients with colorectal cancer aged >80yrs treated at this institution in the years 1993-99. Data on current health and social status were obtained from the patients’ GPs. Mortality was ascertained from the Scottish Central Records. Results: Of 198 patients identified, 180 underwent abdominal resection. Overall complication rate was 55.6% but anastomotic leak rate was only 3.3%. Thirty and 60-day mortality was 8.9% and 13.9%, respectively. Five-year survival, by life-table analysis, was 27%. Dukes’ stage was highly predictive of survival (p=<0.0001). Multivariate analysis identified the following as significant independent predictors of survival outcome: emergency/elective operation (p<0.05), Dukes’ stage (p<0.0001). Site of primary tumour and social status prior to admission did not predict outcome. Fifty percent of patients admitted from home were discharged directly home. Of those alive at the end of the study period, 84.4% were still living at home. Conclusions: These data indicate that cancer-specific survival in octagenarians is comparable with the population as a whole following radical resection for colonic carcinoma. The major determinant of survival is tumour stage. Good social outcome can also be anticipated in many cases. These findings provide support for strategies aimed at early identification and elective surgical treatment of colonic cancer in the elderly.

Dundee experience of near-total thyroidectomy for thyrotoxicosis: an audit of 101 cases
  A.B. CRESSWELL, G.B. HANNA and D.M. SMITH Ninewells Hospital, Dundee, Scotland, U.K.

Background: There is no consensus opinion for the ideal extent of resection in the surgical management of thyrotoxicosis. Traditionally, a sub-total approach has been associated with a lower complication rate, albeit at the expense of a higher rate of disease recurrence. We present an audit of the largest British series of near-total thyroidectomy as the initial surgical management of thyrotoxicosis. Methods: A retrospective review of case notes was performed for 101 consecutive patients treated over a period of two years by a single endocrine surgeon. In addition to demographic data information on reasons for referral, complication rates, operative and postoperative details were recorded. Results: Of 101 patients undergoing operation, there was one case of recurrent disease, which was treated with radioactive iodine. Five patients required short-term supplementation for transient hypocalcaemia, and one patient developed persistent hypocalcaemia secondary to hypoparathyroidism. No patients required re-operation for significant haemorrhage and there was no incidence of recurrent laryngeal nerve dysfunction. Conclusions: This audit of near-total thyroidectomy has shown a reduced rate of recurrence of hyperthyroidism with no increase in complication rate, as compared with subtotal thyroidectomy. Near-total thyroidectomy, therefore, has replaced subtotal resection in our Unit as the optimal surgical management of thyrotoxicosis.

A profile of emergency general surgery in Edinburgh
E.J.DAWSON and S. PATERSON-BROWN University of Edinburgh, Royal Infirmary, Edinburgh, Scotland, U.K.

Objectives: To obtain data regarding the spectrum of emergency admission and operations in Edinburgh and how these differ between hospitals and surgeons of different subspecialties. The accuracy of the data source used was also assessed. Background: Loss of accident and emergency departments and increasing specialisation of surgical services dictate that such data will be necessary in the future planning, provision and delivery of emergency health care. Methods: Using the Lothian Surgical Audit database, all emergency admissions in 1999 were identified. The admitting consultant, clinical diagnosis, operation and consultant under whom any operation was performed were recorded under various categories. Consultants were categorised as Colorectal (CR), Hepatobiliary (HPB)/Upper Gastrointestinal (UGI) or General. The database used was assessed for its accuracy by comparison with other data sources. Chi squared was used for statistical analysis. Results: A broad spectrum of diagnoses and operations was demonstrated at the Royal Infirmary Edinburgh (RIE) and the Western General Hospital (WGH). Specialisation occurred in both hospitals. The RIE UGI/HPB surgeons saw and operated on more patients with biliary disease as a percentage of their total workload. (11.39% compared with 6.41% and 6.02% for the general and CR teams, respectively). (p<0.001) A similar trend was seen regarding emergency colorectal surgery by the colorectal surgeons. The comparison of data sources showed that the reproducibility of the data was excellent for diagnoses but not for operations. Conclusions: 1. The spectrum of emergency admissions and operations in Edinburgh is consistent with previously published data. 2. The trend towards specialisation in the surgical services has extended to involve emergency care. 3. The local database is an accurate representation of the emergency workload.

Outcome of admission with acute right iliac fossa pain to a dedicated emergency surgical unit 
M.A. THAHA, A.B. CRESWELL, G.B. HANNA , K.L. CAMPBELL, R.J.C. STEELE and M.H. LYALL Ninewells Hospital and Medical School, Dundee, Scotland, U.K.

Aims: Despite increased means of investigation, diagnosis of patients with acute right iliac fossa pain (RIF) remains a clinical challenge. This descriptive study focused on the outcome of admissions with acute RIF pain, to a dedicated surgical emergency unit. Methods: Using a prospectively collected database of acute adult surgical admissions, patients referred with RIF pain, during 1996 and 1997 were identified. The database was used to ascertain the management outcomes of these patients. A pathology database was utilised to confirm histologically any diagnosis of acute appendicitis. The diagnosis of non-specific abdominal pain was reached when the patient was discharged without a definitive diagnosis and without needing further investigations as outpatient. Results: A total of 633 referrals with RIF pain were received during the two-year period. These involved 602 patients (age range 12-87, male: female 228:405), with a read-mission rate of 4.7%. A total of 374 (59.1%) patients underwent operative treatment, comprising 248 appendicectomies (70 following diagnostic laparoscopy) and 178 diagnostic laparoscopies. Histological data were complete in 242 (97.6%) cases confirming acute appendicitis in 199 (82.2%) and revealing a normal appendix in 26 (10.7%) cases. Gynaecological pathology was identified in 57 (9%) patients and urological disease in 25 (3.9%). A diagnosis of non-specific abdominal pain was made in 211 (33.3%) cases. Mean hospital stay was 2.5 days with a complication rate of 1.7%. Discussion: Despite modern investigative techniques and use of a dedicated emergency unit, a significant proportion of patients referred with RIF pain leave without a specific diagnosis and a significant number of normal appendices continue to be removed. These data form the basis on which to assess future interventions to address these issues.

Variations in implementation of current national guidelines for the treatment of acute pancreatitis: implications for acute surgical service provision
I.T. VIRLOS, J. BARNARD and A. K SIRIWARDENA Manchester Royal Infirmary, Manchester, England, U.K.

Introduction: In 1998, the British Society of Gastroenterology published guidelines for the initial management of acute pancreatitis. Although several units reported comparisons of individual performance against these guidelines such audits only identify local variations. The aim of this study was to obtain a national overview of clinical performance in comparison with the 1998 guidelines, in order to identify areas of consistent variation from guideline standards. Methods: A search of MEDLINE (using the keywords acute pancreatitis; management; audit) yielded one audit. Published abstracts from major national surgical, pancreatic and gastroenterological meetings held in the United Kingdom, since publication of the guidelines, were then examined. Abstracts from the following annual meetings were studied: Association of Surgeons of Great Britain and Ireland 1999 (no audits identified) and 2000 (two audits); Association of Upper Gastrointestinal Surgeons 1998 (no audits) and 1999 (one audit); Pancreatic Society 1999 (no audits) and the British Society of Gastroenterology 1999 and 2000 (no audits). No audit was included more than once. Audits were included only if a full published extract was available. Data included at the time of presentation but not available for analysis in the published extract were also excluded. Data from these audits were compared with the audit goals defined in the original document. Results: Four audits providing pooled data on 445 patients were identified. Overall mortality from severe disease was 11% (range 4-17%). Definitive treatment of gallstone disease within 4 weeks of the index attack was performed in 49% (16-65%) [guideline recommendation 100% within four weeks]. High Dependency or intensive care facilities were available in 52% (20-100%) [recommendation 100%]. Urgent endoscopic cholangiography was available in 25% (12-41%) [recommendation 100%]. Conclusion: Accepting that pooled data is not a substitute for a national overview this study provides evidence of striking and consistent divergence from guideline standards in the current management of acute pancreatitis in the United Kingdom. While a component may be related to variations in individual practice, the finding of national variation in issues such as the availability of high dependency facilities and access to operating theatres for “indexadmission” cholecystectomy has potential implications for the provision of services for patients with acute pancreatitis.

The role of caudal epidural injections in the management of low back pain P.A. BANASZKIEWICZ, D. KADER and D. WARDLAW Woodend Hospital, Aberdeen, Scotland, UK.

Background: The value of caudal epidural steroid injection (CEI) in treatment of low back pain and sciatica is controversial. It is believed that CEI is mainly effective in treating acute radiculopathy, at intermediate-term follow-up, and has no long-term benefit. Objective: To study the outcome of all orthopaedic patients with low back pain and sciatica treated with CEI in a one-year period (1997) in Aberdeen. To evaluate the benefit of CEI in acute presentations and in patients with predominantly leg pain symptoms. Finally, to assess its cost effectiveness. Method: The case notes of 163 out of a possible 175 patients, (87 female 76 male), mean age of 51 years (17-88 years), were reviewed. Data collected included primary symptoms, presentation time, presumed diagnosis, magnetic resonance imaging diagnosis, grade of the surgeon ordering and performing the procedure, the quality and duration of response and final outcome. Patients were regarded as having an excellent response if they had good or excellent pain relief for >3 months, good (6 weeks-3 months), fair (4-6 weeks), brief (any pain relief for <4 weeks) and no relief. Results: Forty-one percent of patients had either an excellent/good response to CEI, while 34% were no better or worse. Patients presenting acutely or with predominantly leg pain symptoms did not respond any better to the CEI than chronic low back pain. Only 36 of 73 patients who were discharged had an excellent/good response. Therefore, the response to CEI did not influence the decision to discharge half of these patients. The grade of surgeon performing the procedure did not make a difference to the outcome. Conclusions: The outcome of CEI was entirely unpredictable and its clinical value remains unproven. The decision to perform the procedure may well remain a matter of personal choice and experience.

An audit of surgery for cervical spondylosis
J .COOPER and D. CURRIE Aberdeen Royal Infirmary, Foresterhill, Aberdeen, Scotland, U.K.

Objectives: This project entailed a retrospective analysis of the case notes of patients who had undergone surgery for cervical spondylosis during 1997 and 1998. It was performed in order to assess the nature and clinical effectiveness of these operations by comparing results with those published. Methods and Results: In all, there were 35 operations performed for radiculopathy and 32 for myelopathy though notes were not available for review in a small number of instances. In each case, the presenting features, type of operation and post-operative course were carefully noted. For radiculopathy, an anterior cervical discectomy with or without fusion was performed. Arm pain was relieved in 88% and 100% of cases, respectively. A wider variety of operations were under-taken for myelopathy but principally consisted of either anterior cervical decompression and fusion or posterior decompressive laminectomy. Criteria for a successful outcome were harder to quantify in myelopathy and clinical improvement or, at least the absence of deterioration post operatively was considered a positive outcome in accordance with much of the published work. This was achieved in 81% and 100% of patients, respectively. Discussion and Conclusion: Though it would appear that the short-term results of these operations are good, longer follow-up is needed. This audit also raised a number of questions. Does discectomy and fusion have any advantage over discectomy alone in the treatment of radiculopathy? Is magnetic resonance imaging or myelography the optimum investigation in radiculopathy? Does the source of bone graft in fusion operations have any bearing on clinical outcome? What are the best outcome measures in myelopathy and which surgical approach is best? Prospective trials are required in many areas of this subject so as to identify best practice and bring some uniformity to the surgical management of cervical spondylosis.

Is secondary revascularisation with prosthetic graft worthwhile?
P.J. DRISCOLL, C. J. PAYNE, J. KELMAN, J.A.MURIE and R.T.A. CHALMERS Royal Infirmary of Edinburgh, Scotland, U.K.

Background and Aims: The choice of conduit for secondary revascularisation after infra-inguinal bypass failure is often determined by the lack of available autologous vein. Prosthetic conduit (polytetrafluoroethylene (P.T.F.E.) or ‘Dacron’) is thought to be inferior in terms of graft patency and limb salvage rates. A retrospective study was performed of all infra-inguinal secondary bypass procedures performed in the Royal Infirmary of Edinburgh Vascular Unit to determine whether prosthetic bypass is a reasonable alternative to autologous vein in this clinical setting. Method: All patients who underwent repeat infra-inguinal revascularisation for critical limb ischaemia, between 1st January 1990 and 31st December 2000, were studied. Bypass patency, limb salvage and patient survival were assessed using Kaplan-Meier Life Tables and comparisons were made between vein and prosthetic graft using log rank analysis. Results: Forty-two secondary bypass operations were performed on 40 patients (28 male) of median age 67.5 years (range: 29 to 87). The primary bypass procedure was performed between 5 days and 15 years earlier and involved vein (18) or prosthetic (24) conduits. Any operation performed less than 30 days prior was considered to have failed due to technical causes and was, thus, discarded leaving 36 operations for analysis. All were performed for graft occlusion with subsequent limb threat. Secondary conduits used were prosthetic with interposed vein cuff (14), prosthetic alone (12), reversed saphenous vein (7), in-situ saphenous vein (2) and reversed cephalic vein (1). Outflow was to above-knee popliteal (3), below-knee popliteal (17) and crural vessels (16). Median follow-up was 15 months. Of these patients, 32 left hospital with a functioning graft. 80% of these were alive at median follow-up with 80% of vein conduits and 60% of prosthetic conduits patent with an intact limb. Interestingly, patient survival in this cohort was 60% at 5 years. Conclusions: There is no appreciable difference between graft patency and limb salvage for prosthetic and autologous conduits in secondary infra-inguinal revascularisation. Therefore, the use of prosthetic material is a reasonable option when limb threat complicates the failure of primary infra-inguinal bypass grafts, when no vein is available.

Audit of dynamic hip screw tip-apex distance in a District General Hospital 
S.E. CLARKE and W.T. JARVIS Crosshouse and Inverclyde Royal Hospitals , Scotland, U.K.

Background and Aim: Failure of internal fixation of proximal hip extracapsular fractures is a serious complication in a frail population and can be reduced by careful surgical technique. Baumgertner reported the value of the tip-apex distance in predicting failure of fixation of the hip with a sliding hip screw, in the Journal of Bone and Joint surgery in July of 1995. This showed that if the tip-apex distance was greater than 25mm there was a significant risk of cut-out. We reviewed our practice in the Ayr Hospital against the standard set in the above publication. Patients and Methods: We calculated the tip-apex distance on anteroposterior and lateral hip radiographs of 43 patients, between August 1999 and February 2000. The result obtained show that 84% of cases had a tip-apex within 25mm. The standard set was that all sliding hip screws should be placed within 25mm of the apex of the femoral head. The following interventions were put in place; A poster was placed in the trauma theatre, and the orthopaedic team was fully briefed of the audit. We re-audited the radiographs of 33 patients between May 2000 to October 2000. The results obtained showed an increased compliance to 91%. Conclusion: This audit has shown a clinical improvement in our practice and, hopefully, this will decrease our incidence of cut-out in sliding hip screws performed for peritrochanteric fractures of the hip.

Comparison of oblique and longitudinal groin incisions in vascular reconstructive surgery
A.M. PAISLEY, L. MARSON, P. REID, A. HOWD, and A.R. TURNER Queen Margaret Hospital, Dunfermline, Scotland, U.K.

Background and Aim: Lower limb vascular reconstructive procedures have traditionally been approached via a longitudinal groin incision. A recent audit of such incisions in our unit has yielded an infection rate of 37% (89/241 incisions). A previous comparative study has suggested a significant reduction in wound infection if an oblique incision is adopted. The aim of the present prospective randomised trial is to compare the complication rate in oblique and longitudinal incisions.Methods: All patients undergoing a vascular reconstructive procedure involving a groin incision in our unit, from March 2000, have been entered into the study. Prior to surgery, patients are randomised to undergo a longitudinal or oblique incision. All other aspects of surgical procedure are similar in both groups. Bilateral groin incisions, when performed, are included as a single incision and the same incision performed in both groins. Patients with previous groin incisions are excluded. Wounds are inspected post-operatively by a single observer (PR) and presence and severity of wound infection (minor: erythema with or without discharge; intermediate: requiring antibiotics; major: requiring readmission or surgery) or other complication (e.g. sensorimotor deficit, swelling) documented. Inspection occurs daily while an inpatient and on day 7, 14 and 21, if patients have been discharged. Differences in infection rate, infection severity and incidence of other wound complications between the 2 groups are determined using Chi-Square, with a p value of less than 0.05 being regarded as significant. Sixty patients will be recruited into each arm. Results: Forty-one patients have been studied to date, (22 males and 19 females, median age 73 (IQR 65-80) years). Twenty-two patients were randomised to receive longitudinal and 19 to oblique incisions. There was no significant difference between the two patient groups with regard to age, sex, body mass index, incidence of diabetes mellitus, smoking, bilateral incisions, per-operative antibiotic use or graft type. The incidence of wound infection was similar in both groups (longitudinal 50% v. oblique 63.2%, p=0.397). Infection in oblique wounds tended to be more severe than that in longitudinal wounds (oblique: minor 55%, intermediate 45%; longitudinal: minor 83%, intermediate 17% p=0.134). The incidence of additional complications tended to be higher in patients with oblique wounds (oblique: 4/22 (18%) v. 8/19 (42%); p=-.093). Conclusions: Interim results indicate that patients with oblique groin incisions tend to have wound infections of greater severity and a higher incidence of additional wound complications than patients with longitudinal incisions. Further patients are needed to determine whether this trend reaches statistical significance.

Minor amputations on a vascular surgical unit
G. PAMMA, V.L. HEATON, W.P. STUART and R.T.A. CHALMERS Royal Infirmary of Edinburgh , Scotland, U.K.

Background and Aims: Minor (digital and transmetatarsal) amputations are a common component of limb salvage surgery. These amputations are sometimes performed without a simultaneous or preceding revascularisation procedure. The aim of this study was to assess the success rates of these procedures and, furthermore, to identify which are likely to fail. Patients and Methods: The records of 104 patients (66 male, median age 67, range 42-92 years), who underwent 123 procedures on 108 limbs were studied. Twenty-three transmetatarsal (TM) and 100 single or multiple digital amputation procedures were performed. Results: In 74 (60%) cases bypass surgery or endovascular procedure was performed (22 (18%) simultaneously, 34 (28%), less than a month 18 (15%) longer than a month prior to amputation). Twenty-nine (29%) of the digital and 12 (53%) of the TM amputations required further intervention in the form of revision of amputation to a higher level, surgical debridement or further revascularisation. Of these, 24 (24%) of digital and 11 (47%) of TM amputations required conversion to a higher level of amputation. Of the 23 TM amputations, 7 had resulted from a failed digital amputation, only two of these were ever documented as healed. Forty percent of digital amputations performed without previous revascularisation healed, compared with 70% in cases following revascularisation (p<0.05, c2 test). A diagnosis of diabetes mellitis (DM) adversely affected the probability of healing to a significant degree (p<0.05, c2 test). Conclusions: Overall digital amputation is associated with an acceptable success rate. Furthermore, a large proportion of patients in whom no revascularisation is performed and/or possible will heal a digital amputation. Transmetatarsal amputations have a lower success rate. In particular, TM following a failed digital amputation has a particularly high failure rate.

An audit of an integrated care pathway for primary hip arthroplasty
G.H. PROSSER, A.D. TOMS, S.K. BAJAJ, and E.S. IBISTER New Cross Hospital, Wolverhampton, England, U.K.

Background: Clinical pathways are being increasingly used to integrate the care of patients with a specific illness. Clinical pathways are best practice tools that display treatment goals and the sequence and timing of staff activity to achieve these goals. They aim to produce a high quality outcome with increased patient care and decreased length of hospital stay and related costs. Primary hip arthroplasty is suitable for a clinical pathway as it is a common, costly procedure and the patient care process is relatively standardised. Success with such pathways is widely reported. The ‘Integrated Care Pathway’ (ICP) for primary hip arthroplasty was introduced in June 2000 at New Cross Hospital, Wolverhampton. Objectives: The objectives of the study were threefold: a) to determine any reduction in length of stay for patients undergoing primary hip arthroplasty during the 1st 4 months of using the ICP, b) to assess if the ICP is being used properly, and c) to highlight any changes that need to be made to the ICP. Method and Materials: Data was collected retrospectively from 173 patients, who underwent primary hip arthroplasty in the 12 months before the introduction of the ICP, and from 98 patients who underwent primary hip arthroplasty, during the 1st 4-month period of utilising the ICP. A questionnaire was sent to all staff that were using the ICP. Results: Mean length of stay was reduced from 16.28 to 10.1 days. Mode length of stay was reduced from 15 to 9 days. The readmission rate was low (2/98). Overall staff compliance with ICP was good with positive feedback from staff. Some changes to ICP was suggested. Conclusion: The ICP is an effective method for streamlining care and reducing length of stay for patients undergoing primary hip arthroplasty.

Patients with critical carotid artery stenosis or occlusion should be offered routine ophthalmic examination: results of a 12 month prospective audit
  M. SINTLER, H. BROWN, D. RITTOO, M. TSALOUMAS and R. VOHRA University Hospital Birmingham NHS Trust, Selly Oak Hospital, Birmingham, England, U.K.

Aims: To evaluate the carotid arteries for stenosis or occlusion in patients with signs of ocular ischaemic syndrome (OIS), asymmetrical ‘diabetic’ retinopathy, asymptomatic retinal emboli, rubeosis iridis or temporary visual loss (amaurosis fugax). To establish whether routine ophthalmic screening of patients with carotid artery stenosis or occlusion should be implemented. Methods: A retrospective audit was carried out, over a 12 months period, of all patients referred to the ophthalmic department at Selly Oak Hospital. Patients with signs of retinal ischemia, asymmetrical ‘diabetic’ retinopathy or amaurosis fugax were referred for carotid duplex ultrasound. Clinical and demographic data were recorded. Results: Thirty-three patients were identified (19 male, 14 female) with a mean age of 67 years (50-90). Eighteen patients were diabetic, 22 hypertensives, 14 with ischaemic heart disease; 10 patients had previously had a transient ischaemic accident or stroke. Five patients (15%) were identified with 100% occlusion of the carotid artery. Of these, one had asymmetrical ‘diabetic’ retinopathy, 2 had marked retinal ischemic changes and 2 had rubeosis. Four (12%) patients had critical stenosis (70-99%), one with asymmetrical retinopathy and 3 with amaurosis fugax. Eight patients had <70% stenosis and 16 patients had normal scans. From this group of patients, 52% (17 of 33) had carotid disease, 27% (9 of 33) had critical stenosis or occlusion. Of the patients with carotid occlusion, all had ocular changes that could progress to blindness. Conclusion: Patients with critical carotid artery stenosis or occlusion should have a routine ophthalmic examination to identify changes of ocular ischemia. Carotid occlusion may not be a stable, asymptomatic condition and may lead to ocular neovascularization and blindness if not treated.

Impact of laparoscopic appendicectomy on training opportunities for basic surgical trainees 
J. HENDERSON, M.J. FORSHAW and D.M. POWER Addenbrooke’s Hospital, Cambridge, England, U.K.

Introduction: Open appendicectomy (OA) is one of the commonest emergency operations performed by basic surgical trainees (BSTs) as principal operators. However, the introduction of laparoscopic appendicectomy (LA) may impede these opportunities. We audited the impact of laparoscopic appendicectomy on the operative experience of BSTs against previously published experiences. Methods: The study was set in a teaching hospital with 4 senior house officers (SHOs) each attached to a surgical firm. New SHOs started halfway through each study period. Two time periods were examined: 6 months during 1992 and 1998. Operation notes and computerised theatre data were examined for all patients undergoing emergency open and laparoscopic appendicectomy. The operative experience of SHOs in open appendicectomy was recorded in one of three categories: performed (P), performed with assistance (PA) and assisted (A). Appendicectomies performed by locums and other surgical specialities were excluded. Results: By 1998, 21% of appendicectomies were performed laparoscopically. SHOs in 1992 performed a greater proportion of open appendicectomies (67.0%), compared with 1998 (61.4%), but this was not statistically significant (p>0.05). Seven SHOs in 1992, compared with only two in 1998, had previous general surgery experience. The operative experience in each time period compared favourably with previous accounts. Conclusions: Although current BSTs may be more inexperienced, we found no evidence to suggest that they were performing fewer open appendicectomies. However, this may change if laparoscopic appendicectomy is performed more frequently than in this audit.

Auditing the audit!
R.S. JUTLEY, A. McKINLEY, M. HOBELDIN, A. MOHAMED and G.G. YOUNGSON Royal Aberdeen Children’s Hospital, Aberdeen, Scotland, U.K.

Introduction: In an attempt to improve patient care and in keeping with the spirit of clinical governance, all NHS clinicians will be subject to revalidation and re-certification. This will be partly based upon previous and current clinical data, which, however, may be susceptible to error during collection and entry, and thus open to misinterpretation. An audit system (EMAS) has been in place at Royal Aberdeen Children’s Hospital for the last 10 years. All admissions are entered using specific codes for diagnoses, procedures and complications. EMAS produces a discharge summary that is discussed at a weekly meeting by all attending medical staff and any errors are corrected on the database. The aim of our study was to audit the existing audit process by evaluating the accuracy of data collection and entry. Methods: All consecutive admissions for three paediatric surgeons between February and July 2000 were evaluated. Results: There were 655 admissions during the study period. Erroneous data were entered in 62 instances in 60 patients (90.8% accuracy). In 26 (42%), the admission code was wrong. In 13 (21%), the coding for the procedure, although appropriate, could have been more specific. In 3 cases, the procedure was inappropriate to the diagnosis. In 1 case, there was no procedure documented for a condition that necessitated surgery (testicular torsion). In one case, other secondary diagnoses pertinent to the admission were not entered. In five instances, a wrong diagnosis was included in the database while in three others the consultant information was misleading. There was also missing data in 8 entries. Eight of the 62 erroneous entries (13%) were documented during a one-week period during which a locum staff member, unfamiliar with data entry, was appointed. Discussion/Conclusion: Revalidation will be mandatory and partially based upon audit data, yet few units have mechanisms to verify data collection and entry. In our speciality, every tenth entry has been shown to have inaccurate data. Such an error rate is unacceptably high. However, on closer examination and taking into account available fields for data entry on the EMAS form (100), the true accuracy rate may be computed to be hundred-fold better. We highlight, therefore, the fallibility of audit and its potential for misinterpretation, especially when being extrapolated to judge performance-related issues. We also highlight the role of adequate supervision and increased vigilance of staff unaccustomed to data collection and entry.

Consent to surgery in colorectal cancer patients: a prospective audit 
P.S. KANG, X. ESCOFET, A.Z. KHAN, A.P. WETHERALL and K.E. WHEATLEY Sandwell District General Hospital and Kidderminster General Hospital, England, U.K.

Introduction: Patients have the right to information about their condition, the treatment options, complications of the proposed treatment and the competent adult patient has a fundamental right to give, or withhold, consent to surgery. We set out to assess how well patients were being consented for elective and urgent colorectal cancer surgery according to the national guidelines. Patients and Methods: A questionnaire was distributed postoperatively to a total of 40 patients who had undergone surgery for colorectal cancer in two district general hospitals in the West Midlands from May 1999 to January 2000. The questionnaire was designed to assess the type of information given to the patients and test the patients understanding of this information. Data was collected prospectively and measured against nationally accepted guidelines. Results: Thirty-seven patients (92%) had been warned about the possibility of stoma formation. However, only 4 patients (10%) felt that they were informed about late complications and only one patient (3%) was aware of the risk of recurrence. Adequate information about early complications including anastomotic leak, infection and haemorrhage was recalled by 19%, 18% and 15% of patients, respectively. Conclusion: Efforts were made to follow national guidelines. Deficiencies in patients consent were identified in our study, particularly regarding late complications and recurrence of disease. Partly this may be due to patients being unable to recall information given to them postoperatively. Written information distributed amongst patients preoperatively can help address this problem.

An audit of trainees performing clinical audit
A.Z. KHAN, E.D. BABU and A. KASHABA Torbay Hospital, Torquay, England, U.K.

Introduction: Audit is an important tool for comparing one’s practice against set standards and is now an integral part of the health service. It has been stressed that basic and higher trainees need to be aware of the importance of and participate in clinical audit programmes. Aims: To assess the understanding and involvement of junior doctors in clinical audit. Methods: A questionnaire based survey of 167 trainees in six teaching hospitals and three district general hospitals across three deaneries in England. Results: A total of 167 trainees were approached. Twenty-one incomplete questionnaires were excluded. There were 62 specialist registrars (SpRs) and 85 senior house officers (SHOs). Eighty-two percent of respondents had participated in an audit at some point in their careers but only 42% were comfortable with their understanding of the principles of audit. Fifty-two percent of SpRs and 38% of SHOs had received formal training in the principles of audit; however, only 77% of SpRs and 59% of SHOs knew about the audit cycle. Ninety-four percent of SpRs and 82% of SHOs had presented their audit. Sixty-one percent of SpRs and 38% of SHOs found it difficult to obtain data. Conclusion: Junior doctors appear to be interested in performing audit. Further improvement can be made by imparting formal training to junior doctors about the principles of audit, providing them with easier access to data and ensuring consultant supervision. Also, all audit programmes should be supported by the hospital’s information technology department, as well as department of clinical audit, and all doctors involved in audit should be encouraged to present their results.

An audit of colorectal surgery performed by a breast surgeon
S.A. McINTOSH, R.J. MILLER, Y. ABU-OMAR and S.D. HEYS Aberdeen Royal Infirmary, Foresterhill, Aberdeen, Scotland, U.K.

Aims: It has been suggested that outcomes of surgery for malignant disease may be affected by subspecialisation in general surgery, although in the UK, many “general surgeons with a subspecialty inter-est” continue to perform general surgery outwith their subspecialty. This study aimed to audit the outcomes of patients undergoing surgery for colorectal cancer in the hands of a general surgeon with an interest in breast surgery. Methods: A retrospective review of case notes of all new cases of colorectal cancer presenting to a single surgeon between 1992 and 1999 was conducted. Data was recorded on presentation, tumour site, operation details, adjuvant therapy, complications, and recurrence. Principal outcome measures were operative morbidity, mortality, disease-free and overall survival and local recurrence rates. Results: A total of 65 new patients with a diagnosis of colorectal carcinoma were seen, of whom 58 underwent resection - 42 potentially curative procedures (72%) and 16 palliative procedures (27%). Operative morbidity was 28% and 30-day mortality 3.5% (10% in the emergency group and 0% in elective cases). The 5-year survival was 54%, with a median survival of 37 months in patients having potentially curative surgery. Overall local recurrence rates in those patients undergoing potentially curative resection were 9.5%: 37.5% in those patients undergoing rectal surgery and 3% in patients with tumours above the recto-sigmoid junction. Conclusions: Operative mortality and morbidity data and overall survival are comparable with results published in the literature, suggesting that colonic resection for malignancy may be safely undertaken by the adequately trained general surgeon. However, local recurrence rates for patients with rectal tumours are higher than is desirable. This may be due to surgeon factors, although possible confounding factors such as high percentage of emergency patients and use of adjuvant therapy must be considered.

Consultant surgeon supervision of basic surgical trainees
A.M. PAISLEY, P.J. BALDWIN and S. PATERSON-BROWN Royal Infirmary of Edinburgh, Scotland, U.K.

Background: In previous published work we have defined 70 skills that consultant surgeons consider to be essential or important in basic surgical trainees. These skills fall into five domains: patients and relatives, application of knowledge, teamwork, clinical skills and technical skills. If consultant trainers are to provide accurate assessment of trainee progress, they must directly observe the trainees performing such tasks. However, consultants in hard-pressed surgical specialities have many demands on their time and close supervision of their junior trainees in all settings may not always be possible. Aim: To determine whether consultant surgeons directly observe their basic surgical trainees performing the 70 skills, previously defined as necessary for a successful career in surgery. Method: All supervising consultants, drawn from all major surgical sub-specialities, involved in the South East Scotland basic surgical training scheme were surveyed over an 18 month period. At the end of each trainee attachment, an assessment form, asking for evaluation of the 70 skills, was sent by post to each trainee’s supervising consultant. For each task, the consultants were asked to state whether they had or had not directly observed the trainee perform the task in question. Observation levels for the five skill areas were determined. Results: 110 of the 117 forms distributed were returned (response rate 94%). The median percentage of tasks observed in each domain was 89% for patients and relatives, 78% for application of knowledge, 82% for teamwork, 91% for clinical skills and 95% for technical skills. However, some of the consultants surveyed observed their trainees in few areas. In the case of the 9 patients and relatives tasks, three consultants observed zero tasks and 25 observed less than 75%. In the case of the 19 technical skills, three consultants observed zero tasks and 14 observed less than 75%. In the remaining domains of application of knowledge, teamwork and clinical skills, 41, 41 and 30 of the consultants observed under 75%, respectively. Furthermore, low levels of observation were found in some tasks, with 14 of the 70 tasks being observed in under 80% of cases. For example, the teamwork task of ‘able to offer constructive criticism to others’ was observed by only 56% of the consultants, the patients and relatives tasks of ‘able to diffuse anger and hostility’ was observed by 53% and the application of knowledge tasks of ‘can initiate research’ and ‘can complete research’ were observed by 39 and 40% of consultants, respectively. Conclusion: Although overall levels of consultant observation of trainees appeared satisfactory, we have identified areas of concern. Some consultants observe very few of the important tasks and some of the task are observed by few of the consultants. These areas must be addressed if accurate and appropriate assessment of trainees’ competence is to be achieved.

Critical incident reporting: does it make a difference?
E. PATERSON, T.J. VARGHESE and G.G. YOUNGSON Royal Aberdeen Children’s Hospital, Aberdeen, Scotland, U.K.

Introduction: Clinical governance is emerging as a major issue in modern clinical practice. In our multidisciplinary paediatric surgical unit there has been a simple method of problem reporting and solving that has been ongoing from August 1993. This method involves the ‘Incident Book’, a jotter that any member of staff can report problems or incidents for discussion at the weekly departmental meeting. Aims: To categorise and code all entries in the incident book over the seven year period (August 1993 - August 2000) and to analyse the various types of data (communication problems, errors, domestic problems, services, policies/protocols, training and backup), including frequency and timing of incidents as well as trends. Materials and Methods: junior doctors rotate in a 6-month period. There were 14 six-month blocks within the 7-year period. Entries were analysed according to incident category, type and trends examined. These were collated into a single six-month period for analysis. Results: Communication problems attracted 33 entries in the first two months and four in the last two months. There were seven site marking errors in the first month and none in months five and six. Reporting of incidents decreased by 94.8% over the collated six-month period. Fifty-eight incidents were reported in the first month and only three reported in the last. This trend was observed over the majority of the incident categories especially those pertaining to the junior doctors. Communications and resident errors were the most frequent and the most important, respectively. Conclusion: Reduction in the reporting of incidents over the collated six month period is likely to be multi-factorial and may include a decrease in reporting over the 6 month block (reporting bias) or changes that decrease critical incidents and finally the junior doctors’ learning curves.

SIGN guidelines on tonsillectomy: audit of clinical record keeping and adherence to guidelines
A.T. WILLIAMS and A.I.G. KERR Royal Hospital for Sick Children, Edinburgh, Scotland, U.K.

Background: SIGN (Scottish Intercollegiate Guidelines Network) guidelines on the indications for tonsillectomy were published in January 1999. The relevant information must be recorded in clinical records for audit of adherence to the guidelines to be possible. This audit assesses the record keeping of number of attacks of tonsillitis in the previous 12 months in patients admitted for tonsillectomy to a specialist children’s hospital. The results revealed poor record keeping in a first cohort such that audit of adherence to SIGN guidelines was inadequate. A second audit cycle following a change in awareness, showed greatly improved record keeping this facilitating future audit of clinical practice. Adherence to the SIGN guidelines is discussed. Objective: To assess the adequacy of clinical record keeping in patients admitted for tonsillectomy and adherence to SIGN guidelines. No standard currently exists for recording of attacks of tonsillitis in clinical records or adherence to SIGN guidelines but a high figure of >95% should be the target. Evaluation: The clinical records of 60 consecutive admissions for tonsillectomy to a specialist children’s hospital were assessed for recording of number of attacks of tonsillitis in the preceding year. On the basis of this adherence to the SIGN Guidelines, the process was evaluated.

Comparison

Comparison

RECORDING SIGN ADHERENCE GP letter 50% Adherence 85% Clinic notes/letter 74% Average attacks/yr 5.4

Conclusions: This audit illustrates the fact that clinical record keeping is often inadequate for audit of practice, as was the case in the first cycle. Feedback of this resulted in a significant improvement in the second cycle. Adherence to SIGN Guidelines improved also. A third cycle is in process.

RECORDING SIGN ADHERENCE GP letter 25% Adherence 75% Clinic notes/letter 30% Average attacks/yr 5.7

Change in Practice: The results were presented to clinicians in the unit and more rigorous record keeping encouraged. The audit was repeated after an interval of three months, and 61 consecutive patients were evaluated.

A retrospective audit of biliary stenting for malignant biliary structures: the impact of type of stent on patency rate
S. ALISHAHI, M. LAVELLE-JONES and A. CUSCHIERI Ninewells Hospital and Medical School, Dundee, Scotland, U.K.

Aim: To compare the patency rate of a pigtail stent (7Fr) with a straight stent (7Fr) for malignant biliary strictures. Material and Method: Data on 595 patients, who had undergone 714 (ERCPs) over a four -year period at one institute, were collected prospectively. Subsequently, case notes of a subgroup of patients who were stented for malignant biliary strictures, were studied retrospectively. Any lapse of time between successive stent insertions or until time of death or study for single stent insertion was accepted as duration of patency. Two types of stents of the same caliber were compared. Results are shown in median and interquartile range. Results: Seventeen percent (100/595) of patients had malignant biliary strictures; 3.4%(20/595) had advanced malignancy and were not stented. Records of 58 patients were obtained. The overall survival was 114(61,237) days with 22%(13/58) alive and 88%(45/58) dead, at time of examination. Survival rate was not different with respect to the site of stricture (upper versus middle or lower common bile duct). (p >0.05) The overall patency rate was 63(21,91) days. The patency rate for the pigtail stent (7Fr) was 134(76,541) which was over twice the straight stent (7Fr) patency of 59(24,106). (p = 0.003)(Wilcoxon Signed Rank Test). Conclusion: Patency rate of pigtail,7Fr stents are superior to straight,7Fr stents for malignant biliary stricture. Use of pigtail stents should preclude the need for multiple stent insertions.

Open versus laparoscopic pre-peritonial mesh repair for bilateral and recurrent inguinal hernias H.D.E. ATKINSON, J.D. TERRACE, C.F. TAYLOR and S. PATERSON-BROWN Edinburgh Royal Infirmary, Scotland, U.K.

Background: There remains controversy as to the exact place of laparoscopic surgery in the repair of inguinal hernias, although patients with bilateral and recurrent hernias are most likely to benefit. This study has reviewed the outcome following laparoscopic repair in patients with recurrent and bilateral inguinal hernias and compared this to open repair. Methods: Details of all patients undergoing laparascopic pre-peritoneal mesh repair were collected prospectively and subsequent follow-up obtained by postal questionnaire. Patients under-going open repair were identified retrospectively using the hospital information system (ICD9 and OPCS4 coding). Results: Between January 1994 and January 1999, 81 patients, with 126 hernias, under-went laparascopic repair. During the same period, 142 patients underwent open repair for recurrent inguinal hernias and 112 for bilateral hernias. Post-operative stay was significantly shorter following laparoscopic repair of bilateral hernias, compared with the open approach (Mean 1.3 days (range 0-12 days) v 1.7 days (range 0-9 days), p<0.03, Mann Whitney U test). Conclusions: Laparoscopic pre-peritoneal mesh repair results in a shorter hospital stay than the open procedure in patients with bilateral inguinal hernias. Other benefits may be apparent in those patients with recurrent hernias.

Changing trends in the surgical management of inguinal hernias 1977-2000
H.D.E. ATKINSON, S. HOWARD, S. PATERSON-BROWN Edinburgh Royal Infirmary, Scotland, U.K.

Background: For over a century there has remained controversy as to the best way of managing inguinal hernias. There have been many changes in the techniques used, each with varying rates of success, complications and recurrence rates but none, until now, has been adopted so overwhelmingly as the use of the prosthetic mesh in the tension-free repair. Methods: Patients undergoing open inguinal hernia repair between 1977 and 2000 in Edinburgh Royal Infirmary, were identified retrospectively using the hospital information system (OPCS4) and ISD (Scotland) coding. The operative procedures, were then categorised into primary and recurrent repairs and also into whether techniques used a suture repair or a prosthetic mesh repair. Results: Between January 1977 and January 2000, 6212 patients underwent open primary inguinal hernia repair; 4950 of these were suture repairs, while 1262 were prosthetic mesh repairs. Prior to 1994, the vast majority of techniques employed variations in suture repair. From 1994, there was an increase in the use of prosthetic mesh in primary repairs, from 6.2%, to around 44.5% of procedures in 1995. This trend continued and around 94.8% of primary inguinal hernia repairs incorporated the use of prosthetic mesh in 1999. Over the same period, 509 patients underwent open recurrent inguinal hernia repair; 331 of these were suture repairs, while 178 used prosthetic mesh. In 1994, less than 5% of these procedures used prosthetic mesh, rising to around 82% of procedures in 1995. In 1999, 97% of recurrent inguinal hernia repairs made use of prosthetic mesh. Interestingly, the total number of repairs per year increased over the 23-year period. This was not shown to be statistically significant. The rates of recurrent operations as a proportion of primary repairs ranged between 4 and 15%. This did not significantly increased over this period. Only one patient (0.07%) required that his mesh be removed, due a tracking wound infection. Conclusions: The use of prosethetic mesh has simplified the procedure of open inguinal hernia repair, and successfully reinforces the tissues with no tension. Even though there has been a dramatic uptake in this new technique, its long-term efficacy is yet to be determined. In addition, it adds the risk of introducing foreign material into the tissues, which may have adverse consequences. Even though these fears are yet to be confirmed, critics warn of these potential dangers, and there are many surgeons who are still very conscious about the use of prosthetic mesh in younger patients.

Reduction of blood transfusion rates in the unilateral total knee arthroplasty by the introduction of a simple blood transfusion protocol
A. BALLANTYNE and I. BRENKEL Queen Margaret Hospital, Dunfermline, Scotland, U.K.

Introduction and Objective: Unilateral total knee arthroplasty can result in substantial blood loss necessitating post-operative blood transfusion. The routine blood cross matching of patients prior to arthroplasty is commonplace in many orthopaedic departments. This has been shown to result in significant levels of non-utilisation of cross-matched units. In addition, the decision to transfuse a patient post operatively is often based on poorly defined criteria and may, in some cases, result in inappropriate transfusion. When this is considered within the context of the potential health risk associated with the transmission of blood borne infective agents during allogenic blood transfusion, it seems clear that the introduction of a simple transfusion protocol may reduce exposure to such risks. We undertook, therefore, a prospective audit before and after the introduction of a transfusion protocol to determine the effects of such a policy on blood transfusion practices in patients undergoing unilateral total knee arthroplasty (TKA) within a single institution. Methods: A prospective audit of 393 consecutive TKA (group I) was undertaken prior to the introduction of the transfusion protocol. After the establishment of the protocol, a further 296 consecutive TKA (group II) were audited. Both groups were comparable for age, basal metabolic index, preoperative knee score, pre-operative haemoglobin, post-opera-tive haemoglobin drop and tourniquet time. The same surgeons using the same technique and instrumentation carried out all procedures. The blood transfusion practices in each group were compared. Results: The introduction of the transfusion protocol reduced overall transfusion rates from 31% in group I to 12% in group II. The length of stay was reduced from 11.4 days to 9.3 days. There were no adverse effects. Conclusion: The introduction of a simple transfusion protocol can reduce the overall transfusion in patients undergoing unilateral total knee arthroplasty. There was also a substantial reduction in the non-utilisation of cross-matched blood with resulting cost implications.

An audit of patient information leaflets within orthopaedics in Tayside
A.J. BROOKSBANK, A. HAWKINS and M.J. McNICHOLAS Tayside University Hospitals, Scotland, U.K.

Aims: We recognise the importance of our patients being well informed about their healthcare, allowing them to make informed decisions. The benefits of written patient information are well established, however, the standard and presentation of this information is questionable. This was recognised within our region, leading to a document: ‘ Policy for writing and reviewing information for patients’. Our aim is to compare our patient information leaflets against the standard, and close the audit loop by looking at leaflets produced following the implementation of this document. Methods: We collected leaflets relating to orthopaedic patients throughout Tayside, from Ninewells hospital, Perth Royal Infirmary and Stracathro hospital. These were scored on the following subheadings: content (5); presentation (5); style (5); production quality (4); format (1) and illustrations, if they were present. A fleisch readability score was included, a score greater than 60 being taken as adequate for the general public to understand the document, in accordance with Microsoft Word. Results: We analysed 77 leaflets in total, 8 from Ninewells, 28 from Stracathro and 41 from Perth hospitals. Members of staff produced all the leaflets. The average total score of all documents was 48.6%, incorporating a mean fleisch score of 65.9, mean production quality 30.8%, style 38.3%, presentation 63.1% and content 49.0%. Leaflets produced since the introduction of this document had an average total score of 85%. Discussion: The overall standard of patient information is well below that recommended, but those leaflets produced since this time are much improved, hence showing that the audit loop is starting to be closed. The demand for healthcare information is on the increase, unless we take the initiative to supply this, patients will seek alternative sources as found in the media and on the Internet, the accuracy and relevance of which is unproven.

A systematic review of chronic pain after hernia repair surgery
A.S. POOBALAN, J. BRUCE, P.M. KING, Z.H. KRUKOWKSI, W.C.S. SMITH and W.A. CHAMBERS University of Aberdeen, Foresterhill, Aberdeen, Scotland, U.K.

Background and Aim: Until the last decade, chronic pain after inguinal hernia repair was reported as a rare and infrequent postsurgical complication. More recent studies have suggested that up to 30% of patients report persistent pain at one year after hernia surgery. The aim of this study was to conduct a comprehensive review of the reporting and prevalence of chronic pain following inguinal hernia repair. Methods: A systematic search of the English literature published between 1987 to 2000 was undertaken on five bibliographic databases (Medline, Embase, Cinahl, HealthSTAR & Cochrane Library). Case definition of chronic pain was pain lasting for more than three months, beyond expected normal healing time as per IASP. Results: A total of 527 abstracts were identified, 101 studies critically appraised 40 of which fulfilled inclusion criteria. Of 17 RCTs comparing laparoscopic and open repair, 8 (47%) reported less chronic pain and four (24%) reported more after laparoscopic surgery; 5 (29%) reported no difference. Of the three RCTs comparing open (mesh v. non-mesh) repair, two reported less chronic pain after mesh repair. Although included studies followed-up patients for longer than three months and reported overall morbidity, few defined chronic pain or clearly specified the timing of measurement of pain in relation to period of follow-up. Postoperative pain is one of multiple outcomes measured after inguinal herniorrhaphy and estimation of true prevalence of chronic pain can prove difficult. Of seven studies specific to the measurement of chronic pain, frequency ranged from 0 to 53% up to two years after surgery. Risk factors identified from the literature included: recurrent surgery on the same side, preoperative pain, high pain scores in the first postoperative week and day case surgery. Conclusion: Recent studies that specifically addressed chronic pain using a clear definition have reported a much higher frequency than previously reported. This disparity may be due to the definition used and the quality of reporting. Poor quality of reporting can limit estimation of true prevalence of chronic pain. Future research should consider the risk factors for the development of chronic pain identified in this review.

An audit of perforated peptic ulcer disease in West Lothian
M.R. BROADBENT, H.D.E. ATKINSON and B. WALDRON St John’s Hospital, Livingston, Scotland, U.K.

Background and Aim: The epidemiology of perforated peptic ulcer perforations has changed throughout time. The aim of this study was to see if the management, and outcome had changed accordingly. Methods: A retrospective study was performed on patients admitted to St. John’s Hospital, between the 1st January 1994 and the 31st December 1999. Results: One hundred patients, 58 male (median 48 years), 42 female (median 70.5) were studied overall age ranged from 17 to 90 years (median 61.5 years). Overall there were 19 mortalities: 9 were female and 10 male (median age 71 in each group); 7 were directly related to reperforations, 7 directly related to effects of original perforation, 5 directly related to medical complications. Mortality risk according to time between perforation and surgery: 4, < 6 hrs; 1 from 6 - 12 hrs; 2 from 12 - 24 hrs; 9 ³ 24 hrs. Operation risk: 5 partial gastrectomy; 11 simple closure. Overall risk factors: NSAIDS/ steroid use - 17 male, 18 female. The site of the ulcers was: 80 were duodenal; 11 pyloric/ prepyloric; 9 gastric. Diagnosis was achieved by chest radiograph (45) radiographic contrast swallow (15), at time of laparotomy (30). Eighty patients had simple closure at operation. Complications occurred in 16 males and 27 females. Forty-seven patients were given triple therapy; 9 tested positive for H.Pylori, 5 were negative and 86 were not tested. Discussion and Conclusions: From the data it is possible to see that the proportion of women is increasing and is almost equal in numbers to the male population. It is seen, however, that the median age of the two groups are quite different, with the females being older. Proportionally, the mortality and morbidity remains high, but it is demonstrated that the age and proportion of the mortalities in both populations are equal, while the number of morbidities is almost twice as common in females.

CSF shunt infections: an audit and reappraisal of contributing factors
Y.C. GAN, R.A. JACKSON, I. KHAN and M.S. CHOKSEY Walsgrave Hospital, Coventry, England, U.K.

Introduction: The incidence of CSF shunt infection ranges from 2-20% in most studies. Many factors have been implicated in its causation such as the duration of the operation (the shorter the better), the time of the day the operation is performed, the neurosurgeon’s experience, the age of the patients (especially <6 months) and the number of revisions. Objective: To review the incidence of shunt infections in our practice and identify the contributing factors. Patients and Methods: Retrospective audit, from 1994-1999, was carried out. Patients who had shunt insertions during the 6-year period were stud-ied. Outcome measures included the duration of operation, rank of surgeon, age of patient, first-time shunt or a revision, and incidence of shunt infections. Results: One hundred and fifty-one shunts (134 ventriculo-peritonial shunts, 12 lumbo-peritonial shunts, four ventriculo-pleural shunts and one ventriculo-atrial shunt) were inserted in 121 patients (65 females, 56 males). Ages ranged from one month to 83 years of age; 10 patients were under six months old. Ninety-seven (83 VP, 10 LP, three Vpleural, one V Atrial) were first-time shunts and 54 (51 VP, two LP, one Vpleural) were revisions. All were given pre-, intra- and post-operative antibiotics. Duration of operation ranged from 30-235 minutes (mean 107 minutes). Sixty-nine procedures were performed by the consultant and 82 by trainess. There were three infections (two males, one female) - all adults and in first-time shunts. They were - a shunt infection caused by staphylococcus epidermidis (duration - 90 minutes); the other two were ascending infections caused by peritonitis due to ruptured appendix and perforated colon, respectively. Discussion and Conclusion: There was only one true case of shunt infection, giving an incidence of shunt infection of 0.7%. None of the known contributing factors seemed to play a role in our series. In our experience, observation of strict aseptic technique and careful use of prophylactic antibiotics - rather than the duration of operation, number of revisions and experience of the operator - seems to be the main factor in preventing shunt infections.

Should surgeons swab during appendicectomy? An audit
J. HENDERSON, M.J. FORSHAW, D.M. POWER and E. BABU DINKARA Addenbrooke’s Hospital, Cambridge, England, U.K.

Introduction: It is common practice to send microbiological swabs during appendicectomy but little evidence exists for their usefulness. We audited the current role of microbiological swabs and their effect on patient outcome. Methods: All patients undergoing emergency open and laparoscopic appendicectomy during a six month period in a teaching hospital were retrospectively identified from computerised theatre data. Their case notes and laboratory results were then reviewed. Patients whose case notes could not be retrieved after threeattempts were excluded (n=12). Statistical analysis was performed using a Mann Whitney U test. Results: Twenty-three laparoscopic and 88 open appendicectomies were included in this study. A variety of antibiotic prophylaxis regimens were used. These 2 groups had similar operative and histological findings. Only 34 microbiological swabs were sent, all within the open appendicectomy group (39%); 24 were sent by senior house officers operating alone and the commonest indication was pus within the abdomen (n=13). Only 6 swabs (18%) were positive, growing a mixed large bowel flora, in each case covered by the antibiotic prophylaxis given. The results of the swabs were recorded in the patient’s notes in only five cases (three positive/two negative), all in instances of persistent postoperative pyrexia, but no change in antibiotic therapy was instituted as a result. Three patients developed postoperative collections requiring intervention; swabs sent at this stage showed no growth and no antibiotics were changed. A longer hospital stay was associated with swabbing (p=0.05), positive swabs (p=0.05) and recording the results in patient’s notes (p=0.01). Conclusions: Fewer than expected microbiological swabs were performed, usually by junior surgeons. Their results had little impact on patient management or outcome, even in the presence of complications. We suggest that there is no role for routinely sending microbiological swabs during appendicectomy.

An audit of HIV post exposure prophylaxis awareness amongst medical staff in a district general hospital
AZ KHAN*, KM DUNCAN*, X ESCOFET#, N DEBNEY#, WFA MILES*
*Worthing Hospital and # Torbay Hospital, England, U.K.

Background: All doctors practising hospital medicine should be aware of the department of health guidelines regarding post exposure prophylaxis (PEP) following exposure to actual or strongly suspected HIV material. Objectives: To measure awareness of guidelines governing PEP and assess the reaction of medical staff in case of occupational exposure to HIV. Patients and Methods: The audit was carried out in a District General Hospital in the South West of England. A questionnaire was circulated to all medical staff employed by the hospital and used to collate data regarding awareness of guidelines following exposure to HIV. Questions were posed regarding the actual risk of contracting HIV following a high risk exposure, the time interval between exposure and starting therapy, the currently recommended drugs for PEP and the time interval following cessation of PEP and conclusive antibody testing. Results: A total of 222 questionnaires were circulated of which 121 were returned making the overall response rate 55%. Forty-five percent of respondents knew that the risk of contracting HIV following high risk exposure is 0.03 % and only 22% correctly answered that it was imperative to start PEP therapy within one hour of the exposure. Twenty two percent of doctors were aware of all three drugs that are currently recommended for prophylaxis while 49 % got the correct answer for the time interval between cessation of PEP therapy and conclusive antibody testing. Conclusion: Awareness about HIV prophylaxis was low amongst the participants of our survey, particularly amongst house officers. Information regarding PEP prophylaxis should be included in the formal induction programme for newly recruited house officers as well as circulated to existing NHS employees.

Improving transfusion practice in a Scottish District General Hospital K. CASSAR, J.L. DUNCAN and T. FERGUSON Raigmore Hospital, Inverness, Scotland, U.K.

Background: There are widespread concerns regarding the ability of donor services to meet the increasing demand for red cells and the need to improve the management of this scarce resource. One method for improving the efficiency of red cell transfusion is the use of a Maximum Surgical Blood Order Schedule (MSBOS). Patients and Methods: A prospective audit was performed at Raigmore Hospital, Inverness between 1995 and 1999 to determine the effect of introducing the MSBOS on the pattern of packed cell usage. The total number of packed cells requested and used was monitored and the results made available to individual units within the hospital. The percentage red cell use and the ratio of crossmatched to transfused units (CTR) was determined. Results and Conclusions: The number of crossmatched units fell from an average of 1067 per month in 1995 to 675 in 1999, although the actual number of units transfused increased from 457 in 1995 to 523 per month in 1999. The percentage use of blood increased from 42.9% in 1995 to 77.5% in 1999, and CTR fell from 2.33 to 1.29. This shows that the introduction of MSBOS can lead to more efficient and responsible use of red cells.

(This poster was subsequently chosen as the best poster of those exhibited and a prize was awarded to Mr Kevin Cassar)

The management of congenital talipes equinovarus in Scotland: a nation-wide audit
D. CHESNEY, S. BARKER, N. MAFFULLI and R. PORTER University of Aberdeen, Foresterhill, Aberdeen, Scotland, U.K.

Background: The Scottish Talipes Register was set up in 1993, and collected information on all newly diagnosed cases of congenital talipes equinovarus (CTEV) over a four-year period. Information regarding initial management and all subsequent management interventions including surgery was recorded. All 18 hospitals involved in the management of CTEV across Scotland took part and, in total, 216 children were entered into the register. Patients and Results: All the children were initially managed conservatively. The method of conservative management varied across Scotland, but each surgeon managed all the children in his or her care in the same fashion. Thirty-six percent of children were initially managed with adhesive strapping, 34% were initially managed with plaster of Paris casts, and 29% were managed with Denis-Browne splints and boots. Surgery was required in 53% of cases. Of those initially managed with adhesive strapping, 69% required surgery, of those managed with plaster casts, 42% required surgery and of those managed in splints and Denis Browne boots, only 14% have required surgery, to date. Discussion and Conclusion: Despite a lot of research into the management of CTEV, the best method remains unclear. Across Scotland there are clearly differences in the use of surgery to correct the deformity, and these differences may reflect variable success with different methods of conservative management. A prolonged period of conservative management appears to reduce the need for surgery. A multicentre trial is required to further assess the different methods of conservative management, and their effect on the natural history of CTEV.

Survival in patients undergoing surgery for chronic critical limb ischaemia
K. KRISHNA, S. LAW, C. BEAMAN, J.N. DAVIES, K.R. WOOD-BURN Royal Cornwall Hospital, Truro, England, U.K.

Aims: To examine the long-term survival and cause of death in all patients undergoing surgery for chronic critical limb ischaemia in Cornwall between January 1998- December 1999. Methods: All patients undergoing surgery for chronic critical limb ischaemia were identified from the prospective vascular database and these data were then cross-referenced with the Office of National Statistics to obtain survival status and certified cause of death. Results: Seventy-one revascularisation procedures and 51 amputations were undertaken for chronic critical limb ischaemia in 112 patients. Thirty-day mortality following amputation (50patients) was 14%, compared with 7% following revascularisation (66 patients). Cumulative survival was similar following both amputation (40% 3-year survival), and revascularisation (48% 3-year survival). There was no significant difference in the cause of death between the two groups, the majority dying of cardiovascular causes. All analysis were undertaken on an intention to treat basis. Conclusions: Revascularisation surgery for chronic critical limb ischaemia does not confer any long-term survival benefit, when compared with major limb amputation. Long term survival in patients with chronic critical limb ischaemia is comparable with that of patients with malignant disease.

Chronic pain and discomfort after prosthetic mesh repair of groin hernia: laparoscopic versus open repair 
S. KUMAR, S.J. NIXON, R.G. WILSON and I.M.C. MACINTYRE Western General Hospital, Edinburgh, Scotland, U.K.

Introduction: The aim of this study was to compare the incidence of chronic groin pain or discomfort after laparoscopic or open mesh repair of groin hernia. Patients and Methods: A questionnaire was sent to 441 patients, of whom 403 responded; mean age 64 years (range 20 -94 years). Two hundred and eight patients had laparoscopic total extra-peritoneal repair, and 195 had open mesh repair between Jan 1998 and Dec 1999. The questionnaire was sent between 6 - 24 months after the operation. Results: The two groups of patients were comparable in age and sex distribution. Chronic pain was reported by 33/403 (8%) of patients. 9/208 (4.3%) patients reported pain after laparoscopic repair and 24/195 (12.3%) after open mesh repair of groin hernia (p < .01 ). Chronic pain was reported by 81/403 (20 %) patients, 34/208 (16.3%) patients after laparoscopic repair and 47/195 (24.1%) patients after open mesh repair (p > .05). Conclusion: Chronic discomfort was reported by 20% and chronic groin pain by 8% of the patients after groin hernia repair. Chronic groin pain was significantly more common after open than laparoscopic repair.

Surgical management of upper limb ischaemia: a 12 year experience
A. MAHMOOD, R. SALAMAN, M. SINTLER, J. LUSCOMBE, A. EDWARD, S.R.G. SMITH, M.H. SIMMS and R.K. VOHRA University Hospital Birmingham NHS Trust, Selly Oak, England, U.K.

Aims: The aims were to assess the diversity and aetiology as well as to evaluate the management, outcome and morbidity of upper limb ischaemia. Methods: A retrospective analysis of all patients admitted to the University Hospital Birmingham NHS Trust with upper limb ischaemia, between 1988 and 2000, was carried out. A standard proforma was used to record presentation, aetiology/pathology, treatment and outcome along with routine clinical and demographic data. Results: A total of 86 patients with a mean age of 64 years (range 2-92) presented with upper-limb ischaemia. Sixty-two presented with acute ischaemia and 18 had chronic ischaemia and 6 resulted from trauma to the upper limb. The aetiology was cardiac/ degenerative in 46%, iatrogenic in 14%, inflammatory in 8% and thoracic outlet syndrome in 5%. The surgical procedures carried out were embolectomies in 45 patients, bypass reconstruction to the great vessels in 13 and distal upper limb reconstruction in seven. Five patients had thrombolysis of which three failed. There were four graft occlusions two of which required further surgery. In the embolectomy group, there were four early and one late thrombosis who required surgical intervention. There were three limb losses; two patients had above elbow amputation and one hand amputation and five re-occlusions for the embolectomies at four year follow-up. Two patients required above elbow amputations and one required a hand amputation. The limb salvage rate was 96.6% and the 30-day mortality for the series was 7%. Only one patient had ischaemic symptoms at follow-up. Conclusion: Upper limb ischaemia results from a multitude of underlying pathologies. Referral to a vascular specialist centre, offering a full range of diagnostic and interventional procedures, is recommended. We recommend surgical intervention in these patients to preserve limb function and prevent sequelae.

Audit on bowel surveillance for inflammatory bowel disease patients who developed cancer
E. ONG and C. MAKIN Wirral Hospitals NHS Trust, Merseyside, England, U.K.

Background: Patients with inflammatory bowel disease are at increased risk of developing colorectal cancer. The usual recommendation for cancer surveillance is regular colonoscopy with multiple biopsies after 8 - 10 years of colitis. There was no specific bowel surveillance protocol in our hospital and this audit was to ascertain the current recommendation and implement any required changes. Methods: Review of the last 5 years’ pathology database at Wirral Hospitals NHS Trust, identified 11 colorectal cancer patients with a history of inflammatory bowel disease. The case notes were carefully reviewed. A review of the literature was performed to ascertain the current recommendation. Results: Only 4 cases had some form of bowel surveillance and there was no specific follow-up protocol for these patients in our hospital. On literature review, no randomised controlled trial has demonstrated the effectiveness of colonoscopic bowel surveillance for early detection of cancer. There was significant heterogeneity shown on the available studies. Studies showed that one cancer was detected for every 360 surveillance colonoscopies and, in 41% of cases, they were diagnosed at Dukes C or higher. Newer methods (eg molecular screening tools) for colorectal cancer screening in these patients are available but they are at an early stage of development. Discussion and Conclusion: No optimal strategy has been identified. For the time-being, bowel surveillance is recommended until newer and better methods are available. To reduce the delay in diagnosis of colorectal cancer, the main key is patients’ education and their understanding of the disease and its associated risk of colorectal cancer. It is also important for them to realise that the sensitivity of the current available surveillance is limited.

An audit on the blood transfusion requirements for revision hip arthroplasty
S. SHARMA, H. COOPER and J.P. IVORY Princess Margaret Hospital, Swindon, England, U.K.

Background: The hospital transfusion committee of Swindon and Marlborough NHS Trust had formulated a maximum surgical blood ordering schedule (MSBOS) which included the standard practice of six units of blood for revision hip replacement. An audit was under-taken by the orthopaedic department to identify current practice, to assess compliance with the standard and to ensure that the standard is adequate for patient safety. Patients and Results: All patients under-going revision hip replacement from April 1998 to March 1999 were included in the audit. The sample consisted of 73 patients, 66 of whom underwent single stage revision and 7 of whom underwent two-stage revision. Information regarding number of units requested, number of units transfused, preoperative haemoglobin (Hb), lowest postoperative Hb, number of additional units of blood requested within three days post-operatively and number of these additional units of blood transfused were collected retrospectively from patients casenotes. We had a 91% compliance with standard in the ordering of 6 units of blood for our single-stage revisions, 100% compliance in the first stage of a two stage revision and 85.7% compliance in the second stage of a two-stage revision hip arthroplasty. Based on preoperative Hb, blood usage was analysed. 92.3% of crossmatched units were used when preoperative Hb was less than 12 g/dl, 64.4% were used when Hb was between 12.1 - 13.0 g/dl, 54.3% were used when the Hb was between 13.1 - 14.0 g/dl, 38.9% were used when Hb was between 14.1 - 15 g/dl and 39.7% used with preoperative Hb of >15 g/dl in single-stage revisions. Similarly, for the first stage of a two-stage revision, 100% of cross-matched units were used when preoperative Hb was less than 12 g/dl, 55.6% of crossmatched blood used when Hb was between 13.1 - 15 g/dl and 37.5% of blood used when Hb was >15 g/dl. For all patients undergoing single-stage revision, the change in Hb from the preoperative value ranged from +1.0 to -7.5 g/dl after transfusion. Fourteen patients had a postoperative Hb of less than 9 g/dl, for whom additional units of blood were ordered and given to achieve a Hb of between 10.1 and 14.2 g/dl prior to discharge. In the single-stage revision, 80.3% of patients received less than six units, 12.2% received 6 units and 7.5% received more than six units. Conclusion: This audit suggests that in patients with a pre-operative Hb of 13 g/dl or more the crossmatch could be four units instead of six units for revisions.

The role of upper tract imaging in men with lower urinary tract symptoms A.K. BHUVANAGIRI, P. NAIK, B. NAIR, A.S. JAMES, R.J.S. WEBBER, R. HILL Hairmyres Hospital, East Kilbride, Scotland, U.K.

Introduction and Objectives: Routine upper tract imaging forms an integral component of many Prostate Assessment Clinics in the evaluation of patients with lower urinary tract symptoms. We aim to determine the usefulness of this investigation on a routine basis and to identify a subgroup of patients who may benefit from upper tract imaging. Material and Methods: We collected data on 250 men (range 38-90 years, median -66) who attended our prostate assessment clinic on the basis of lower urinary tract symptoms between October 1999 and October 2000. All of the patients underwent trans-abdominal ultrasonography to assess the upper tract, along with prostate volume and residual bladder volume. Serum urea and creatinine levels were collected simultaneously. Results: Imaging detected five patients (2.0%) with evidence of dilatation of the pelvi-cael system with no hydronephrosis. Elevated serum creatinine were found in two of the above five patients, however these did not differ significantly between the two groups (p=0.463). High residual volumes was a consistent feature in this sub-group (median 200 ml versus 30ml in those without dialatation) (p=0.013). Prostate volume measurements in this sub-group did not differ significantly from those without dilatation (p=0.379). Conclusions: Upper tract ultrasonography is not mandatory in patients with lower urinary tract symptoms in the Prostate Assessment Clinic unless the patient has clear evidence of high bladder residual volumes and, especially so, in patients with raised serum urea and creatinine levels.

Tympanoplasty: an audit of outcomes
O.J. HILMI, P. BOLTON, M. YANN LIM and D.A. NUNEZ Aberdeen Royal Infirmary, Foresterhill, Aberdeen, Scotland, U.K.

Aim: The aim of this audit was to determine outcome of tympanic membrane reconstructive surgery in Aberdeen Royal Hospitals and compare this with published UK audit data. Methods: Of patients who underwent tympanoplasty or myringoplasty, between July 1998 and July 1999, were reviewed. Details of the perforation site, presence of infection, grade of surgeon, technique for closure and outcomes were recorded on to a Microsoft access database. Any patients with concurrent mastoid disease were excluded. Results: A total of 104 cases, 83 adult and 21 paediatric, underwent treatment. An overall average of

17.6 cases per consultant; Surgeons-in-Training performed 10% of operations. Fifty-seven patients had follow-up data to 12 months; 25 patients were discharged prior to 12 months, all of which had intact tympanic membranes. The overall graft take rate at 3-6 months was 83%, falling to 67% for those followed up for 12 months. Discussion: Overall, these figures show a high number of operations per consultant and overall success rates that compare favourably with other retrospective published data. There is a need to increase the number of operations performed by the trainees, as part of the units training requirements, and a need to prospectively audit outcomes.

Audit of patient satisfaction with cleft lip palate services in a regional centre
S.L.A. JEFFERY and J.G. BOORMAN The Queen Victoria Hospital, East Grinstead, England, U.K.

Introduction: The management of a child born with a cleft lip and palate requires the services of many professionals over an extended period of time. We audited our patient population, to determine their views and obtain any suggestions for possible improvements. Patients and Methods: We sent a questionnaire to the parents of 478 children who are under our care, and received 341 replies. Questions were asked about means of transport to the clinic, time to get there, specialists at the clinic, satisfaction with the service, level of adequacy of information given and level of involvement in decision making. Results: Satisfaction levels were very high, with 96% of parents either satisfied or very satisfied with the level of care received. Eighty-nine percent of parents found it quite easy or very easy to talk about their concerns with the specialists. Eighty-two percent of parents felt either fairly or very involved in recent treatment planning decisions, but 7% felt no involvement at all. Thirty percent of parents would like to be more involved in treatment planning decisions. Discussion: Patient dissatisfaction and inefficient or inconvenient clinics have previously been identified as being important factors in non-compliance with medical advice and treatment. Our study showed a surprising number of parents, many with children with a cleft palate, who felt that speech therapists did not need to be present in the combined clinics. Disturbingly, one third of the parents of our patients thought that they had either not enough knowledge about clefts or none at all. Every maternity unit in our catch-ment area has recently been provided with information folders to be given to the parents of newborn babies with cleft lip or palate, which should hopefully improve their level of knowledge.

An audit of salivary gland surgery in a District General Hospital
D.P. MURRAY and S.W. DENHOLM Raigmore Hospital, Inverness, Scotland, U.K.

Six years experience of salivary gland operations in a District General Hospital is presented. From June 1994, all patients undergoing salivary gland surgery were entered into a prospective database. One hundred and eleven parotidectomies were performed; 87( 75 superficial, 12 total or “near total”) were for benign disease and 24 (12 total, 12 superficial) for malignant disease. There were 24 cases of temporary facial nerve weakness in the benign group (27.6%) and two in the malignant group (8.3%). In addition, two cases required deliberate sacrifice of a branch of the facial nerve to remove the tumour and these were the only cases of permanent facial nerve weakness. There were three cases of wound haematoma requiring exploration (2.7%). The average hospital stay for the benign and malignant groups was 4.6 and 12 days, respectively. Thirty-nine submandibular gland operations were performed, including five for malignant tumours. There was one case of temporary weakness of the marginal mandibular nerve. One patient developed a wound haematoma requiring exploration and a temporary tracheostomy. The average hospital stay was 3.9 days for benign and 4.8 for malignant cases, respectively.

An audit of epistaxis management
D.P. MURRAY and W. McKERROW Raigmore Hospital, Inverness, Scotland, U.K.

Background: Epistaxis is the most common emergency referral to an ENT department. In order to evaluate our standard of practice and highlight any areas that could be improved we performed a prospective audit of all patients referred to our ENT ward with epistaxis. Methods: A questionnaire (see copy) with details of patient demographics, risk factors, management and outcome was filled out by the attending ENT doctor. If the patient was admitted to the ward the sheet was updated on patient discharge. Results: Between September 2000 and January 2001 50 patients (aged 12 to 97 years) were referred to the ENT ward with epistaxis. Thirty-nine patients had an identifiable bleeding point cauterised with silver nitrate. Ten of these subsequently required nasal packing (failure rate of 26%). Six patients had no identifiable source of bleeding and were packed without cautery. Three patients were treated with Naseptin cream only. Thirty-one (62%) patients required admission. Thirty two percent of patients referred required nasal packing [52% of those admitted]. Packs remained in-situ for an average of 33 hours (12-72). Antibiotics were not used routinely. Eighty-four percent of patients admitted had a known risk factor(s) for epistaxis, compared with 58% of those not requiring admission. Of those admitted all had a full blood count, 20 had a coagulation screen, and 19 had a Group and Save done. Ten patients had a prolonged INR and all of those were taking Warfarin. The average hospital stay was 2 days (1-4 days). Two patients required referral to a specialist centre for control of their bleeding. Seventy-four percent of patients were discharged with a 2-week course of Naseptin cream. The readmission rate was 2%. Discussion: We have completed the first part of the audit cycle and identified a number of areas in our management that could be improved upon. We have now drawn up a protocol of epistaxis management for our department and are currently monitoring our change of practice.

Audit of management of acute urinary retention
M. NICOL, G. BANERJEE and N. COHEN Aberdeen Royal Infirmary, Foresterhill, Aberdeen, Scotland, U.K.

Background: A recent randomised study showed that it is advisable for patients presenting in acute retention of urine, secondary to benign prostatic enlargement, to be given an alpha-blocker prior to a trial removal of catheter (TROC). This increases the chances of a successful TROC and this, in turn, would lead to a reduction in the morbidity and inconvenience associated with prolonged uretheral catheterisation. Materials and Methods: We have conducted a retrospective audit of patients presenting with acute urinary retention to the Urology Department of Aberdeen Royal Infirmary between April 1999 and March 2000. One hundred and eight patients presented with retention and, of these, 77 were deemed not suitable to undergo TROC nor receive an alpha-blocker (due to previous prostatic surgery, deranged renal function, etc). Of the remaining 31 patients, 20 were given one or more doses of Tamsulosin, 400 micrograms, prior to a TROC. The final group of 11 patients did not receive an alpha-blocker, with no specific reason recorded in the case notes. Results: In addition to auditing our management, we were able to look at whether or not the giving of alpha-blockers prior to TROC, is as efficacious as has been claimed. Our results have shown that out of the group receiving alpha-blockers prior to TROC (20), 13 passed urine successfully, compared with one out of 11 in the group that did not receive an alpha-blocker (p<0.01). Discussion: We have been able to produce a departmental protocol for the management of acute retention of urine, secondary to benign prostatic enlargement, confident that our suggested management appears appropriate.

A retrospective audit of patients admitted as emergencies with suspected renal colic
  S.M. YONG, S.A. McNEILL, C. GOODMAN and D. SHEPHERD Ninewells Hospital and Medical School, Dundee, Scotland, U.K.

Background and Objectives: Unenhanced helical computed tomography (CT) is considered to be the most sensitive and specific test for the detection of ureteric calculi in patients with renal colic. The most frequent diagnosis is that of ureteric calculi of which the incidence is 50-55%, as diagnosed by CT. Computerised Tomography was not routinely used in our department in the investigation of patients with loin pain. A retrospective audit was carried (January to May 2000) to further define the demographics of the patients admitted and the eventual outcome of admission. Results and Discussion: A total of 111 consecutive patients were seen in this department within this time period. There were 68 male (61.3%) and 43 female (38.7%) admissions (p=0.0063). Sixty-nine patients (62.2%) had a urological cause for the renal colic. Fifteen patients (13.5%) had a non-urological cause for their pain. There was no diagnosis achieved in 27 patients (24.3%). The proportion of ureteric calculi diagnosed in this department was 37 patients (33.3%). This value falls below the expected proportion of patients with ureteric calculi, as diagnosed by CT in other centres. Follow-up was performed on the 27 patients discharged without any diagnosis; four patients reported the passage of calculi per urethra after discharge. The sensitivity and specificity of the plain radiograph (KUB) and intravenous urogram (IVU) used were evaluated and found to be lower than that of CT (as reported in the literature); IVU was found to be more expensive than CT. Conclusion: Investigation with KUB and IVU misdiagnosed ureteric calculi in patients with loin pain. CT should be considered for use as first line investigation in these patients, where available, and considering radiation exposure in individual cases.

Scottish urology cancer audit
L.E.F. MOFFAT The Scottish Urology Cancer Audit, c/o Aberdeen Royal Infirmary, Scotland, U.K.

The Scottish Urology Cancer Audit has been set up with a £360,000 grant from the Scottish Executive to audit the provision of urological cancer services throughout Scotland. Following a successful pilot programme in three units; Aberdeen, Edinburgh and Ayr, a nationwide audit will be starting on the 29th of January. This will provide a resource of unparalleled information throughout the world and has the opportunity to link both with observational studies and other research.

For details of the 2002 Audit Symposium, e-mail education@rcsed.ac.uk