Emergency general surgery and the implications for specialisation

E.J. Dawson, S. Paterson-Brown
Department of Surgery, Royal Infirmary, Little France, Edinburgh, EH16 4SA, UK

Correspondence to: S. Paterson-Brown, Department of Surgery, Royal Infirmary, Little France, Edinburgh, EH16 4SA, UK

Introduction Patients and methods Results

Discussion References

Keywords: Emergency general surgery, specialisation 
Surg J R Coll Surg Edinb Irel., 2 June 2004, 165-170

Background and Purpose: To examine the overall spectrum of emergency general surgical admissions and operations in Edinburgh, to identify the influence of an Accident and Emergency (A&E) department and observe the current practice of sub-specialisation. Patients and Methods: Data for all general surgical admissions and operations in the two main Edinburgh hospitals are recorded prospectively using the Lothian Surgical Audit system. These data were examined for 1999. Results: 5346 patients were admitted to the two hospitals with acute surgical conditions. Head injuries (n=1069, 20%) and Non Specific Abdominal pain (NSAP) (n= 855, 16%) made up a third of all emergency surgical admissions. The most common single category of operations were those done on the appendix (n = 348, 15%). The Royal Infirmary, with the only A&E department had more acute surgical admissions (n=4071) than the Western General Hospital (n=1275), surgeons in the Royal Infirmary also operated on a much lower percentage of patients (30% v 55%). In the Royal Infirmary, upper gastrointestinal surgeons treated a significantly higher proportion of patients with upper gastro-intestinal and hepatobiliary/pancreatic conditions than either the general or colorectal surgeons and, similarly, the colorectal surgeons treated a higher proportion of patients with colorectal conditions than either the general or upper gastro-intestinal surgeons. Conclusion: The spectrum of emergency admissions and operations in Edinburgh is consistent with previously published data. An A&E department alters the spectrum of diagnoses and, therefore, the overall workload. Specialisation in emergency surgery is already quite advanced. These results all have important implications in future healthcare planning

INTRODUCTION 
Emergency surgical care is of high priority within the National Health Service and has been estimated to account for as much as 50% of all general surgical admissions.^1-3 With the continuing problems of limited resources, now combined with the reduced experience and availablitiy of the junior doctors following the introduction of the Calman training programme and the New Deal, many hospitals have re-examined how best this emergency service might be delivered without compromising patient care.^4-5 In some regions this has resulted in some hospitals closing their A&E departments, while others considered further steps towards greater specialisation. However, there are little data currently available to demonstrate the influence of an A&E department per se on emergency workload and the degree of specialisation which might already exist within the overall spectrum of emergency general surgery, considering the changes already taking place in elective surgery. If the improved outcomes following specialisation in elective surgery can also be shown for emergency surgery, there will be important implications for the delivery of emergency surgical services throughout the UK.^6-10 There are some suggestions that this might well be true for patients undergoing emergency colorectal surgery, who appear to be less likely to have a stoma if their surgeons are attached to units with a special interest in colorectal surgery, as compared with those with other interests.¹¹ Indeed, a recent audit carried out in the Royal Infirmary Edinburgh (David Bartolo, personal communication) demonstrated a significantly higher morbidity and stoma rate for patients undergoing emergency colorectal procedures when the operations were carried out by surgeons without a specialist colorectal interest. Further specialisation would of course have other implications for surgical training and the operative experiences for the trainee surgeon.¹²

This study examines the overall spectrum of emergency general surgical admissions and operations in Edinburgh in order to identify the influence of an A&E department and the current practice of sub-specialisation.

PATIENTS AND METHODS
There are two hospitals in Edinburgh which provide acute surgical care for a population of approximately 800 000. The Royal Infirmary of Edinburgh (RIE) has the only A&E department in the city and also serves as the regional trauma unit. The Western General Hospital (WGH) is the other hospital and although it does not have and A&E department, it is the home of the regional cancer centre and shares the emergency general surgical receiving for three days each week to the four days receiving of the RIE.

Data for all general surgical admissions and operations, in both the general surgical units at the RIE and WGH are recorded prospectively using the Lothian Surgical Audit (LSA) system. The data for all emergency admissions and operations were analysed from the 1 January to 31 December 1999. Vascular admissions were excluded from the study as there is a separate team which deals with these patients. Data recorded included the final consultant at the point of discharge, final clinical diagnosis, operation and consultant under which this was carried out. The data were then categorised according to the diagnostic and operative categories. Conditions involving the appendix, small bowel, non-specific abdominal pain (NSAP), urinary tract disorders and gynaecological disorders were categorised for analysis as general, whilst spleen, head injury and non-specific trauma were all catergorised as trauma. Consultants were grouped according to surgical specialty: oesophago-gastric and hepatobiliary-pancreatic were considered together (UGI/HPB) and colorectal (CR) surgeons and general surgeons separately. Operative procedures were catergorised as general (renal, other, appendix, small bowel, spleen, laparoscopy/laparotomy) upper GI (oesophageal/ gastric/duodenal/endoscopy), hepatobiliary, anorectal and colonic (colonic, sigoidoscopy/colonoscopy). 

STATISTICAL ANALYSIS
Statistical analysis was carried out using the Chi squared test and significance only considered for P values <0.01.

RESULTS
A total of 5346 acute surgical admissions were treated in Edinburgh during the calendar year of 1999 (1275 at the WGH and 4071 at the RIE). The broad spectrum of diagnoses and operations are shown in Tables 1 and 2. It can be seen that head injuries (n= 1069, 20%) and NSAP (n= 855, 16%) together make up a third of all emergency surgical admissions. The most common single category of operations were those done on the appendix (n = 348, 15%). Endoscopic procedures were only considered if they were carried out in theatre, not in the endoscopic unit, as these procedures were not recorded on the LSA system.

Overall the RIE had more acute surgical admissions than the WGH, not only because of the extra day receiving each week, but also because of the presence of the A&E department. However, it only carried out 156 more operations than the WGH and as a result the WGH (n=1085 operations) operated on 55% of the emergency admissions compared with only 30% at the RIE (n= 1241 operations). Overall, 2326/5346 (44%) patients underwent surgery. Analysis of all the emergency surgical admissions shows the clear differences in the spectrum of emergency diagnoses at the two hospitals and is almost certainly related to the source of patients and referral pattern differences between the two hospitals. There were more head injuries and NSAP admissions to the RIE. The spectrum of operations carried out between the two hospitals was similar.

Tables 3 and 4 illustrate the spectrum of diagnoses treated by the different surgical teams at the RIE and WGH. When considered as a proportion of overall workload, both the CR and general surgeons treated significantly more patients with colorectal disorders than the UGI/HPB surgeons at the RIE (X2 = 9.32, p<0.01 and X2 = 7.72, p<0.01) and CR surgeons also treated significantly more patients with anorectal disease than either general or UGI/HPB surgeons (X2 = 13.1, p<0.01 and X2 = 6.57, p<0.01). Interestingly, although the UGI/HPB surgeons treated more patients with HPB disorders than either of the other two groups of surgeons, the differences were not significant. The same was not true at the WGH where both the general and UGI surgeons treated significantly more patients with HPB disorders than the CR surgeons (X2 = 13.2, p<0.01 and X2 = 33.2, p<0.0001). At the WGH, CR surgeons also treated a significantly higher number of patients with colorectal disorders than the UGI surgeons.

It was not surprising, considering the degree of specialisation already seen in the diagnoses between UGI/HPB, general and CR surgeons, that similar differences were also seen in the spectrum of operations performed (Tables 5 and 6). At the RIE, HPB surgery constituted a much greater proportion (12%) of the UGI/HPB surgeons workload than for either the general (7%) or the CR (4%) teams (X2 =6.5, p<0.01 and X2 =15.7, p<0.0001). The CR surgeons at the RIE carried out significantly more CR surgery than the UGI/HPB surgeons (X2 = 9.18, p<0.01).   

There were no differences in the proportion of anorectal procedures carried out between the three groups. Again, similar patterns were observed in the WGH.^18,19 

DISCUSSION 
The overall spectrum of emergency surgical admissions in Edinburgh, with the most common diagnoses of head injuries (20%) and NSAP (16%) is not dissimilar to other studies.^13,14 The difference in numbers and proportion of various diagnoses between the two hospitals in our study reflects, to an extent, the influence of an A&E department. However, the presence of tertiary referral units for both colorectal and upper GI/HPB disorders has undoubtedly also had some influence. The large number of head injuries seen in the RIE also makes comparison of proportions of diagnoses between the two hospitals difficult. It cannot be ignored, however, that head injuries in the RIE represent a significant proportion of the consultants’ workload and with the recent publication of the SIGN guidelines (2000) for the management of head injuries, it is likely that as a proportion of the total number of admissions they will increase, with major resource implications.¹⁵ Although the data for the study were compiled from the calender year of 1999, little has changed in Edinburgh in either 2000 or 2001, to suggest that either of these years should have been studied instead.

The range of operations carried out in Edinburgh demonstrates the broad spectrum of general surgery and were comparable between the two hospitals.^16,17 However, when comparing the number of WGH operations, as a percentage of the total number of emergency admissions with those from the RIE, the WGH was shown to operate on a much higher percentage. This is obviously related to the large number of patients admitted through the A&E department in the RIE who did not require surgery. Indeed, with increasing resource restrictions this is a cause for concern and might be reduced by providing an experienced surgical opinion at an early stage in the patient’s assessment.^18,19 This is something which we have increasingly tried to provide in Edinburgh, with the development of an Emergency Team, where the Consultant and junior staff have no elective commitments and, therefore, can focus on emergency care.²⁰      

With the ongoing trend towards specialisation in all areas of medicine which has generated considerable attention, both in the UK and North America, it is interesting to note the significant variation in the spectrum of emergency admissions and operations carried out between surgeons of the different sub-specialties in this study.^1,21,22 Although some of this may be due to tertiary referrals, the biggest driving force is almost certainly the inter-unit referral which has become so common in many hospitals throughout the UK.   

This study has confirmed what many surgeons have probably observed in their own hospitals for some time: firstly, the presence of an A&E department significantly alters the spectrum of disease and provides a large non-surgical workload for the emergency surgeon; secondly, and perhaps of more importance in the further development of emergency surgical services and surgical training, there is now already a significant degree of sub-specialisation within general surgery in relation to the management of emergency admissions. With further reductions in the time available for training the general surgeons of the future and the introduction of the European Working Time directive for the current Consultants, these changes have significant implications for the delivery of emergency surgical services in the future.

Following the results of this study and the various arguments discussed already, general surgical services in Edinburgh were re-organised in August 2002. All CR surgeons moved to the WGH where they receive their share of emergency general surgery for the city along with all the colorectal emergencies. The remaining General and UGI/HPB surgeons now work at the RIE receiving their share of general surgical emergencies and all the UGI/HPB emergencies along with the trauma. The early results from this new service are currently being evaluated and will be reported in due course.

REFERENCES
1. The Scottish Office. Acute Services Report 2000.
2. Senate of the Royal Surgical Colleges of Great Britain and Ireland. Consultant Practice and Surgical Training in the UK. London; 1994.
3. Ellis BW, Rivett RC, Dudley HAF. Extending the use of clinical audit data: a resource planning model. Br Med J 1990; 301: 159-62.
4. Calman K.Hospital doctors:training for the future. The report of the working group on specialist medical training. London: HMSO, 1993.
5. NHS Management Executive. Junior doctors, the New Deal. Working arrangements for hospital doctors and dentists in training. London: Department of Health, 1991.
6. Lane RHS, Thompson MR., Smith J. Effect of Specialisation on outcome in patients having surgery for rectal Cancer Gut 1999; 44: T63.
7. Dorrance HR, Docherty GM, O’Dwyer PJ. Effect of surgeon speciality interest on patient outcome after potentially curative colorectal cancer surgery Dis Colon Rectum 2000; 43: 492-98.
8. Velanovich V. Should general surgeons perform speciality procedures? An outcome experience with reduction mammoplasty. Am Sur 1996; 62: 156- 58.
9. Sutton CD, Gilmour JP, Berry DP, Lewis MH. The evolution of a vascular surgeon at a district general hospital, is sub sub-specialisation inevitable? Ann R Coll Surg Engl 2000; 82: 272-74.
10. Consultant Surgeons of the Lothian and Borders Health Boards. Lothian and Borders large bowel cancer project: immediate outcome after surgery. Br J Surg 1995; 82: 888-90.
11. Darby CR, Berry AR, Mortensen N. Management variability in surgery for colorectal emergencies. Br J Surg 1992; 79: 206-10.
12. Nash GF, Reddy KM. Blooom IT. A regional survey of emergency surgery: the trainees’ perspective Ann R Coll Surg Engl 2000; 82: 95-6.
13. Williams RJL, Hittinger R, Glazer G. Resource implications of Head Injuries on a surgical unit. J Roy Soc Med 1994; 87: 83-6.
14. Irvin T T. Abdominal pain: A surgical audit of 1190 emergency admissions. Br J Surg 1989; 76: 1121-25.
15. Scottish Intercollegiate Guidelines NetworkGuidelines for the management of head Injuries; 2000.
16. Chiasson PM, Henshaw JD, Roy PD. General Surgical Practice Patterns in Nova Scotia: The role of the ‘generalist’ general surgeon. Can J Surg 1994; 37: 285-88.
17. Loefler IJP. The drawbacks of Overspecialisation. J R Coll Surg Edinb 1999; 44: 11-12.
18. Dookeran KA, Bain I., Moshakis V(1996), Audit of General practioner referrals to a surgical assessment unit: New methods to improve the efficiency of the acute surgical service. Br J Surg 1996; 83:1544-47.
19. Bowrey DJ, Wheeler JMD, Evans RON, Figelstone LJ, Vellacott KD. Can Emergency general surgical referrals be reduced? A prospective study. J R Coll Surg Edinb1997; 42: 381-82.
20. Addison PDR, Getgood A, Paterson-Brown S. Separating elective and emergency surgical care (the emergency team). Scot Med J 2001; 46: 48-50.
21. Schinbein JE. The need for General Surgeons. Can J Surg 1993; 36: 17.
22. Jordan GL. The Future of General Surgery. Am J Surg 1991; 161: 194-202.

Copyright: 11 March 2004

ERRATUM

J R Coll Surg Edinb Irel 2:2;107-111 Prospective audit of quality of colonoscopy in a surgical coloproctology unit

J.S. Varma, T. Fasih and M.A. Tabaqchali

should have read

Prospective audit of quality of colonoscopy in a surgical coloproctology unit

T. Fasih, J.S. Varma and M.A. Tabaqchali