The U.K. small aneurysm trial: Design, methods and progress

Eur J Vasc Endovasc Surg 9, 4248 (1995)

The U.K. Small Aneurysm Trial: Design, Methods and Progress

The UK Small Aneurysm Trial Participants

Objectives: To study the outcome of surgery versus ultrasound surveillance for small aortic aneurysms. Design: Multicentre, random allocation, open trial. Setting: University and general hospitals across the UK. Materials, Patients aged 60-76 years with small asymptomatic abdominal aortic aneurysms (4-5.5 cm in diameter). Chief Outcome Measures, The mortality rate, costs, usage of health service resources and quality of life of each treatment will be compared. Conclusions: From patients randomised to ultrasound surveillance, the natural history of small aneurysms will be determined. The design, methods and progress of this trial are reported here.

Key Words: Abdominal aortic aneurysm; Mortality; Growth rates.

Introduction

Background

Abdominal aortic aneurysm is a common condition. There is evidence from Britain, America and Australia that the number of aortic aneurysms presenting, both electively and for rupture, is increasing in absolute terms. 1-3 Population screening studies indicate that the prevalence of abdominal aortic aneurysm in men over 65 years is about 5%. 4"5 This "epidemic" of aneurysms together with the aging population is likely to impose increasing burdens on health resources in Europe.

Over the past decade surgical and anaesthetic techniques have improved rapidly, such that 30-day mortality rates of <5% are reported for elective aneurysm repair. 6, 7 However, overall national mortality rates for elective aneurysm surgery may be higher, s Nevertheless, in contrast with the improvement of mortality associated with elective surgery, mortality rates associated with grafting for ruptured aneurysms remain high ( - 40%). 9 Despite the increasing numbers of elective aneurysm repairs, the number of patients admitted with ruptured aortic aneurysm also con- tinues to increase. 1 The dilemma for the vascular surgeon, is whether to operate early and electively on

Please address all correspondence to: J. T. Powell, Department of Surgery, Charing Cross & Westminster Medical School, Fulham Palace Road, London W6 8RF, U.K.

All vascular surgeons consider elective surgery when an abdominal aortic aneurysm reaches 6 cm in diameter. Some surgeons have advocated elective surgery for a patient of any age with an aneurysm of 4 cm or more in diameter. ~2 For most there is a grey area of uncertainty concerning the best management of aneurysms between 4 and 6 cm in diameter. After vigorous discussion at the Vascular Surgical Society of Great Britain and Ireland a group of surgeons agreed upon their grey area of uncertainty: asymptomatic aneurysms of 4.0-5.5 cm in diameter in fit patients aged 60-76 years. These surgeons agreed to participate in a random allocation trial of early elective surgery or ultrasound surveillance in such patients.

Clearly, the establishment of a multicentre trial provided the opportunity to answer other important questions, aside from whether all cause mortality was reduced by early elective surgery of 4.0-5.5 cm diameter aneurysms. The trial will provide an opportunity to investigate the growth and rupture rates of small aneurysms, factors associated with accelerated

small asymptomatic abdominal aortic aneurysms or to observe the aneurysm for increase in size. The risk of rupture is very low in small aneurysms but increases with increasing diameter. 1°'11

Rationale

1078-5884/95/010042 + 07 $08"00/0 © 1995 W. B. Saunders Company Ltd.

U.K. Small Aneurysm Trial 43

Steering Committee Monitoring Committee

Regional trial Trial statistician directors

Trial coordinators < ~ Participating surgeon

" ' " - " - - ' ~ Patient ~ . / /

Fig. 1. Main organisat ional s t ructure of the trial.

aneurysmal growth, costings and patients' quality of life for the two treatments and how 24-hour and 30-day mortality rates for elective aneurysm surgery across Britain, vary according to age, sex, aneurysm size and concomitant ischaemic heart disease.

Organisational Structure

The Steering Committee has responsibility for fund- ing, running and reporting the trial. 13 Consecutive patients are recruited from all participating surgeons (Fig. 1). The surgeon sees patients together with the local trial co-ordinator (five co-ordinators cover the five regions in Britain). Each trial co-ordinator is based in one of five hospitals, where they are supervised by a surgeon who functions as Regional Trial Director (in Bath, Edinburgh, Leicester, London and Manchester). The trial co-ordinators collect and forward the patient data to the trial statistician in London. The Steering Committee arranges regular meetings for trial co- ordinators and participating surgeons and reviews

Patient (aged 60-76 years) with an asymptomatic infrarenal aneurysm

I A-P ultrasound diameter

L I J

3.0-3.9 cm 4.0-5.5 cm >5.5 cm

Could patient be fit for surgery?

If no 4-- If yes, randemise

I

Observation Surgery Observation Surgery

I o r j S Enlargement development of symptoms

Fig. 2. Outl ine of trial design.

recruitment at regular intervals. The Steering Commit- tee appointed a Monitoring Committee, which ensures the proper running of the trial and also functions as a Stopping Committee. The Monitoring Committee receives regular reports from the trial statistician, particularly with respect to deaths. All others, including the Steering Committee and trial participants, have no access to information on deaths in the trial.

The outline design for the trial is given in Fig. 2. The recruitment of patients derives from referrals to the participating vascular surgeons. These referrals normally come from general practitioners, other hospital departments and from hospital or community based screening programmes. The trial co-ordinators have been trained together to use portable ultrasound equipment to optimise standardisation of aortic diameter assessment. 14 We have chosen the Aloka SSD-500 equipped with a 3.5 mHz probe for these investigations. Random cross-checking between operators can be performed by measurement of hard copy image. The trial co-ordinators also meet at regular intervals to standardise their readings of blood pressure and lung function.

Apart from the initial patient recruitment form, which is completed by the surgeon, the remainder of the paperwork is completed by the trial co-ordinator. All patients aged between 60-76 years with an asymptomatic aneurysm are included. All patients with asymptomatic aneurysms of < 4 cm are followed up by the trial co-ordinators at either 6 or 12 monthly intervals to monitor the aneurysm for growth. All patients with an aneurysm diameter between 4.0-5.5 cm who are fit for surgery are considered for entry into the trial. Those unwilling or considered unfit for surgery enter the study for observation of aneurysm growth rates.

The fate of those patients with asymptomatic aneurysms >5.5 cm is also being documented. Patients who are unfit for surgery, where possible are also followed to monitor aneurysm growth and from those fit for surgery 30-day operative mortality rate is determined.

Those patients who are considered fit for elective surgery and agree to enter the trial are randomised using a telephone computer link line. For patients who are randomised to elective surgery, the operation should be scheduled within the following 3 months. For patients who are randomised to ultrasound surveillance, there are regular follow-up appointments.

Eur J Vasc Endovasc Surg Vol 9, January 1995

44 Study Group

Recruitment, Randomisation and Sample Size

Recruitment

Recruitment of patients is through the consultant trialists. When agreeing to the study it is emphasised that all patients with a non-ruptured infra-renal aortic aneurysm who are between the ages of 60 and 76 years should be registerex~. At the first visit following registration the patient with an aneurysm 4.0-5.5 cm in diameter is seen by the co-ordinator together with the surgeon to discuss entry to the trial. Patients who are not suitable for entry into the trial are offered follow-up by the trial co-ordinator.

Randomisation

If a patient is considered suitable for entry into the trial the co-ordinator contacts the randomisation computer established in collaboration with British Telecom plc, which is held at the Central Co-ordinating Office. This method of randomisation is unique in that the computer responds to the telephone call and understands the responses to a series of questions used to confirm the patient's, suitability for the trial. The computer records details of the patient, hospital and consultant surgeon. This information can then be accessed by the Central Co-ordinating Office. The treatment allocated is immediately available to the co- ordinator, surgeon and patient. This service is available 24 hours a day together with information on treatment allocation of previous patients.

Sample size

The required sample size was estimated to be 1000 patients (500 in each arm of the trial). This estimate was dependent on the predicted mortality in the surgical and surveillance groups. However, in keeping with most trials, the incidence of the outcome measure was not known precisely and was based on the best available evidence when the trial was designed. We assumed a 30-day mortality of 2% associated with elective surgery and a 2% annual mortality in the observation group due to aneurysm rupture or sub- sequent surgery, if the aneurysm enlarges rapidly or symptoms developed. Mortality due to other causes in a typical population with small abdominal aortic aneurysm was estimated to be about 6% per annum. With recruitment of 1000 patients, the trial will, over a 5-year period of follow-up, have a

predictability of 80% to detect a 7% difference in mortality at p < 0.05. The 95% confidence limits of mortality in either group will be + 3%. The sample size will also be adequate to detect differences in quality of life and other outcome measures and, in the surveillance arm of the trial will detect differences in the prognostic factors related to aneurysm growth.

Eligibility and Exclusion Criteria

Eligibility criteria and exclusions

All patients between the ages of 60 and 76 with an unruptured aneurysm are entered into the aneurysm study. Randomisation into the trial requires that the maximum antero-posterior diameter should not be less than 4.0 cm and not exceeding 5.5 cm as measured by the trial co-ordinator.

Specific exclusions for this trial are: (i) patients who refuse to be randomised; (ii) patients who are unfit for surgery; (iii) patients with symptoms refer- rable to the aneurysm; (iv) patients who are unable to give informed consent; (v) patients who are unable to attend for regular follow-up.

Informed consent

A patient who meets the entry criteria is informed fully about the trial and provided with an information sheet, a copy of which is sent to the patient's family practitioner. Patients may, if they ask, be given a period of time, usually around one month, to consider the options before being asked for consent to randomisation. Where there is doubt about fitness for surger~ further assessment and discussion with the surgeon and anaesthetist ensues before seeking consent.

Fitness for surgery

There are no specific investigations prescribed for surgeons to decide on suitability for surgery. The surgeons follow their normal routine for preoperative assessment. Some patients require other treatments, e.g. for hypertension, before being considered fit for surgery.

The Steering Committee did not seek to define criteria for fitness for surger~ since it was considered unlikely that universal agreement among surgeons and anaesthetists was achievable and the wish was to reflect practice as it pertained nationwide.



Treatment Groups, Visits and Procedures

At randomisation the patients are assigned to either early elective surgery or regular ultrasound surveillance.

performs quality of life assessments, records changes in cardiovascular details (including recent cardiovascular events) a n d details of use of health service resources by the patient. The only further follow-up required by the trial is data on mortality which are obtained from the NHS central registry.

Entry to the trial - - baseline data

Included in the initial interview are questions on previous medical history, particularly with respect to history of respiratory disease, diabetes and hypertension. The Rose angina and claudication ques- tionnaires are administered, together with a questionnaire on smoking history. ~s Family history of aortic aneurysm and family history of early predis- position to atherosclerosis are assessed. All drugs currently taken are recorded and categorised according to their groupings in the British National For- mulary. 16 Cardiovascular and respiratory drugs are coded individually.

Measurements are taken of height, weight, ankle- brachial systolic pressure indices using hand held Doppler, systolic and diastolic blood pressure and respiratory functions of vital capacity and forced expiratory volume using a hand held spirometer. Widest anteroposterior infra-renal aortic diameter is taken as previously described. ~4 A non-fasting blood sample is taken for analysis of total cholesterol HDL- cholesterol triglycerides, fibrinogen, creatinine, hae- matocrit and white cell count. HDL-cholesterol and fibrinogen are measured in a central laboratory at Charing Cross, other variables are measured locally. Patients who are suitable for the trial undergo a chest X-ray from which the cardiothoracic ratio is determined. An electrocardiograph is taken which is coded accorded to the Minnesota Classification.

Surgery

After a patient has been randomised to early elective surgery the surgeon trialist should try to schedule the operation within the next three months. Surgical details including type of graft, surgeon and anaesthetist rank, findings at operation, need for reopera- tion and 30-day mortality are recorded by the trial co- ordinator.

After discharge patients are seen at 1 month by the surgeon trialist, and at 6 months and i year by the trial co-ordinator. At the 6-month and 1-year visits patients are seen by the trial co-ordinator who

Ultrasound surveillance

Patients who are assigned to the surveillance arm of the trial are seen by the trial co-ordinator on a regular basis as described previouslj6 usually at the surgeon's outpatient clinic. Patients (aneurysm diameter 4.0-5.0 cm) who are randomised to ultrasound surveillance are reviewed at 6-monthly intervals. However, if the aneurysm diameter is between 5.0 and 5.5 cm the interval between visits is shortened to 3 months. Should the aneurysm grow to beyond 5.5 cm, rapidly increase in diameter (> 1 cm/year), or the patient develop symptoms, surgery may be considered. At this time the surgeon is informed of the change in status and makes a decision on the patient's suitability for surgery.

The follow-up procedures are similar to those for the group assigned to surgery except that the aneurysm diameter and patient's blood pressure are also measured. Information is also collected about changes in medication and smoking habit. These surveillance visits continue until either the patient dies, the trial ends or surgery is considered and effected. Surgery is considered only when the aneurysm reaches 5.5 cm, grows rapidl)~ or becomes symptomatic. At the 12-month follow-up visit a further blood sample is obtained for the analysis of plasma cotinine.

Outcome Measures

Mortality

The principal end point for this trial is mortality. Operative mortality is assessed at 30 days and all causes of death recorded. All patients are given a card which gives information on the trial and which is carried at all times. This includes a number to ring in the event of a serious event or death. Other mortality data are obtained from hospital records and family practitioner notes. All patients entered into the trial are "flagged" with the NHS Central Registry which provides a copy of the death certificate when a patient dies.


46 Study Group

Quality of life

The Medical Outcomes Study short form general health survey is being used throughout this trial. 18 This provides information on a number of self assessed scales for physical and social functioning, mental health, health perceptions and pain, with scores from 0 to 100. These assessments are made when the initial questionnaire is completed and at all follow-up visits made by the nurse co-ordinator.

Growth rates

Information on growth rates is derived from serial measurements of the aortic diameter in patients randomised to ultrasound surveillance. There is addi- tional information on growth rates provided from patients outside the trial: those with 3-3.9 cm aneurysms and those with larger aneurysms (> 5.5 cm) who are not fit or will not consent to elective surgery.

Cost effectiveness

In order to determine the cost effectiveness of surgery compared to observation, the health service costs of both treatment options will be determined. The costs of admission for elective aneurysm repair are being determined for 50 patients in a specialist vascular unit in a teaching hospital and in a district general surgical unit. This costing includes a detailed assessment of the costs of such resources as staffing, investigations, drugs, and overheads consumed during the pre- and post-operative periods. In addition, the costs of the operation are being determined for 15 operations. All operations are recorded and coded according to a standard classification. 19 The costs of an ultrasound examination are also being estimated, based on staff time, use of equipment, consumables and overheads.

For each of the costings, sensitivity analyses will be carried out so that estimates may be adjusted according to differing use of major resources in different centres. For example, the effects on costs of different average lengths of postoperative stay can be estimated. The extent to which the use of major resources might vary between centres will be determined by means of a questionnaire sent to all surgeons participating in the trial.

In addition to these costing exercises, the nurse co-ordinators are collecting data on the use of health service resources by each patient. At 6-monthly intervals, the number and type of health service

contacts in relation to the aneurysm (e.g. in-patient admission, out-patient attendance and family practitioner consultation) are being obtained from each patient in the trial. This information is obtained using a short questionnaire administered by the co-ordinator at the 6-monthly ultrasound examination for the observation group or at the 6-month and 1-year postoperative follow-up on the surgical group.

The cost estimates outlined above will be com- bined to generate a total programme cost for the surgical and observation arms of the trial. The relative cost of these alternatives will be examined in terms of their association with survival rates and the corre- sponding number of life years gained.

Results and Statistical Analyses

Interim analysis - - stopping rules

Operative (30 day) mortality. Operative (30 day) mortality is expected to be of the order of 2%. The monitoring committee considers that a 5% (30d) mortality is unacceptable. Surgical mortality will be analysed cumulatively after each 100 patients randomised to surgery have had aortic aneurysm surgery. Confidence intervals have been calculated that give the lower bound for stopping the trial at 5%. This gives the cumulative number of deaths for stopping the trial at 11 for the first 100 patients, 18 for 200, 25 for 300, 30 for 400, and 36 for 500 patients operated.

Interim comparisons of groups. It is anticipated that four interim analyses and one final analysis will take place throughout the trial. The O'Brien Fleming boundaries will be used as the basis for stopping the trial. 2° The overall probabilities for a treatment benefit will be set at p = 0.05, with an adverse effect set at p = 0.10. Since 350 deaths are expected in this trial the interim analyses will be after 70, 140, 210, 280 and finally after 350 deaths.

Final analysis. As given above the final analysis will be expected to occur after 350 patients have died in this trial. The survival analysis will be performed using the Cox proportional hazards model with time to death as the principal end point for the trial, as The survival model will be adjusted for potential con- founding variables which are expected to influence mortality. These will include age, sex, history of ischaemic heart disease, peripheral arterial disease, hypertension and chronic obstructive airways disease. Proportionality will be assured using the methods of Kalbfleisch and Prentice. 2a


Progress to Date 30

Conclusion

The first pat ients were assessed for suitabil i ty to enter the trial in A u g u s t 1991. Since that t ime more than =~

• ~ 20 2,469 pat ients have been regis tered a nd of these 2,176 have h a d a decis ion m a d e on their t rea tment (Fig. % 3).

The half w a y po in t for en t ry into the trial, the = 10 r a n d o m i s a t i o n of 500 pat ients w a s ach ieved in Octo- ber 1993. The r a n d o m i s e d pa t ien t p o p u l a t i o n com- pr i sed 85% m e n a nd 15% w o m e n , wi th a m e a n age of 69.1 years (S.D. = 4.5). The d is t r ibut ion of a n e u r y s m 0 d iameters is s h o w n in Fig. 4. 4.0-4.2 4.3-4.5

The m a n a g e m e n t of small a b d o m i n a l aortic aneu- r y s m s is like m a n y o ther i m p o r t a n t issues of surgical practice, m o s t app rop r i a t e ly inves t iga ted b y m e a n s of a mul t icent re r a n d o m al locat ion trial. In o rder to de te rmine w h e t h e r ear ly elective su rge ry or ultra- s o u n d survei l lance is the bet ter m a n a g e m e n t m o d a l i t y for smal l a b d o m i n a l aort ic aneu rysms , wi th in a re la t ively shor t t ime frame, we have chosen a trial s t ructure w h i c h focuses on h igh qual i ty data t h r o u g h the e m p l o y m e n t of ded ica ted trial co-ordinators . The U.K. Small A n e u r y s m Trial is s chedu led to close in July 1996 after r a n d o m i s a t i o n of 1,000 patients. The f r a m e w o r k of the trial has been des igned so that several o ther i m p o r t a n t issues can be addressed , These o ther issues inc lude the costs of t reatment , the qua l i ty of life for each t r ea tment a nd the ident i f icat ion of factors associa ted wi th the r ap id g r o w t h of small

Key

[ ] unfit

[ ] 3.0-3.9 cm

~]~> 5.5cm

• ....... "'~:~ ~ refused

~] symptomatic

[ ] other I

[ ] randomised

n = 2176

Fig. 3. Distribution of 2,176 decisions made on patients entered into the U.K. Aneurysm Trial. Of the total, 21% met the inclusion criteria for the trial and were randomised.


4.6-4.8 4.9-5.1 5.2-5.5

Aneurysm diameter (cm)

Fig. 4. Distribution of antero-posterior aneurysm diameter in patients randomised in the U.K. Small Aneurysm Trial.

aneurysms . The par t ic ipa t ing su rgeons are listed in the appendix : the success of this trial will be attrib- uted, in large part , to their en thus iasm. With barr iers d i s appea r ing in Europe we h o p e that o ther i m p o r t a n t issues in vascular su rge ry can be add re s sed b y similar large trials recrui t ing pat ients f r o m several countr ies .

Acknowledgements

This work is being funded by research grants from the Medical Research Council of Great Britain and the British Heart Foundation.

References

1 FOWKES FGR, MACINTYRE CCA, RUCKLEY CV. Increasing incidence of aortic aneurysms in England and Wales. BMJ 1989; 298: 33-35.

2 MELTON LJ, BICKERSTAFF LK, HOLLIER LH, et aI. Changing incidence of abdominal aortic aneurysms: a population based study. Am J Epidemiol 1984; 120: 379-386.

3 CASTLEDEN WM, MERCER JC, and Members of the West Austra- lian Vascular Service. Abdominal aortic aneurysms in Western Australia: descriptive epidemiology patterns of rupture. Br ] Surg 1985; 72: 109-112.

4 COLLIN J, ARAUJO L, WALTON J, J~INDSELL D. Oxford screening programme for abdominal aortic aneurysm in men aged 65 to 74 years. Lancet 1988; ii : 613-615.

5 SCOTT RAP, ASHTON H, SUTTON G L J . Ultrasound screening of a general practice population for aortic aneurysm. Br J Surg 1986; 73: 318.

6 JOHNSTON kqA/, SCOBIE TK. Multicenter prospective study of non ruptured abdominal aortic aneurysms: population and operative management. J Vasc Surg 1988; 7: 69-81•

7 GREENHALGH RI~. Prognosis of abdominal aortic aneurysm. BMJ 1990; 301: 136.

8 JENKINS AMcL, RUCKLEY CV, NOLAN B. Ruptured abdominal aortic aneurysm. Br J Surg 1986; 73: 395-398.

9 CAMPBELL WB. Mortality statistics for elective aortic aneurysms. Eur ]Vasc Surg 1991; 5: 111-113.

10 NEvrrw MP, BALLARD DJ, HALLETT JW. Prognosis of abdominal aortic aneurysms. N Engl ] Med 1989; 321: 1009-1014.


48 Study Group

11 DARLING RC. Ruptured arteriosclerotic abdominal aortic aneurysms. A pathologic and clinical study. Am ] Surg 1970; 119: 397-401.

12 COLLIN J. Elective surgery for small abdominal aortic aneurysms. Lancet 1987; i: 909.

13 POWELL JT, GREENHALGH RM, RUCKLEy CV, FOWKES FGR. Pro- logue to a surgical trial. Lancet 1993; 342: 1473-1474.

14 ELLIS M, POWELL JT, GREENHALGH RM. The limitations of ultrasonography in surveillance of abdominal aortic aneurysms. Br J Surg 1991; 78: 614-616.

15 ROSE G, MCCARTNEY P, REID, DD. Self-administration of a questionnaire on chest pain and intermittent claudication. Br J Prevent Med 1977; 31: 42-48.

16 British National Formulary (No 22). London. BMA & RPS 1991. 17 BLACKBURN H, KEYS A, SIMONSON E, et al. The electrocardiogram

in population study. Circulation 1960; 21: 1160-1175. 18 STEWART AL, HaYs RD, WARE JE. MOS short form general health

survey. Reliability and validity in a patient population. Med Care 1988: 26: 724-735.

19 OPCS. Tabular List of the Classification of Surgical Operations & Procedures (4th edition). London. HMSO 1990.

20 O'BRIEN PC, FLEMING TR. A multiple testing procedure for clinical trials. Biometrics 1979; 35: 549-556.

21 Cox DR. Regression models and life tables. J R Statist Soc B 1972; 34: 187-220.

22 KALBFLEISCH JD, PRENTICE RL. The Statistical Analysis of Failure Time Data. New York. Wiley 1980.

Accepted 24 April 1994

Appendix

Steering Committee

R.M. Greenhalgh, J.F. Forbes, F.G.R. Fowkes, J.T. Powell, C.V. Ruckley.

Monitoring Committee

P.A. Poole-Wilson (Chairman), N.L. Browse, C.J. Bul- pitt, E.C. Coles, A.E. Fletcher.

Regional Centres

Bath ' Regional Trial Director: M. Horrocks. Trial Co-ordinator: S. Logan. Consultant Trialists: R.N. Baird, A.R. Baker, R.L. Brackett, B.P. Bliss, RJ. Clarke, J. Earnshaw, D.R.A. Finch, A.D.S. Flowerdew, C.P. Gibbons, D.R. Harvey, A.R. Hedges, P.M. Lamont, P. Lear, P. Lewis, S.D. Parvin, D.C. Wilkins.

Edinburgh Regional Trial Director: C.V. Ructdey.

Trial Co-ordinator: A. Rattray. Consultant Trialists: R. Baker, J. Chamberlain, G.G Cooper, J.G.R. Cumming, A.I.G. Davidson, J.L. Dun- can, J. Engeset, J.C. Forrester, J.L.R. Forsythe, D.R. Harper, A. Howd, A.McL. Jenkins, N.A.G. Jones, D. Lambert, G.C. MacBain, P.T. McCollum, J.St.C. McCormick, A.J. McKay, I. Muir, J.A. Murie, G. Proud, R.C. Smith, G. Stewart, P. Stonebridge, A.R. Turner.

Leicester Regional Trial Director: P.R.F. Bell. Trial Co-ordinator: R. Meer, S. Blair (resigned). Consultant Trialists: P.R. Armistead, A. Aukland, E.T. Bainbridge, W.W. Barrie, J. Beard, J. Black, K.G. Callum, R. Downing, D.B. Hamer, B.R. Hopkinson, R.E. Jenner, D.S. MacPherson, G.S. Makin, J.R. Nash, S. Powis, D.A. Ratliff, R.D. Stewart, C.L. Welsh, P.W. Wenham.

London Regional Trial Director: R.M. Greenhalgh. Trial Co-ordinator: K. Ferguson, R. Clark (resigned). Consultant Trialists: J.S. Ackroyd, M. Adiseshiah, J.W.P. Bradley, S. Brearley, K.G. Burnand, C. Butler, A.E.P. Cameron, J.M.F. Clarke, J.F. Colin, J.A. Dor- mandy, A.E.B. Giddings, G. Hamilton, R.W. Hoile, R.H.S. Lane, R.J. MacFarland, A.R.L. May, A.J. Mclr- vine, D. Negus, D.M. Nott, M.K. O'Malley, T. Paes, B.D. Pardy, P.H. Pattisson, M. Pietroni, M.R. Pittam, D.T. Reilly, J.M. Rutter, C.J.L. Strachan, P.R. Taylor, R.S. Taylor, M. Thomas.

Manchester Regional Trial Director: C.N. McCollum. Trial Co-ordinator: M. Mossa. Consultant Trialists: M.I. Aldoori, J. Clarke, L. de Cossart, L.A. Donaldson, R.G.M. Duffield, D. Durrans, n .J . Gough, M.G. Greaney, B.R. Gwynn, R. Hall, A.R. Hearn, M.C. Holbrook, G.B. Hopkinson, R. Hughes, N. Hulton, J.F. Kelly, O.E. Klimach, M.E. Lambert, S.H. Leveson, D.M. Matheson, A.R. Maudsley, J. Mosley, T. O'Brien, W.G. Paley, I.G. Schraibman, M.G. Walker, A.R. Wilkinson, C.K. Yeung.

Central Co-ordinating Office Clinical Data Management: P.J. Franks.

Randomisation Computer Service (British Telecom): N. Forse, S.J. Barrington.

This report was compiled and written by: P.J. Franks, C.V. Ruckley, F.G.R. Fowkes, R.M. Green- halgh and J.T. Powell.


Documents

The U.K. small aneurysm trial: Design, methods and progress