Intended for healthcare professionals


Lessons from the Bristol case

BMJ 1998; 316 doi: (Published 06 June 1998) Cite this as: BMJ 1998;316:1685

More openness on risks and on individual surgeons' performance

  1. Tom Treasure, Professor of cardiothoracic surgery
  1. St George's Hospital, London SW17 0QT

    News p 1691

    Papers pp 1697, 1701, 1705

    Education and debate pp 1734, 1736, 1740

    Personal view p 1756

    Medicine and the media p 1757

    Cardiac surgery has changed within living memory from desperate attempts to achieve miracles for a few to the present situation where there is high expectation of a good result for tens of thousands of patients each year. It is easy to recall the surgeons who performed the first heart operations, who used cardiopulmonary bypass while it was still in its infancy, or who started transplantation — all undertaken with a high initial mortality. They worked on doggedly, in the face of doubt, scepticism, and often widely publicised criticism. They are now remembered with respect as having had “the courage to fail.” Many others, equally determined, did fail and are not remembered.1

    Some of that determination, in the face of possible failure, is necessary in every surgeon. But the congenital heart surgery undertaken in Bristol in the past 10 years — the subject of the General Medical Council's most recent, and arguably most important, disciplinary case (p 1691, 1740) 2 3 — was within the realms of routine practice, for which there are known and well established standards. In heart surgery accountability supplemented doggedness a long time ago. The present arrangements, however — which rely on local monitoring of results, clinician based judgments about acceptable standards, and continuing referrals — failed to avert the situation in Bristol, which we have seen spelt out in distressing detail before the General Medical Council and the nation's media (p 1757).4

    Irrespective of the long awaited conclusions of the GMC in the Bristol case, it was already abundantly clear that British cardiac surgeons, themselves well in advance of other surgical specialties in keeping a record of their results, would have to be audited in a more explicit way. The United Kingdom Cardiac Surgery Register, to which cardiac surgeons voluntarily submit their annual figures, has been run by the Society of Cardiothoracic Surgeons of Great Britain and Ireland since 1977.5 It has provided a useful benchmark against which to discuss variations in the provision of services and for individual surgeons to monitor their own mortality figures against a national average. At the time it was set up, and until recently — when there have been two high profile cases of unacceptable mortality for cardiac surgery 6 7 — there was a tacit assumption that the patients and surgeons in the dataset were anonymous and would remain so. Furthermore, the use of that benchmark to assess one's own practice was a matter of honour and personal reflection.

    The first steps in changing that have already been undertaken. All cardiac and thoracic surgeons will now have to submit for inspection their individual figures in specified areas of practice. These will be coded, but any unsatisfactory results can be easily traced back to the surgeon and poor performance investigated. Before long central recording of full data on all cases, with appropriate risk stratification, is likely to be the norm. The anonymity offered by coding is notional and may be the last vestige of a belief in confidentiality for surgical results that for years was held to be sacrosanct.

    The arterial switch operation for transposition of the great arteries, central to the Bristol case, has presented a particular dilemma for surgeons in balancing risk and benefit. “The switch” replaced well established operations (those devised by Senning in 1959 and Mustard in 1963) which provided very effective palliation by redirecting the blood flow in the atria, so that the physiology was corrected. The ventricles continued to do each other's work but only for as long as the right ventricle could withstand the systemic load. During the 1980s more and more surgeons turned to the technically exacting, but in the long term more satisfactory, arterial switch operation, with the objective of restoring normal expectation of life and function, rather than providing palliation of uncertain duration. The transition entailed the possibility of an increase in operative mortality for this condition during the “learning curve.” The operation became the standard of care, but precise preoperative assessment, impeccable surgical technique, and skilled perioperative care are needed for consistently good results. An analysis of a cluster of deaths for this operation in an otherwise excellent series at Great Ormond Street is an exemplar of honest self appraisal.8 Well in advance of the conclusions reached by the GMC in the Bristol case, a meticulous national registry for this operation has been established. We already know that for all 23 surgeons performing the operation (in 15 units) the mortality in just over 200 operations performed within the past two years is 6.5% (DeLeval M, British Cardiac Society meeting, May 1998).

    A major issue in the Bristol case has been the nature of the information given to the parents. The estimates of risk of death were substantially less than the true risk of surgery in that unit. There may be a place for giving an optimistic outlook to a patient judged to have no choice but to undergo high risk emergency surgery to save life, but the circumstances where that approach is justified are limited. There was no justification for a rosy glow in this case, where the operations were elective, could be performed elsewhere, and the difference between success and failure was potentially many years of life. It appears to be self evident that parents have a right to know the truth from both referring cardiologist and the surgeon.9 Why are doctors ever economical with it? Is truth thought to contaminate the trust in a relationship? A frank presentation of the risks and benefits to the family should include sympathy and compassion, but this should not supplant frankness.

    The hearings and deliberations at the GMC into the Bristol paediatric surgery case have stretched over many months and explored complex issues. It is often the case with a disaster (and this has been a disaster not only for these families but for many others who work in and around heart surgery) that there is no one isolated gross and culpable error. Instead a sequence of more minor faults, errors, omissions, poor procedures, failure to follow protocol, and unheard warnings together lead to the eventual tragedy. In this case the unheard warnings are particularly worrying. In 1989 Professor David Hamilton's paediatric cardiac surgery working party, exploring the provision of supraregional services, included data which might have raised questions about Bristol's continuation as a centre for paediatric cardiac surgery. Quite separately, a consultant anaesthetist in Bristol “blew the whistle” but was disregarded (p 1739).10 The UK register, to which Bristol contributed data, was available for comparison throughout this time. We have to ask why these warnings, and the questions and doubts that clearly must have surrounded the practice of paediatric cardiac surgery in Bristol for several years, were not heard.

    It has now been agreed that the Royal College of Surgeons of England and the Society of Cardiothoracic Surgeons will provide a “rapid response group” so that a member of the council of the college and a senior cardiac surgeon can be on site within 48 hours, to listen and advise on action. This is an attempt by the profession to protect patients from continuing poor performance and also to safeguard surgeons from inappropriate fault finding (since cardiac surgeons now feel very much under scrutiny and vulnerable in a climate of criticism and blame). It remains to be seen whether this initiative by the college and the surgeons' own society can be implemented effectively and whether it will be seen to be open enough to allay anxieties about the profession supposedly monitoring itself, but not doing it well enough. If we do not monitor ourselves effectively there is little doubt that it will be imposed upon us.11


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    View Abstract