|Home | About | Journals | Submit | Contact Us | Français|
Thresholds for elective surgery may be too high
In this week's BMJ, Roberts and colleagues used record linkage analysis to compare mortality rates after elective colectomy, emergency colectomy, and no colectomy in people admitted with inflammatory bowel disease.1 The results are important and indicate that thresholds for undertaking elective colectomy in clinical practice are too high.
The chronic inflammatory bowel diseases—ulcerative colitis and Crohn's disease—affect 1.4 million people in the United States and 2.2 million people in Europe.2 They result in substantial morbidity, lost days at work, and reduced quality of life. Despite a variety of advances in medical treatment, 20-30% of patients with ulcerative colitis need surgery at some time,3 while the lifetime risk of surgery for Crohn's disease is as high as 80%.4 In people with severe inflammatory bowel disease the long term effectiveness of drugs, such as immunomodulators, rarely exceeds 40%. Surgery can be regarded as a cure for ulcerative colitis, albeit an imperfect one. Similarly, surgery can be highly effective in Crohn's disease but is offset by the recurring nature of the disease.
The lack of consensus on how to manage these patients means that doctors and surgeons spend much time debating the optimal timing of surgery. In general, surgery is undertaken to improve symptoms and quality of life, to offset or resolve complications (such as abscess or perforation), or for the salvage of acute severe disease. Although gastroenterologists know the potential complications associated with acute severe disease, most are reassured by overall mortality data, which consistently show that people with inflammatory bowel disease have near normal standardised mortality ratios. A prospective cohort of 692 people with inflammatory bowel disease in Olmsted County, Minnesota showed an overall standardised mortality ratio of 0.8 (95% confidence interval 0.6 to 1.0) for ulcerative colitis and 1.2 (0.9 to 1.6) for Crohn's disease over a 44 year follow-up.5 A European cohort of 796 patients with the disease across 10 centres in north and south Europe showed a standardised mortality ratio of 0.69 (0.21 to 0.91) for men with inflammatory bowel disease and 1.18 (0.54 to 2.25) for women.6 In a 10 year follow-up of the same cohort, 8.7% underwent colectomy. Interestingly, wide differences in colectomy rates were seen between north and south Europe (10.4% v 3.9%; P<0.001).
Despite these reassuring statistics, Roberts and colleagues' findings appear alarming at first glance. In England and Wales between 1998 and 2000, three year mortality for a general admission to hospital was 12.2% (11.5% to 12.9%) for ulcerative colitis and 9.1% (8.5% to 9.7%) for Crohn's disease. For ulcerative colitis, three year mortality was much higher for emergency operations (13.2%, 11.0% to 15.8%) than for elective operations (3.7%, 2.7% to 4.9%).
Although these figures seem high, they cover a three year period, and the one month mortality figures for elective and emergency operations of 0.8% and 5.7% are similar to other published studies. Perhaps most worrying is the mortality in people admitted for ulcerative colitis and Crohn's disease who had no surgery (13.6% (12.8% to 14.5%) and 10.1% (9.4% to 10.8%)). Most deaths occurred after the acute admission—more than half occurred between six and 36 months later. Unfortunately, because of the nature of the study, exactly how these patients died is not clear.
The recent national inflammatory bowel disease audit (a collaboration of the British Society of Gastroenterologists, Royal College of Physicians, National Association for Colitis and Crohn's Disease, and Association of Coloproctology of Great Britain and Ireland) gives some additional information about cause of death in patients with inflammatory bowel disease.7 The audit documents 2767 admissions for ulcerative colitis, of which 397 were elective. Forty five patients (1.6%) died in hospital—25 deaths were directly attributable to ulcerative colitis—and operative mortality was 2.1% (15/715). A further 47 deaths (mortality of 1.7%) occurred after discharge, in a median follow-up period of 80 days; only eight (17%) were attributed to ulcerative colitis. In Crohn's disease 2914 patients were admitted, 556 for elective surgery; 36 (1.2%) died in hospital, 25 as a direct result of Crohn's disease. Operative mortality was 1.2% (14/1092), and 36 people died after discharge (mortality of 1.2%), only 13 (36%) as a direct result of Crohn's disease. Although the audit had a shorter follow-up than that used in the paper by Roberts and colleagues, and mortality may have been under-reported because of difficulty in retrieving notes, these unadjusted mortality figures are reassuringly closer to those reported in the above cohort studies.
Roberts and colleagues infer that people with inflammatory bowel disease who have planned elective surgery fare much better than those in hospital for emergency treatment. This may be true, but it does not follow that patients having elective surgery would otherwise end up in hospital for intensive medical treatment or emergency surgery. Equally, patients having emergency treatment may have had no previous opportunity for elective treatment.
Are there missed opportunities for elective surgery in the emergency admission group? The decision to operate remains a difficult one, that should involve the doctor, surgeon, patient, and family. Decision making is not helped by the lack of clear data on surgical outcomes. The idea that surgery for inflammatory bowel disease should be the last resort is flawed, and the data presented by Roberts and colleagues—even allowing for interpretation—should be a word of caution to those who promote it.
This article was posted on bmj.com on 30 October 2007
Competing interests: None declared.
Provenance and peer review: Commissioned; not externally peer reviewed.