COMMENTARIES 317 Colon cancer sigmoidoscope. This additional examina- ................................................................................... tion above that of conventional flexible sigmoidoscopy resulted in a further 3% of patients being offered full colonoscopy Are proximal colorectal cancers and three proximal carcinomas being Gut: first published as 10.1136/gut.52.3.318 on 1 March 2003. Downloaded from detected in subjects who would have always associated with distal otherwise been misclassified as having no neoplasia. The authors conclude that adenomas? the finding of any adenoma at flexible sigmoidoscopy should trigger a full A J M Watson colonoscopy. They recommend that the initial examination should be an unsed- ................................................................................... ated examination with a colonoscope Only half of proximal colon cancers are associated with adenomas after a simple enema. This study is consistent with previous in the distal colon. This has important implications for the selection findings that 46–52% of PAN are not of the initial investigation for colorectal cancer screening accompanied by distal polyps.10 11 Addi- tion of faecal occult blood testing to flex- f all the common cancers, colo- association of distal adenomas with proxi- ible sigmoidoscopy does not significantly rectal cancer is the best suited to mal colorectal neoplasia [see page 398]. increase the detection of advanced Oprevention through screening as The investigators took advantage of the neoplasia.12 One is therefore left with the it is derived from benign adenomas Norwegian Colorectal Cancer Prevention conclusion that colonoscopy remains the which can be easily detected and re- study (NORCCAP) in which 20 780 indi- most sensitive screening tool and, if per- moved. The best screening investigation viduals aged 54–64 years, selected randomly formed by a skilled operator, is reason- remains much debated. Many argue that from the population registry of Oslo and ably safe. No screening technique will colonoscopy is superior to other tech- Telemark County, were offered a once only entirely eliminate the risk of colorectal niques because it has the highest sensi- examination by flexible sigmoidoscopy or a cancer. Risk reduction is all that can be tivity (>90%) and examines the whole combination of flexible sigmoidoscopy and achieved and this must be carefully colon.1 However, it has a number of faecal occult blood testing. Individuals explained to patients. Flexible sig- important disadvantages. Firstly, it is diagnosed as having an adenoma of any size moidoscopy is safer, cheaper, and more potentially dangerous. Perforation rates were offered full colonoscopy. The current convenient for patients than colonoscopy of 1 in 1000–1 in 20 000 have been found study examined the risk of proximal adeno- but at the cost of lower efficacy for in large studies from the USA and mas and carcinomas in the 2154 individuals preventing and detecting cancer. 23 Germany. Colonoscopy is also expen- (17% of the total screenees) who were Gut 2003;52:317–318 sive and requires highly skilled operators found to have distal neoplasms. Of these, who are in short supply.4 For these 1833 individuals were studied. Twenty one ..................... reasons investigators have sought a per cent of subjects had colonic neoplasms http://gut.bmj.com/ screening strategy that reduces the proximal to the level reached by flexible sig- Author’s affiliation A J M Watson, Department of Medicine, number of colonoscopies undertaken. A moidoscopy and a further 5% of subjects University of Liverpool, Daulby St, Liverpool study from St Mark’s Hospital of the long had proximal advanced neoplasms (PAN) L69 3GA, UK; [email protected] term risk of colorectal cancer in patients defined as high risk adenomas or carcino- with rectosigmoid adenomas found that mas. The risk of PAN increased threefold in REFERENCES 88% of cancers developed in patients subjects with distal adenoma >10 mm in 1 Ransohoff DF, Sandler RS. Clinical practice. with high risk (villous, tubulovillus diameter or containing villous components. Screening for colorectal cancer. N Engl J Med histology, or >10 mm in diameter) The investigators then calculated the 2002;346:40–4. on October 5, 2021 by guest. Protected copyright. rectosigmoid adenomas.5 This study led number of PAN that would have been 2 Tran DQ, Rosen L, Kim R, et al. Actual colonoscopy: what are the risks of Atkin et al to propose that a single exam- missed depending on the threshold criteria perforation? Am Surg 2001;67:845–7. ination with a flexible sigmoidoscopy for offering colonoscopy. If the threshold 3 Sieg A, Hachmoeller-Eisenbach U, Eisenbach leading to full colonoscopy in patients criteria for colonoscopy had been more than T. Prospective evaluation of complications in outpatient GI endoscopy: a survey among with high risk rectosigmoid adenomas one adenoma or a single high risk adenoma German gastroenterologists. Gastrointest would be a cost effective and safe (as defined by a diameter >10 mm or Endosc 2001;53:620–7. screening protocol.6 This strategy is now villous components or showing severe dys- 4 Detsky AS. Screening for colon cancer—can we afford colonoscopy? N Engl J Med being tested in a randomised controlled plasia) then 38% of PAN would have been 2001;345:607–8. clinical trial. Baseline findings have missed, including 17% of proximal carcino- 5 Atkin WS, Morson BC, Cuzick J. Long-term already established the perforation rate mas. Furthermore, the tendency to miss risk of colorectal cancer after excision of rectosigmoid adenomas. N Engl J Med of diagnostic flexible sigmoidoscopy to PAN was found to increase with the age of 1992;326:658–62. be considerably lower than that of colon- the subject. On the other hand, implemen- 6 Atkin WS, Cuzick J, Northover JM, et al. oscopy at 1 in 40 000.7 tation of these strict threshold criteria Prevention of colorectal cancer by once- only sigmoidoscopy. Lancet 1993;341:736–40. The crucial assumption for the use of would have resulted in 66% fewer colono- 7 Single flexible sigmoidoscopy screening to flexible sigmoidoscopy as the initial screen- scopies being undertaken. prevent colorectal cancer: baseline findings of ing test is that all proximal cancers are A particularly interesting feature of a UK multicentre randomised trial. Lancet associated with distal adenomas. If this is the study was that colonoscopes were 2002;359:1291–300. 8 McCallion K, Mitchell RM, Wilson RH, et al. not true then the findings at flexible used to perform many of the flexible sig- Flexible sigmoidoscopy and the changing sigmoidoscopy will not trigger the colonos- moidoscopic examinations with the ex- distribution of colorectal cancer: implications copy required to make the diagnosis of tent of examination limited by the for screening. Gut 2001;48:522–5. 9 Gondal G, Grotmal T, Hofstad B, et al. proximal colorectal cancer. This assumption degree of bowel cleansing from a single Grading of distal colorectal adenomas as is brought into sharper focus by the sorbitol enema. In this way the investiga- predictors for proximal colonic neoplasia and increasing proportion of colorectal cancers tors were able to examine a greater pro- choice of endoscope in population screening: 8 experience from the Norwegian Colorectal arising in the right colon. In this issue of portion of the colon than is usually pos- Cancer Prevention study (NORCCAP). Gut Gut, Gondal and colleagues9 investigated the sible with a conventional flexible 2003;52:398–403. www.gutjnl.com 318 COMMENTARIES 10 Imperiale TF, Wagner DR, Lin CY, et al. Risk 11 Lieberman DA, Weiss DG, Bond JH, et al. 12 Lieberman DA, Weiss DG. One-time of advanced proximal neoplasms in Use of colonoscopy to screen asymptomatic screening for colorectal cancer with combined asymptomatic adults according to the distal adults for colorectal cancer. Veterans Affairs fecal occult-blood testing and examination of colorectal findings. N Engl J Med Cooperative Study Group 380. N Engl J Med the distal colon. N Engl J Med 2000;343:169–74. 2000;343:162–8. 2001;345:555–60. Gut: first published as 10.1136/gut.52.3.318 on 1 March 2003. Downloaded from Liver disease The presence of a base pair change ................................................................................... (R6S) is not synonymous with a major metabolic effect on iron metabolism. Further studies will be required on this Lessons from liver transplantation: polymorphism to determine if functional changes occur in the HFE protein, and it flip, flop, and why? would be uncommon for heterozygotes of any HFE mutation to have significant P C Adams changes in iron metabolism. For exam- ple, the H63D mutation does not lead to ................................................................................... iron overload in H63D heterozygotes. A case of hereditary haemochromatosis developing in a To further complicate the clinical pic- ture, recently a ferroportin mutation has non-hereditary haemochromatosis recipient following been described leading to moderate to transplantation of a C282Y heterozygous donor liver severe iron overload in patients. Many of the patients in the original report had a he widespread application of liver trans- had mild iron overload secondary to normal transferrin saturation and plantation has been the single most alcoholic siderosis and was found to have ferritin.8 If iron tests cannot predict the Timportant therapy to extend long term a new polymorphism in the HFE gene development of iron overload, there may survival in patients with a variety of acute (R6S). Four years later the recipient was be cases of occult iron overload in the and chronic liver diseases. A fringe benefit of found to have severe hepatic siderosis. general population or post liver transplant liver transplantation has been identification, The tantalising aspect of this unusual that have not come to medical attention. confirmation, and cure of the metabolic case report is the suggestion that two The unusual case described in this basis of diseases.1 Haemophilia patients independent mutations could be synergistic report may never occur again but this transplanted for chronic hepatitis C are to result in iron overload.