Cancers of the colon and rectum are the second most common cancers in New Zealand (NZ). The most recent figures released by the NZ Ministry of Health show 2716 new cases were registered in 20051 (66 cases / 100,000 population2). The majority of cases are sporadic but 8-12% are associated with genetic syndrome. The most common of the known colorectal cancer (CRC) predisposing syndromes is Lynch syndrome,3 also known as hereditary non-polyposis colorectal cancer (HNPCC). This is an autosomal dominant syndrome, which confers a predisposition to colonic and other cancers. Lynch syndrome is responsible for 2-3% of the colorectal cancer burden. Typically, families with Lynch syndrome are diagnosed with colonic cancers at an earlier age than sporadic colonic cancers and multiple tumours (either synchronous or metachronous) are more frequent than in sporadic colon cancer patients. Other tumours associated with the syndrome include endometrial carcinoma with a lifetime risk of 40-60%4 as well as a modestly increased risk of ovarian, stomach, small bowel, hepatobiliary, renal pelvic, and ureteric tumours.5 A personal or family history of any of these malignancies in a patient with colon cancer should raise the suspicion of Lynch syndrome. Lynch syndrome is caused by germline mutations in one of several DNA mismatch repair genes which results in faulty DNA repair. Laboratory techniques including microsatellite instability (MSI) testing and immunohistochemistry (IHCwhich tests for the expression of DNA repair proteins suggesting the possibility of MSI) can help identify patients with suspected Lynch syndrome. Thus genetic counseling, genetic testing and appropriate surveillance can be offered to families. The revised Bethesda criteria6 (Table 1) are the currently accepted guidelines used when deciding which patients to screen for potential Lynch syndrome. Table 1. Revised Besthesda criteria Lynch syndrome should be suspected and MSI testing of the tumour carried out if any of the following criteria apply: 1 Colorectal cancer diagnosed in a patient under 50 years of age 2 Presence of synchronous, metachronous colorectal cancer or other HNPCC associated tumours*, regardless of age 3 Colorectal cancer with MSI-H like histology diagnosed in a patient less than 60 years of age 4 Colorectal cancer diagnosed in a patient with one or more first degree relatives with an HNPCC associated tumour*, with one of the tumours being diagnosed under 50 years of age 5 Colorectal cancer diagnosed in a patient with two or more, first or second degree relatives with HNPCC associated tumours*, regardless of age * HNPCC associated tumours are; endometrial; ovarian; stomach; small bowel; hepatobiliary; renal pelvis; and ureteric Surgeons who treat patients with colorectal cancer need to be alert to the possibility of Lynch syndrome and arrange appropriate testing in response to clinical triggers. Previously this required identification of those that fit the Bethesda criteria and subsequent referral onto a genetic service. The genetic service would then take a detailed family history and determine if genetic testing was required. IHC staining is a relatively rapid method of screening for those that may benefit from genetic testing. IHC looks for the presence or absence of certain proteins with the results influencing the decision making process about the need to refer for a genetics opinion. We questioned the current knowledge and practice of general and sub-specialty colorectal surgeons in NZ with regard to IHC testing and indications for referral to genetic services. Methods Using the revised Bethesda criteria a questionnaire (Fig 1) was designed. In designing this tool we consulted geneticists in our institution. There was concern regarding the misrepresentation of IHC as a test for MSI, which it is not. IHC shows loss of expression of the protein product of one or more mismatch repair genes. Positive IHC staining merely indicates the possibility of MSI in the DNA and hence the chance of finding Lynch syndrome mutations. In the interest of trying to keep the questionnaire at a length likely to be responded to we agreed not to have lengthy explanations before any questions and use the word possible in question 2. We felt that this was an acceptable compromise as the question therefore did not state categorically IHC was a test for micro satellite instability but an abnormal result meant that MSI was a possibility. Our finalised questionnaire was then sent to all NZ surgeons identified by the Royal Australasian College of Surgeons as general surgeons, with or without a sub-specialty interest in colorectal surgery (CRS). The questionnaire was sent out by the college to preserve confidentiality. The only identifier was a small 'c' on the questionnaire of those surgeons who had declared sub-specialty interest in colorectal surgery to the college. One month after the original mailing a follow-up letter was distributed to improve the response rate. Figure 1. Questionnaire The returned questionnaires were analysed against the revised Bethesda criteria and the results tabulated. Statistical significance was set at p<0.05. Chi-squared tests were used to test for this except when a cell had <5 observed responses Fisher's exact test was used. Results 183 questionnaires were distributed and after reminder letters 94 (51%) were returned. Of the 94 responses, 77 surgeons were involved in treating patients with CRC, 23 of these were identified as having a sub-specialty interest in colorectal surgery. The remaining 17 were no longer practicing or not involved with CRC. As this was a study of current practice the responses of the 17 surgeons no longer practicing or not treating CRC were excluded from the final analysis. Overall, 68/77 (88%) surgeons knew about IHC testing. Predictably, those surgeons with a specialist interest in colorectal surgery had a greater awareness of the availability of IHC testing than those with no declared interest (96% vs 85%, p=0.3) The results for the practice based questions are shown in Table 2. Forty-five percent of the respondents correctly identified that a colorectal cancer diagnosed before the age of 50 was one of the Besthesda criteria. A similar proportion (40%) thought a patient should be younger than 45 years of age. Table 2. Correct responses (raw numbers and (%)) to the practice based questionnaire Criteria to suspect Lynch syndrome All surgeons n=77 (%) General surgeons n=54 (%) Colorectal interest n=23 (%) CRC diagnosed <50 years (Yes) 35 (45) 23 (43) 12 (52) Right sided CRC (no) 49 (64) 33 (61) 16 (70) Metachronous CRC (yes) 54 (70) 40 (74) 14 (61) One 1st degree relative CRC (no) 36 (47) 25 (46) 11 (48) FHx bladder cancer (no) 67 (87) 46 (85) 21 (91) FHx ovarian cancer (yes) 37 (48) 29 (54) 8 (35) FHx prostate cancer (no) 74 (96) 51 (94) 23 (100) FHx renal tract cancer (yes) 25 (32) 18 (33) 7 (30) FHx endometrial cancer (yes) 43 (56) 31 (57) 12 (52) Hx IBD (no) 72 (94) 49 (91) 23 (100) For the remaining questions the correct response rate ranged from 32-96%. The questions with the lowest correct response rate are those about Lynch syndrome associated cancers. Only 32% knew renal tract cancers were associated with Lynch syndrome , 48% knew about ovarian cancer and 56% about endometrial cancer. There is no statistically significant difference between the response rates for those with and without a declared interest in colorectal surgery. After receiving a positive IHC result 63/77 (83%) surgeons would discuss screening/surveillance with the patient and their family, 15/77(19%) would refer the case onto a tertiary surgical unit and 70/77(93%) would refer to the regional genetic service on the basis of such a result. Discussion IHC staining of tumour tissue for the presence or absence of proteins indicating the possibility of a mismatch repair gene mutation is quick, inexpensive and efficient, being 100% specific and 94% sensitive7 compared to genetic testing. Therefore it is reasonable to regard IHC as an initial screening test for Lynch syndrome. IHC testing and results should be regarded as genetic information and thus appropriate consent should be obtained from patients prior to testing. Ideally this should be done by the operating surgeon. However, to obtain valid consent the surgeon needs to be aware of whom to offer the test to. Our results show that 40% of surgeons would not offer testing to patients older than 45 years and, depending on the other types of cancer in the family, 50-75% or potentially eligible patients would not be identified and tested. The Royal College of Pathologists of Australasia released a position statement in 2006,8 which recommends no additional consent is required to perform IHC. However, with the exception of age, pathologists are not routinely given the necessary clinical information to make a decision about IHC.9 The surgical team cannot hope to provide this information on specimen forms if suspicious details in the patients history are overlooked or not enquired about. The survey results show that New Zealand surgeons know IHC is available but knowledge about who to test is lacking. It could be argued that the results are simply a reflection of the low response rate to the questionnaire. Surveys of New Zealand Surgeons have in the past achieved response rates of 70%.10 However, these previous surveys are of all fellows of the college and not just general surgeons. When surveying General Surgeons in NZ this may be the best we can hope to achieve. Even if we had had a 100% response rate New Zealand surgeons show gaps in their knowledge of the Besthesda criteria. The 89 non-responding surgeons were included and a sensitivity analysis (not shown here) was undertaken. The results showed that there would still be approximately 30% of patients with suspicious criteria who would go unidentified. It is possible that some more knowledgeable surgeons found the stems of question 3 harder to answer (with a potential skewing effect on the results) as the main question could be taken to imply that IHC tests directly for MSI, which as we have noted, is incorrect. As the focus of this paper is knowledge and education we have made efforts to clarify this point in our introduction and methodology. Conclusion This study has shown a lack of awareness about aspects of the Besthesda criteria amongst New Zealand general surgeons. Any delay in the diagnosis of Lynch syndrome may have a negative impact on patients and their families. For those treating colorectal cancer it is prudent, if not essential, to be aware of the classic features of Lynch syndrome and test appropriately.