腔内水囊磁共振成像对直肠癌术前分期的诊断

目的评估腔内水囊磁共振成像（MRI）预测直肠癌术前分期的准确程度。方法回顾性分析19例直肠癌患者的临床资料，对照术前MRI和术后病理结果．评估MRI能否准确预测直肠癌T和系膜淋巴结转移（N）分期。结果MRI正确T分期15例（78．9％），错误4例，其中2例T1-2期报告为T3期，2例T3期报告为T1-2期；T1-2期的准确率为80％（8／10），B期的准确率为7／9。MRI正确N分期11例（57．9％），错误分期8例；其中高估2例，低估6例，灵敏度为3／9，特异度为80．0％（8／10）。结论术前腔内水囊MRI不能对直肠癌系膜淋巴结转移做出准确预测，可以对T分期进行一般预测。     

高分辨率MRI显示直肠周围筋膜结构

目的观察直肠周围筋膜高分辨率MRI成像表现。方法对50名正常志愿者行盆腔MRI常规T1WI、T2WI及高分辨率T2WI扫描,观察直肠周围筋膜结构,比较各序列对直肠周围筋膜的显示情况。结果高分辨率T2WI序列对前、后、左、右方直肠系膜筋膜显示率分别98%、100%、94%、90%;对Denonvilliers筋膜、骶前筋膜及腹膜返折显示率分别为58%、54%、70%。高分辨率T2WI对前、左、右方筋膜、Denonvilliers筋膜及腹膜返折的显示率明显高于盆腔常规T1WI、T2WI(P均0.05);对后方直肠系膜筋膜、骶前筋膜显示率略高于常规T1WI、T2WI,但差异无统计学意义(P均0.05)。结论高分辨率T2WI显示直肠周围筋膜结构优于盆腔常规MRI序列。     

Intersphincteric resection for low rectal cancer: laparoscopic vs open surgery approach

Aim Laparoscopic sphincter‐saving surgery has been investigated for rectal cancer but not for tumours of the lower third. We evaluated the feasibility and efficacy of laparoscopic intersphincteric resection for low rectal cancer. Method From 1990 to 2007, patients with rectal tumour below 6 cm from the anal verge and treated by open or laparoscopic curative intersphincteric resection were included in a retrospective comparative study. Surgery included total mesorectal excision with internal sphincter excision and protected low coloanal anastomosis. Neoadjuvant treatment was given to patients with T3 or N+ tumours. Recurrence and survival were evaluated by the Kaplan–Meier method and compared using the Logrank test. Function was assessed using the Wexner continence score. Results Intersphincteric resection was performed in 175 patients with low rectal cancer: 110 had laparoscopy and 65 had open surgery. The two groups were similar according to age, sex, body mass index, ASA score, tumour stage and preoperative radiotherapy. Postoperative mortality (zero) and morbidity (23%vs 28%; P = 0.410) were similar in both groups. There was no difference in 5‐year local recurrence (5%vs 2%; P = 0.349) and 5‐year disease‐free survival (70%vs 71%; P = 0.862). Function and continence scores (11 vs 12; P = 0.675) were similar in both groups. Conclusion Intersphincteric resection did not alter long‐term tumour control of low rectal cancer. The safety and efficacy of the laparoscopic approach for intersphincteric resection are suggested by a similar short‐ and long‐term outcome as obtained by open surgery.     

MRI of extraperitoneal rectal carcinoma

Last years technological developments in imaging field have made a substantial contribution to diagnosis and staging of rectal cancer. Endorectal ultrasound and MRI with endorectal coil are very useful in rectal cancer initial staging thanks to their ability to distinguish between the rectal wall layers. Major ultrasound limitations are presence of inflammations, desmoplastic reaction and small field of view which limits evaluation of perirectal invasion. MRI with phased-array coils, instead, allows depiction of mesorectum and to assess the distance between tumor and mesorectal fascia. Unfortunately CT shows low accuracy compared to MRI in local staging because it fails to distinguish the rectal wall layers. The criterion used in assessing nodal involvement remains unfortunately still the dimensional one even if new contrast media based on nano-iron particles look promising in this regard On reassessment after chemo-radiotherapy treatment, MRI proved to be a very accurate tool thanks to its ability to detect tumor downstaging, disappearance of mesorectal fascia infiltration or even to show a complete response. The presence of recurrence can be studied by contrast enhanced perfusion-MRI or with good accuracy using PET which, however, presents major technical limitations at present.     

A critical appraisal of endorectal ultrasound and transanal endoscopic microsurgery and decision-making in early rectal cancer

Aim Transanal endoscopic microsurgery (TEM) for early rectal cancer (ERC) gives results similar to major surgery in selected cases. Endorectal ultrasound (ERUS) is an important part of the preoperative selection process. This study reports its accuracy and impact for patients entered on the UK TEM database. Method The UK TEM database comprises prospectively collected data on 494 patients. This data set was used to determine the prevalence of ERUS in preoperative staging and its accuracy by comparing preoperative T‐stage with definitive pathological staging following TEM. Results ERUS was performed in 165 of 494 patients who underwent TEM for rectal cancer. It inaccurately staged rectal cancer in 44.8% of tumours: 32.7% were understaged and 12.1% were overstaged. There was no significant difference in the depth of TEM excision or R1 rate between the patients who underwent ERUS before TEM and those who did not (P = 0.73). Conclusion The data show that ERUS is employed in a minority of patients with rectal cancers undergoing TEM in the UK and its accuracy in this ‘Real World’ practice is disappointing.     

Preoperative Staging of Rectal Cancer With MRI: Accuracy and Clinical Usefulness

Background: Preoperative staging is essential for planning of optimal therapy for patients with rectal cancer. Recently, magnetic resonance imaging (MRI) is used frequently because of its benefits of clear pelvic image are better than other diagnostic methods. The purpose of this study was to determine accuracy rates and clinical usefulness of MRI in preoperative staging of rectal cancer.Methods: Between February, 1997, and December, 1999, 217 patients with histologically proven rectal cancer were staged preoperatively and had surgical resections performed. MRI criteria for depth of invasion was determined by the degree of disruption of the rectal wall. Metastatic perirectal lymph nodes were considered to be present if they showed heterogenous texture, irregular margin, and enlargement (.10 mm).Results: The accuracy of the MRI for determining depth of invasion was 176/217 (81%) and regional lymph node invasion was 110/217 (63%). In the T stage, accuracy rate of T1 was 3/4 (75%), T2 was 20/37 (54%), T3 was 141/162 (87%), and T4 was 12/14 (86%), respectively. The specificity of lymph node invasion was 45/110 (41%) and the sensitivity was 91/107 (85%). The accuracy rate of regional lymph node involvement was 136/217 (63%). T1 and T2 were overstaged in 1/4 (25%) and 17/37 (46%), respectively, and T3 was understaged in 15/162 (9.2%). The accuracy rate to detect metastatic lateral pelvic lymph node was 4/14 (29%) after lateral pelvic lymph node dissection was done in 14 patients under MRI. The accuracy rate in assessing levator ani muscle tumor involvement was 8/11 (72%).Conclusions: MRI showed a good, comparable accuracy rate for determining depth of tumor invasion, compared with transrectal ultrasonography, which still has a low accuracy rate for detecting metastatic lymph node. MRI with endorectal coil may increase the accuracy rate of T1 and T2 lesions. In addition, clear sagittal and coronal sectional pelvic images can give a lot of information about adjacent organ invasion or any invasion of levator ani muscle. MRI can be useful for choosing an appropriate extent of lymph node dissection and type of surgery.     

Positron‐emission tomography/computed tomography (PET/CT) in the initial staging of primary rectal cancer

Objective The aim of this study was to assess the role of ¹⁸flourodeoxyglucose positron‐emission tomography/computed tomography (PET/CT) in the initial staging of primary rectal adenocarcinoma. Method A total of 20 patients with adenocarcinoma of the rectum were assessed with both PET/CT and conventional staging (CT chest/abdomen/pelvis, MRI rectum). Discordance with conventional imaging and incidental findings on PET were recorded and the patients presented to a colorectal cancer multidisciplinary team to assess management changes. Patients were followed up so that discordant or incidental findings could be verified by intra‐operative examination, imaging or histology where possible. Results Positron‐emission tomography/computed tomography correctly identified the primary tumour in all 20 patients. Comparing PET/CT with conventional staging modalities, there were 11 discordant or incidental findings in nine patients (45%). This resulted in a potential change in stage in 30% (four patients downstaged and two upstaged). PET/CT suggested additional neoplastic pathology in three patients and excluded the same in two patients. The incidental neoplastic findings were of minor clinical significance and one was eventually deemed false positive. While PET/CT resulted in potential management changes in five patients (25%), no changes in surgical management occurred. When tumours were grouped according to conventional stage, PET/CT resulted in fewer changes in stage in stage I (0%), compared with stages II to IV (43%) (P = 0.08). Conclusion Positron‐emission tomography/computed tomography provides additional information to conventional staging in primary rectal cancer. This information produced minor management changes in this study and did not effect surgical management. PET/CT may be most appropriately used selectively in more advanced stages and where indeterminate findings exist with conventional staging.     

Preoperative staging of rectal cancer: pelvic MRI plus abdomen and pelvic CT. Does extrahepatic abdomen imaging matter? A case for routine thoracic CT

Objective To assess the quality of preoperative magnetic resonance imaging (MRI) staging of rectal cancer, and the clinical significance of abdomen and pelvic computed tomogram (CT) scans in preoperative staging of rectal cancer in a district general hospital. We postulated that the ‘metastatic yield’ of extrahepatic abdominal imaging is poor, and rarely altered management of rectal cancer. Methods This is a retrospective study of preoperative MRI, CT scans and postoperative histology results of patients who had definitive surgery for rectal cancer at the Mid‐Staffordshire General Hospitals NHS Trust over a 36‐month period. Preoperative multiplanar pelvic MRI locoregional staging was compared with eventual histology. The incidence of and significance of abdomen and pelvic CT detected pathology (including metastasis) in the management of rectal cancers was also assessed. Results Preoperative pelvic MRI correctly predicted ‘clear’ Circumferential resection margins, in 28 of 29 patients who had primary surgery. This is comparable with many published studies. Significant CT detected pathology (including metastasis) on preoperative abdomen and pelvic CT scans was uncommon, and did not influence management of any rectal cancer patient in our study. Discussion Given that exclusive CT detected significant pathology caudal to the liver (extrahepatic abdomen) is rare, can full abdomen and pelvic CT scans be justified for preoperative staging of rectal cancers? – especially where chest X rays are employed for lung staging. Preoperative thoracic and upper abdomen CT scan may be a more productive use of resources. Full abdominal scans may be more appropriate for selection of rectal cancer patients with isolated liver metastasis for metastasectomy.     

Clinical significance of macroscopic completeness of mesorectal resection in rectal cancer

Aim Local recurrence after resection of rectal cancer is usually regarded as being due to a ‘failure’ of surgery. The completeness of resection of the mesorectum has been proposed as an indicator of the ‘quality’ of the resection. We determined the prognostic value of macroscopic evaluation of rectal cancer resection specimens and the circumferential resection margin (CRM) after curative surgery. Method From 1999 to 2006, the macroscopic quality of the mesorectum and the CRM were prospectively assessed in 127 patients who underwent rectal cancer resection with curative intent (R0+R1). Chemoradiotherapy was administered for 61 tumours staged as locally advanced tumours (T3, T4 and N+). Univariate analysis of time to local recurrence and cancer‐free survival were tested (Kaplan–Meier) and multivariate analysis calculated with a Cox regression model. Results The mesorectum was incomplete in 34 (26.8%) patients. At a median follow up of 34 months (range, 9–96 months), in the group with an adequate mesorectal excision, the cumulative risk of local recurrence at 5 years was 10%. This was 25% if the mesorectum was incomplete (P < 0.01). Five‐year cancer‐free survival was 65% if the mesorectal excision was adequate and 47% if it was not (P < 0.05). Multivariate analysis identified T status, the CRM and the mesorectal score as independent factors for local recurrence, and T and N status and the mesorectal score as independent factors for disease‐free survival. Conclusion The outcome of surgical treatment of rectal cancer is related to the completeness of mesorectal excision. It is a more discriminative prognostic factor than the classic tumour–node–metastasis (TNM) system.     

术前高分辨率MRI评估T3期直肠癌新辅助治疗效果

目的观察术前高分辨率MRI评估新辅助治疗用于T3期直肠癌效果的价值。方法 165例T3期直肠癌患者于新辅助治疗前接受高分辨率MR检查,观察直肠系膜受累情况、肿瘤下缘距肛缘距离、肿瘤直径,评估T分期、N分期;之后手术切除肿瘤,术后以结直肠癌TNM分期标准行病理学再分期及肿瘤退缩评分(TRG),系统评估新辅助治疗效果。采用单因素及多因素Logistic回归分析观察新辅助治疗效果的影响因素。结果新辅助治疗后,根据TNM分期评价标准,63例疗效好,102例效果差。单因素分析结果显示,治疗效果的影响因素包括患者年龄和肿瘤N分期、T3亚分期及肿瘤直径(P均0.05);多因素分析结果显示,患者年龄及肿瘤T3亚分期、肿瘤直径是TNM分期评估疗效的独立影响因素(P均0.05)。以TRG系统评估新辅助治疗效果,患者年龄、肿瘤标记物及肿瘤T3亚分期、N分期、肿瘤直径、直肠系膜受累情况等均与之无明显相关(P均0.05)。结论高分辨率MRI可用于术前预测T3期直肠癌患者接受新辅助治疗的效果。     

Apical node metastasis independently predicts poor survival in Dukes C colorectal cancer

Aim The prognostic significance of apical node metastasis in node‐positive colorectal cancer (CRC) is disregarded by the Fourth American Joint Committee on Cancer and the International Union Against Cancer (AJCC/UICC) TNM classification system. The influence of apical node metastases on overall 5‐year survival among patients with Dukes stage C CRC was examined. Method Patients who underwent operative resection for CRC between 1999 and 2003 were reviewed. Results Two‐hundred and ninety patients were included in the study, including 203 with Dukes C apical node‐negative cancers, 39 with Dukes C apical node‐positive cancers and 48 with Dukes D cancers. The respective prevalence of extramural vascular invasion was 35%vs 64%vs 56% (P = 0.0005), T4‐stage 24%vs 38%vs 48% (P = 0.013), positive resection margin 16%vs 41%vs 23% (P = 0.001), more than three positive nodes harvested 28%vs 85%vs 52% (P < 0.0001) and poorer tumour differentiation grade 9%vs 21%vs 23% (P = 0.009). Multivariate analyses of all Dukes C cancer patients (n = 242) showed a positive apical node to be a highly significant independent predictor of mortality (hazard ratio 2.281, 95% confidence interval 1.421–3.662, P = 0.0006). Extramural vascular invasion and a positive resection margin were also independent predictors of poor survival. Patients with Dukes C apical node‐positive cancers had a significantly poorer overall 5‐year survival compared to patients with Dukes C apical node‐negative cancers (P < 0.0001) but survival was not significantly different compared to patients with distant metastases at initial presentation (P = 0.504). Conclusion Apical node metastasis appears to be a strong independent, negative prognostic factor of poor survival in Dukes C CRC.     

Neoplastic Mesorectal Microfoci (MMF) following Neoadjuvant Chemoradiotherapy: Clinical and Prognostic Implications

Background Neoplastic microfoci have frequently been found in the mesorectum, with poor outcome. In this study, incidence and clinical significance of mesorectal microfoci (MMF) were analyzed in patients operated on for rectal cancer following neoadjuvant chemoradiation. Methods A case series of 68 patients with extraperitoneal rectal cancer treated with neoadjuvant chemoradiation and surgery (including total mesorectal excision) were investigated for presence of neoplastic MMF. Results MMF were found in 26 cases (38.2%). Increasing incidence of microfoci was statistically related to pathologic involvement of the bowel wall (P = 0.0006), Mandard’s tumor regression grading (P = 0.0006), and pathologic neoplastic mesorectal involvement (P < 0.00001). None of the nine patients with complete tumor disappearance displayed both microfoci and lymph node metastasis. Only one local recurrence developed in a patient with multiple MMF. One out of nine pT0 or TRG1 patients (11.1%) had distant metastases compared with 15 out of 59 pT1–4 or TRG2–5 (25.4%, P = 0.70). Conclusions A remarkable incidence of MMF was found following chemoradiation. However, when this therapy induced complete regression of primary tumor (pT0–TRG1), we found that node metastases and neoplastic MMF also disappeared. These features should be confirmed to assess the impact of these microfoci in treatment decision making in rectal cancers.     

Prognostic implications of circumferential location of distal rectal cancer

Aim This study evaluated the prognostic importance of circumferential tumour position of mid and low rectal cancers. Method All uT2, uT3 and uT4 tumours of the middle and lower rectum that underwent total mesorectal excision (TME) with curative intent between 1996 and 2006 were included. The predominant circumferential tumour position (anterior, posterior or circumferential) was defined on preoperative endorectal ultrasound examination (ERUS). The relationships between tumour position and other characteristics and recurrence were explored. Results Two hundred and five patients with distal rectal cancer were operated on for a uT2‐T4 tumour. Median follow up was 49 months. The location of the tumour was predominantly anterior, posterior or circumferential in 128, 49 and 27 patients, respectively. Anterior tumours were more likely to receive neoadjuvant therapy (P = 0.016) and perioperative blood transfusion (P = 0.012). No significant differences were observed between circumferential position and pT or pN stage, circumferential resection margin involvement or mesorectal excision quality. Sixty‐three (30.7%) patients developed recurrence, which was local only in 16 (7.8%). Although tumours involving 360° of the rectal wall had a higher risk of local recurrence (P = 0.048), those with a predominant anterior or posterior position were not related to a higher risk of local or overall recurrence. Conclusion Anterior rectal tumours do not differ in pathological characteristics from posterior tumours, and their prognosis is no worse when circumferential resection is complete.     

Changes in the management and outcome of rectal cancer over a 10‐year period in Oxford

Aim The changes in the management and outcome of rectal cancer in Oxford were studied over a 10‐year period. Method Rectal cancer data using a prospectively collected data base were divided into curative (global) and palliative groups. The global curative group was further divided into those with and without (selected group) the following features: emergency cases, local excision, salvage surgery for recurrence or incomplete local excision, metastatic disease, perioperative death, hereditary cancer, inflammatory bowel disease–related cancer, and synchronous cancer. Results Between 1994 and 2003, 709 cases of rectal cancer were treated, 532 for cure and a selected group of 393 after removing patients with the aforementioned exclusions. For the selected group, the average follow‐up was 51.2 months, overall survival 65.4% and cancer‐specific survival 75.3%. There was no 2‐year survival difference between each of the 10‐year periods of study. Two‐year local recurrence was 5.6% for the first 5‐year period and 2.3% for the second (P = 0.11). MRI staging increased during the 10 years (0% in 1994; 66.7% in 2003) as did use of definitive chemoradiotherapy (dCRT) (0% in 1994; 64.7% in 2003). The anastomotic leakage rate was significantly higher in the second 5‐year period (2.6%vs 9.6%; P = 0.01). Conclusion Despite increasing use of MRI and dCRT, 2‐year survival and local recurrence were not significantly different within the 10 years studied.     

Transverse coloplasty pouch and colonic J-pouch for rectal cancer – a comparative study

Objectives The introduction of the colonic J‐pouch has markedly improved the functional outcome of restorative rectal cancer surgery. However colonic J‐pouch surgery can be problematic and may present some late evacuatory problems. To overcome these limitations a novel pouch has been proposed: the transverse coloplasty pouch. The purpose of our study was to compare the functional outcomes of these two different types of pouches – the transverse coloplasty pouch (TCP) and the colonic J‐pouch (CJP) – during the first 12 months postoperatively. Patients and methods A prospective randomized trial was conducted in which a total of 30 patients with mid and low rectal cancer were submitted either to a transverse coloplasty pouch or a colonic J‐pouch. Clinical defaecatory function was assessed and anorectal physiological assessment was carried out, pre‐operatively and at 3, 6 and 12 months postoperatively, by means of a standard clinical questionnaire and by anorectal manometry. Results No statistically significant differences were found between the two groups regarding bowel function. The postoperative frequency of daily bowel movements was lower in the TCP group in all the phases of the study (3.9 vs. 4.1 at 3 months; 3.1 vs. 3.4 at 6 months; 2.1 vs. 2.8 at 12 months), the same occurring with fragmentation (33%vs. 40% at 3 months; 26.6%vs. 33.3% at 6 months; 7.1%vs. 14.3% at 12 months). Less urgency was also seen in the TCP group during the first 6 months (20%vs. 26.7%), with identical values at 12 months (14.3%vs. 14.3%). No significant differences were also found concerning incontinence grading and scoring, with TCP patients having less nocturnal leaks. At one year two CJP patients (14.3%) needs the use of enemas to evacuate the pouch and provoke defaecation, a problem never seen in TCP patients. The anorectal manometry data was similar in both types of pouches. The local complication rates were also identical in the two groups (20%); more anastomotic leaks were seen in TCP patients (13.2%vs. 6.6%), without reaching a statistical significance. Conclusion The transverse coloplasty pouch has similar functional results but fewer evacuation problems than the J‐Pouch, making it a safe and reliable alternative to the colonic J‐pouch.     

CT和MRI对直肠癌术前T、N分期价值的Meta分析

目的比较CT和MRI对直肠癌术前T、N分期的价值。方法收集公开发表的对CT与MRI进行术前直肠癌T、N分期价值比较的国内外所有前瞻性和回顾性研究,按照Meta分析的要求对检索到的原始研究的质量进行评估,对符合条件的所有研究结果进行Meta分析,分别计算在T1～T4、N分期上,CT与MRI的灵敏度和特异度并进行比较。结果符合纳入标准的共9篇文章,总样本量347例。对于T分期,CT的灵敏度为0.772,特异度为0.880;MRI的灵敏度为0.910,特异度为0.935。对于N分期,CT的灵敏度为0.636,特异度为0.734;MRI的灵敏度为0.563,特异度为0.849。结论对于T分期,MRI在灵敏度、特异度均优于CT;对于N分期,CT灵敏度优于MRI,MRI特异度优于CT。另外,直肠癌术前放化疗能显著影响CT和MRI术前T分期的灵敏度和特异度,对于N分期,则影响不明显。     

Outcome of right‐ and left‐sided colonic and rectal cancer following surgical resection

Aim To determine the outcome of surgery for colorectal cancer from a single region and to see whether location of the primary cancer influences prognosis. Method Patients with colorectal cancer diagnosed from January 2002 to December 2006, entered into a prospective database were followed until death or to December 2008. Right‐sided (caecum to transverse colon) and left‐sided (splenic flexure to rectosigmoid junction) colonic cancers and rectal cancers (distal to rectosigmoid junction to the anus) were identified. Statistical analysis was performed using Pearson’s chi‐square test, Kaplan–Meier (log‐rank statistic) and Cox regression analysis with a P‐value < 0.05 denoting significance. Results Of 841 patients with solitary colorectal cancers identified (median age 72 [30–101] years; 53% male), 283 (33.7%) were right‐sided colonic, 330 (39.2%) were left‐sided colonic and 228 (27.1%) were rectal. Respective resection rates were 82.7%, 77.9% and 91.6%, and curative resection rates were 79.9%, 82.9.0% and 85.7%, respectively. There was no significant difference in recurrence rates between right‐ (16.1%), left‐sided (23.0%) colonic and rectal (20.7%) cancers (P = 0.207). Respective mean survival rates were 54.4, 59.8 and 63.6 months (P = 0.007). Conclusion Right‐sided colorectal cancers had a worse prognosis than left‐sided and rectal cancers, possibly because of more advanced staging and fewer curative resections.     

Accuracy of radiological staging in identifying high‐risk colon cancer patients suitable for neoadjuvant chemotherapy: a multicentre experience

Aim A pilot study was undertaken to determine the accuracy of computed tomography (CT) staging in identifying patients with high‐risk colon cancers who would be considered as candidates for a neoadjuvant therapy trial (FOxTROT) and those at low risk (T1/T2) who would be excluded. Method Participating radiologists from 19 centres attended workshops for standardization of image interpretation according to previously defined prognostic criteria: good prognosis tumours, including, T1/T2; intermediate prognosis, T3 < 5 mm tumour invasion beyond the muscularis propria (MP); and poor prognosis tumours, including T3 with tumour extension ≥ 5 mm beyond the MP or T4. The CT findings were compared with histopathology as the reference standard. Results Of 94 patients with radiological and pathological data, 71% were categorized by CT as having a poor prognosis. The sensitivity and specificity of CT in identifying these tumours were 87% (95% CI, 74–94) and 49% (95% CI, 33–65). Sensitivity and specificity for tumour infiltration beyond the MP (T3/T4 vs T1/T2) were 95% (95% CI, 87–98) and 50% (95% CI, 22–77), respectively. Including all CT‐staged T3 and T4 patients in the trial would have increased the proportion eligible for entry to 89% (n = 84) without affecting the false‐positive rate of 7%. Some 20% of T3/T4 patients would have been ineligible for FOxTROT because of synchronous metastases. Conclusion In a multicentre setting, CT scanning identified high‐risk (T3/4) colon cancers with minimal overstaging of T1/T2 tumours, thus establishing the feasibility of radiologically guided neoadjuvant chemotherapy.     

Preoperative loco-regional imaging in rectal cancer

Preoperative staging of rectal cancer can influence the choice of surgery and the use of neoadjuvant therapy. This review evaluates the use of endorectal ultrasound (ERUS) and magnetic resonance imaging (MRI) in the local staging of rectal cancer. Staging for distant metastases is beyond the scope of this review. A MEDLINE search for published work in English between 1984-2004 was carried out by entering the key words of ERUS, MRI and preoperative imaging and rectal cancer. Initially, 867 articles were retrieved. Abstracts were reviewed and papers selected according to the inclusion criteria of a minimum of 50 patients and papers published in English. Papers focusing on preoperative chemoradiotherapy and distal metastases were excluded. Thirty-one papers were included in the systematic review. The examination techniques and images obtained are discussed and the respective accuracy is reviewed. ERUS and MRI have complementary roles in the assessment of tumour depth. Ultrasound has an overall accuracy of 82% (T1, 2, 40-100%; T3, 4, 25-100%) and is particularly useful for early localized rectal cancers. MRI has an accuracy of 76% (T1, 2, 29-80%; T3, 4, 0-100%) and is useful in more advanced disease by providing clearer definition of the mesorectum and mesorectal fascia. Both methods have similar accuracy in the assessment of nodal metastases. Ultrasound is more operator dependent and accuracies improve with experience, but it is more portable and accessible than MRI. Improvements in technology and increased operator experience have led to more accurate preoperative staging. ERUS and MRI are complementary and are most accurate for early localized cancers and more advanced cancers, respectively.     

Prospective comparison of endorectal ultrasound, three-dimensional endorectal ultrasound, and endorectal MRI in the preoperative evaluation of rectal tumorsrid=""

Background: The aim of this study was to compare the value of endorectal ultrasound (EUS), three-dimensional (3D) EUS, and endorectal MRI in the preoperative staging of rectal neoplasms. Methods: Thirty consecutive patients with rectal tumors were assessed by EUS and endorectal MRI. Additionally, three-dimensional ultrasound was performed in a subgroup of 25 patients. EUS data were obtained with a bifocal multiplane transducer (10 MHz) and processed on a 3D ultrasound workstation. MR imaging was carried out with a 1.5 T superconducting unit using an endorectal surface coil. Results: EUS was carried out successfully in all 30 patients, whereas endorectal MRI was not feasible in two patients. Compared with the histopathological classification, EUS and endorectal MRI correctly determined the tumor infiltration depth in 25 of 30 and 28 patients, respectively. The comparative accuracy of EUS, 3D EUS, and endorectal MRI in predicting tumor invasion was 84%, 88%, and 91%, respectively. EUS, three-dimensional EUS, and endorectal MRI enabled us to assess the lymph node status correctly in 25, 25, and 24 patients, respectively. Both three-dimensional EUS and endorectal MRI combined high-resolution imaging and multiplanar display options. Assessment of additional scan planes facilitated the interpretation of the findings and improved the understanding of the three-dimensional anatomy. Conclusion: The accuracy of three-dimensional EUS and endorectal MRI in the assessment of the infiltration depth of rectal cancer is comparable to conventional EUS. One advantage of both methods is the ability to obtain multiplanar images, which may be helpful for the planning of surgery in the future. Received: 4 April 2000/Accepted: 25 August 2000/Online publication: 27 October 2000