Efficacy comparison among high risk factors questionnaire and Asia-Pacific colorectal screening score and their combinations with fecal immunochemical test in screening advanced colorectal tumor / 中华胃肠外科杂志

ABSTRACT

Objective: To investigate the effects of high risk factors questionnaire (HRFQ), Asia-Pacific colorectal screening (APCS) score and their combinations with fecal immunochemical test (FIT) in screening advanced colorectal neoplasia, in order to provide an evidence for further optimization of cancer screening program. Methods: A retrospective cohort study method was used to summarize and analyze the results of colorectal tumor screening in Jiashan County, Zhejiang Province from March 2017 to July 2018. Those with severe diseases that were not suitable for colonoscopy and those with mental and behavioral abnormalities who can not cooperate with the screening were excluded. Those who met any one or more of the followings in the HRFQ questionnaire were classified as high-risk people of HRFQ (1) first-degree relatives with a history of colorectal cancer; (2) subjects with a history of cancer or any other malignant tumor; (3) subjects with a history of intestinal polyps; (4) those with two or more of the followings chronic constipation (constipation lasted for more than 2 months per year in the past two years), chronic diarrhea (diarrhea lasted for more than 3 months in the past two years, and the duration of each episode was more than one week), mucus and bloody stools, history of adverse life events (occurring within the past 20 years and causing greater trauma or distress to the subject after the event), history of chronic appendicitis or appendectomy, history of chronic biliary disease or cholecystectomy. In this study, those who were assessed as high risk by HRFQ were recorded as "HRFQ (+)", and those who were not at high risk were recorded as "HRFQ (-)". The APCS questionnaire provided risk scores based on 4 risk factors including age, gender, family history and smoking (1) age 2 points for 50-69 years old, 3 points for 70 years old and above; (2) gender 1 point for male, 0 point for women; (3) family history 2 points for first-degree relatives suffering from colorectal cancer; (4) smoking 1 point for current or past smoking, 0 point for non-smokers. The population was divided into low-risk (0-1 point), intermediate-risk (2-3 points), and high-risk (4-7 points). Those who were assessed as high risk by APCS were recorded as "APCS (+)", and those with intermediate and low risk were recorded as "APCS (-)". The hemoglobin threshold for a positive FIT was set to 100 μg/L. Those who were assessed as high risk by APCS with positive FIT were recorded as "APCS+FIT (+)". Those who were assessed as high risk by APCS with negative FIT, those who were assessed by APCS as low-middle risk with positive FIT, and those who were assessed by APCS as low-middle with negative FIT were all recorded as "APCS+FIT(-)". Observation indicators in this study were as follows (1) the screening compliance rate of the cohort and the detection of advanced colorectal tumors; (2) positive predictive value, negative predictive value, sensitivity and specificity of HRFQ and APCS and their combination with FIT for screening advanced colorectal tumors; (3) comparison of the detection rate between HRFQ and APCS questionnaire for different colorectal lesions. Using SPSS 21.0 software, the receiver operating characteristic (ROC) curve was drawn to evaluate the clinical value of HRFQ and APCS combined with FIT in screening advanced colorectal tumors. Results: From 2017 to 2018 in Jiashan County, a total of 53 268 target subjects were screened, and 42 093 people actually completed the questionnaire, with a compliance rate of 79.02%. A total of 8145 cases underwent colonoscopy. A total of 3607 cases among HRFQ positive population (5320 cases) underwent colonoscopy, and the colonoscopy compliance rate was 67. 80%; 8 cases were diagnosed with colorectal cancer and 88 cases were advanced colorectal adenoma. A total of 2977 cases among APCS positive population (11 942 cases) underwent colonoscopy, and the colonoscopy compliance rate was 24.93%; 17 cases were diagnosed with colorectal cancer and 148 cases were advanced colorectal adenoma. The positive rate of HRFQ screening was lower than that of APCS [12.6% (5320/42 093) vs. 28.4% (11 942/42 093), χ2=3195. 547, P<0.001]. In the FIT positive population (6223 cases), a total of 4894 cases underwent colonoscopy, and the colonoscopy compliance rate was 78.64%; 34 cases were diagnosed with colorectal cancer and 224 cases were advanced adenoma. The positive predictive values of HRFQ and APCS and their combination with FIT for screening advanced colorectal tumors were 2.67%, 5.54%, 5.44%, and 8.56%; negative predictive values were 94.89%, 96.85%, 96.11% and 96.99%; sensitivity was 29.27%, 50.30%, 12.20 % and 39.02%; specificity was 55.09%, 64.03%, 91.11% and 82.51%, respectively. The ROC curves constructed by HRFQ, APCS, FIT, HRFQ+FIT and APCS+FIT indicated that APCS+FIT presented the highest efficacy in screening advanced colorectal tumors (AUC 0.608, 95%CI 0.574-0.642). The comparison of the detection rates of different colorectal lesions between HRFQ and APCS questionnaires showed that there were no significant differences in detection rate of inflammatory polyps and hyperplastic polyps between the two questionnaires (both P>0.05). However, as compared to HRFQ questionnaire, APCS questionnaire had higher detection rates in non-advanced adenomas [26.10% (777/2977) vs. 19.43% (701/3607), χ2=51.228, P<0.001], advanced adenoma [4.97% (148/2977) vs. 2.44% (88/3607), χ2=30.249, P<0.001] and colorectal cancer [0.57% (17 /2977) vs. 0.22% (8/3607), χ2=5.259, P=0.022]. Conclusions: APCS has a higher detection rate of advanced colorectal tumors than HRFQ. APCS combined with FIT can further improve the effectiveness of advanced colorectal tumor screening.