Objective: To explore the correlation between different nutritional status and depressive symptoms among adolescents, so as to provide a theoretical basis for targeted improvement of nutritional status and depressive symptoms in adolescents. Methods: A total of 8 102 adolescents aged 12-17 years in Shanghai, Urumqi, Changsha and Kunming were selected by random cluster sampling. Depressive symptoms were assessed by the Center for Epidemiologic Studies Depression Scale(CES-D). Height and weight were measured. Kruskal Wallis H test, Chi square test and ordinal Logistic regression were used to analyze the relationship between nutritional status and depressive symptoms in adolescents. Results: The overall detection rate of depressive symptoms in adolescents was 25.6%, with girls (30.1%) higher than that of boys (21.1%) ( χ 2=87.60, P <0.01). There were statistically significant differences in the scores of depressive symptoms among girls with different nutritional status ( Z =8.34, P <0.05). The detectable rate of depressive symptoms increased with BMI among girls( χ 2 trend =6.04, P <0.05). After controlling for confounding factors, ordinal Logistic regression showed that the obese girls had higher risks in depressive symptoms detection compared with normal weight girls ( OR=1.45, 95%CI=1.10-1.91, P <0.05). Conclusion The nutritional status of girls is better than boys, but the prevalence of depressive symptoms is higher than boys. The risk of depressive symptoms in girls increases with BMI.

Objective: To explore the correlation between non suicidal self injury(NSSI) behaviors and depressive symptoms in adolescents and to provide a theoretical basis for improvement of adolescent physical and mental health. Methods: A total of 8 102 adolescents aged 12-17 years were selected by random cluster sampling method in Shanghai, Urumqi, Changsha and Kunming, using the Center for Epidemiologic Studies Depression Scale (CES-D) and the Adolescent Non Suicidal Self Injurious Behavior Assessment Questionnaire. Depressive symptoms and the occurrence of NSSI behaviors were assessed. The relationship between NSSI behaviors and depressive symptoms was analyzed using Chi square test, one way analysis and Logistic regression analysis. Results: The overall detection rate of NSSI in adolescents, 24.6%, with a significantly lower detection rate in boys (23.1%) than in girls (26.2%) ( χ 2=10.97, P <0.05), and 24.7%, 24.6% in the middle school and high school segments, respectively, but the difference was not statistically significant ( χ 2=0.01, P >0.05). The overall detection rate of adolescent depressive symptoms was 49.7 % in the group with NSSI and 17.7% in the group without NSSI, with statistically significant differences ( χ 2=808.80, P < 0.01 ). The detection rates of adolescent depressive symptoms in the group with NSSI (male:44.0%, female:54.7%, middle school:49.6%, high school:49.8%) and the group without NSSI (male:14.2%, female:21.4%, middle school:14.3%, high school: 21.0 %) were statistically significant by sex and age groups ( χ 2=385.58, 412.44, 520.60, 313.78, P ＜0.01). Logistic regression models, after adjusting for gender and age, the risk for depressive symptoms was 2.65 times (95% CI =2.27-3.09) and 7.28 times (95% CI = 6.34- 8.37) more frequently in adolescents with episodic self injurious behavior and frequent self injurious behavior, respectively, compared to those without self injurious behavior. The association between NSSI status and depressive symptoms did not show sex differences ( P >0.05). Conclusion Adolescent NSSI behavior is positively associated with depressive symptoms, the attention should be paid to adolescent physical and mental health to reduce the occurrence of NSSI.

Objective: To analyze the correlation between physical activity and depressive symptoms in adolescents, so as to provide a reference for the early prevention of depression among adolescents. Methods: Using a random cluster sampling method, 8 102 adolescents from Shanghai, Urumqi, Changsha and Kunming were investigated via the Physical Activity Assessment Questionnaire of Chinese Children and Adolescents aged 7 to 18 and the Center for Epidemiologic Studies Depression Scale (CES-D) for physical activity and depressive symptoms. Multivariate Logistic regression analysis was used to analyze the correlation between physical activity and depressive symptoms in adolescents. Results: The proportion of adolescents who met moderate to vigorous physical activity (MVPA) standard was 37.0%, and the detection rate of depressive symptoms was 25.6%. The detection rate of depressive symptomsin compliant with MVPA standard group was significantly lower than that in non compliant group among different age and gender( χ 2=7.62-34.54, P <0.05). Adolescents with severe depressive symptoms had high light intensity physical activity(LPA) duration and low vigorous intensity physical activity(VPA) duration, while adolescents with mild depressive symptoms or asymptomatic adolescents had higher MVPA duration. There were differences in LPA and VPA among adolescents with different levels of depressive symptoms in the overall sample, and the differences were still significant among girls, middle school and statistically significant differences in LPA, moderate intensity physical activity(MPA), MVPA were also found among the high school subgroup. However, differences only in LPA were observed among boys with different levels of depression( H=46.48, 10.31, 25.02, 29.54, 30.25, 19.80, 22.20, 9.54, 8.62, 18.06, P <0.05). Multivariate Logistic regression model showed that gender and duration of MVPA were significantly related to risk of depressive symptoms( OR=1.56, 1.42, P <0.05). Conclusion Physical activity among adolescents is significantly related to depressive symptoms. MVPA may be an effective measure to reduce the degree of depressive symptoms.To improve the physical health and prevent depression among adolescents, it is important to choose appropriate physical activity types and effectively increase the duration of MVPA.

Objective:To investigate the short-term efficacy of laparoscopic surgery after short-course radiotherapy followed by sequential chemotherapy combined with anti-programmed death-1 (PD-1) antibody therapy for locally advanced rectal cancer.Methods:The prospective study was conducted. The clinicopathological data of 30 locally advanced rectal cancer patients who were admitted to the Union Hospital of Tongji Medical College of Huazhong University of Science and Technology from November 2019 to September 2020 were selected. Patients underwent laparos-copic surgery after short-course radiotherapy followed by sequential chemotherapy combined with anti-PD-1 antibody therapy. Observation indicators: (1) situations of the enrolled patients; (2) situations of short-course radiotherapy followed by sequential chemotherapy combined with anti-PD-1 antibody therapy and adverse events; (3) preoperative evaluation and surgical situations; (4) postoperative situations and pathological examinations; (5) postoperative adjuvant chemo-therapy and follow-up. Follow-up was conducted using outpatient examination and telephone interview up to March 2022. Patients were followed up once every 3 weeks during the period of short-course radiotherapy followed by sequential chemotherapy combined with anti-PD-1 antibody therapy to detect the adverse events and patients were followed up once every 3 months during the first postoperative 2 years and once every 6 months thereafter to detect tumor recurrence and survival of patients. Measurement data with normal distribution were represented as Mean± SD, and measurement data with skewed distribution were represented as M(range). Count data were described as absolute numbers or percentages. The Kaplan-Meier method was used to calculate survival rates and draw survival curves. Results:(1) Situations of the enrolled patients. A total of 30 patients were selected for eligibility. There were 17 males and 13 females, aged (57±16)years. Cases with preoperative primary tumor in stage cT3 and cT4 were 22 and 8, respectively. Cases with preoperative clinical lymph node metastasis in stage cN0, cN1, cN2 were 4, 16, 10, respectively. Cases in preoperative clinical stage Ⅱ and Ⅲ were 4 and 26, respectively. Of the 30 patients, there were 21 cases with positive circumferential margin and 12 cases with vascular invasion in extramural of rectum in the preoperative imaging evaluation. Distance from the distal margin of tumor to anal margin and tumor diameter of the 30 patients were 4.7(range, 1.9?9.0)cm and 5.4(range, 2.1?10.0)cm, respectively. There were 28 cases with mismatch repair proficient and 1 case with mismatch repair deficiency in tumor tissues. There was 1 case missing the data of mismatch repair in tumor tissues as failed in biopsy of pathological examination before the treatment. (2) Situations of short-course radiotherapy followed by sequential chemotherapy combined with anti-PD-1 antibody therapy and adverse events. All the 30 patients completed preoperative short-course radiotherapy successfully. Of the 30 patients, there were 3 cases not undergoing the sequential chemotherapy combined with anti-PD-1 antibody therapy and there were 24 cases undergoing 2 courses of the sequential chemotherapy combined with anti-PD-1 antibody therapy and 3 cases undergoing 1 course of the sequential chemotherapy combined with anti-PD-1 antibody therapy. The time interval between ending of radiotherapy and starting of chemotherapy combined with anti-PD-1 antibody therapy of the 27 patients was 12(range, 4?18) days. Cases with leukopenia, cases with endothelial hyperplasia of skin capillaries, cases with radiation proctitis, cases with anemia, cases with peripheral neurotoxicity, cases with neutropenia, cases with thrombocytopenia, cases with fatigue, cases with anorexia, cases with abnormal liver function, cases with hypothyroidism were 24, 22, 21,20, 18, 16, 16, 13, 10, 9, 2 in the 30 patients during the preoperative short-course radiotherapy followed by sequential chemotherapy combined with anti-PD-1 antibody therapy. Cases with the above adverse events were improved after symptomatic treatment. (3) Preoperative evaluation and surgical situations. Seven of the 30 patients were in clinical complete remission after preoperative multidisciplinary evaluation and the other 23 patients were not in clinical complete remission. Twenty-seven of the 30 patients underwent laparoscopic radical resection of rectal cancer and 3 patients not undergoing the sequential chemotherapy combined with anti-PD-1 antibody therapy did not undergo surgery. The time interval between ending of chemotherapy combined with anti-PD-1 antibody therapy and the surgery of the 27 patients were 14(range, 5?141)days. Of the 27 cases, there were 13 cases and 14 cases with 0 and 1 of the preoperative Eastern Cooperative Oncology Group score, respectively, and there were 24 cases undergoing low anterior proctectomy and 3 cases undergoing abdominoperineal excision. The operation time and volume of intra-operative blood loss of the 27 cases were (182±36)minutes and 30(range, 10?150)mL, respectively. Of the 27 cases, there were 16 cases with protective ileostomy and 24 cases with anal preservation. (4) Postoperative situations and pathological examinations. The time to postoperative first flatus, time to postoperative initial liquid food intake and duration of postoperative hospital stay of the 27 patients undergoing surgery were 2(range, 1?4)days, 3(range, 2?5)days and 8(range, 7?16)days, respectively. Five of the 27 patients had postoperative grade Ⅰ?Ⅱ complications, including 2 cases with incision infection, 1 case with abdominal infection, 1 case with incision hemorrhage and 1 case with venous thrombosis in left lower limb intermuscular. Cases with postoperative complica-tions were improved after symptomatic treatment. Results of postoperative pathological examina-tion showed that the rate of pathologic complete response in 27 patients was 48.1%(13/27). Of the 27 cases, cases in grade 0, grade 1, grade 2, grade 3 of the tumor regression grading were 13, 5, 7, 2, respectively, cases in stage T0, stage Tis, stage T2, stage T3 of the tumor T staging were 13, 1, 5, 8, respectively, cases in stage N0, stage N1, stage N2 of the tumor N staging were 19, 6, 2, respectively, cases in stage 0, stage Ⅰ, stage Ⅱ, stage Ⅲ of the tumor TNM staging were 14, 0, 5, 8, respectively. The number of lymph node dissected of the 27 patients was 15(range, 3?29). Of the 27 patients, there was 1 case with positive circumferential margin and 26 cases achieving R 0 resection. None of the 27 patients underwent secondary operation or perioperative death. (5) Postoperative adjuvant chemotherapy and follow-up. Of the 27 patients undergoing surgery, 21 cases underwent post-operative adjuvant chemotherapy, with the cycles of 4(range, 1?6). All the 27 patients were followed up for 20(range, 20?29)months. During the follow-up, 3 cases not achieving pathological complete response had tumor recurrence and no patient died. The disease free survival rate of the 27 patients was 88.9%. Conclusion:Laparoscopic surgery after short-course radiotherapy followed by sequential chemotherapy combined with immunotherapy for locally advanced rectal cancer is safe and feasible, with satisfied short-term efficacy.

Objective:To explore the value of Caprini risk assessment scale and serum D-dimer in early prediction of postoperative lower extremities deep vein thrombosis (DVT) in patients with gastrointestinal malignant tumor.Methods:A total of 240 patients with gastrointestinal malignant tumors treated in Union Hospital of Tongji Medical College of Huazhong University of Science and Technology from Jan to Oct 2020 were analyzed retrospectively.Results:Caprini score was 4 in 8 cases, 5-7 in 217 cases, and 8 in 15 cases. Sixty-seven patients developed lower extremity DVT after operation. No patients with Caprini score of 4 had DVT, 57 cases (26.3%) with a score of 5-7 had DVT; 10 cases whose score were ≥8 points (66.7%) developed DVT. There was a higher incidence of lower extremity DVT in patients ≥8 points than those of 5-7 points after surgery ( P<0.01). The postoperative Caprini score of the DVT group was higher than that of the non-DVT group (6.37±1.01 vs. 5.80±0.94, t=4.108, P<0.001). D-dimer on the first day after operation in DVT group (4.08±2.27 vs. 2.01±1.04, t=7.715, P<0.001) and the level of serum D-dimer (2.93±1.81 vs. 2.30±1.21, t=2.631, P<0.001) on day 3 was higher than that in the non-DVT group. According to the ROC curve, the best cut-off value for serum D-dimer to predict lower extremity DVT on the first postoperative day was 2.84 mg/L, the sensitivity was 70.1%, the specificity was 87.3%, and the area under the curve (AUC) was 0.815. The best cut-off value of D-dimer for predicting lower limb DVT on day 3 after surgery was 1.67 mg/L, sensitivity was 85.1%, specificity was 34.7%, and AUC was 0.611. Conclusions:Patients with gastrointestinal malignant tumors have a high incidence of postoperative lower extremity DVT. When the serum D-dimer exceeds 2.84 mg/L on the first postoperative day, the likelihood of postoperative lower extremity DVT is higher.

Objective:To explore the risk factors, clinical features, endoscopic characteristics and the efficacy of antiviral therapy in ulcerative colitis (UC) patients complicated with cytomegaloviremia (CMV) and Epstein-Barr (EB) viremia.Methods:From April 1, 2014 to January 31, 2019, at The Second Hospital of Hebei Medical University, a total of 320 UC patients hospitalized at the Department of Gastroenterology were enrolled. According to the pathogens, the patients were divided into four groups: complicated with CMV and EB viremia group ( n=35), only complicated with CMV viremia group ( n=33), only complicated with EB viremia group ( n=52) and without CMV and EB viremia group ( n=200). Clinical features and the efficacy of antiviral therapy of the patients were retrospectively analyzed. Multivariate logistic regression was used to analyze the risk factors of UC complicated with CMV and EB viremia. Kruskal-Wallis H test, Chi-square test and Fisher exact test were used for statistical analysis. Results:The proportion of patients of age>60 years old (42.86%, 15/35), the rate of glucocorticoid use (51.43%, 18/35) within three months before onset and the inefficacy rate of glucocorticoid treatment (22.86%, 8/35) of UC complicated with CMV and EB viremia group were all higher than those of UC without CMV and EB viremia group (14.00%, 28/200; 24.50%, 49/200; 1.00%, 2/200), and the differences were statistically significant ( χ2=17.062, 10.598 and 29.769; all P<0.01). However, there were no statistically significant differences between UC complicated with CMV and EB viremia group and UC without CMV and EB viremia group in gender, and treatment of 5-aminosalicylic acid (5-ASA), azathioprine and infliximab within three months before onset (all P>0.05). The proportion of patients with fever (54.29%, 19/35), abdominal pain (91.43%, 32/35), hematochezia (94.29%, 33/35), weight loss (28.57%, 10/35), severe disease activity (94.29%, 33/35), total colon involvement (91.43%, 32/35), serum albumin less than 30 g/L (71.43%, 25/35) and hemoglobin less than 100 g/L (48.57%, 17/35) of UC complicated with CMV and EB viremia group were all higher than those of UC without CMV and EB viremia group (13.50%, 27/200; 43.00%, 86/200; 44.00%, 88/200; 13.50%, 27/200; 38.00%, 76/200; 65.00%, 130/200; 18.00%, 36/200 and 18.50%, 37/200), and the differences were statistically significant ( χ2=31.475, 27.945, 32.930, 5.100 and 40.194, Fisher exact test, χ2=44.242 and 15.220, all P<0.01). However, there were no statistically significantl differences in clinical classification and disease course (all P>0.05). The incidence rates of deep ulcer (45.71%, 16/35), irregular ulcer (42.86%, 15/35) and longitudinal ulcer (8.53%, 3/35) under endoscopy of UC complicated with CMV and EB viremia group were significantly higher than those of UC without CMV and EB viremia group (1.50%, 3/200; 3.50%, 7/200 and 1.00%, 2/200), and the differences were statistically significant ( χ2=72.521 and 49.837, Fisher exact test, all P<0.01). The incidence rates of deep ulcer and irregular ulcer under endoscopy of UC complicated with CMV and EB viremia group were higher than those of UC only complicated with EB viremia group (15.38%, 8/52 and 11.54%, 6/52), and the differences were statistically significant ( χ2=9.663 and 11.206, P=0.002 and 0.001). The results of Multivariate Logistic regression analysis showed that severe disease activity, serum albumin level less than 30 g/L, and deep ulcer and irregular ulcer under endoscopy were risk factors of UC patients complicated with CMV and EB viremia (odds ratio=48.519, 44.352, 53.432 and 39.989, 95% confidence interval 9.057 to 587.669, 4.499 to 437.245, 3.302 to 864.670 and 3.418 to 467.910, all P<0.05). The improvement rate of antiviral therapy in UC complicated with CMV and EB viremia group (73.53%, 25/34) was significantly lower than those of UC only complicated with CMV group (96.88%, 31/32) and UC only complicated EB viremia group (95.65%, 44/46), and the differences were statistically significant ( χ2=6.989 and 6.310, P=0.008 and 0.012). Conclusions:UC patients with severe disease activity, serum albumin level less than 30 g/L, and deep ulcer and irregular ulcer under endoscopy are more likely to develop CMV and EB viremia. The more severe the disease, the worse the treatment response, so it is necessary to strengthen the screening to CMV and EB virus infection in UC patients.

Objective:To explore the differences of attention bias performance between patients with heroin dependence in methadone maintenance treatment(MMT) and healthy controls.Methods:A total of 38 heroin dependent patients in the Fourth People's Hospital of Huainan from January 2018 to September 2018 were selected as MMTgroup, and 32 gender-and age-matched healthy controls were selected as healthy control(HC) group.The attention bias was evaluated by Emotional Stroop Task, and the difference between the two groups was compared.Results:The error number of the herion-related words in Emotional Stroop Task of MMT group was lower than that of HC group[0.75(2, 5) vs.1(0, 2)], the difference was statistically significant( Z=-2.184, P=0.029). The number of errors in point and neutral words and reaction time of point, neutral word and clue word had no statistically significant differences between MMT group and HC group[1(0, 4) vs.1(0, 2); 3(2, 5) vs.2(2, 3); (1 055.14±303.50)ms vs.(985.40±173.71)ms, (1 126.89±347.82)ms vs.(1 022.76± 173.62)ms; (1 188.53±371.80)ms vs.(1 113.23±225.11)ms]( Z=-1.331, P=0.183; Z=-1.723, P=0.085; t=1.150, P=0.254; t=1.539, P=0.129; t=1.001, P=0.320). There were no statistically significant differences in attention bias influenced by the dosage and duration of methadone between MMT group and HC group(all P>0.05). Conclusion:These results demonstrate that heroin dependence patients in MMT have attentional bias in heroin-related words.Methadone has no effect on attention bias in patients with heroin dependence during maintenance treatment intervention.

Objective: To analyze the clinical features and risk factors of ulcerative colitis (UC)complicated with Epstein-Barr(EB)-viremia and the effect of antiviral therapy on the remission of the symptoms. Methods: From April 2014 to January 2018, data of 239 UC patients hospitalized at the Department of Gastroenterology of Second Hospital of Hebei Medical University were collected. The patients were divided into EB-viremia group (trial group, n=43) and non-EB-viremia group (control group, n=196) according to EB virus-DNA detection. The general condition, clinical characteristics and therapeutic efficacy of the two groups were compared. The risk factors and the effect of antiviral therapy on the remission of symptoms of UC complicated with EB-viremia were analyzed. Chi-square test and logistic analysis were used for statistical analysis. Results: There were no significant differences in gender, age, clinical type, lesion range, the proportion of patients treated with 5-aminosalicylic acid or corticosteroids, the percentage of patients with diarrhea and bloody stool, the proportion of patients with spontaneous bleeding, platy ulcer and longitudinal ulcer under colonoscopy, the course of disease or Mayo score between the trial group and control group (all P>0.05). The proportions of patients with smoking history and severe disease, treatment with azathioprine and 6-mertocapurine (6-MP), treatrnent with infliximab, symptoms of fever or abdominal pain and deep and large ulcer under colonoscopy in the trial group were all higher than those in the control group, and the differences were all statistically significant (χ²=5.304, 6.608, 6.718, 6.939, 8.783, 4.493 and 5.425, all P<0.05). The results of multivariate logistic regression analysis showed that smoking history and treatment with azathioprine and 6-MP were risk factors of UC complicated with EB-viremia (OR=2.801 and 9.343, 95%CI 1.170 to 6.703 and 1.749 to 49.922, P=0.021 and 0.009). The improvement rates of the trial group and control group were 79.1%(34/43) and 95.4%(187/196), respectively. There was a significant difference in the improvement rate between the two groups (χ²=10.551, P=0.001). In the trial group, 12 patients (27.9%) received ganciclovir treatment, fonr patients (9.3%) had foscarnet sodium treatment, 21 patients (48.8%) were treated with the combination of these two medications and six cases (14.0%) did not received any antiviral treatment. After three weeks, the improvement cases of the above regimens were 8, 4, 16 and 6, respectively. There were no statistically significant differences in the improvement rate of patients with and without antiviral treatment, further more, no difference was found in the improvement rate of patients with different antiviral treatments (all P>0.05). Conclusions Smoking history and purine treatment are risk factors of UC complicated with EB-viremia.

Objective: To analyze the consistency of gastroscopic biopsy in the diagnosis of high grade intraepithelial neoplasia(HGIN) and postoperative pathological diagnosis, and explore the risk factors associated with missed diagnosis of HGIN. Methods: From January 2012 to December 2018, the clinical data of 63 patients who were diagnosed with HGIN by gastroscopic biopsy prior to operation and underwent complete resection in the Union Hospital, Tongji Medical College, Huazhong University of Science and Technology were retrospectively analyzed. There were 55 males and 8 females, with a median age of 60 (35 to 76) years old. The gender, age, endoscopic lesion shape, longest diameter, CT image and inflammatory markers were analyzed, to investigate the correlation between them and pathological upgrading after operation. Receiver operating characteristic (ROC) curve was drawn to analyze the cut off value of measurement data, and the comparison of count data was performed by chi-square test or Fisher exact probability method. Univariate analysis was used to screen potential risk factors, and multivariate logistic regression analysis was futher utilized to analyze the independent risk factors of postoperative pathological upgrading. Results: A total of 63 patients were enrolled, including 47 cases underwent surgical resection and 16 cases underwent endoscopic submucosal dissection(ESD). Among them, 19 patients(30.2%) were pathologically diagnosed with HGIN, while 44 patients(69.8%) were pathologically diagnosed with invasive cancer after resection. Preoperative contrast-enhanced CT showed that 11 patients(17.5%) with perigastric fat spiculation around the lesion, all of which were confirmed as invasive carcinoma after operation. Univariate analysis showed that the longest diameter of the lesion ≥2 cm (P=0.002), ulcer lesions under gastroscopy (P=0.013), platelet to lymphocyte ratio (PLR) ≥103 (P=0.030), lymph node enlargement (P=0.046) and spiculation of the perigastric fat (P=0.025) were significant differences between the groups. Logistic regression analysis showed that the longest diameter of the lesion ≥2 cm (P=0.033) and ulcer lesions under gastroscopy (P=0.007) were independent risk factors for predicting postoperative pathological upgrading. Conclusions Gastroscopy biopsy in the diagnosis of high-grade intraepithelial neoplasia of the stomach has poor consistency with pathological diagnosis after operation. Clinicians should take active surgical intervention, especially for cases with the longest diameter of the lesions ≥2 cm, ulcer or perigastric fat spiculation on CT.

Objective: To explore the risk factors of cytomegalovirus (CMV) infection or reactivation in ulcerative colitis (UC) patients. Methods: We performed a search at the databases of Pubmed, Cochrane, Embase, CNKI, Wanfang, SinoMede and VIP up to March 2017. A search strategy was constructed by using a combination of the following words: "inflammatory bowel disease or IBD" or "ulcerative colitis or UC" and "cytomegalovirus or CMV" . Literature was screened according to the inclusion and exclusion criteria and statistics was analyzed using RevMan 5.3 software provided by Cochrane collaboration network and analyzed using Stata 12.0 software to evaluate publication bias. Results: After searching and screening, we included 18 case-control studies finally. Meta-analysis showed that the risk of CMV infection or reactivation in severe UC was 1.45 times that in mild to moderate UC and the risk in whole colon was 1.54 times that of patients with left colon and rectum with the pooled OR values as 1.45 (1.02-2.05) and 1.54 (1.05-2.27). The risk of CMV infection in middle-aged patients with UC was higher than that in young patients with the pooled MD values of 4.60(2.13-7.07). The therapies with glucocorticoid and immunosuppressive agents such as azathioprine, cyclosporine and methotrexate were high risk factors of CMV infection or reactivation in UC patients, with the pooled OR values as 3.87 (2.70-5.53) and 2.07 (1.29-3.32), but the duration of UC, treatment with 5-Amino salicylic acid/salazosulfapyridine (5-ASA/SASP) and infliximab had no statistically significant correlation with the CMV infection or reactivation in UC patients, with the pooled OR values as -1.20 (-2.64-0.24), 0.94 (0.61-1.42) and 1.50 (0.65-3.44). Conclusions In patients with severe UC, whole colon lesions and the therapy with glucocorticoid and immunosuppressive agents were the risk factors of CMV infection or reactivation. The risk of CMV infection or reactivation in middle-aged patients was higher than that in young patients.