| Background and ObjectionInflammatory bowel disease(IBD)is a nonspecific intestinal inflammatory disease,including ulcerative colitis(UC)and Crohn's disease(CD),which etiology and pathogenesis have not been fully known yet.As an important immune mediator and proinflammatory factor,numerous studies show that tumor necrosis factor alpha(TNF-?)plays a central role in the inflammatory cascade of IBD,and anti-TNF-? biologics have become the most effective drugs to treat IBD.However,there are many evidences also indicate that TNF-a can play a key role in the defense mechanism of Mycobacterium tuberculosis(M tuberculosis)infection,it's involved in the formation of tuberculous granuloma,promoting macrophage apotosis and so on.As a result,most of the patients with tuberculosis infection do not show active tuberculosis(TB),but in the form of a latent tuberculosis infection(LTBI),which without any clinical manifestation and radiographic changes,and its diagnosis mainly rely on the positive tuberculin test(TST)or positive interferon gamma release test(IGRAs).While LTBI can be reactivated into active TB when the body's immune defense mechanism is reduced.Therefore,though anti-TNF-? therapy can effectively improve the condition of IBD patients,it will also weaken the body's resistance to M tuberculosis infection,making it easy to get active TB at last.With the increasing incidence of IBD in the global scope,and the effectiveness of anti-TNF-a biological agents for the treatment of the disease has been more and more recognized,the use od such drugs is also increasing.So,it's of great significance to provide an assessment of TB risk associated with anti-TNF-a therapy for IBD in clinical.At present,there are four kinds of anti-TNF-a biologics for IBD have been listed in the world,including human mouse chimeric TNF-? IgGl monoclonal antibody infliximab(IFX),humanized TNF-? IgG monoclonal antibody adalimumab(ADA),polyethylene glycol humanized Fab fragment of TNF-?monoclonal antibody certolizumabpegol(CZP)and humanized TNF-? monoclonal antibody golimumab(GLM).Up to date now,from the point of evidence-based medicine,there are few reports about the risk assessment of TB induced by above four kinds of anti-TNF-? biologics in treatment of IBD.Besides,more than 90%of M tuberculosis infection is LTBI,in which 5%?10%LTBI can be developed into active TB in the future,and this ratio will be higher when the protective immunity of the body is reduced.For example,both of the World Health Organnization(WHO)latest management guidelines of LTBI and the US Center for Disease Control(CDC)regarde the LTBI population who will receive anti-TNF-?treatment as such high-risk persons.Although China is one of the 22 high countries of TB and has 550 million LTBI people,the investigation of LTBI screening before IFX using is almost empty,which is the mainly anti-TNF-? biologics for IBD in our country.In summary,in order to provide evidence-based assessment of TB risk induced by anti-TNF-? therapy in IBD,which can guide clinical drug use,and fill the domestic blank of the investigation results of LTBI screening before IFX using,we have carried out the study from two parts.In the first part,the meta analysis was used to comprehensively evaluate the randomized controlled trials(RCTs)of IFX,ADA,CZP and GLM for the treatment of adult IBD.In the second part,through a retrospective cohort study,from January 2013 to December 2015,all IBD patients who were admitted to Nangfang Hospital of Southern Medical University and meet all research requirements were enrolled,then preliminary explored a series of issues about LTBI screening before using IFX to treat IBD.Method1.Meta-analysis was performed.Firstly,we conducted a structured search using a series of related search terms.Secondly,according to the purpose of the study,the inclusion and exclusion criteria were established.Then articles which were met the criteria were preliminary selected by reading titles and abstracts.Finally,eligible randomized controlled trials(RCTs)were picked by reading full text of preliminary selected articles,and data extraction,risk of bias,publication bias,heterogeneity test and so on were evaluated among these RCTs.2.Through a retrospective cohort study,from January 2013 to December 2015,all IBD patients who were admitted to Nangfang Hospital of Southern Medical University were enrolled.Then the inclusion and exclusion criteria were established according to the purpose of the study,patients who were met the criteria were included by careful examination of each one's medication and medical records.when data was collected,in addition to general information,which mainly include gender,age,body mass index(BMI),type of IBD,other high risk factors for LTBI(TB patient contact history,glucocorticoid hormone or other immunosuppressive agents use history,history of diabetes,severe malnutrition status and smoking history)and the duration of IFX treatment from first time to the last.We should pay more attention on the history of LTBI screening before IFX was used for IBD patients,including whether there was a history of LTBI screening,which LTBI screening program was selected,what was the result of LTBI screening,whether there was a history of preventive anti-TB treatment and active TB was happened during the IFX treatment.At last,the statistical analysis was conducted by SPSS version 20.0.Differences between measurement data were assessed with the t test.Comparison between count data were assessed with the Chi-square test.Comparison between paired count data using McNemar and kappa coefficient tests.A two-tailed P value<0.05 was considered significant.Result1.Outcomes of meta-analysis1.1 Risk of tuberculosis induced by anti-TNF-a therapyIn 25 RCTs which were included in the meta-analysis,8 cases of TB were reported,all of them came from anti-TNF-? treatment group.The occurrence of TB was 0.23%(4/1754)in IFX treatment group,0.15%(2/1313)in ADA treatment group,0.12%(1/833)in CZP treatment group and 0.12%(1/818)in GLM treatment group.Funnel plot symmetry was detected and there was no significant publication bias.The Cochran Q test had a P value of 1.00,and the corresponding I2 statistic was 0%,both indicating little variability between studies,so the fixed effects model was used to analyze the data.Although the results did not reach statistical significance,exposure to anti-TNF-? biologic agents appeared to be associated with increased risk ratio(RR)of TB[0.17%(8/4718)vs 0%(0/2991),P=0.20,RR=2.17,95%CI=0.66-7.09].The RR of TB with IFX treatment was higher than those with placebo treatment,though there was no statistically significance[0.23%(4/1754)vs 0%(0/1169),P=0.36,RR=2.08,95%CI=0.43-10.06].The RR of TB with ADA treatment was higher than those with placebo treatment,though there was no statistically significance[0.15%(2/1313)vs 0%(0/776),P=0.55,RR=2.53,95%CI=0.12-52.49].The RR of TB with CZP treatment was higher than those with placebo treatment,though there was no statistically significance[0.12%(1/833)vs 0%(0/639),P=0.51,RR=2.94,95%CI=0.12-71.88].The RR of TB with GLM treatment was higher than those with placebo treatment,though there was no statistically significance[0.12%(1/818)vs 0%(0/407),P=0.80,RR=1.52,95%CI=0.06-37.20].The risk of TB among above four kinds of anti-TNF-a biologic agents group were also no statistically significance(P=0.99).1.2 Risk of tuberculosis induced by anti-TNF-? therapy in each subgroupSubgroup analysis were performed.The occurrence of TB was 0.13%(6/4548)in CD patients during anti-TNF-? therapy and 0.06%(2/3161)in UC patients.The RR of TB with anti-TNF-a therapy for CD patients(RR=2.47,95%CI=0.61-10.06)was higher than those UC patients(RR=1.51,95%CI=0.61-10.06),though there was no statistically significance(P=0.72).The occurrence of TB was 0.17%(8/4820)in trials with>8w treatment and no case in trials with ?8w treatment.The RR of TB with anti-TNF-? therapy in trials with>8w treatment was2.17(95%CI=0.66-7.09).The occurrence of TB was 0.19%(5/2654)in trials with placebo patients exposed to anti-TNF-? therapy during induction of remission and 0.06%(3/5055)in trials with no placebo patients exposed to anti-TNF-? therapy during induction of remission.The RR of these two group was no significant difference[(RR=2.06,95%CI=0.43-9.83)vs(RR=2.32,95%CI=0.34-14.37),P=0.92].Furthermore,our assessment of risk of bias(RoB)indicated low RoB in 5 RCTs,15 RCTs were rated high RoB,and RoB of 5 RCTs was unclear.The occurrence of TB was 0.13%(2/1566)in trials with low RoB and 0.10%(6/6143)in trials with high and unclear RoB.There was also no significant difference in the RR of TB between there two group[(RR=2.03,95%CI=0.22-18.97)vs(RR=2.22,95%CI=0.55-9.00),P=0.95].2.Results of retrospective cohort study2.1 Basic characteristics of patients from cohort studyIn total,115 IBD patients were considered eligible and enrolled.The mean age of the study population was(32.8 ± 12.7)years.The population of male patients was 57.4%(66/115)and female patients was 42.6%(49/115).The population consisted of 74.8%(86/115)CD and 25.2%(29/115)UC.According to whether there was a LTBI screening before IFX treatment,the study population were divided into two groups:LTBI screening group and no LTBI screening group.The population of LTBI screening group was 58.3%(67/115)and no LTBI screening group was 41.7%(48/115).The mean age of LTBI screening group was younger than no LTBI screening group,and the difference between them was statistically significant[(29.8±11.2)vs(37.0 ± 13.6),P=0.002].In the LTBI screening group,the number of CD patients were more than UC patients,and not less than the CD patients of no LTBI screening group[95.5%(64/67)vs 4.5%(3/67),74.4%(64/86)vs 25.6%(22/86)].At the same time,in the no LTBI screening group,the number of UC patients were more than CD patients,and not less than the UC patients of LTBI screening group[54.2%(26/48)vs 45.8%(22/48),86.7%(26/29)vs 10.3%(3/29)].Above difference was statistically significant(P<0.001).However,there was no significant difference between the two groups in gender,BMI,other high risk factors for LTBI and the dutation of IFX treatment from first time to the last(P values wre greater than 0.05).2.2 Screening for LTBI before IFX therapyThe sceening outcomes prior to IFX therapy were summarized.In the LTBI screening group,there were total of 67 patients.Among these patients,53.7%(36/67)underwent TST,22.4%(15/67)used T-SPOT.TB and 23.9%(16/67)choose the program of TST combined with T-SPOT.TB.The results of McNemar and kappa coefficient tests of the TST and T-SPOT.TB is shown that there was no significance between these two methods on diagnosis of LTBI(P=0.25),but the degree of anastomosis is weak(?=0.333,P=0.074).According to the diagnosis of LTBI,a total od 7 patients with LTBI were detected,the detection rate was 10.4%(7/67).In the 7 patients with LTBI,4 cases were detected by TST and 3 cases were detected by TST combined with T-SPOT.TB.Among these 7 patients,only 3 cases received orally isoniazid(0.3g/d)for 4 months as prophylactic anti-TB treatment accompanied with IFX therapy.2.3 Incidence and clinical features of active TB during treatment with IFXAmong 115 patients with IFX therapy,a total of 3 patient with chest imaging were considered the occurrence of active TB,the incidence was 2.6%(3/115),and all of them were from the no LTBI screening group.The mean age of these patients was(46.3 ± 11.0)years.The population of male patients was two and UC patients was 2.The first one was a male patient with CD,before his third IFX treatment(since the initial IFX treatment time is about 12 weeks),chest computed tomography(CT)examination showed diffuse pulmonary nodules in both lungs which considered the sub acute military pulmonary TB.The second one was a male patient with UC,befror his fifth IFX treatment(since the initial IFX treatment time is about 21 weeks),chest CT examination showed secondary pulmonary TB of right upper lung and right tuberculous pleurisy.The third one was a female with UC,before her fifth IFX treatment(since the initial IFX treatment time is about 21 weeks),chest CT examination showed diffuse pulmonary nodules in both lungs which considered the chronic hematogenous disseminated pulmonary TB.Besides,there were multiple enlarged lymph nodes in the mediastinum and bilateral hilar area which considered the TB of lymph nodes.None of the 3 patients died due to active TB.Conclusion1.Although the risk of TB between anti-TNF-a therapy and the placebo group was no statistically significant difference,the results showed that all TB occurred in the anti-TNF-a treatment group.Therefore,the clinical use of anti-TNF-? biological agents should still pay attention to the risk of TB induced by such drugs,especially in the countries with high burden of M tuberculosis infection;2.The occurrence of TB in trials with>8w anti-TNF-? treatment was higher than those tirals with<8w anti-TNF-a treatment.It was suggested that the risk of TB induced by anti-TNF-a therapy may increase with the prolongation of treatment time.So for patients who need long-term use of these drugs should be more attention to the risk of TB;3.The importance of LTBI screening before the IFX treatment for IBD patients in our country need to be imporved,especially for patients with UC;4.In regions with a high incidence of TB and widely use of BCG,and under the premise of patients with moderate and severe malnutrition due to IBD and the use of immunosuppressive agents were common,the results of LTBI screening which only used TST were poor reliability. |
PDF Full Text Request