Accuracy of optical diagnosis with narrow band imaging in the surveillance of ulcerative colitis: a prospective study comparing Kudo, Kudo-IBD and NICE classifications.

PURPOSE: The diagnostic accuracy of Narrow Band Imaging (NBI) in the endoscopic surveillance of ulcerative colitis (UC) has been disappointing in most trials which used the Kudo classification. We aim to compare the performance of NBI in the lesion characterization of UC, when applied according to three different classifications (NICE, Kudo, Kudo-IBD). METHODS: In a prospective, real-life study, all visible lesions found during consecutive surveillance colonoscopies with NBI (Exera-II CV-180) for UC were classified as suspected or non-suspected for neoplasia according to the NICE, Kudo and Kudo-IBD criteria. The sensitivity (SE), specificity (SP), positive (+LR) and negative (-LR) likelihood ratios of the three classifications were calculated, using histology as the reference standard. RESULTS: 394 lesions (mean size 6 mm, range 2-40 mm) from 84 patients were analysed. Twenty-one neoplastic (5%), 49 hyperplastic (12%), and 324 inflammatory (82%) lesions were found. The diagnostic accuracy of the NICE, Kudo and Kudo-IBD classifications were, respectively: SE 76%-71%-86%; SP 55-69%-79% (p < 0.05 Kudo-IBD vs. both Kudo and NICE); +LR 1.69-2.34-4.15 (p < 0.05 Kudo-IBD vs. both Kudo and NICE); -LR 0.43-0.41-0.18. CONCLUSION: The diagnostic accuracy of NBI in the differentiation of neoplastic and non-neoplastic lesions in UC is low if used with conventional classifications of the general population, but it is significantly better with the modified Kudo classification specific for UC.

Noninvasive Monitoring After Azathioprine Withdrawal in Patients With Inflammatory Bowel Disease in Deep Remission.

BACKGROUND & AIMS: There is uncertainty regarding the optimal duration of treatment with azathioprine (AZA) in ulcerative colitis (UC) and Crohn's disease (CD). We analyzed the clinical course and predictors of relapse after AZA withdrawal in patients in sustained deep remission. METHODS: A prospective study was performed on patients who stopped their treatment with AZA while being in steroid-free, extended deep remission (normal clinical, endoscopic, and histologic indexes, C-reactive protein, and fecal calprotectin [FC]). Standard biochemical tests and FC were measured at 3 and 6 months, then every 6 months. Bowel ultrasounds and ileocolonoscopy were performed every 6 and 12 months, respectively. Multivariate analysis for predictors of relapse was performed using a Cox proportional hazards model and hazard ratios were calculated. Spearman nonparametric correlation test was also used. The accuracy of significant predictors was calculated. RESULTS: Fifty-seven patients with inflammatory bowel disease stopped AZA after median 7 years (range, 5-19) and were followed up for median 50 months (range, 25-85). Twenty-six patients (18/31 UC, 8/26 CD; P = .003) relapsed, within a median 15 months (range, 2-37). FC was the only variable significantly correlated with later relapse of both diseases (UC: hazard ratio, 3.3; 95% confidence interval, 1.2-10; CD: hazard ratio, 4.5; 95% confidence interval, 1.4-12.5). The sensitivity, specificity, and positive and negative predictive values of FC were 50%, 100%, 100%, and 59% in UC and 50%, 94%, 80%, and 81% in CD. CONCLUSIONS: More than half patients with UC and one-third of patients with CD relapse after AZA withdrawal despite previous deep remission. FC positivity is associated with high risk of relapse, allowing early correction of the therapeutic strategy.

[New European guidelines for the management of dyslipidaemias: their aggressiveness is not legitimated by current evidence]. / Nuove linee guida europee sulla gestione delle dislipidemie: aggressività non giustificata dalle prove esistenti.

Le linee guida 2019 delle Società europee di cardiologia e dell'aterosclerosi sulla gestione delle dislipidemie hanno aumentato l'aggressività diagnostico-terapeutica e non distinguono prevenzione primaria e secondaria, con parziale eccezione per gli ultra75enni. Raccomandano nuovi target di cLDL: per pazienti a rischio molto alto si raccomanda una riduzione a <55 mg/dl e >=50% rispetto al basale; per pazienti a rischio alto una riduzione a <70 mg/dl e >=50% del basale; per pazienti a rischio moderato una riduzione a <100 mg/dl; per pazienti a rischio basso una riduzione a <116 mg/dl. In base alle carte SCORE e ai dati di mortalità cardiovascolare in Italia, quasi tutti i maschi dai 70 anni e le donne dai 70-75 anni risulterebbero ad alto rischio per solo effetto dell'età. Anche quasi tutti i 60enni sarebbero a rischio moderato, con target di cLDL <100, e spesso necessità di aggiungere costosi ipolipemizzanti per chi già assume statine. Le prove supportano ben poco tale aggressività. Infatti, anche negli studi randomizzati controllati (RCT) i benefici su esiti cardiovascolari non fatali subiscono esagerazioni sistematiche, quello meno distorto e di maggior interesse per gli assistiti è la mortalità totale. Questa, con terapie ipocolesterolemizzanti più intensive, non si è ridotta nelle metanalisi di RCT con pazienti con cLDL tra 80 e <100 mg/dl al basale; con inibitori di PCSK9, pazienti con questi valori mostrano persino tendenza all'aumento della mortalità totale. Dunque, abbassare il cLDL a <80 mg/dl può non giovare neppure ad anziani coronaropatici. A oggi, ciò vale ancora di più per anziani nella popolazione generale, in cui una revisione sistematica di studi di coorte ha mostrato relazioni nulle o più spesso inverse tra cLDL e mortalità. Le nuove linee guida europee forzano le prove disponibili, trascurano il principio di precauzione e non possono esser base per un equo consenso informato.

[Rt or RDt, that is the question!] / Rt or RDt, that is the question!

The article compares two of the most followed indices in the monitoring of COVID-19 epidemic cases: the Rt and the RDt indices. The first was disseminated by the Italian National Institute of Health (ISS) and the second, which is more usable due to the lower difficulty of calculation and the availability of data, was adopted by various regional and local institutions.The rationale for the Rt index refers to that for the R0 index, the basic reproduction number, which is used by infectivologists as a measure of contagiousness of a given infectious agent in a completely susceptible population. The RDt index, on the other hand, is borrowed from the techniques of time series analysis for the trend of an event measurement that develops as a function of time. The RDt index does not take into account the time of infection, but the date of the diagnosis of positivity and for this reason it is defined as diagnostic replication index, as it aims to describe the intensity of the development of frequency for cases recognized as positive in the population.The comparison between different possible applications of the methods and the use of different types of monitoring data was limited to four areas for which complete individual data were available in March and April 2020. The main problems in the use of Rt, which is based on the date of symptoms onset, arise from the lack of completeness of this information due both to the difficulty in the recording and to the absence in asymptomatic subjects.The general trend of RDt, at least at an intermediate lag of 6 or 7 days, is very similar to that of Rt, as confirmed by the very high value of the correlation index between the two indices. The maximum correlation between Rt and RDt is reached at lag 7 with a value of R exceeding 0.97 (R2=0.944).The two indices, albeit formally distinct, are both valid; they show specific aspects of the phenomenon, but provide basically similar information to the public health decision-maker. Their distinction lies not so much in the method of calculation, rather in the use of different information, i.e., the beginning of symptoms and the swabs outcome.Therefore, it is not appropriate to make a judgment of preference for one of the two indices, but only to invite people to understand their different potentials so that they can choose the one they consider the most appropriate for the purpose they want to use it for.

Hepatobiliary phase in cirrhotic patients with different Model for End-stage Liver Disease score: comparison of the performance of gadoxetic acid to gadobenate dimeglumine.

OBJECTIVES: The purpose of this study was to compare the performance of gadobenate dimeglumine-enhanced MRI and gadoxetic acid-enhanced MRI in the hepatobiliary phase (HBP) in cirrhotic patients with different degrees of liver dysfunction. METHODS: In this retrospective cross-sectional study, we analyzed the unenhanced phase and the HBP of 131 gadobenate dimeglumine-enhanced MRI examinations (gadobenate dimeglumine group) and 127 gadoxetic acid-enhanced MRI examinations (gadoxetic acid group) performed in 249 cirrhotic patients (181 men and 68 women; mean age, 64.8 years) from August 2011 to April 2017. For each MRI, the contrast enhancement index of the liver parenchyma was calculated and correlated to the Model For End-Stage Liver Disease (MELD) score (multiple linear regression analysis). A qualitative analysis of the adequacy of the HBP, adjusted for the MELD score (logistic regression analysis), was performed. RESULTS: The contrast enhancement index was inversely related (r = - 0.013) with MELD score in both gadoxetic acid and gadobenate dimeglumine group. At the same MELD score, the contrast enhancement index in the gadoxetic acid group was increased by a factor of 0.23 compared to the gadobenate dimeglumine group (p < 0.001), and the mean odds ratio to have an adequate HBP with gadoxetic acid compared to gadobenate dimeglumine was 3.64 (p < 0.001). The adequacy of the HBP in the gadoxetic acid group compared to the gadobenate dimeglumine group increased with the increase of the MELD score (exp(b)interaction = 1.233; p = 0.011). CONCLUSION: In cirrhotic patients, the hepatobiliary phase obtained with gadoxetic acid-enhanced MRI is of better quality in comparison to gadobenate dimeglumine-enhanced MRI, mainly in patients with high MELD score. KEY POINTS: • In cirrhotic patients, the adequacy of the hepatobiliary phase with gadoxetic acid-enhanced MRI is better compared to gadobenate dimeglumine-enhanced MRI. • Gadoxetic acid-enhanced MRI should be preferred to gadobenate dimeglumine-enhanced MRI in cirrhotic patients with MELD score > 10, if the hepatobiliary phase is clinically indicated. • In patients with high MELD score (> 15), the administration of the hepatobiliary agent could be useless; even though, if it is clinically indicated, we recommend to use gadoxetic acid given the higher probability of obtaining clinically relevant information.

Virtual Chromoendoscopy With FICE for the Classification of Polypoid and Nonpolypoid Raised Lesions in Ulcerative Colitis.

GOALS: The aim of this study was to analyze the performance of Fuji Intelligent Color Enhancement (FICE) using the classification of Kudo in the differentiation of neoplastic and non-neoplastic raised lesions in ulcerative colitis (UC). BACKGROUND: The Kudo classification of mucosal pit patterns is an aid for the differential diagnosis of colorectal polyps in the general population, but no systematic studies are available for all forms of raised lesions in UC. STUDY: All raised, polypoid and nonpolypoid, lesions found during consecutive surveillance colonoscopies with FICE for long-standing UC were included. In the primary prospective analysis, the Kudo classification was used to predict the histology by FICE. In a post hoc analysis, further endoscopic markers were also explored. RESULTS: Two hundred and five lesions (mean size, 8 mm; range, 2 to 30 mm) from 59 patients (mean age, 56 y; range, 21 to 79 y) were analyzed. Twenty-three neoplastic (11%), 18 hyperplastic (9%), and 164 inflammatory (80%) lesions were found. Thirty-one lesions (15%), none of which were neoplastic, were unclassifiable according to Kudo. After logistic regression, a strong negative association resulted between endoscopic activity and neoplasia, whereas the presence of a fibrin cap was significantly associated with endoscopic activity. Using FICE, the sensitivity, specificity, and positive and negative likelihood ratios of the Kudo classification were 91%, 76%, 3.8, and 0.12, respectively. The corresponding values by adding the fibrin cap as a marker of inflammation were 91%, 93%, 13, and 0.10, respectively. CONCLUSIONS: FICE can help to predict the histology of raised lesions in UC. A new classification of pit patterns, based on inflammatory markers, should be developed in the setting of UC to improve the diagnostic performance.

Ultrasonography for endoleak detection after endoluminal abdominal aortic aneurysm repair.

BACKGROUND: People with abdominal aortic aneurysm who receive endovascular aneurysm repair (EVAR) need lifetime surveillance to detect potential endoleaks. Endoleak is defined as persistent blood flow within the aneurysm sac following EVAR. Computed tomography (CT) angiography is considered the reference standard for endoleak surveillance. Colour duplex ultrasound (CDUS) and contrast-enhanced CDUS (CE-CDUS) are less invasive but considered less accurate than CT. OBJECTIVES: To determine the diagnostic accuracy of colour duplex ultrasound (CDUS) and contrast-enhanced-colour duplex ultrasound (CE-CDUS) in terms of sensitivity and specificity for endoleak detection after endoluminal abdominal aortic aneurysm repair (EVAR). SEARCH METHODS: We searched MEDLINE, Embase, LILACS, ISI Conference Proceedings, Zetoc, and trial registries in June 2016 without language restrictions and without use of filters to maximize sensitivity. SELECTION CRITERIA: Any cross-sectional diagnostic study evaluating participants who received EVAR by both ultrasound (with or without contrast) and CT scan assessed at regular intervals. DATA COLLECTION AND ANALYSIS: Two pairs of review authors independently extracted data and assessed quality of included studies using the QUADAS 1 tool. A third review author resolved discrepancies. The unit of analysis was number of participants for the primary analysis and number of scans performed for the secondary analysis. We carried out a meta-analysis to estimate sensitivity and specificity of CDUS or CE-CDUS using a bivariate model. We analysed each index test separately. As potential sources of heterogeneity, we explored year of publication, characteristics of included participants (age and gender), direction of the study (retrospective, prospective), country of origin, number of CDUS operators, and ultrasound manufacturer. MAIN RESULTS: We identified 42 primary studies with 4220 participants. Twenty studies provided accuracy data based on the number of individual participants (seven of which provided data with and without the use of contrast). Sixteen of these studies evaluated the accuracy of CDUS. These studies were generally of moderate to low quality: only three studies fulfilled all the QUADAS items; in six (40%) of the studies, the delay between the tests was unclear or longer than four weeks; in eight (50%), the blinding of either the index test or the reference standard was not clearly reported or was not performed; and in two studies (12%), the interpretation of the reference standard was not clearly reported. Eleven studies evaluated the accuracy of CE-CDUS. These studies were of better quality than the CDUS studies: five (45%) studies fulfilled all the QUADAS items; four (36%) did not report clearly the blinding interpretation of the reference standard; and two (18%) did not clearly report the delay between the two tests.Based on the bivariate model, the summary estimates for CDUS were 0.82 (95% confidence interval (CI) 0.66 to 0.91) for sensitivity and 0.93 (95% CI 0.87 to 0.96) for specificity whereas for CE-CDUS the estimates were 0.94 (95% CI 0.85 to 0.98) for sensitivity and 0.95 (95% CI 0.90 to 0.98) for specificity. Regression analysis showed that CE-CDUS was superior to CDUS in terms of sensitivity (LR Chi2 = 5.08, 1 degree of freedom (df); P = 0.0242 for model improvement).Seven studies provided estimates before and after administration of contrast. Sensitivity before contrast was 0.67 (95% CI 0.47 to 0.83) and after contrast was 0.97 (95% CI 0.92 to 0.99). The improvement in sensitivity with of contrast use was statistically significant (LR Chi2 = 13.47, 1 df; P = 0.0002 for model improvement).Regression testing showed evidence of statistically significant effect bias related to year of publication and study quality within individual participants based CDUS studies. Sensitivity estimates were higher in the studies published before 2006 than the estimates obtained from studies published in 2006 or later (P < 0.001); and studies judged as low/unclear quality provided higher estimates in sensitivity. When regression testing was applied to the individual based CE-CDUS studies, none of the items, namely direction of the study design, quality, and age, were identified as a source of heterogeneity.Twenty-two studies provided accuracy data based on number of scans performed (of which four provided data with and without the use of contrast). Analysis of the studies that provided scan based data showed similar results. Summary estimates for CDUS (18 studies) showed 0.72 (95% CI 0.55 to 0.85) for sensitivity and 0.95 (95% CI 0.90 to 0.96) for specificity whereas summary estimates for CE-CDUS (eight studies) were 0.91 (95% CI 0.68 to 0.98) for sensitivity and 0.89 (95% CI 0.71 to 0.96) for specificity. AUTHORS' CONCLUSIONS: This review demonstrates that both ultrasound modalities (with or without contrast) showed high specificity. For ruling in endoleaks, CE-CDUS appears superior to CDUS. In an endoleak surveillance programme CE-CDUS can be introduced as a routine diagnostic modality followed by CT scan only when the ultrasound is positive to establish the type of endoleak and the subsequent therapeutic management.

The architecture of diagnostic research: from bench to bedside--research guidelines using liver stiffness as an example.

UNLABELLED: The diagnostic research process can be divided into five phases, designed to establish the clinical utility of a new diagnostic test--the index test. The aim of the present review is to illustrate the study designs that are appropriate for each diagnostic phase, using clinical examples regarding liver fibrosis diagnosed with transient elastography, when possible. Phase 0 is the preclinical pilot phase during which the validity, reliability, and reproducibility of the index test are assessed in healthy and diseased people. Phase I is designed to describe the distribution of the index test results in healthy people and its normal values. Phase IIA comprises studies designed to estimate the accuracy (sensitivity and specificity) of the index test in discriminating between diseased and nondiseased people in a clinically relevant population. Phase IIB studies allow the comparison of the accuracy of different index tests; Phase IIC studies aim to evaluate the possible harms of incorporating the index test in a diagnostic-therapeutic strategy. In phase III, diagnostic test-therapeutic randomized clinical trials aim to assess the benefits and harms of the new diagnostic-therapeutic strategy versus the present strategy. Phase IV comprises large surveillance cohort studies that aim to assess the effectiveness of the new diagnostic-therapeutic strategy in clinical practice. CONCLUSION: As common in clinical research, giving excessive weight to the results of single studies and trials is likely to divert from the totality of evidence obtained through the systematic reviews of these studies, conducted with rigorous methodology and statistical methods. (Hepatology 2014;60:408-418).

Priorities for emergency department syncope research.

STUDY OBJECTIVES: There is limited evidence to guide the emergency department (ED) evaluation and management of syncope. The First International Workshop on Syncope Risk Stratification in the Emergency Department identified key research questions and methodological standards essential to advancing the science of ED-based syncope research. METHODS: We recruited a multinational panel of syncope experts. A preconference survey identified research priorities, which were refined during and after the conference through an iterative review process. RESULTS: There were 31 participants from 7 countries who represented 10 clinical and methodological specialties. High-priority research recommendations were organized around a conceptual model of ED decisionmaking for syncope, and they address definition, cohort selection, risk stratification, and management. CONCLUSION: We convened a multispecialty group of syncope experts to identify the most pressing knowledge gaps and defined a high-priority research agenda to improve the care of patients with syncope in the ED.

Mucosal healing predicts late outcomes after the first course of corticosteroids for newly diagnosed ulcerative colitis.

BACKGROUND & AIMS: It is uncertain whether mucosal healing after the first course of corticosteroids therapy predicts outcome in patients with ulcerative colitis (UC). We evaluated whether early clinical and endoscopic responses to this therapy are associated with late outcomes in UC. METHODS: Patients with newly diagnosed UC who were prescribed corticosteroid therapy (n = 157) were followed up for 5 years. They were evaluated using clinical (Powel-Tuck [PT]) and endoscopic (Baron) indexes after 3 and 6 months, then every 6 months. Outcomes at month 3 (early response) were used to identify patients with complete (group A: PT, 0-1; Baron, 0), partial (group B: PT, 0-1; Baron, 1-3), or no response (group C: persistence of clinical and endoscopic activity). The association between early and late outcomes was assessed. RESULTS: After 5 years, there were significant differences between complete and partial responders in the rates of hospitalization (25% in group A vs 48.7% in group B; P = .0152; odds ratio [OR], 2.85; 95% confidence interval [CI], 1.21-6.72), immunosuppression therapy (5% in group A vs 25.6% in group B; P = .0030; OR, 6.55; 95% CI, 1.67-25.67), colectomy (3.3% in group A vs 18.0% in group B; P = .0265; OR, 6.34; 95% CI, 1.24-32.37), and their combination (26.7% in group A vs 48.7% in group B; P = .0249; OR, 2.61; 95% CI, 1.12-6.11). After multivariate analysis, lack of mucosal healing was the only factor associated with negative outcomes at 5 years (immunosuppressors: hazard risk [HR], 10.581; 95% CI, 2.193-51.039; P = .0033; hospitalization: HR, 3.634; 95% CI, 1.556-8.485; P = .0029; colectomy: HR, 8.397; 95% CI, 1.278-55.186; P = .0268). CONCLUSIONS: No mucosal healing after corticosteroid therapy is associated with a more aggressive disease course.

Correction: Personalized risk stratification through attribute matching for clinical decision making in clinical conditions with aspecific symptoms: The example of syncope.

[This corrects the article DOI: 10.1371/journal.pone.0228725.].

Evaluation of Uterine Artery Doppler and Estrogen Milieu in Oocyte Donation Pregnancies-A Pilot Study.

Oocyte donations (OD) represent 4.5% of all in vitro fertilization (IVF) cycles. While OD pregnancies face increased risks of obstetrical complications, especially pregnancy-induced hypertension and pre-eclampsia (PE), little is known about the physiology and the physiopathology of placentation. We performed a prospective case-control study to analyze uterine artery Doppler pulsatility index (UtA-PI) and serum maternal 17ß-estradiol (17ß-E) at 11 + 0 to 13 + 6 weeks' gestation in singleton pregnancies with different modes of conception. Study groups were: 55 OD, 48 IVF with autologous oocytes from fresh cycles (Autologous-Fresh IVF), 10 IVF with autologous oocytes from frozen cycles (Autologous-Frozen IVF) and 122 spontaneously conceived pregnancies (SC). The mean UtA-PI and serum maternal 17ß-E at 11 to 13 + 6 weeks were significantly lower in OD as compared to SC and autologous IVF, either from fresh or frozen cycles. Oocyte donation presents lower UtA-PI and lower serum 17ß-E in the first trimester of pregnancy. The etiology of these particularr differences is likely multifactorial and deserves further investigation.

Modified Kudo classification can improve accuracy of virtual chromoendoscopy with FICE in endoscopic surveillance of ulcerative colitis.

Background and study aims Virtual chromoendoscopy with Fuji Intelligent Color Enhancement (FICE) has never been studied in prospective trials of endoscopic surveillance for ulcerative colitis (UC). We compared FICE and white light endoscopy (WLE) in differentiation of visible lesions in UC. Patients and methods In a prospective parallel study, we compared consecutive outpatients with UC submitted to surveillance colonoscopy with FICE or WLE. At least one visible polypoid or non-polypoid lesion for each patient was required. Random biopsies from normal mucosa, targeted biopsies or removal of suspected neoplastic lesions and targeted biopsies of unsuspected lesions were performed. In the FICE arm, neoplasia was suspected according to a modified Kudo classification (FICE-KUDO/inflammatory bowel disease [IBD]). Sensitivity (SE), specificity (SP), positive and negative likelihood ratios (LR) and negative predictive value (NPV) were analyzed. Results One hundred patients were submitted to FICE (n = 46) or WLE (n = 54). Twenty-two patients (11 in WLE, 11 in FICE) had a least one neoplastic lesion. No neoplasia was found in random biopsies. Among 275 lesions, 17 of 136 by FICE and 27 of 139 by WLE were suspected neoplasia, but 28 (14 in each arm) were true neoplastic lesions. The accuracy of FICE-KUDO/IBD vs WLE (per lesion) was: SE 93 % vs 64 % ( P  = 0.065), SP 97 % vs 86 % ( P  = 0.002), positive-LR 28.3 vs 4.5 ( P  = 0.001), negative-LR 0.07 vs 0.42 ( P  = 0.092), NPV 99 % vs 96 % ( P  = 0.083). FICE-KUDO/IBD detected more non-polypoid lesions than WLE ( P  = 0.016). Conclusions Targeted biopsies of polypoid and non-polypoid lesions, using the modified Kudo classification with FICE are more accurate than WLE in UC surveillance.

Personalized risk stratification through attribute matching for clinical decision making in clinical conditions with aspecific symptoms: The example of syncope.

BACKGROUND: Risk stratification is challenging in conditions, such as chest pain, shortness of breath and syncope, which can be the manifestation of many possible underlying diseases. In these cases, decision tools are unlikely to accurately identify all the different adverse events related to the possible etiologies. Attribute matching is a prediction method that matches an individual patient to a group of previously observed patients with identical characteristics and known outcome. We used syncope as a paradigm of clinical conditions presenting with aspecific symptoms to test the attribute matching method for the prediction of the personalized risk of adverse events. METHODS: We selected the 8 predictor variables common to the individual-patient dataset of 5 prospective emergency department studies enrolling 3388 syncope patients. We calculated all possible combinations and the number of patients in each combination. We compared the predictive accuracy of attribute matching and logistic regression. We then classified ten random patients according to clinical judgment and attribute matching. RESULTS: Attribute matching provided 253 of the 384 possible combinations in the dataset. Twelve (4.7%), 35 (13.8%), 50 (19.8%) and 160 (63.2%) combinations had a match size ≥50, ≥30, ≥20 and <10 patients, respectively. The AUC for the attribute matching and the multivariate model were 0.59 and 0.74, respectively. CONCLUSIONS: Attribute matching is a promising tool for personalized and flexible risk prediction. Large databases will need to be used in future studies to test and apply the method in different conditions.

Maintenance treatment with azathioprine in ulcerative colitis: outcome and predictive factors after drug withdrawal.

OBJECTIVES: Whether the duration of maintenance treatment with azathioprine (AZA) affects the outcome of ulcerative colitis (UC) is unclear. We investigated clinical outcomes and any predictive factors after withdrawal of AZA in UC. METHODS: In this multicenter observational retrospective study, 127 Italian UC patients, who were in steroid-free remission at the time of withdrawal of AZA, were followed-up for a median of 55 months or until relapse. The frequency of clinical relapse or colectomy after AZA withdrawal was analyzed according to demographic, clinical, and endoscopic variables. RESULTS: After drug withdrawal, a third of the patients relapsed within 12 months, half within 2 years and two-thirds within 5 years. After multivariable analysis, predictors of relapse after drug withdrawal were lack of sustained remission during AZA maintenance (hazard ratio, HR 2.350, confidence interval, CI 95% 1.434-3.852; P=0.001), extensive colitis (HR 1.793, CI 95% 1.064-3.023, P=0.028 vs. left-sided colitis; HR 2.024, CI 95% 1.103-3.717, P=0.023 vs. distal colitis), and treatment duration, with short treatments (3-6 months) more disadvantaged than >48-month treatments (HR 2.783, CI 95% 1.267-6.114, P=0.008). Concomitant aminosalicylates were the only predictors of sustained remission during AZA therapy (P=0.009). The overall colectomy rate was 10%. Predictors of colectomy were drug-related toxicity as the cause of AZA withdrawal (P=0.041), no post-AZA drug therapy (P=0.031), and treatment duration (P<0.0005). CONCLUSIONS: Discontinuation of AZA while UC is in remission is associated with a high relapse rate. Disease extent, lack of sustained remission during AZA, and discontinuation due to toxicity could stratify relapse risk. Concomitant aminosalicylates were advantageous. Prospective randomized controlled trials are needed to confirm whether treatment duration is inversely associated with outcome.