Impact of Clostridioides difficile infection in patients admitted with ulcerative colitis

BACKGROUND

Patients with ulcerative colitis (UC) are at increased risk of Clostridioides difficile infection (CDI), which is the principal causative agent of nosocomial diarrhoea in western countries. This has been related to complications such as need of colectomy and mortality among these patients. The aim of this study was to assess the incidence and impact of CDI in patients hospitalised with UC.

METHODS

Case-control retrospective study including patients admitted due to a UC flare from January 2000 to September 2018. Porpensity score matching (PSM) was performed to minimise selection bias taking into account the small number of cases compared to controls.

RESULTS

339 patients were included; CDI in 35 (10.3%) patients. After PSM, 35 (33.33%) cases and 70 (66.67%) controls were analysed. Patients with CDI presented higher rates of readmission (52.9% vs. 21.4%, p = .001), increased mortality within the first 3 months post-discharge (5.9% vs. 0%, p = .042) and increased need of therapy intensification in the first year after admission (20.7% vs. 12.5%, p = .001). No risk factors for CDI were identified. Multivariable cox regression showed that treatment with 5-aminosalycilates at baseline (HR 0.42, 95% CI 0.18-0.92) and albumin <3.5 g/dL (HR 3.11, 95% CI 1.21-8.03) were associated with worse outcomes.

CONCLUSIONS

CDI is a prevalent situation in hospitalised UC patients related to higher mortality within the first 3 months after the infection, need for therapy intensification within the first year and readmission. Our results underline the importance of CDI detection in patients with a flare of UC.

Effectiveness and Safety of the Switch from Remicade® to CT-P13 in Patients with Inflammatory Bowel Disease

BACKGROUND AND AIMS

To evaluate the clinical outcomes in patients with IBD after switching from Remicade® to CT-P13 in comparison with patients who maintain Remicade®.

METHODS

Patients under Remicade® who were in clinical remission with standard dosage at study entry were included. The 'switch cohort' [SC] comprised patients who made the switch from Remicade® to CT-P13, and the 'non-switch' cohort [NC] patients remained under Remicade®.

RESULTS

A total of 476 patients were included: 199 [42%] in the SC and 277 [58%] in the NC. The median follow-up was 18 months in the SC and 23 months in the NC [p < 0.01]. Twenty-four out of 277 patients relapsed in the NC; the incidence of relapse was 5% per patient-year. The cumulative incidence of relapse was 2% at 6 months and 10% at 24 months in this group. Thirty-eight out of 199 patients relapsed in the SC; the incidence rate of relapse was 14% per patient-year. The cumulative incidence of relapse was 5% at 6 months and 28% at 24 months. In the multivariate analysis, the switch to CT-P13 was associated with a higher risk of relapse (HR = 3.5, 95% confidence interval [CI] = 2-6). Thirteen percent of patients had adverse events in the NC, compared with 6% in the SC [p < 0.05].

CONCLUSIONS

Switching from Remicade® to CT-P13 might be associated with a higher risk of clinical relapse, although this fact was not supported in our study by an increase in objective markers of inflammation. The nocebo effect might have influenced this result. Switching from Remicade® to CT-P13 was safe.

Extracolonic Cancer in Inflammatory Bowel Disease: Data from the GETECCU Eneida Registry

OBJECTIVES

The objective of this study was (a) To know the prevalence and distribution of extracolonic cancer (EC) in patients with inflammatory bowel disease (IBD); (b) To estimate the incidence rate of EC; (c) To evaluate the association between EC and treatment with immunosuppressants and anti-tumor necrosis factor (TNF) agents.

METHODS

This was an observational cohort study.

INCLUSION CRITERIA

IBD and inclusion in the ENEIDA Project (a prospectively maintained registry) from GETECCU.

EXCLUSION CRITERIA

Patients with EC before the diagnosis of IBD, lack of relevant data for this study, and previous treatment with immunosuppressants other than corticosteroids, thiopurines, methotrexate, or anti-TNF agents. The Kaplan-Meier method was used to evaluate the impact of several variables on the risk of EC, and any differences between survival curves were evaluated using the log-rank test. Stepwise multivariate Cox regression analysis was used to investigate factors potentially associated with the development of EC, including drugs for the treatment of IBD, during follow-up.

RESULTS

A total of 11,011 patients met the inclusion criteria and were followed for a median of 98 months. Forty-eight percent of patients (5,303) had been exposed to immunosuppressants or anti-TNF drugs, 45.8% had been exposed to thiopurines, 4.7% to methotrexate, and 21.6% to anti-TNF drugs. The prevalence of EC was 3.6%. In the multivariate analysis, age (HR=1.05, 95% CI=1.04-1.06) and having smoked (hazards ratio (HR)=1.47, 95% confidence interval (CI)=1.10-1.80) were the only variables associated with a higher risk of EC.

CONCLUSIONS

Neither immunosuppressants nor anti-TNF drugs seem to increase the risk of EC. Older age and smoking were associated with a higher prevalence of EC.

[Pigment dispersion syndrome in a patient with megalocornea]

CASE REPORT

A case of a patient of 12 years old with megalocornea and bilateral pigment dispersion syndrome is reported. An ultrasound biomicroscopic study was performed with the UBM 840 (Zeiss-Humphrey, 50 MHz) and an iris concavity with an iridolenticular and iridozonular contact was observed. A new exam after pilocarpine instillation drops showed the resolution of the iris concavity, zonular and lenticular contact.

DISCUSSION

The anterior chamber configuration of this patient analyzed with ultrasound biomicroscopy allowed us to understand the mechanism of the pigment dispersion syndrome.

Neural networks with a continuous squashing function in the output are universal approximators

In 1989 Hornik as well as Funahashi established that multilayer feedforward networks without the squashing function in the output layer are universal approximators. This result has been often used improperly because it has been applied to multilayer feedforward networks with the squashing function in the output layer. In this paper, we will prove that also this kind of neural networks are universal approximators, i.e. they are capable of approximating any Borel measurable function from one finite dimensional space into (0,1)" to any desired degree of accuracy, provided sufficiently many hidden units are available.