Standardized mortality ratios in multiple sclerosis: Systematic review with meta-analysis

Time-Dependent Deep Learning Prediction of Multiple Sclerosis Disability

The majority of deep learning models in medical image analysis concentrate on single snapshot timepoint circumstances, such as the identification of current pathology on a given image or volume. This is often in contrast to the diagnostic methodology in radiology where presumed pathologic findings are correlated to prior studies and subsequent changes over time. For multiple sclerosis (MS), the current body of literature describes various forms of lesion segmentation with few studies analyzing disability progression over time. For the purpose of longitudinal time-dependent analysis, we propose a combinatorial analysis of a video vision transformer (ViViT) benchmarked against traditional recurrent neural network of Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) architectures and a hybrid Vision Transformer-LSTM (ViT-LSTM) to predict long-term disability based upon the Extended Disability Severity Score (EDSS). The patient cohort was procured from a two-site institution with 703 patients' multisequence, contrast-enhanced MRIs of the cervical spine between the years 2002 and 2023. Following a competitive performance analysis, a VGG-16-based CNN-LSTM was compared to ViViT with an ablation analysis to determine time-dependency of the models. The VGG16-LSTM predicted trinary classification of EDSS score in 6 years with 0.74 AUC versus the ViViT with 0.84 AUC (p-value < 0.001 per 5 × 2 cross-validation F-test) on an 80:20 hold-out testing split. However, the VGG16-LSTM outperformed ViViT when patients with only 2 years of MRIs (n = 94) (0.75 AUC versus 0.72 AUC, respectively). Exact EDSS classification was investigated for both models using both classification and regression strategies but showed collectively worse performance. Our experimental results demonstrate the ability of time-dependent deep learning models to predict disability in MS using trinary stratification of disability, mimicking clinical practice. Further work includes external validation and subsequent observational clinical trials.

Comparison of 2 Methods for Estimating Multiple Sclerosis-Related Mortality

BACKGROUND AND OBJECTIVES

Determining whether multiple sclerosis (MS) causes death is challenging. Our objective was to contrast 2 frameworks to estimate probabilities of death attributed to MS (PMS) and other causes (Pother): the cause-specific framework (CSF), which requires the causes of death, and the excess mortality framework (EMF), which does not.

METHODS

We used data from the Observatoire Français de la Sclérose en Plaques (OFSEP, n = 37,524) and from a comparative subset where causes of death were available (4,004 women with relapsing-onset MS [R-MS]). In CSF, the probabilities were estimated using the Aalen-Johansen method. In EMF, they were estimated from the excess mortality hazard, which is the additional mortality among patients with MS as compared with the expected mortality in the matched general population. PMS values were estimated at 30 years of follow-up, (1) with both frameworks in the comparative subset, by age group at onset, and (2) with EMF only in the OFSEP population, by initial phenotype, sex, and age at onset.

RESULTS

In the comparative subset, the estimated 30-year PMS values were greater using EMF than CSF: 10.9% (95% CI 8.3-13.6) vs 8.7% (6.4-11.8) among the youngest and 20.4% (11.3-29.5) vs 16.2% (8.7-30.2) for the oldest groups, respectively. In the CSF, probabilities of death from unknown causes ranged from 1.5% (0.7-3.0) to 6.4% (2.5-16.4), and even after their reallocation, PMS values remained lower with CSF than with EMF. The estimated probabilities of being alive were close using the 2 frameworks, and the estimated POther (EMF vs CSF) was 2.6% (2.5-2.6) vs 2.1% (1.2-3.9) and 18.1% (16.9-19.3) vs 26.4% (16.5-42.2), respectively, for the youngest and oldest groups. In the OFSEP population, the estimated 30-year PMS values ranged from 7.5% (6.4-8.7) to 24.0% (19.1-28.9) in patients with R-MS and from 25.4% (21.1-29.7) to 36.8% (28.3-45.3) in primary progressive patients, depending on sex and age.

DISCUSSION

EMF has the great advantage of not requiring death certificates, their quality being less than optimal. Conceptually, it also seems more relevant because it avoids having to state, for each individual, whether death was directly or indirectly caused by MS or whether it would have occurred anyway, which is especially difficult in such chronic diseases.

Association of multiple-sclerosis-related mortality with COVID-19 and other common infections: a multiple causes of death analysis

BACKGROUND AND PURPOSE

People with multiple sclerosis (MS) suffer from higher infection-related mortality compared to the general population; however, sparse data are available on the increased risk of death associated with coronavirus disease 2019 (COVID-19) and other common types of infections.

METHODS

All mortality records and multiple-cause-of-death data in 2010-2021 of residents in the Veneto region (northeastern Italy) were extracted. Mention of specific infections was compared between death certificates reporting MS or not. Odds ratios (OR) with 95% confidence intervals (95% CI) were estimated by conditional logistic regression matching by age, sex and calendar year. The bimonthly averages of MS-related deaths in 2010-2019 were compared with those registered during the pandemic (2020-2021).

RESULTS

Of 580,015 deaths through 2010-2021, MS was mentioned in 850 cases (0.15%), 59.3% women. Influenza and pneumonia were reported in 18.4% of MS-related compared to 11.0% non-MS-related deaths (OR 2.72, 95% CI 2.28-3.25). The odds of mention of urinary tract infections was significantly greater in MS-related deaths of men (OR 8.16, 95% CI 5.23-12.7) than women (OR 3.03, 95% CI 1.82-5.02). Aspiration pneumonia, pressure ulcers/skin infections and sepsis were also significantly associated with MS-related deaths. Reporting of COVID-19 as a cause of death did not significantly differ between deaths with and without mention of MS (approximately 11% of both). However, compared to 2010-2019, peaks in MS-related deaths were observed during the pandemic waves.

CONCLUSIONS

Infections continue to play a significant role in MS-related deaths, underlying the need to improve prevention and management strategies.

Improving risk prediction for target subpopulations: Predicting suicidal behaviors among multiple sclerosis patients

Several recent studies have applied machine learning techniques to develop risk algorithms that predict subsequent suicidal behavior based on electronic health record data. In this study we used a retrospective cohort study design to test whether developing more tailored predictive models-within specific subpopulations of patients-would improve predictive accuracy. A retrospective cohort of 15,117 patients diagnosed with multiple sclerosis (MS), a diagnosis associated with increased risk of suicidal behavior, was used. The cohort was randomly divided into equal sized training and validation sets. Overall, suicidal behavior was identified among 191 (1.3%) of the patients with MS. A Naïve Bayes Classifier model was trained on the training set to predict future suicidal behavior. With 90% specificity, the model detected 37% of subjects who later demonstrated suicidal behavior, on average 4.6 years before the first suicide attempt. The performance of a model trained only on MS patients was better at predicting suicide in MS patients than that a model trained on a general patient sample of a similar size (AUC of 0.77 vs. 0.66). Unique risk factors for suicidal behavior among patients with MS included pain-related codes, gastroenteritis and colitis, and history of smoking. Future studies are needed to further test the value of developing population-specific risk models.

Cost-effectiveness analysis of hydrophilic-coated catheters in long-term intermittent catheter users in the UK

OBJECTIVE

To estimate the cost-effectiveness of single-use hydrophilic-coated intermittent catheters (HCICs) versus single-use uncoated intermittent catheters (UICs) for urinary catheterization.

METHODS

The evaluation took a UK national health service (NHS) perspective. The population of interest were people using intermittent catheters, with either a spinal cord injury or multiple sclerosis. A Markov model was developed that estimated costs and clinical evidence over the lifetime of a hypothetical cohort and applied health-related quality-of-life estimates. Model inputs were sourced from published evidence, including a network meta-analysis to inform the treatment effect (reduction in catheter-associated urinary tract infections [CAUTIs]), and were supported by expert opinion. The model outputs included per-patient lifetime costs, quality-adjusted life years (QALYs), and the incremental cost effectiveness ratio (ICER). Event counts were also produced.

RESULTS

Using HCICs instead of UICs could prevent seven CAUTI events per patient over a lifetime horizon (1.8 requiring secondary care). Overall, lifetime use of HCICs is £3,183 more expensive than use of UICs per patient. However, for these additional costs, 0.55 QALYs are gained. The ICER is £5,755 per additional QALY gained. Key drivers of the model results were identified and subject to sensitivity analyses. The results were found to be robust to parameter uncertainty.

CONCLUSION

HCICs are likely to be a cost-effective alternative to UICs, a result driven by substantial reductions in the number of CAUTIs. Their adoption across clinical practice could avoid a substantial number of infections, freeing up resources in the NHS and reducing antibiotic use in urinary catheter users.

Mortality and cause of death in multiple sclerosis in western Norway 1950-2021: a registry-based linkage study

BACKGROUND

Persons with multiple sclerosis (pwMS) have higher risk of mortality compared with the general population. Longitudinal studies are important for understanding the evolution of survival in pwMS.

OBJECTIVE

Examine changes in mortality among pwMS during the past seven decades.

METHODS

We followed pwMS from Hordaland and Møre and Romsdal in Western Norway, with disease onset from before 1950, identified from population-based epidemiological surveys and the Norwegian MS Registry and Biobank, until 1 January 2021. Data were linked to the Norwegian Cause of Death Registry to obtain underlying cause of death. We examined all-cause, and cause-specific mortality using standardised mortality ratios (SMR) and excess death rates (EDR). We calculated life expectancies and assessed survival stratified by sex, age and disease phenotype at onset. We compared hazard ratios (HRs) for mortality, in pwMS diagnosed before and after the era of disease-modifying treatment (DMT).

RESULTS

Of 3624 pwMS, 964 (55.5% women) had died, predominantly of multiple sclerosis (49.0%). Median life expectancy for pwMS was 74.3 years (95% CI 73.3 to 75.3), compared with 83.1 years for the general population (p<0.001). From disease onset, pwMS survived 14.6 years shorter than the general population (p<0.001). Overall, SMR was 2.3 (95% CI 2.13 to 2.42) and EDR was 6.8 (95% CI 6.42 to 7.09) for pwMS. Treatment-eligible pwMS diagnosed in the DMT era had the lowest risk of mortality, HR 0.49 (95% CI 0.34 to 0.70,p<0.001).

CONCLUSION

Excess mortality among pwMS declined during the past seven decades, possibly due to improved diagnostics, better symptomatic treatment and access to DMTs.

Association of multiple sclerosis with risk of mortality among a nationally representative sample of adults in the United States

Background

Multiple sclerosis (MS) has been associated with increased mortality ratios, but few studies have investigated the independent association of MS with mortality.

Objective

To examine the prospective association of MS with risk of mortality in a nationally representative sample of U.S. adults.

Methods

This prospective study included 23,053 adults aged 45–79 years who participated in the National Health Interview Survey in 2002 and 2008. Physician-diagnosed MS was reported by participants during household interviews. These participants were linked to death records from survey date through December 31, 2015.

Results

Among the 23,053 participants included in this study, 120 reported a physician’s diagnosis of MS, with a higher prevalence in females (0.85%) than in males (0.31%). During on average 9.4 years (maximum 13.8 years) of observation, 4208 deaths occurred. After adjustment for age, sex, race/ethnicity, socioeconomic factors, lifestyle factors, and BMI, participants with MS had an 80% higher risk of mortality (HR 1.80; 95% CI, 1.11–2.92), compared with those without MS. The association remained significant (HR 1.75; 95% CI, 1.07–2.87) after further adjustment for baseline diabetes, cardiovascular disease, chronic lung disease, and cancer.

Conclusion

In this nationally representative sample of U.S. adults, MS was associated with an increased risk of mortality.

How to choose initial treatment in multiple sclerosis patients: a case-based approach

BACKGROUND

Immunotherapy dramatically changed the natural history of multiple sclerosis (MS), which was classically associated with severe disability. Treatment strategies advocate that early control of disease activity is crucial to avoid progressive disability, and the use of high efficacy drugs may be beneficial, but safety is a concern. Choosing the disease-modifying therapy is challenging in clinical practice and should be further discussed.

OBJECTIVE

To discuss the state of art of selecting the initial therapy for relapsing MS patients.

METHODS

We used a case-based approach followed by clinical discussion, exploring therapeutic options in different MS settings.

RESULTS

We presented clinical cases profile compatible with the use of MS therapies, classified into moderate and high efficacy. In the moderate efficacy group, we discussed interferons, glatiramer acetate, teriflunomide and dimethyl fumarate, while in the high efficacy group we discussed fingolimod, cladribine, natalizumab, ocrelizumab, alemtuzumab and ofatumumab.

CONCLUSION

Advances in MS treatment are remarkable. Strong evidence supports the use of early high efficacy therapy. However, biomarkers, clinical and radiologic prognostic factors, as well as patients' individual issues, should be valued and considered for a personalized treatment decision.