Enriching the FIDEO ontology with food-drug interactions from online knowledge sources

The increasing number of articles on adverse interactions that may occur when specific foods are consumed with certain drugs makes it difficult to keep up with the latest findings. Conflicting information is available in the scientific literature and specialized knowledge bases because interactions are described in an unstructured or semi-structured format. The FIDEO ontology aims to integrate and represent information about food-drug interactions in a structured way. This article reports on the new version of this ontology in which more than 1700 interactions are integrated from two online resources: DrugBank and Hedrine. These food-drug interactions have been represented in FIDEO in the form of precompiled concepts, each of which specifies both the food and the drug involved. Additionally, competency questions that can be answered are reviewed, and avenues for further enrichment are discussed.

Efficacy of carbapenem vs non carbapenem β-lactam therapy as empiric antimicrobial therapy in patients with extended-spectrum β-lactamase-producing Enterobacterales urinary septic shock: a propensity-weighted multicenter cohort study

BACKGROUND

The rise in antimicrobial resistance is a global threat responsible for about 33,000 deaths in 2015 with a particular concern for extended-spectrum beta-lactamase-producing Enterobacterales (ESBL-E) and has led to a major increase in the use of carbapenems, last-resort antibiotics.

METHODS

In this retrospective propensity-weighted multicenter observational study conducted in 11 ICUs, the purpose was to assess the efficacy of non carbapenem regimen (piperacillin-tazobactam (PTZ) + aminoglycosides or 3rd-generation cephalosporin (3GC) + aminoglycosides) as empiric therapy in comparison with carbapenem in extended-spectrum β-lactamase-producing Enterobacterales (ESBL-E) urinary septic shock. The primary outcome was Day-30 mortality.

RESULTS

Among 156 patients included in this study, 69 received a carbapenem and 87 received non carbapenem antibiotics as empiric treatment. Baseline clinical characteristics were similar between the two groups. Patients who received carbapenem had similar Day-30 mortality (10/69 (15%) vs 6/87 (7%), OR = 1.99 [0.55; 5.34] p = 0.16), illness severity, resolution of septic shock, and ESBL-E infection recurrence rates than patients who received an empiric non carbapenem therapy. The rates of secondary infection with C. difficile were comparable.

CONCLUSIONS

In ESBL-E urinary septic shock, empiric treatment with a non carbapenem regimen, including systematically aminoglycosides, was not associated with higher mortality, compared to a carbapenem regimen.

Acute respiratory distress syndrome after SARS-CoV-2 infection on young adult population: International observational federated study based on electronic health records through the 4CE consortium

PURPOSE

In young adults (18 to 49 years old), investigation of the acute respiratory distress syndrome (ARDS) after severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection has been limited. We evaluated the risk factors and outcomes of ARDS following infection with SARS-CoV-2 in a young adult population.

METHODS

A retrospective cohort study was conducted between January 1st, 2020 and February 28th, 2021 using patient-level electronic health records (EHR), across 241 United States hospitals and 43 European hospitals participating in the Consortium for Clinical Characterization of COVID-19 by EHR (4CE). To identify the risk factors associated with ARDS, we compared young patients with and without ARDS through a federated analysis. We further compared the outcomes between young and old patients with ARDS.

RESULTS

Among the 75,377 hospitalized patients with positive SARS-CoV-2 PCR, 1001 young adults presented with ARDS (7.8% of young hospitalized adults). Their mortality rate at 90 days was 16.2% and they presented with a similar complication rate for infection than older adults with ARDS. Peptic ulcer disease, paralysis, obesity, congestive heart failure, valvular disease, diabetes, chronic pulmonary disease and liver disease were associated with a higher risk of ARDS. We described a high prevalence of obesity (53%), hypertension (38%- although not significantly associated with ARDS), and diabetes (32%).

CONCLUSION

Trough an innovative method, a large international cohort study of young adults developing ARDS after SARS-CoV-2 infection has been gather. It demonstrated the poor outcomes of this population and associated risk factor.

MRI and rectal endoscopy sonography performance to diagnose the digestive depth infiltration of pelvic endometriosis

INTRODUCTION

The main objective of this study was to evaluate the performances of MRI and rectal endoscopy sonography (RES) in predicting the depth of bowel wall infiltration by deep infiltrating endometriosis (DIE).

MATERIAL AND METHOD

We conducted a single center retrospective study from April 2014 to March 2020 including all patients who had undergone digestive tract resection (discoid or segmental) for DIE removal and who had benefited from full preoperative imaging workup based on both pelvic MRI and RES.

RESULTS

Fifty two patients were enrolled in the study. Median age was 35.8 years (26.1-44.5 years). Indications for surgery mainly comprised chronic pelvic pain (94.2%) and infertility (36.5%). Overall, pathological examination showed digestive involvement in 92.3% of patients, while transmural infiltration was found in 38.4% of cases. In contrast, both MRI and RES suspected transmural involvement in 42 patients (80.8%). Corresponding sensitivity and specificity were 0.95 [95% CI (0.751-0.999)] and 0.28 [95% CI (0.137-0.467)], respectively. Our results revealed agreement between MRI and RES in 85% of cases with a kappa at 0.5 [95% CI (0.207-0.803), moderate agreement]. Subgroup analysis in patients with transmural MRI lesions showed a sensitivity of 0.95 [95% CI (0.740-0.999)] and a specificity of 0.13 [95% CI (0.028-0.336)].

CONCLUSION

Our study suggests that performing a second-line examination is not useful if there is no transmural impairment in MRI or RES. Nevertheless, the combination of these two preoperative examinations seems to be essential for the evaluation of the depth of digestive involvement of endometriosis to guide surgical management as effectively as possible. The constitution and training of multidisciplinary expert groups must be developed to be able to offer optimal patient management.

The benefit of augmenting open data with clinical data-warehouse EHR for forecasting SARS-CoV-2 hospitalizations in Bordeaux area, France

Objective

To develop an accurate regional forecast algorithm to predict the number of hospitalized patients and to assess the benefit of the Electronic Health Records (EHR) information to perform those predictions.

Materials and Methods

Aggregated data from SARS-CoV-2 and weather public database and data-warehouse of the Bordeaux hospital were extracted from 2020-05-16 to 2022-01-17. The outcomes were the number of hospitalized patients in the Bordeaux Hospital at 7 and 14 days. We compared the performance of different data sources, feature engineering and machine learning models.

Results

During the period of 88 weeks, 2561 hospitalizations due to COVID19 were recorded at the Bordeaux Hospital. The model achieving the best performance was an elastic-net penalized linear regression using all available data with a median relative error at 7 and 14 days of 0.136 [0.063; 0.223] and 0.198 [0.105; 0.302] hospitalizations, respectively. Electronic health records (EHRs) from the hospital data-warehouse improved median relative error at 7 and 14 days by 10.9 and 19.8%, respectively. Graphical evaluation showed remaining forecast error was mainly due to delay in slope shift detection.

Discussion

Forecast model showed overall good performance both at 7 and 14 days which were improved by the addition of the data from Bordeaux Hospital data-warehouse.

Conclusion

The development of hospital data-warehouse might help to get more specific and faster information than traditional surveillance system, which in turn will help to improve epidemic forecasting at a larger and finer scale.

LAY SUMMARY

The objective of this work was to develop a forecast algorithm to predict the number of hospitalized patients at Bordeaux Hospital. In addition, we assessed the benefit of the Electronic Health Records (EHRs) information to perform those predictions.

To perform this task, we used data between 2020-05-16 and 2022-01-17 from national database on SARS-CoV-2 epidemics, public database on weather and the data-warehouse of the Bordeaux hospital. The outcomes were the number of hospitalized patients in the Bordeaux Hospital at 7 and 14 days.

During the period of 88 weeks, 2561 hospitalizations due to COVID19 were recorded at the Bordeaux Hospital. The best model had an error of 13.6% at 7 days and 19.8% at 14 days. EHRs from the hospital data-warehouse improved the performance by 10% at 7 days and 20% at 14 days. Graphical evaluation showed remaining forecast error was mainly due to delay in slope shift detection.

Forecast model showed overall good performance which were improved by the addition of EHRs data. The development of hospital data-warehouse might help to get more specific and faster information than traditional surveillance system, which in turn will help to improve epidemic forecasting at a larger and finer scale.

SurvMaximin: Robust federated approach to transporting survival risk prediction models

OBJECTIVE

For multi-center heterogeneous Real-World Data (RWD) with time-to-event outcomes and high-dimensional features, we propose the SurvMaximin algorithm to estimate Cox model feature coefficients for a target population by borrowing summary information from a set of health care centers without sharing patient-level information.

MATERIALS AND METHODS

For each of the centers from which we want to borrow information to improve the prediction performance for the target population, a penalized Cox model is fitted to estimate feature coefficients for the center. Using estimated feature coefficients and the covariance matrix of the target population, we then obtain a SurvMaximin estimated set of feature coefficients for the target population. The target population can be an entire cohort comprised of all centers, corresponding to federated learning, or a single center, corresponding to transfer learning.

RESULTS

Simulation studies and a real-world international electronic health records application study, with 15 participating health care centers across three countries (France, Germany, and the U.S.), show that the proposed SurvMaximin algorithm achieves comparable or higher accuracy compared with the estimator using only the information of the target site and other existing methods. The SurvMaximin estimator is robust to variations in sample sizes and estimated feature coefficients between centers, which amounts to significantly improved estimates for target sites with fewer observations.

CONCLUSIONS

The SurvMaximin method is well suited for both federated and transfer learning in the high-dimensional survival analysis setting. SurvMaximin only requires a one-time summary information exchange from participating centers. Estimated regression vectors can be very heterogeneous. SurvMaximin provides robust Cox feature coefficient estimates without outcome information in the target population and is privacy-preserving.

Evaluating the Relevance of Virtual Drugs

Information about drugs is numerous and varied, and many drugs can share the same information. Grouping drugs that have common characteristics can be useful to avoid redundancy and facilitate interoperability. Our work focused on the evaluation of the relevance of classes allowing this type of grouping: the "Virtual Drug". Thus, in this paper, we describe the process of creating this class from the data of the French Public Drug Database, which is then evaluated against the codes of the Anatomical Therapeutic Chemical classification associated with the drugs. Our evaluation showed that 99.55% of the "Virtual Drug" classes have a good intra-class consistency.

Data Element Mapping in the Data Privacy Era

Secondary use of health data is made difficult in part because of large semantic heterogeneity. Many efforts are being made to align local terminologies with international standards. With increasing concerns about data privacy, we focused here on the use of machine learning methods to align biological data elements using aggregated features that could be shared as open data. A 3-step methodology (features engineering, blocking strategy and supervised learning) was proposed. The first results, although modest, are encouraging for the future development of this approach.

Maternal and neonatal morbidity after forceps or spatulas-assisted delivery in preterm birth

OBJECTIVE

The aim of this study was to assess perinatal morbidity associated with spatulas or forceps assisted delivery in preterm birth.

STUDY DESIGN

This is a retrospective cohort study including all women with assisted deliveries on singleton pregnancy in cephalic presentation, before 37 weeks of gestation, in two tertiary care centers. We compared forceps-assisted deliveries with spatula-assisted deliveries. The main outcome was the rate of neonatal birth trauma. Secondary outcomes included other neonatal parameters, maternal outcomes and obstetric anal sphincter injuries.

RESULTS

Out of 37 002 deliveries, 59 (0.2 %) preterm assisted deliveries with forceps and 111 (0.3%) preterm spatulas deliveries were included. The rate of neonatal birth trauma was low for both devices, without significant difference (3.4% in Forceps group vs 0.9% in Spatulas group, p = 0.28). The rate of episiotomy was 79.7% after forceps-assisted delivery and 48.6% for spatulas (p < 0.001). The rate of obstetric anal sphincter injuries was 1.7% and 2.7% respectively (p = 0,9).

CONCLUSION

The rate of birth trauma was low in both forceps-assisted deliveries and spatula-assisted deliveries and was not significantly different between the two groups.

Antiplatelet therapy increases symptomatic ICH risk after thrombolysis and thrombectomy

BACKGROUND AND PURPOSE

The influence of chronic treatment by antiplatelet drug (APD) at stroke onset on the outcomes of patients with acute ischemic stroke (AIS) treated with combined intravenous thrombolysis (IVT) and endovascular therapy (EVT) is unclear. We investigated whether prior APD use influences the risk of symptomatic intracranial hemorrhage (sICH) and functional outcome in AIS patients treated with combined reperfusion therapy.

METHODS

A single-center retrospective analysis of AIS patients with proximal intracranial occlusion who underwent IVT and EVT between January 2015 and May 2017. The main outcomes were the incidence of sICH using the Heidelberg Bleeding Classification and patients' functional status at 90 days, as defined by the modified Rankin scale (mRS). Outcomes were evaluated according to daily exposure to APD, and associations were assessed using multivariate logistic regression analysis.

RESULTS

This study included 204 patients: 71 (34.8%) were taking APD before AIS. Patients with chronic treatment by APD at stroke onset had a higher rate of sICH (26.7% vs. 3.7%; p< .001) and worse functional outcome (mRS >2) at 90 days (69% vs. 36.8%; p < .001). Prior APD use was associated with an increased likelihood of sICH (OR 9.8; 95%CI [3.6-31.3], p < .05) and of functional dependence at 90 days (OR 5.72; 95%CI [2.09-1.72], p < .001), independent of confounders on multivariate analysis.

CONCLUSIONS

Chronic treatment by APD at stroke onset in AIS patients with proximal intracranial occlusion treated using IVT and EVT increases the risk of sICH and worsens the functional prognosis. Further investigation to refine acute revascularization strategies in this population might be required.

The response to TNF blockers depending on their comparator in rheumatoid arthritis clinical trials: the lessebo effect, a meta-analysis

OBJECTIVE

To compare the effect of the biological reference agents (infliximab, etanercept, adalimumab) in rheumatoid arthritis (RA) in pivotal superiority placebo-controlled trials (reference agent vs placebo) vs their effect in equivalence active comparator-controlled trials (reference agent vs biosimilar).

METHODS

The PubMed, EMBASE, Cochrane, databases were searched for randomized, double-blind, controlled trials up to March 2020 comparing a biological reference agent vs placebo or biosimilar. The study assessed the American College of Rheumatology (ACR) 20/50/70 responses of the reference agent in these groups (Reference-pbo and Reference-bs, respectively). The effect of the reference agent in both groups was estimated with 95% confidence intervals (95%CI), pooled using random-effects models and then compared using a meta-regression model.

RESULTS

We included 31 trials. The main characteristics of the population (disease duration and activity, % seropositivity and methotrexate dose) of the population in both groups were similar. The meta-analysis found a better ACR20 response to the biological originator in the Reference-bs group with a global rate of 70% (95%CI, 66-74) compared with 59% (95%CI, 55-62) in the reference-pbo group (p= 0.001). A significant difference was also found for ACR 50 [44% (95%CI, 39-50) vs 35% (95%CI, 31-39) respectively, p< 0.01].

CONCLUSION

Effect of the reference biologic agent was better when compared with an active drug to a placebo. This could be linked to an increased placebo effect in active comparator-controlled studies or a nocebo effect in placebo-controlled studies. This effect can be called the Lessebo effect.

Aligning an interface terminology to the Logical Observation Identifiers Names and Codes (LOINC®)

Objective

Our study consists in aligning the interface terminology of the Bordeaux university hospital (TLAB) to the Logical Observation Identifiers Names and Codes (LOINC). The objective was to facilitate the shared and integrated use of biological results with other health information systems.

Materials and Methods

We used an innovative approach based on a decomposition and re-composition of LOINC concepts according to the transversal relations that may be described between LOINC concepts and their definitional attributes. TLAB entities were first anchored to LOINC attributes and then aligned to LOINC concepts through the appropriate combination of definitional attributes. Finally, using laboratory results of the Bordeaux data-warehouse, an instance-based filtering process has been applied.

Results

We found a small overlap between the tokens constituting the labels of TLAB and LOINC. However, the TLAB entities have been easily aligned to LOINC attributes. Thus, 99.8% of TLAB entities have been related to a LOINC analyte and 61.0% to a LOINC system. A total of 55.4% of used TLAB entities in the hospital data-warehouse have been mapped to LOINC concepts. We performed a manual evaluation of all 1-1 mappings between TLAB entities and LOINC concepts and obtained a precision of 0.59.

Conclusion

We aligned TLAB and LOINC with reasonable performances, given the poor quality of TLAB labels. In terms of interoperability, the alignment of interface terminologies with LOINC could be improved through a more formal LOINC structure. This would allow queries on LOINC attributes rather than on LOINC concepts only.

Linkage of Hospital Records and Death Certificates by a Search Engine and Machine Learning

Introduction

Vital status is of central importance to hospital clinical research. However, hospital information systems record only in-hospital death information. Recently, the French government released a publicly available dataset containing death-certificate data for over 25 million individuals. The objective of this study was to link French death certificates to the Bordeaux University Hospital records to complete the vital status information.

Materials and Methods

Our linkage strategy was composed of a search engine to reduce the number of comparisons and machine-learning algorithms. The overall pipeline was evaluated by assembling a file containing 3,565 in-hospital deaths and 15,000 alive persons.

Results

The recall and precision of our linkage strategy were 97.5% and 99.97% for the upper threshold and 99.4% and 98.9% for the lower threshold, respectively.

Conclusion

In this study, we demonstrated the feasibility of accurately linking hospital records with death certificates using a search engine and machine learning.

International Comparisons of Harmonized Laboratory Value Trajectories to Predict Severe COVID-19: Leveraging the 4CE Collaborative Across 342 Hospitals and 6 Countries: A Retrospective Cohort Study

Objectives

To perform an international comparison of the trajectory of laboratory values among hospitalized patients with COVID-19 who develop severe disease and identify optimal timing of laboratory value collection to predict severity across hospitals and regions.

Design

Retrospective cohort study.

Setting

The Consortium for Clinical Characterization of COVID-19 by EHR (4CE), an international multi-site data-sharing collaborative of 342 hospitals in the US and in Europe.

Participants

Patients hospitalized with COVID-19, admitted before or after PCR-confirmed result for SARS-CoV-2.

Primary and secondary outcome measures

Patients were categorized as "ever-severe" or "never-severe" using the validated 4CE severity criteria. Eighteen laboratory tests associated with poor COVID-19-related outcomes were evaluated for predictive accuracy by area under the curve (AUC), compared between the severity categories. Subgroup analysis was performed to validate a subset of laboratory values as predictive of severity against a published algorithm. A subset of laboratory values (CRP, albumin, LDH, neutrophil count, D-dimer, and procalcitonin) was compared between North American and European sites for severity prediction.

Results

Of 36,447 patients with COVID-19, 19,953 (43.7%) were categorized as ever-severe. Most patients (78.7%) were 50 years of age or older and male (60.5%). Longitudinal trajectories of CRP, albumin, LDH, neutrophil count, D-dimer, and procalcitonin showed association with disease severity. Significant differences of laboratory values at admission were found between the two groups. With the exception of D-dimer, predictive discrimination of laboratory values did not improve after admission. Sub-group analysis using age, D-dimer, CRP, and lymphocyte count as predictive of severity at admission showed similar discrimination to a published algorithm (AUC=0.88 and 0.91, respectively). Both models deteriorated in predictive accuracy as the disease progressed. On average, no difference in severity prediction was found between North American and European sites.

Conclusions

Laboratory test values at admission can be used to predict severity in patients with COVID-19. Prediction models show consistency across international sites highlighting the potential generalizability of these models.

Comparison of maternal and neonatal outcomes after forceps or spatulas-assisted delivery

OBJECTIVE

The aim of this study was to compare the perinatal outcomes associated with spatulas or forceps assisted delivery.

STUDY DESIGN

This is a bicentric retrospective cohort study including all assisted deliveries in cephalic presentation after 37 weeks of gestation, performed on singleton pregnancy with forceps and with spatulas in two tertiary centers. The main outcome was the rate of episiotomy. Secondary outcomes included obstetric anal sphincter injuries (OASIS), maternal outcomes and neonatal parameters.

RESULTS

Out of 37 002 deliveries, the overall rate of assisted delivery was 11.4 %, and 1 041 (2.8 %) assisted deliveries with forceps and 2 462 (6.7 %) spatulas deliveries were included. The rate of episiotomy was 90.3 % after forceps-assisted delivery and 70.9 % for spatulas (p < 0.001). The rate of OASIS was 7.2 % and 5.6 % respectively (p = 0.06). A slight but significant decrease in neonatal trauma after spatulas was observed.

CONCLUSION

In this retrospective cohort study, the episiotomy rate was higher with forceps assisted deliveries than with spatulas. Both instruments have low neonatal morbidity and are similar regarding OASIS.

Factors predicting the spontaneous regression of cervical high-grade squamous intraepithelial lesions (HSIL/CIN2)

PURPOSE

To determine clinical, pathological and virological factors predicting the spontaneous regression of HSIL/CIN2.

METHODS

This retrospective study included 73 patients with HSIL/CIN2 diagnosed by biopsy between 2012 and 2016 and followed-up without treatment in the department of gynecology at Bordeaux University Hospital. All biopsies sampled inside or outside our department were reviewed and immunolabelled for p16 and Ki67. The response rate was the regression or the disappearance of HSIL/CIN2 as defined by the regression or the disappearance of initial colposcopic findings, cytological and/or histological results.

RESULTS

The diagnosis of CIN2 was confirmed in 63 of 70 biopsies available for review. The Cohen's kappa coefficient was κ = 90%, indicating almost perfect inter-observer agreement. The lesion spontaneously regressed or disappeared in 36 of 60 patients (60%) with confirmed CIN2 during a median follow-up of 20 months (range 6-55). Baseline factors influencing the response rate were colposcopic findings (69% with minor change vs 31% with major change, p = 0.033), cytological results (72% with ASCUS/LSIL vs 28% with ASC-H/HSIL, p = 0.018), and HPV genotyping (71% with HPV not 16 vs 42% with HPV-16, p = 0.027). The other factors (age, smoking, surface area of the lesion, p16 and Ki67 expressions) did not significantly influence the outcome.

CONCLUSION

Colposcopic findings, cytological results, and HPV genotyping were baseline factors predicting spontaneous regression of HSIL/CIN2.

International electronic health record-derived COVID-19 clinical course profiles: the 4CE consortium

We leveraged the largely untapped resource of electronic health record data to address critical clinical and epidemiological questions about Coronavirus Disease 2019 (COVID-19). To do this, we formed an international consortium (4CE) of 96 hospitals across five countries (www.covidclinical.net). Contributors utilized the Informatics for Integrating Biology and the Bedside (i2b2) or Observational Medical Outcomes Partnership (OMOP) platforms to map to a common data model. The group focused on temporal changes in key laboratory test values. Harmonized data were analyzed locally and converted to a shared aggregate form for rapid analysis and visualization of regional differences and global commonalities. Data covered 27,584 COVID-19 cases with 187,802 laboratory tests. Case counts and laboratory trajectories were concordant with existing literature. Laboratory tests at the time of diagnosis showed hospital-level differences equivalent to country-level variation across the consortium partners. Despite the limitations of decentralized data generation, we established a framework to capture the trajectory of COVID-19 disease in patients and their response to interventions.

International electronic health record-derived COVID-19 clinical course profiles: the 4CE consortium

We leveraged the largely untapped resource of electronic health record data to address critical clinical and epidemiological questions about Coronavirus Disease 2019 (COVID-19). To do this, we formed an international consortium (4CE) of 96 hospitals across five countries (www.covidclinical.net). Contributors utilized the Informatics for Integrating Biology and the Bedside (i2b2) or Observational Medical Outcomes Partnership (OMOP) platforms to map to a common data model. The group focused on temporal changes in key laboratory test values. Harmonized data were analyzed locally and converted to a shared aggregate form for rapid analysis and visualization of regional differences and global commonalities. Data covered 27,584 COVID-19 cases with 187,802 laboratory tests. Case counts and laboratory trajectories were concordant with existing literature. Laboratory tests at the time of diagnosis showed hospital-level differences equivalent to country-level variation across the consortium partners. Despite the limitations of decentralized data generation, we established a framework to capture the trajectory of COVID-19 disease in patients and their response to interventions.

Romedi: An Open Data Source About French Drugs on the Semantic Web

The W3C project, "Linking Open Drug Data" (LODD), linked several publicly available sources of drug data together. So far, French data, like marketed drugs and their summary of product characteristics, were not integrated and remained difficult to query. In this paper, we present Romedi (Référentiel Ouvert du Médicament), an open dataset that links French data on drugs to international resources. The principles and standard recommendations created by the W3C for sharing information were adopted. Romedi was connected to the Unified Medical Language System and DrugBank, two central resources of the LODD project. A SPARQL endpoint is available to query Romedi and services are provided to annotate textual content with Romedi terms. This paper describes its content, its services, its links to external resources, and expected future developments.

Neurodegeneration of the Substantia Nigra after Ipsilateral Infarct: MRI R2* Mapping and Relationship to Clinical Outcome

Background The substantia nigra (SN) is suspected to be affected after remote infarction, in view of its large array of connections with the supratentorial brain. Whether secondary involvement of SN worsens overall clinical outcome after a supratentorial stroke has not previously been studied. Purpose To assess longitudinal changes in SN R2* by using MRI in the setting of ipsilesional supratentorial infarct and the relationship of SN signal change to clinical outcome. Materials and Methods Participants prospectively included from 2012 to 2015 were evaluated at 24-72 hours (baseline visit) and at 1 year with MRI to quantify R2*. The SN was segmented bilaterally to calculate an R2* asymmetry index (SN-AI); greater SN-AI indicated greater relative R2* in the ipsilateral compared with contralateral SN. The 95th percentile of R2* (hereafter, SN-AI₉₅) was compared according to infarct location with mixed linear regression models. We also conducted voxel-based comparisons of R2* and identified individual infarcted voxels associated with high SN-AI₉₅ through voxel-based lesion-symptom mapping. Multivariable regression models tested the association between SN-AI₉₅ and clinical scores. Results A total of 181 participants were evaluated (127 men, 54 women; mean age ± standard deviation, 64.2 years ± 13.1; 75 striatum infarcts, 106 other locations). Visual inspection, SN-AI₉₅, and average maps consistently showed higher SN R2* at 1 year if ipsilateral striatum was infarcted than if it was not (SN-AI₉₅, 4.25 vs -0.88; P < .001), but this was not observed at baseline. The striatal location of the infarct was associated with higher SN-AI₉₅ at 1 year independently from infarct volume, SN-AI₉₅ at baseline, microbleeds, age, and sex (β = 4.99; P < .001). Voxel-based lesion-symptom mapping confirmed that striatum but also insula, internal capsule, and external capsule were associated with higher SN-AI₉₅ at 1 year. SN-AI₉₅ was an independent contributor of poor motor outcome (Box and Block Test, β = -.62 points; P = .01). Conclusion In patients with stroke, greater substantia nigra R2*, likely reflective of greater iron content, can be observed at 1 year ipsilateral from remote infarcts of specific location, which is associated with worse motor function. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Vernooij in this issue.