Claudia Spies’s research while affiliated with Charité Universitätsmedizin Berlin and other places

Some patients undergoing surgical procedures display long-term post-surgery cognitive impairment (post-operative cognitive dysfunction; POCD), which may precipitate progression to dementia. We investigated whether preoperative cognitive impairment defined using specific cognitive tests (Paired-Associates Learning and Spatial-Span from the Cambridge Neuropsychological Test Automated Battery, (CANTAB) was associated with increased risk of POCD. N = 590 patients >65years and a matched control group n=114 comprised the final sample. Patients were classified as impaired if a composite memory-score derived from two tests from the CANTAB test battery (spatial working memory and paired-associate learning) scored 1 SD below norms derived from a normative database. Risk of developing POCD 3 months post-surgery was higher [odds ratio 2.048 (95% CI 1.027 – 4.087)] for those with pre-surgical cognitive impairment compared to those with no impairment. This suggests that impairment on hippocampus-based tasks spatial-span memory and paired-associates learning is associated with increased risk for POCD in older surgical patients.

Background: Five fulfilled hemophagocytic lymphohistiocytosis (HLH)-2004 criteria, and the HScore are widely used and recommended by international expert consensus to diagnose secondary HLH. Both diagnostic scores have never been validated in heterogeneous patient cohorts of secondary HLH patients. We aimed to systematically optimize and validate diagnostic criteria of secondary HLH using a multicenter approach. Methods: We developed optimized criteria in our cohort of critically ill patients as a first step. We next validated these new criteria together with the original and modified HLH-2004 criteria as well as the HScore using original data of 13 published cohorts, which were identified by a systematic literature search. Results: The best performing HLH diagnostic criteria sets over all 13 validation cohorts were the original HLH-2004 criteria with a decreased cut-off (cut-off 4, mean sensitivity 86.5%, mean specificity 86.1%), followed by the revised HLH-2004 criteria (natural killer cell activity removed; cut-off 4, mean sensitivity 83.8%, mean specificity 87.8%) and the HScore (cut-off 169, mean sensitivity 82.4%, mean specificity 87.6%). Our newly developed HLH diagnostic criteria showed inferior performance. Ferritin ≥500 µg/L had 94.0% mean sensitivity over all cohorts. Conclusions: In this first multicenter validation study, four fulfilled HLH-2004 criteria and an HScore of 169 were suitable to diagnose secondary HLH, which will lead to rapid diagnosis and improved patient outcomes. Ferritin proved as a reliable HLH screening marker. Our results should be taken into account in clinical recommendations and in designing new studies.

Background: Postoperative delirium (POD) is a critical and distressing condition, making identification of high-risk patients essential. Yet, complexity of existing predictive models limits their practical use. This study evaluates fast-and-frugal decision trees (FFTrees)—simple binary prediction algorithms—and compares them to complex algorithms using routine clinical data from a large German hospital.Objectives: To evaluate the utility of FFTrees in predicting POD and compare performance against complex algorithms.Design: Retrospective cohort study.Setting: Routine clinical data from three campuses of a German hospital.Participants: 61,150 patients who underwent surgery between 2017 and 2020.Interventions: Two FFTrees—previously established for pre- and peri-operative settings using research data—were tested, compared against complex algorithms (logistic regression, classification and regression trees, random forests, support vector machines), and two novel FFTrees were developed for the clinical dataset.Main Outcome Measures: Predictive performance assessed by balanced accuracy (mean of sensitivity and specificity).Results: Established FFTrees achieved balanced accuracies of 54% pre-operatively and 59% peri-operatively after thresholds were adjusted to the routine clinical dataset. FFTrees outperformed complex comparators in the pre- and peri-operative setting. The novel pre-operative FFTree identified age, American Society of Anesthesiologists status, and type of surgery as key predictors, achieving a balanced accuracy of 58%. The novel peri-operative FFTree highlighted age, anesthesia duration, and pre-existing conditions as key predictors, achieving a balanced accuracy of 60%.Conclusion: FFTrees serve as practical diagnostic tools for adhering to POD risk assessment guidelines and offer a feasible alternative to more complex models in routine clinical practice.

Background In 1962, the idea emerged that medical students' tolerance of uncertainty could determine their specialty choice. While some studies supported this claim, others refuted it, often using independently developed instruments. We explored whether the reported link between specialty choice and uncertainty tolerance is more myth than evidence by employing established instruments to investigate whether specialty choice could be explained by variance in uncertainty tolerance. Method We conducted a cross‐sectional online survey at two periods of time. From February to June 2023, we queried 563 final‐year medical students from 34 German medical universities (1) on their uncertainty tolerance using three validated tools (the modified tolerance for ambiguity scale, the physicians' reaction to uncertainty scale and the uncertainty intolerance scenario method) and (2) on their intended specialty choice. In a follow‐up 1 year later (May to June 2024), 263 of those medical students responded to our query on their final specialty choice and again on their uncertainty tolerance. Results Participants' ( N = 563) median age was 26.0 years (mean: 27.2; SD = 3.8), and 70% ( n = 396) were female. Originally reported differences and rank orders in uncertainty tolerance among medical students with different intended specialty choices could not be replicated for any of the three scales. Instead, our results suggest different rank orders of uncertainty tolerance by different tools, as well as nonsignificant differences between intended medical specialties. Intercorrelation coefficient analyses demonstrated that, depending on the scale, only 0.3% to 1.5% of the variance in uncertainty tolerance could be attributed to specialty choice. Follow‐up data using actual instead of intended medical choices left findings unchanged. Discussion Our findings suggest that the presumed link between uncertainty tolerance and specialty choice is more myth than evidence. Instead of teaching this link or using it as an admissions criterion, medical schools should equip students with the skills needed to navigate uncertainty across their careers.

Background Clinical practice guidelines are important tools for clinical decision support, but monitoring guideline adherence manually is highly resource-intensive. Therefore, we developed an automated system for evaluating guideline adherence based on computer-interpretable representations of guidelines. We implemented the system across multiple university hospitals and assessed its validity and performance by comparing its guideline adherence evaluations to those conducted by medical professionals. Methods We selected six representative clinical guideline recommendations from across 41 intensive care guidelines and translated these text-based recommendations into a computer-interpretable, Fast Healthcare Interoperability Resources (FHIR)-based format using an iterative consensus approach. Clinical data from five university hospitals were transformed into the Observational Medical Outcomes Partnership (OMOP) common data model. A decision support system was developed to interpret FHIR-encoded recommendations and apply them to OMOP-formatted patient data. We evaluated the system retrospectively on intensive care data covering 3.5 years and validated its performance by comparing system-generated decisions with human decisions in three hospitals. We created and iteratively refined a user interface for individual and ward-level adherence visualization. Findings We expert-reviewed more than 18,000 patient days to assess the applicability of and adherence to the recommendations. The system demonstrated 97.0% accuracy in identifying guideline applicability and adherence, with significantly higher accuracy than human reviewers (accuracy 86.6%, p<0.001, McNemar's Test). The automated system processed more than 2000 patient days per second for a total of 2,200,000 patient days across 82,000 intensive care episodes, compared to humans' two patient days per minute. Interpretation We demonstrate that an automated adherence monitoring system outperforms human reviewers in both accuracy and time efficiency. Using FHIR-encoded recommendations enables flexibility and scalability across hospitals with different data infrastructures. Future work should focus on integrating unstructured patient data and expanding the range of encoded recommendations. Funding Federal Ministry of Education and Research of Germany.

Postoperative cognitive dysfunction (POCD) is a detrimental complication after surgery with a lasting impact on the patients' daily life. It is most common after postoperative delirium. Dopaminergic dysfunction has been suggested to play a role in delirium, but little knowledge exists regarding its relevance for POCD. We hypothesized that POCD is associated with altered resting-state functional connectivity of the ventral tegmental area (VTA) and the substantia nigra pars compacta (SNc) in functional magnetic resonance imaging (fMRI) before surgery and at postoperative follow-up after three months. Patients >65 years old underwent resting-state fMRI and neuropsychological assessment before major elective surgery and at follow-up three months later. POCD was determined as the reliable change index. Connectivity between VTA or SNc and 132 regions were calculated. Principal component analysis (PCA) was used for dimensionality reduction in preoperative connectivity, and components explaining >5% of variance were retained for analysis. To study postoperative changes in patients with POCD, we applied the same transformation to postoperative connectivity, and multi-factor analysis. Regression analyses were used to describe connectivity alterations while adjusting for age, sex, MMSE, surgery and anaesthesia. Of 214 patients, 26 (12%) developed POCD. Among 132 principal components, four components for VTA- and SNc-FC were selected for further analysis. For both VTA and SNc connectivity, one component was significantly associated with POCD. No postoperative alterations of dopaminergic networks were associated with POCD. Altered preoperative functional connectivity of the dopaminergic system predisposes to POCD. Particularly relevant regions are associated with sensory, especially spatial perception and higher cognitive functions.

IMPORTANCE: Postoperative delirium (POD) is a multietiological condition and affects 20% of older surgical patients. It is associated with poor clinical outcome and increased mortality. OBJECTIVE: We aimed to develop and validate a risk prediction algorithm for POD based on a multimodal biomarker database exploiting preoperative data (predisposing factors) and procedural factors as well as perioperative molecular changes associated with POD (precipitating factors). DESIGN: BioCog is a prospective cohort study conducted from November 2014 to April 2017. Patients were followed up for seven postoperative days after surgery for POD. Gradient-boosted trees (GBT) with nested cross-validation was used for POD prediction. SETTING: Patients aged ≥65 years were enrolled at the anesthesiologic departments of two tertiary care centers. EXPOSURE: All patients underwent surgery with an expected duration of at least 60min. Clinical, neuropsychological, neuroimaging data and blood were collected and clinically well established as well as non-established biomarkers (e.g., gene expression profiling) were measured pre- and postoperatively. MAIN OUTCOME: POD according to DSM 5 until the seventh postoperative day RESULTS: 184 of 929 (20%) patients experienced POD. A GBT algorithm using both preoperative data, characteristics of the intervention and postoperative changes in laboratory parameters achieved the highest area under the curve (0.83, [0.79; 0.86]) with a Brier score of 0.12 (0.12; 0.13). CONCLUSIONS AND RELEVANCE: Models combining predisposing factors with precipitating factors predict POD best. Non-routine laboratory data provide useful information for POD risk prediction, providing relevant results for future studies on the molecular factors of POD. In addition, possibly relevant molecular mechanisms contributing to the development of POD were identified, mostly indicating a dysregulated postoperative immune response. This study constitutes the basis for future hypothesis-driven analyses or implementation of prediction expert system for clinical practice.

Background Deficits in cholinergic function are assumed to cause cognitive decline. Studies have demonstrated that changes in serum cholinesterase activities are associated with a higher incidence of delirium in critically ill patients. Additionally, basic research indicates that the cholinergic and circadian systems are interconnected, with each system influencing the functionality of the other. This data analysis of a proof-of-concept pilot study investigates whether modification in ICU design, including dynamic light therapy, may influence the circadian oscillation of serum cholinesterase activities. Methods We enrolled adult critically ill patients who were on mechanical ventilation and had an anticipated ICU stay of at least 48 h. The patients were treated in either modified or standard ICU rooms. The modified rooms received extensive architectural modifications, including a new dynamic lighting system. Serum acetylcholinesterase and butyrylcholinesterase activities were measured every four hours for up to three 24-h assessment periods. Results We included 64 patients in the data analysis ( n = 34 patients in modified rooms, n = 30 in standard rooms). The median values of serum acetylcholinesterase and butyrylcholinesterase activities showed different patterns. Acetylcholinesterase activities differed significantly between the groups during the first assessment period ( p = 0.04) and the second assessment period ( p = 0.045). The intensity of light, as quantified by the effective circadian irradiance, significantly influenced the activities of acetylcholinesterase and butyrylcholinesterase throughout all assessment periods for patients in both groups ( p < 0.001). The analysis showed significant interaction ( p < 0.001), indicating that the differences in acetylcholinesterase and butyrylcholinesterase activities between the groups were inconsistent over time but apparent during specific periods of the day. Conclusion Implementing a comprehensive set of changes to the design of ICU rooms, including a dynamic lighting system, may influence the course of the activity patterns of acetylcholinesterase and butyrylcholinesterase in critically ill patients. Modifications to environmental factors could potentially offer neuroprotective benefits and facilitate the realignment of circadian rhythms within the cholinergic system. Clinical trial registration ClinicalTrials.gov: NCT02143661. Registered May 21, 2014.