Applying Machine Learning to Blood Count Data Predicts Sepsis with ICU Admission.

BACKGROUND: Timely diagnosis is crucial for sepsis treatment. Current machine learning (ML) models suffer from high complexity and limited applicability. We therefore created an ML model using only complete blood count (CBC) diagnostics. METHODS: We collected non-intensive care unit (non-ICU) data from a German tertiary care centre (January 2014 to December 2021). Using patient age, sex, and CBC parameters (haemoglobin, platelets, mean corpuscular volume, white and red blood cells), we trained a boosted random forest, which predicts sepsis with ICU admission. Two external validations were conducted using data from another German tertiary care centre and the Medical Information Mart for Intensive Care IV database (MIMIC-IV). Using the subset of laboratory orders also including procalcitonin (PCT), an analogous model was trained with PCT as an additional feature. RESULTS: After exclusion, 1 381 358 laboratory requests (2016 from sepsis cases) were available. The CBC model shows an area under the receiver operating characteristic (AUROC) of 0.872 (95% CI, 0.857-0.887). External validations show AUROCs of 0.805 (95% CI, 0.787-0.824) for University Medicine Greifswald and 0.845 (95% CI, 0.837-0.852) for MIMIC-IV. The model including PCT revealed a significantly higher AUROC (0.857; 95% CI, 0.836-0.877) than PCT alone (0.790; 95% CI, 0.759-0.821; P < 0.001). CONCLUSIONS: Our results demonstrate that routine CBC results could significantly improve diagnosis of sepsis when combined with ML. The CBC model can facilitate early sepsis prediction in non-ICU patients with high robustness in external validations. Its implementation in clinical decision support systems has strong potential to provide an essential time advantage and increase patient safety.

Evidence-based cutoffs for total and adjusted calcium: a major factor in detecting severe hypo- and hypercalcemia.

OBJECTIVES: Severe hypo- and hypercalcemia are common and urgent treatment is recommended. Free calcium (fCa) is the gold standard but needs blood gas tests with challenging preanalytics. Total calcium (tCa) and calculated adjusted calcium (aCa) are readily available, but their interpretation is hampered by identical tCa and aCa cutoffs, laborious local aCa calculation and difficult comparability of calcium biomarkers. METHODS: Laboratory results from University Medicine Leipzig were evaluated over a five-year period (236,274 patients). A local aCa equation was derived by linear least squares regression, the agreement between fCa, tCa and aCa assessed with Cohen's κ and decision thresholds derived by this indirect method. RESULTS: The local aCa equation was created from data of 9,756 patients, each with one paired measurement of tCa, fCa and albumin. Derived aCa cutoffs (1.95/3.15 mmol/L) differ markedly from derived tCa cutoffs (1.6/2.9 mmol/L) and severe hypo- and hypercalcemia can be more accurately assessed by aCa (κ=0.489, 0.812) than by tCa (κ=0.445, 0.744). Comparing our approach to standard care (tCa, literature cutoff), a total 3,250 of 3,680 (88.3 %) misclassified measurements were correctly classified when using aCa with evidence-based cutoffs. CONCLUSIONS: Optimized cutoffs for aCa and tCa hold great potential for improved patient care. Locally derived aCa equations differ mostly in the chosen mean normal calcium and provide minimal overall improvement, but entail a close examination of the used cutoffs before application.

Development, Design and Utilization of a CDSS for Refeeding Syndrome in Real Life Inpatient Care-A Feasibility Study.

BACKGROUND: The refeeding syndrome (RFS) is an oftentimes-unrecognized complication of reintroducing nutrition in malnourished patients that can lead to fatal cardiovascular failure. We hypothesized that a clinical decision support system (CDSS) can improve RFS recognition and management. METHODS: We developed an algorithm from current diagnostic criteria for RFS detection, tested the algorithm on a retrospective dataset and combined the final algorithm with therapy and referral recommendations in a knowledge-based CDSS. The CDSS integration into clinical practice was prospectively investigated for six months. RESULTS: The utilization of the RFS-CDSS lead to RFS diagnosis in 13 out of 21 detected cases (62%). It improved patient-related care and documentation, e.g., RFS-specific coding (E87.7), increased from once coded in 30 month in the retrospective cohort to four times in six months in the prospective cohort and doubled the rate of nutrition referrals in true positive patients (retrospective referrals in true positive patients 33% vs. prospective referrals in true positive patients 71%). CONCLUSION: CDSS-facilitated RFS diagnosis is possible and improves RFS recognition. This effect and its impact on patient-related outcomes needs to be further investigated in a large randomized-controlled trial.

Higher troponin T serum concentrations in hospital patients without diagnosed cardiac diseases compared to a population-based cohort.

OBJECTIVES: Upper reference limits of high-sensitivity cardiac troponin T (hs-cTnT) are derived from healthy, population-based cohorts, and are frequently exceeded in hospitalized patients. In this study we aim to systematically examine the differences between in-hospital patients with no diagnosed cardiac diseases and a population-based cohort. METHODS: Retrospective analyses were performed in two independent cohorts. We included 5,652 participants of the prospective population-based LIFE cohort as well as 9,300 patients having been treated at our hospital between 2014 and 2021. In both cohorts, subjects with diagnosed or suspected cardiac diseases were excluded. We used Spearman's rank correlation for correlation analyses of hs-cTnT serum concentrations and age. Sex- and age-adjusted 99th percentiles for hs-cTnT in subjects with preserved renal function were obtained in both cohorts. RESULTS: In both cohorts, hs-cTnT serum concentrations positively correlated with age. Male sex was associated with higher hs-cTnT serum concentrations. Persons treated in hospital showed significantly higher hs-cTnT concentrations in females and males aged above 50. While in the population-based cohort only 99th percentile hs-cTnT results of females aged above 70 and males aged above 60 years exceeded the assay's upper reference limit, the 99th percentiles of in-hospital females over 40 years and males of all age groups exceeded this threshold. CONCLUSIONS: Besides age and sex, hospitalization per se is correlated with higher serum concentrations of hs-cTnT in most age groups. Our results indicate, that unconditionally applying current hs-cTnT cut-offs to inpatients might overestimate myocardial infarction and potentially lead to overdiagnosis.

Toward a Systematic Assessment of Sex Differences in Cystic Fibrosis.

(1) Background: Cystic fibrosis (CF) is a disease with well-documented clinical differences between female and male patients. However, this gender gap is very poorly studied at the molecular level. (2) Methods: Expression differences in whole blood transcriptomics between female and male CF patients are analyzed in order to determine the pathways related to sex-biased genes and assess their potential influence on sex-specific effects in CF patients. (3) Results: We identify sex-biased genes in female and male CF patients and provide explanations for some sex-specific differences at the molecular level. (4) Conclusion: Genes in key pathways associated with CF are differentially expressed between sexes, and thus may account for the gender gap in morbidity and mortality in CF.

Revising the MELD Score to Address Sex-Bias in Liver Transplant Prioritization for a German Cohort.

(1) Background: Prioritization of patients for liver transplantation in Germany relies on the MELD (model for end-stage liver disease) scoring system that does not consider the patient's sex. Many studies have shown that women are disadvantaged by the MELD score. Using a large patient cohort from a German liver transplant centre, we investigated options to reduce gender inequality in the patient prioritization for liver transplantation. (2) Methods: We calculated female-as-male MELD scores in our cohort by substituting the serum creatinine of a female patient with that of their male equivalent to test for the fairness of the scores. We investigated the effects of the female-as-male scores compared to the original MELD score of 1759 patients listed for liver transplantation. (3) Results: Serum creatinine sex correction (female-as-male) for MELD scores added up to 5.4 points in females, while the median changed by +1.6 points for females. We identified 72 females with an original MELD score < 20, for whom the adjusted female-as-male MELD score would be >20, thus giving them a better chance to receive a liver transplant. (4) Conclusions: Mathematical conversion of female to male creatinine concentrations identified disadvantages in liver transplantation prioritization for females and ascertained MELD 3.0 as having high potential to compensate for these inequalities.

Detecting Pre-Analytically Delayed Blood Samples for Laboratory Diagnostics Using Raman Spectroscopy.

In this proof-of-principle study, we systematically studied the potential of Raman spectroscopy for detecting pre-analytical delays in blood serum samples. Spectra from 330 samples from a liver cirrhosis cohort were acquired over the course of eight days, stored one day at room temperature, and stored subsequently at 4 °C. The spectra were then used to train Convolutional Neural Networks (CNN) to predict the delay to sample examination. We achieved 90% accuracy for binary classification of the serum samples in the groups "without delay" versus "delayed". Spectra recorded on the first day could be distinguished clearly from all subsequent measurements. Distinguishing between spectra taken in the range from the second to the last day seems to be possible as well, but currently, with an accuracy of approximately 70% only. Importantly, filtering out the fluorescent background significantly reduces the precision of detection.

Laboratory Diagnostic of Acute Kidney Injury and Its Progression: Risk of Underdiagnosis in Female and Elderly Patients.

Acute kidney injury (AKI) is a common disease, with high morbidity and mortality rates. In this study, we investigated the potential influence of sex and age on laboratory diagnostics and outcomes. It is known that serum creatinine (SCr) has limitations as a laboratory diagnostic parameter for AKI due to its dependence on muscle mass, which may lead to an incorrect or delayed diagnosis for certain patient groups, such as women and the elderly. Overall, 7592 cases with AKI, hospitalized at the University of Leipzig Medical Center (ULMC) between 1st January 2017 and 31st December 2019, were retrospectively analyzed. The diagnosis and staging of AKI were performed according to the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines, based on the level and dynamics of SCr. The impact of sex and age was analyzed by the recalculation of a female to male and an old to young SCr using the CKD-EPI equation. In our study cohort progressive AKI occurred in 19.2% of all cases (n = 1458). Female cases with AKI were underrepresented (40.4%), with a significantly lower first (-3.5 mL/min) and last eGFR (-2.7 mL/min) (p < 0.001). The highest incidence proportion of AKI was found in the [61-81) age group in female (49.5%) and male (52.7%) cases. Females with progressive AKI were underrepresented (p = 0.04). By defining and staging AKI on the basis of relative and absolute changes in the SCr level, it is more difficult for patients with low muscle mass and, thus, a lower baseline SCr to be diagnosed by an absolute SCr increase. AKIN1 and AKIN3 can be diagnosed by a relative or absolute change in SCr. In females, both stages were less frequently detected by an absolute criterion alone (AKIN1 ♀ 20.2%, ♂ 29.5%, p < 0.001; AKIN3 ♀ 13.4%, ♂ 15.2%, p < 0.001). A recalculated SCr for females (as males) and males (as young males) displayed the expected increase in AKI occurrence and severity with age and, in general, in females. Our study illustrates how SCr, as the sole parameter for the diagnosis and staging of AKI, bears the risk of underdiagnosis of patient groups with low muscle mass, such as women and the elderly. A sex- and age-adapted approach might offer advantages.

Challenge in hyponatremic patients - the potential of a laboratory-based decision support system for hyponatremia to improve patient's safety.

OBJECTIVES: Hyponatremia is the most frequent electrolyte disorder in hospitalized patients with increased mortality and morbidity. In this study, we evaluated the follow-up diagnostic, the risk of inadequate fast correction and the outcome of patients with profound hyponatremia (pHN), defined as a blood sodium concentration below 120 mmol/L. The aim was to identify a promising approach for a laboratory-based clinical decision support system (CDSS). METHODS: This retrospective study included 378,980 blood sodium measurements of 83,315 cases at a German tertiary care hospital. Hospitalized cases with pHN (n=211) were categorized into two groups by the time needed for a follow-up measurement to be performed (time to control, TTC) as either <12 h (group 1: "TTC≤12 h", n=118 cases) or >12 h (group 2: "TTC>12 h", n=93 cases). Length of hospital stay, sodium level at discharge, ward transfers, correction of hyponatremia, and risk of osmotic demyelination syndrome (ODS) due to inadequate fast correction were evaluated with regard to the TTC of sodium blood concentration. RESULTS: pHN was detected in 1,050 measurements (0.3%) in 211 cases. Cases, in which follow-up diagnostics took longer (TTC>12 h), achieved a significantly lower sodium correction during their hospitalization (11.2 vs. 16.7 mmol/L, p<0.001), were discharged more frequently in hyponatremic states (<135 mmol/L; 58 (62.4%) vs. 43 (36.4%), p<0.001) and at lower sodium blood levels (131.2 vs. 135.0 mmol/L, p<0.001). Furthermore, for these patients there was a trend toward an increased length of hospital stay (13.1 vs. 8.5 days, p=0.089), as well as an increased risk of inadequate fast correction (p<0.001). CONCLUSIONS: Our study shows that less frequent follow-up sodium measurements in pHN are associated with worse outcomes. Patients with a prolonged TTC are at risk of insufficient correction of hyponatremia, reduced sodium values at discharge, and possible overcorrection. Our results suggest that a CDSS that alerts treating physicians when a control time of >12 h is exceeded could improve patient care in the long term. We are initiating a prospective study to investigate the benefits of our self-invented CDSS (www.ampel.care) for patients with pHN.

Refeeding Syndrome.

BACKGROUND: Refeeding syndrome (RFS) can occur in malnourished patients when normal, enteral, or parenteral feeding is resumed. The syndrome often goes unrecognized and may, in the most severe cases, result in death. The diagnosis of RFS can be crucially facilitated by the use of clinical decision support systems (CDSS). METHODS: The literature in PubMed was searched for current treatment recommendations, randomized intervention studies, and publications on RFS and CDSS. We also took account of insights gained from the development and implementation of our own CDSS for the diagnosis of RFS. RESULTS: The identification of high-risk patients and the recognition of manifest RFS is clinically challenging due to the syndrome's unspecific symptoms and physicians' lack of awareness of the risk of this condition. The literature shows that compared to patients without RFS, malnourished patients with RFS have significantly greater 6-month mortality (odds ratio 1.54, 95% confidence interval: [1.04; 2.28]) and an elevated risk of admission to intensive care (odds ratio 2.71 [1.01; 7.27]). In a prospective testing program, use of our own CDSS led to correct diagnosis in two thirds of cases. CONCLUSION: RFS is difficult to detect and represents a high risk to the patients affected. Appropriate CDSS can identify such patients and ensure proper professional care.

Invasive Growth of Ailanthus altissima Trees is Associated with a High Rate of Sensitization in Atopic Patients.

Purpose: Ailanthus altissima is one of the world's most invasive species with a globally problematic spread. Pollen is dispersed locally and partially airborne. We aimed at investigating if (i) A. altissima pollen can be detected in relevant quantity in the air and if (ii) sensitization to A. altissima can be detected in patients with seasonal exacerbation of atopic diseases. Patients and Methods: We recorded distribution of A. altissima in Leipzig, Germany. In 2019 and 2020, pollen was collected with a Hirst-type pollen trap placed on the roof of the University Hospital. Specific IgE investigations were performed in children and adults with history of atopic diseases with deterioration between May and July. We analysed specific IgE for A. altissima, Alternaria sp., birch, grasses, profilins, polcalcins and crossreacting carbohydrates. Results: We found abundant growth of A. altissima and pollen was detected from early June to mid-July with a maximum pollen concentration of 31 pollen/m3. Out of 138 patients (63 female, 69 children/adolescents), 95 (69%) had seasonal allergic rhinitis, 84 (61%) asthma, and 43 (31%) atopic dermatitis. Sensitization to A. altissima was shown in 59 (42%). There were no significant differences between age groups. In 59% of patients sensitized (35/59), there was no sensitization to possibly cross-reacting structures. Conclusion: Sensitization to A. altissima pollen could be detected in 42% of our patients with atopic diseases, suggesting allergenic potential of this neophyte. In the context of further spread with climate change, eradication strategies and population-based sensitization studies are needed.

Influencing Factors and Differences in Born Aggregometry in Specialized Hemostaseological Centers: Results of a Multicenter Laboratory Comparison.

Introduction Light transmission aggregometry (LTA) is regarded as the gold standard in platelet function diagnostics. However, there is a relevant degree of interlaboratory variability in practical applications. Objective The aim of the present study was to develop a practicable laboratory comparison on LTA and to analyze differences and influencing factors in regard to standardization in five specialized hemostaseological centers. Methods The study was performed on 30 patients in total. Each center performed LTA on blood samples from six healthy volunteers (three men and three women) using the inductors collagen (Col), adenosine diphosphate (ADP), arachidonic acid (ARA), and ristocetin. The LTA was performed three times using different methods as follows: (1) International Society on Thrombosis and Haemostasis recommendations with identical reagents, (2) in-house protocols and the identical reagents; and (3) in-house protocols and in-house reagents. Results A total of 396 measurements of 30 probands were performed. Even after standardization of the protocol and using identical reagents, there were significant differences between the centers regarding the final and maximum aggregation ( p = 0.002 and <0.001) and further significant differences in the maximum and final aggregation according to the wavelength of the device used to measure the LTA (PAP-8: 430 nm, APACT 4004: 740 nm [ p < 0.001 each]). Using identical reagents but individual inductor concentrations and laboratory protocols also resulted in different maximum and final aggregation. The largest differences were seen with Col and ristocetin; there were significant influences from the reagents' manufacturers in the results of aggregometry for the inductor Col ( p < 0.01) but not for ADP, ARA, and ristocetin. Conclusion In this study, we proved that there are significant influences from the used aggregometers, inductors concentrations, and manufacturers. These results illustrate the challenges and importance of standardization of LTA.

Biomarkers for Non-Invasive Stratification of Coronary Artery Disease and Prognostic Impact on Long-Term Survival in Patients with Stable Coronary Heart Disease.

Knowledge about cardiac and inflammatory biomarkers in patients with stable coronary artery disease (CAD) is limited. To address this, we analyzed 3072 patients (36% female) with a median follow-up of 10 years in the Leipzig LIFE Heart Study with suspected CAD with coronary angiography. Selected biomarkers included troponin T (hsTNT), N-terminal pro B-type natriuretic peptide (NT-proBNP), copeptin, C-reactive protein (hsCRP), and interleukin-6 (IL-6). Patients were stratified by CAD severity: CAD0 (no sclerosis), CAD1 (non-obstructive, i.e., stenosis < 50%), and CAD2 (≥one stenosis ≥ 50%). Group comparison (GC) included GC1: CAD0 + 1 vs. CAD2; GC2: CAD0 vs. CAD1 + 2. CAD0, CAD1, and CAD2 were apparent in 1271, 631, and 1170 patients, respectively. Adjusted for classical risk factors, hs-cTnT, NT-proBNP, and IL-6 differed significantly in both GC and hsCRP only in GC2. After multivariate analysis, hs-cTnT, NT-proBNP, and IL-6 remained significant in GC1. In GC2, hs-cTnT (p < 0.001) and copeptin (p = 0.014) reached significance. Ten-year survival in groups CAD0, CAD1, and CAD2 was 88.3%, 77.3%, and 72.4%. Incorporation of hs-cTnT, NT-proBNP, copeptin, and IL-6 improved risk prediction (p < 0.001). The studied cardiac and inflammatory biomarkers enable fast and precise non-invasive identification of mortality risk in CAD patients, allowing the tailoring of primary and secondary CAD prevention.

Body fat composition determines outcomes before and after liver transplantation in patients with cirrhosis.

Cachexia occurs in late stages of liver cirrhosis, and a low-fat mass is potentially associated with poor outcome. This study compared different computed tomography (CT)-derived fat parameters with respect to its prognostic impact on the development of complications and death before and after liver transplantation. Between 2001 and 2014, 612 patients with liver cirrhosis without hepatocellular carcinoma listed for liver transplantation met the inclusion criteria, including abdominal CT scan (±200 days to listing). A total of 109 patients without cirrhosis served as controls. The subcutaneous fat index (SCFI), the paraspinal muscle fat index, and the visceral fat index were assessed at L3/L4 level and normalized to the height (cm2 /m2 ). Data were collected and analyzed retrospectively. Low SCFI was associated with a higher rate of ascites and increased C-reactive protein levels (p < 0.001). In addition, multivariate Cox regression analysis adjusting for sex, age, body mass index (BMI), and Model for End-Stage Liver Disease showed that decreasing SCFI was also associated with an increased risk of cirrhosis-related complications (p = 0.003) and death on the transplant wait list (p = 0.013). Increased paraspinal and visceral fat were not only positively correlated with creatinine levels (p < 0.001), BMI, and metabolic comorbidities (all p < 0.001) before transplantation, but also predictive for 1-year mortality after transplantation. Conclusion: The distribution of body fat is a major determinant for complications and outcome in cirrhosis before and after liver transplantation.

Revisited Upper Reference Limits for Highly Sensitive Cardiac Troponin T in Relation to Age, Sex, and Renal Function.

(1) Background: Highly sensitive cardiac troponin T (hs-cTnT) plays an essential role in the diagnosis of myocardial injury. The upper reference limit of the respective assay is generally applied, irrespective of age, renal function, or sex. We aimed to identify age-adjusted and sex-adjusted upper reference limits in relation to renal function in a large population-based cohort without cardiac diseases. (2) Methods: We included 5428 subjects of the population-based LIFE-Adult cohort, free of diagnosed cardiac diseases. Sex-adjusted and age-adjusted 99th percentiles for hs-cTnT in subjects with preserved renal function were obtained. (3) Results: The hs-cTnT values were higher in men of all age groups. In both sexes, an increasing age positively correlated with higher hs-cTnT values. Hs-cTnT weakly correlated with serum creatinine. The three-dimensional analysis of age, creatinine, and hs-cTnT showed no relevant additional effect of creatinine on hs-cTnT. In men aged above 60 and women above 70, the calculated 99th percentiles clearly exceeded the commonly applied thresholds. (4) Conclusion: Age and sex have a major impact on the serum concentration of hs-cTnT, while renal function does not. We propose to consider age-adjusted and sex-adjusted reference values.

Raman spectroscopy on blood serum samples of patients with end-stage liver disease.

Raman spectroscopy has shown to be a promising method for the examination of biomedical samples. However, until now, its efficacy has not been established in clinical diagnostics. In this study, Raman spectroscopy's potential application in medical laboratories is evaluated for a large variety (38) of biomarkers. Given 234 serum samples from a cohort of patients with different stages of liver disease, we performed Raman spectroscopy at 780nm excitation wavelength. The Raman spectra were analyzed in combination with the results of routine diagnostics using specifically developed complex mathematical algorithms, including fluorescence filtering, frequency subset selection and several overfitting circumventing strategies, such as independent validation. With the results of this cohort, which were validated in 328 independent samples, a significant proof-of-concept study was completed. This study highlights the need to prevent overfitting and to use independent data for validation. The results reveal that Raman spectroscopy has high potential for use in medical laboratory diagnostics to simultaneously quantify multiple biomarkers.

Experiences from six years of quality assured Model of End Stage Liver Disease (MELD) diagnostics.

BACKGROUND: The model of end-stage liver disease (MELD) score was established for the allocation of liver transplants. The score is based on the medical laboratory parameters: bilirubin, creatinine and the international normalized ratio (INR). A verification algorithm for the laboratory MELD diagnostic was established, and the results from the first six years were analyzed. METHODS: We systematically investigated the validity of 7,270 MELD scores during a six-year period. The MELD score was electronically requested by the clinical physician using the laboratory system and calculated and specifically validated by the laboratory physician in the context of previous and additional diagnostics. RESULTS: In 2.7% (193 of 7,270) of the cases, MELD diagnostics did not fulfill the specified quality criteria. After consultation with the sender, 2.0% (145) of the MELD scores remained invalid for different reasons and could not be reported to the transplant organization. No cases of deliberate misreporting were identified. In 34 cases the dialysis status had to be corrected and there were 24 cases of oral anticoagulation with impact on MELD diagnostics. CONCLUSION: Our verification algorithm for MELD diagnostics effectively prevented invalid MELD results and could be adopted by transplant centers to prevent diagnostic errors with possible adverse effects on organ allocation.

Acute kidney injury and its progression in hospitalized patients-Results from a retrospective multicentre cohort study with a digital decision support system.

In this retrospective multicentric cohort study, we evaluate the potential benefits of a clinical decision support system (CDSS) for the automated detection of Acute kidney injury (AKI). A total of 80,389 cases, hospitalized from 2017 to 2019 at a tertiary care hospital (University of Leipzig Medical Center (ULMC)) and two primary care hospitals (Muldentalkliniken (MTL)) in Germany, were enrolled. AKI was defined and staged according to the Kidney disease: improving global outcomes (KDIGO) guidelines. Clinical and laboratory data was automatically collected from electronic patient records using the frameworks of the CDSS. In our cohort, we found an overall AKI incidence proportion of 12.1%. We identified 6,393/1,703/1,604 cases as AKI stage 1/2/3 (8.0%/2.1%/2.0%, respectively). Administrative coding with N17 (ICD-10-GM) was missing in 55.8% of all AKI cases with the potential for additional diagnosis related groups (DRG) reimbursement of 1,204,200 € in our study. AKI was associated with higher hospital mortality, increased length of hospitalisation and more frequent need of renal replacement therapy. A total of 19.1% of AKI cases (n = 1,848) showed progression to higher AKI stages (progressive AKI) during hospitalization. These cases presented with considerably longer hospitalization, higher rates of renal replacement therapy and increased mortality (p<0.001, respectively). Furthermore, progressive AKI was significantly associated with sepsis, shock, liver cirrhosis, myocardial infarction, and cardiac insufficiency. AKI, and especially its progression during hospitalization, is strongly associated with adverse outcomes. Our automated CDSS enables timely detection and bears potential to improve AKI outcomes, notably in cases of progressive AKI.

The Clinical Decision Support System AMPEL for Laboratory Diagnostics: Implementation and Technical Evaluation.

BACKGROUND: Laboratory results are of central importance for clinical decision making. The time span between availability and review of results by clinicians is crucial to patient care. Clinical decision support systems (CDSS) are computational tools that can identify critical values automatically and help decrease treatment delay. OBJECTIVE: With this work, we aimed to implement and evaluate a CDSS that supports health care professionals and improves patient safety. In addition to our experiences, we also describe its main components in a general manner to make it applicable to a wide range of medical institutions and to empower colleagues to implement a similar system in their facilities. METHODS: Technical requirements must be taken into account before implementing a CDSS that performs laboratory diagnostics (labCDSS). These can be planned within the functional components of a reactive software agent, a computational framework for such a CDSS. RESULTS: We present AMPEL (Analysis and Reporting System for the Improvement of Patient Safety through Real-Time Integration of Laboratory Findings), a labCDSS that notifies health care professionals if a life-threatening medical condition is detected. We developed and implemented AMPEL at a university hospital and regional hospitals in Germany (University of Leipzig Medical Center and the Muldental Clinics in Grimma and Wurzen). It currently runs 5 different algorithms in parallel: hypokalemia, hypercalcemia, hyponatremia, hyperlactatemia, and acute kidney injury. CONCLUSIONS: AMPEL enables continuous surveillance of patients. The system is constantly being evaluated and extended and has the capacity for many more algorithms. We hope to encourage colleagues from other institutions to design and implement similar CDSS using the theory, specifications, and experiences described in this work.

Sensitivity of SARS-CoV-2 antibody tests with late convalescent sera.

SARS-CoV-2-specific IgM antibodies wane during the first three months after infection and IgG antibody levels decline. This may limit the ability of antibody tests to identify previous SARS-CoV-2 infection at later time points. To examine if the diagnostic sensitivity of antibody tests falls off, we compared the sensitivity of two nucleoprotein-based antibody tests, the Roche Elecsis II Anti-SARS-CoV-2 and the Abbott SARS-CoV-2 IgG assay and three glycoprotein-based tests, the Abbott SARS-CoV-2 IgG II Quant, Siemens Atellica IM COV2T and Euroimmun SARS-CoV-2 assay with 53 sera obtained 6 months after SARS-CoV-2 infection. The sensitivity of the Roche, Abbott SARS-CoV-2 IgG II Quant and Siemens antibody assays was 94.3% (95% confidence interval (CI) 84.3-98.8%), 98.1 % (95% CI: 89.9-100%) and 100 % (95% CI: 93.3-100%). The sensitivity of the N-based Abbott SARS-CoV-2 IgG and the glycoprotein-based Euroimmun ELISA was 45.3 % (95% CI: 31.6-59.6%) and 83.3% (95% CI: 70.2-91.9%). The nucleoprotein-based Roche and the glycoprotein-based Abbott receptor binding domain (RBD) and Siemens tests were more sensitive than the N-based Abbott and the Euroimmun antibody tests (p = 0.0001 to p = 0.039). The N-based Abbott antibody test was less sensitive 6 months than 4-10 weeks after SARS-CoV-2 infection (p = 0.0001). The findings show that most SARS-CoV-2 antibody assays correctly identified previous infection 6 months after infection. The sensitivity of pan-Ig antibody tests was not reduced at 6 months when IgM antibodies have usually disappeared. However, one of the nucleoprotein-based antibody tests significantly lost diagnostic sensitivity over time.