Claudia Amato

Optimal convection volume for improving patient outcomes in an international incident dialysis cohort treated with online hemodiafiltration

Online hemodiafiltration (OL-HDF), the most efficient renal replacement therapy, enables enhanced removal of small and large uremic toxins by combining diffusive and convective solute transport. Randomized controlled trials on prevalent chronic kidney disease (CKD) patients showed improved patient survival with high-volume OL-HDF, underlining the effect of convection volume (CV). This retrospective international study was conducted in a large cohort of incident CKD patients to determine the CV threshold and range associated with survival advantage. Data were extracted from a cohort of adult CKD patients treated by post-dilution OL-HDF over a 101-month period. In total, 2293 patients with a minimum of 2 years of follow-up were analyzed using advanced statistical tools, including cubic spline analyses for determination of the CV range over which a survival increase was observed. The relative survival rate of OL-HDF patients, adjusted for age, gender, comorbidities, vascular access, albumin, C-reactive protein, and dialysis dose, was found to increase at about 55 l/week of CV and to stay increased up to about 75 l/week. Similar analysis of pre-dialysis β2-microglobin (marker of middle-molecule uremic toxins) concentrations found a nearly linear decrease in marker concentration as CV increased from 40 to 75 l/week. Analysis of log C-reactive protein levels showed a decrease over the same CV range. Thus, a convection dose target based on convection volume should be considered and needs to be confirmed by prospective trials as a new determinant of dialysis adequacy.

Managing Complexity at Dialysis Service Centers across Europe

Introduction: Dialysis is probably one of the areas of medicine with more guidelines than any other. Issues such as dialysis dose are dealt with in those guidelines, and minimum values to be reached are defined. A target has to be set and reached by using a data-driven continuous quality improvement (CQI) approach. Data collection must be programmed and structured from the beginning. Methods: Fresenius started its activities as a dialysis provider in 1996, following the merger of its dialysis business with the leading service provider in the US, National Medical Care. Currently Fresenius Medical Care’s European activities involve more than 320 dialysis centers located in 15 countries and treating more than 24,000 patients. Management is based on a bi-dimensional organization where line managers can rely on international functional departments. Under this framework, the CQI techniques are applied in conjunction with benchmarking in a system driven by quality targets. In order to combine clinical governance with management targets, the Balanced ScoreCard system was selected. The Balanced ScoreCard monitors the efficiency of each dialysis center compared to an ideal model, targeting maximum possible efficiency whilst having a unique target for patient outcomes. Conclusion: A clear definition of targets is fundamental and activities need to be monitored and continuously improved; scientific collection of clinical data is the key.

Use of Self-Organizing Maps for Balanced Scorecard analysis to monitor the performance of dialysis clinic chains

The Balanced Scorecard (BSC) is a validated tool to monitor enterprise performances against specific objectives. Through the choice and the evaluation of strategic Key Performance Indicators (KPIs), it provides a measure of the past company’s outcome and allows planning future managerial strategies. The Fresenius Medical Care (FME) BSC makes use of 30 KPIs for a continuous quality improvement strategy within its dialysis clinics. Each KPI is monthly associated to a score that summarizes the clinic efficiency for that month. Standard statistical methods are currently used to analyze the BSC data and to give a comprehensive view of the corporate improvements to the top management. We herein propose the Self-Organizing Maps (SOMs) as an innovative approach to extrapolate information from the FME BSC data and to present it in an easy-readable informative form. A SOM is a computational technique that allows projecting high-dimensional datasets to a two-dimensional space (map), thus providing a compressed representation. The SOM unsupervised (self-organizing) training procedure results in a map that preserves similarity relations existing in the original dataset; in this way, the information contained in the high-dimensional space can be more easily visualized and understood. The present work demonstrates the effectiveness of the SOM approach in extracting useful information from the 30-dimensional BSC dataset: indeed, SOMs enabled both to highlight expected relationships between the KPIs and to uncover results not predictable with traditional analyses. Hence we suggest SOMs as a reliable complementary approach to the standard methods for BSC interpretation.

Prediction of the hemoglobin level in hemodialysis patients using machine learning techniques

Patients who suffer from chronic renal failure (CRF) tend to suffer from an associated anemia as well. Therefore, it is essential to know the hemoglobin (Hb) levels in these patients. The aim of this paper is to predict the hemoglobin (Hb) value using a database of European hemodialysis patients provided by Fresenius Medical Care (FMC) for improving the treatment of this kind of patients. For the prediction of Hb, both analytical measurements and medication dosage of patients suffering from chronic renal failure (CRF) are used. Two kinds of models were trained, global and local models. In the case of local models, clustering techniques based on hierarchical approaches and the adaptive resonance theory (ART) were used as a first step, and then, a different predictor was used for each obtained cluster. Different global models have been applied to the dataset such as Linear Models, Artificial Neural Networks (ANNs), Support Vector Machines (SVM) and Regression Trees among others. Also a relevance analysis has been carried out for each predictor model, thus finding those features that are most relevant for the given prediction.

An international observational study suggests that artificial intelligence for clinical decision support optimizes anemia management in hemodialysis patients

Managing anemia in hemodialysis patients can be challenging because of competing therapeutic targets and individual variability. Because therapy recommendations provided by a decision support system can benefit both patients and doctors, we evaluated the impact of an artificial intelligence decision support system, the Anemia Control Model (ACM), on anemia outcomes. Based on patient profiles, the ACM was built to recommend suitable erythropoietic-stimulating agent doses. Our retrospective study consisted of a 12-month control phase (standard anemia care), followed by a 12-month observation phase (ACM-guided care) encompassing 752 patients undergoing hemodialysis therapy in 3 NephroCare clinics located in separate countries. The percentage of hemoglobin values on target, the median darbepoetin dose, and individual hemoglobin fluctuation (estimated from the intrapatient hemoglobin standard deviation) were deemed primary outcomes. In the observation phase, median darbepoetin consumption significantly decreased from 0.63 to 0.46 μg/kg/month, whereas on-target hemoglobin values significantly increased from 70.6% to 76.6%, reaching 83.2% when the ACM suggestions were implemented. Moreover, ACM introduction led to a significant decrease in hemoglobin fluctuation (intrapatient standard deviation decreased from 0.95 g/dl to 0.83 g/dl). Thus, ACM support helped improve anemia outcomes of hemodialysis patients, minimizing erythropoietic-stimulating agent use with the potential to reduce the cost of treatment.

Performance of a Predictive Model for Long-Term Hemoglobin Response to Darbepoetin and Iron Administration in a Large Cohort of Hemodialysis Patients

Anemia management, based on erythropoiesis stimulating agents (ESA) and iron supplementation, has become an increasingly challenging problem in hemodialysis patients. Maintaining hemodialysis patients within narrow hemoglobin targets, preventing cycling outside target, and reducing ESA dosing to prevent adverse outcomes requires considerable attention from caregivers. Anticipation of the long-term response (i.e. at 3 months) to the ESA/iron therapy would be of fundamental importance for planning a successful treatment strategy. To this end, we developed a predictive model designed to support decision-making regarding anemia management in hemodialysis (HD) patients treated in center. An Artificial Neural Network (ANN) algorithm for predicting hemoglobin concentrations three months into the future was developed and evaluated in a retrospective study on a sample population of 1558 HD patients treated with intravenous (IV) darbepoetin alfa, and IV iron (sucrose or gluconate). Model inputs were the last 90 days of patients’ medical history and the subsequent 90 days of darbepoetin/iron prescription. Our model was able to predict individual variation of hemoglobin concentration 3 months in the future with a Mean Absolute Error (MAE) of 0.75 g/dL. Error analysis showed a narrow Gaussian distribution centered in 0 g/dL; a root cause analysis identified intercurrent and/or unpredictable events associated with hospitalization, blood transfusion, and laboratory error or misreported hemoglobin values as the main reasons for large discrepancy between predicted versus observed hemoglobin values. Our ANN predictive model offers a simple and reliable tool applicable in daily clinical practice for predicting the long-term response to ESA/iron therapy of HD patients.

Atrial fibrillation in dialysis patients: time to abandon warfarin?

Atrial fibrillation (AF) is a frequent clinical complication in dialysis patients, and warfarin therapy represents the most common approach for reducing the risk of stroke in this population. However, current evidence based on observational studies, offer conflicting results, whereas no randomized controlled trials have been carried out so far. Additionally, many clinicians are wary of the possible role of warfarin as vascular calcification inducer and its potential to increase the high risk of bleeding among patients on dialysis. Ideally the most promising therapy would be based on direct inhibitors of factor Ila or Xa; however, at the moment, none of these drugs can be safely prescribed in dialysis patients, because of their potentially dangerous accumulation, and the lack of sufficient experience with apixaban or rivaroxaban, two drugs showing a favorable pharmacokinetic profile in end-stage renal disease. Hence, the use of vitamin K inhibitors is currently the only pharmacological option for stroke prevention in dialysis patients with atrial fibrillation, leaving the clinicians in a management conundrum. This review discusses the trade-offs implicated in warfarin use for this population, the promises of newly developed drugs, the role of dialysis as atrial fibrillation trigger, as well as potential non-pharmacological management options suitable in selected clinical situations.

The modern haemodialysis factory: Must quality improvement trump personalized care?

Sir, We would like to add to the debate on continuous quality improvement in haemodialysis raised by the Editorial Comment of Mendelssohn and Benaroia [1] who stress the issue of data collection but would appear to be unaware of good practice in Europe. The European dialysis network of Fresenius Medical Care (FMC) is the largest in Europe, encompassing more than 350 centres in 18 countries. The network deals with several healthcare systems with different reimbursement structures, practice patterns and patient characteristics. The main challenge of such a complex network is to combine clinical governance (quality and safety) with management efficiency [2]. Mendelssohn stresses the variations in performance against quality indicators between facilities and countries. Stated goals of the FMC network are the reduction of such variation, improvement of performance against indicators (European Best Practice Guidelines [3]) and the delivery of high-quality care. FMC developed a standard information management tool called EuCliD [2]. Since its inception the scope has expanded, from an initial database, supporting CQI processes, to the current management tool, fulfilling the requirements of a dialysis unit, from planning the dialysis schedule to managing the warehouse and generating reports on outcomes, in real time, to support the clinical governance process driving up quality whilst reducing variation. We believe this satisfies the needs outlined by Mendelssohn. The results, published last year [4], concluded that such an approach led to a reduction in variation, an improvement against the quality indicators and in the quality of the treatment delivered. The paper was granted by an editorial prepared by the Nephroquest team [5], stressing the importance of data collection, standardization and benchmarking with the aim to measure, analyse and report quality indicators. Lastly, Mendelssohn suggests the use of an aggregated quality index as a way to facilitate comparison of quality between units. We believe that utilizing the standard management theory of the balanced scorecard provides a much more holistic approach. Our scorecard, generated from the EuCliD system, includes indicators of treatment quality and safety, patient satisfaction, personnel management and development, financial stability and resource (water and electricity) use and waste produced. This tool is more appropriate when managing facilities located in the complex European region, providing a more complete overview of performance. In conclusion, we hope to have convinced Mendelssohn that the scenario for patients in need of a dialysis treatment is better than described and that whilst we would agree with their analogy of the factory we would stress that it is important to look at its functioning as a whole rather than just one component of output.

A comparative study of the risk profile of hemodialysis patients in a for profit network and in two regional registries of the Italian Society of Nephrology

In 2013, the Italian Society of Nephrology joined forces with Nephrocare-Italy to create a clinical research cohort of patients on file in the data-rich clinical management system (EUCLID) of this organization for the performance of observational studies in the hemodialysis (HD) population. To see whether patients in EUCLID are representative of the HD population in Italy, we set out to compare the whole EUCLID population with patients included in the regional HD registries in Emilia-Romagna (Northern Italy) and in Calabria (Southern Italy), the sole regions in Italy which have systematically collected an enlarged clinical data set allowing comparison with the data-rich EUCLID system. An analysis of prevalent and incident patients in 2010 and 2011 showed that EUCLID patients had a lower prevalence of coronary heart disease, peripheral vascular disease, heart failure, valvular heart disease, liver disease, peptic ulcer and other comorbidities and risk factors and a higher fractional urea clearance (Kt/V) than those in the Emilia Romagna and Calabria registries. Accordingly, survival analysis showed a lower mortality risk in the EUCLID 2010 and 2011 cohorts than in the combined two regional registries in the corresponding years: for 2010, hazard ratio (HR) EUCLID vs. Regional registries: 0.80 [95% confidence interval: 0.71–0.90]; for 2011, HR: 0.76 [0.65–0.90]. However, this difference was nullified by statistical adjustment for the difference in comorbidities and risk factors, indicating that the longer survival in the EUCLID database was attributable to the lower risk profile of patients included in that database. This preliminary analysis sets the stage for future observational studies and indicates that appropriate adjustment for difference in comorbidities and risk factors is needed to generalize to the Italian HD population analyses based on the data-rich EUCLID database.

Preface to data mining in biomedical informatics and healthcare

In the last decade, healthcare institutions, pharmaceutical companies as well as other organizations started to aggregate biomedical and clinical data in electronic databases. Mining these databases gives promising new threads of knowledge that could lead to a variety of beneficial outcomes for the entire community, from improving patients’ quality of life towards saving public healthcare costs or increasing efficiency of private healthcare companies. Given the complexity of biomedical and clinical information, it is important to make use of the proper tools to gain valuable insights from the available data.