Optimization of anemia treatment in hemodialysis patients via reinforcement learning

OBJECTIVE

Anemia is a frequent comorbidity in hemodialysis patients that can be successfully treated by administering erythropoiesis-stimulating agents (ESAs). ESAs dosing is currently based on clinical protocols that often do not account for the high inter- and intra-individual variability in the patient's response. As a result, the hemoglobin level of some patients oscillates around the target range, which is associated with multiple risks and side-effects. This work proposes a methodology based on reinforcement learning (RL) to optimize ESA therapy.

METHODS

RL is a data-driven approach for solving sequential decision-making problems that are formulated as Markov decision processes (MDPs). Computing optimal drug administration strategies for chronic diseases is a sequential decision-making problem in which the goal is to find the best sequence of drug doses. MDPs are particularly suitable for modeling these problems due to their ability to capture the uncertainty associated with the outcome of the treatment and the stochastic nature of the underlying process. The RL algorithm employed in the proposed methodology is fitted Q iteration, which stands out for its ability to make an efficient use of data.

RESULTS

The experiments reported here are based on a computational model that describes the effect of ESAs on the hemoglobin level. The performance of the proposed method is evaluated and compared with the well-known Q-learning algorithm and with a standard protocol. Simulation results show that the performance of Q-learning is substantially lower than FQI and the protocol. When comparing FQI and the protocol, FQI achieves an increment of 27.6% in the proportion of patients that are within the targeted range of hemoglobin during the period of treatment. In addition, the quantity of drug needed is reduced by 5.13%, which indicates a more efficient use of ESAs.

CONCLUSION

Although prospective validation is required, promising results demonstrate the potential of RL to become an alternative to current protocols.

Effects of fluid overload on heart rate variability in chronic kidney disease patients on hemodialysis

Background

While fluid overload (FO) and alterations in the autonomic nervous system (ANS) such as hypersympathetic activity, are known risk factors for cardiovascular morbidity and mortality in patients on chronic hemodialysis (HD), their relationship has not been thoroughly studied.

Methods

In this observational study involving 69 patients on chronic HD, FO was assessed by whole body bioimpedance measurements before the midweek HD session and ANS activity reflected by Heart Rate Variability (HRV) was measured using 24-hour Holter electrocardiogram recordings starting before the same HD treatment. In total, 13 different HRV indices were analyzed, comprising a mixture of time domain, frequency domain and complexity parameters. A correlation analysis was performed between the HRV indices and hydration status indices. Successively, patients were retrospectively assigned to a high FO (H, FO > 2.5 L) or low FO (L, FO ≤ 2.5 L) group and these were further compared also after stratification by diabetes mellitus. Finally, a small number of patients without diabetes with significant and persistent FO were followed up for 3 months post-study to investigate how normalization of fluid status affects HRV.

Results

SDANN, VLF, LZC and HF% parameters significantly correlate with FO (correlation coefficients were respectively r = –0.40, r = –0.37, r = –0.28 and r = 0.26, p-value < 0.05). Furthermore, LF% and LF/HF were inversely correlated with hydration status (correlation coefficients were respectively r = –0.31 and r = -0.33, p-value < 0.05). These results indicate an association between FO and reduced HRV, higher parasympathetic activation and reduced sympathetic response to the HD session. Indeed, group H tended to have lower values of SDANN, VLF and LZC, and higher values of HF% than patients in the L group. Finally, there was a trend towards lower LF% measured during the last 30 minutes of HD for the H group versus the L group. Reduction in FO achieved over 3 months by implementation of a strict fluid management plan resulted in an increase of HRV.

Conclusions

Our results suggest that depressed HRV is associated with fluid overload and that normalization of hydration status is accompanied by improved HRV.

Computational intelligence for the Balanced Scorecard: studying performance trends of hemodialysis clinics

OBJECTIVES

The Balanced Scorecard (BSC) is a general, widely employed instrument for enterprise performance monitoring based on the periodic assessment of strategic Key Performance Indicators that are scored against preset targets. The BSC is currently employed as an effective management support tool within Fresenius Medical Care (FME) and is routinely analyzed via standard statistical methods. More recently, the application of computational intelligence techniques (namely, self-organizing maps) to BSC data has been proposed as a way to enhance the quantity and quality of information that can be extracted from it. In this work, additional methods are presented to analyze the evolution of clinic performance over time.

METHODS

Performance evolution is studied at the single-clinic level by computing two complementary indexes that measure the proportion of time spent within performance clusters and improving/worsening trends. Self-organizing maps are used in conjunction with these indexes to identify the specific drivers of the observed performance. The performance evolution for groups of clinics is modeled under a probabilistic framework by resorting to Markov chain properties. These allow a study of the probability of transitioning between performance clusters as time progresses for the identification of the performance level that is expected to become dominant over time.

RESULTS

We show the potential of the proposed methods through illustrative results derived from the analysis of BSC data of 109 FME clinics in three countries. We were able to identify the performance drivers for specific groups of clinics and to distinguish between countries whose performances are likely to improve from those where a decline in performance might be expected. According to the stationary distribution of the Markov chain, the expected trend is best in Turkey (where the highest performance cluster has the highest probability, P=0.46), followed by Portugal (where the second best performance cluster dominates, with P=0.50), and finally Italy (where the second best performance cluster has P=0.34).

CONCLUSION

These results highlight the ability of the proposed methods to extract insights about performance trends that cannot be easily extrapolated using standard analyses and that are valuable in directing management strategies within a continuous quality improvement policy.

Seeking an optimal renal replacement therapy for the chronic kidney disease epidemic: the case for on-line hemodiafiltration

The prevalence of chronic kidney disease (CKD) can be expected to increase dramatically in the foreseeable future, with suggestions that it has already reached epidemic proportions. The inadequate supply of donor organs, aggravated by an aging patient population, necessitates provision of sustainable dialysis treatment modalities. These treatment modalities must not only be of established clinical efficacy and effectiveness, but must simultaneously circumvent any potential treatment disparities due to geographical, social or other concurring factors. Home therapies might represent a partial solution to the complex issue of seeking optimal strategies to cope with the CKD epidemic. However, self-care renal replacement therapy (RRT), such as peritoneal dialysis (PD) and home therapies, can only be applied to a limited portion of the CKD population. Consequently, in preparation for coping with this CKD epidemic, specific large-scale plans need to be made that involve optimization of treatments already in use for the majority of the population requiring RRT, e.g. hemodialysis (HD). Extracorporeal chronic HD relies heavily on technology for its clinical success. Like the choice of the treatment modality and the complete medical approach to CKD patient care, the particular selection of the various components of the extracorporeal circuit has a significant impact on the well-being and survival of the patients. We present a medical-technological assessment of how best to treat vast numbers of dialysis patients under the financial restraints that are predicted to become even more severe as CKD entrenches itself as a more 'permanent epidemic'. A treatment modality is proposed that optimally addresses--and resolves--the debilitating effects of uremia, as well as of key clinical conditions closely linked to it. This treatment modality successfully tackles the issues of patient well-being, efficacy, effectiveness, safety and patient-nursing staff convenience--all in relation to the overall costs incurred by payers of renal care. In short, optimal care needs to be provided with shrinking resources and without compromising the medical appropriateness of the therapy. Additionally, we believe ensuring improved quality of life is just as important as prolonging patient survival. Therefore, a balanced compromise between optimal and affordable technology is required in order to reach the targets of achieving good medical care and meeting the expectations of patients, their families, healthcare providers, and society as a whole. Under these premises, and focusing on the aforementioned targets, we believe that on-line hemodiafiltration (HDF) represents the most advanced and clinically appropriate RRT modality available to best cope with the CKD epidemic. Together with the guidance and recommendations of those taking care of CKD patients on dialysis therapy, the contribution of industry is indispensable for the availability of highly reliable and affordable solutions to the impending dilemma. As representatives of the academic-medical community and of industry, we present a joint case for the application of on-line HDF towards meeting the challenge of large-scale provision of dialysis under an increasingly restrictive financial climate.

The role of microvesicles derived from mesenchymal stem cells in tissue regeneration; a dream for tendon repair?

Tendon injuries represent even today a challenge as repair may be exceedingly slow and incomplete. Regenerative medicine and stem cell technology have shown to be of great promise. Here, we will review the current knowledge on the mechanisms of the regenerative potential of mesenchymal stem cells (MSCs) obtained from different sources (bone marrow, fat, cord blood, placenta). More specifically, we will devote attention to the current use of MSCs that have been used experimentally and in limited numbers of clinical cases for the surgical treatment of subchondral-bone cysts, bone-fracture repair and cartilage repair. Based on the recently emerging role in regenerative mechanisms of soluble factors and of extracellular vesicles, we will discuss the potential of non-cellular therapies in horse tendon injuries.

Clock gene variation is associated with breeding phenology and maybe under directional selection in the migratory barn swallow

Background

In diverse taxa, photoperiodic responses that cause seasonal physiological and behavioural shifts are controlled by genes, including the vertebrate Clock orthologues, that encode for circadian oscillator mechanisms. While the genetic network behind circadian rhythms is well described, relatively few reports exist of the phenological consequences of and selection on Clock genes in the wild. Here, we investigated variation in breeding phenology in relation to Clock genetic diversity in a long-distance migratory bird, the barn swallow (Hirundo rustica).

Methodology/Principal Findings

In a sample of 922 adult barn swallows from a single population breeding in Italy we found one very common (Q₇) and three rare (Q₅, Q₆, Q₈) length variants of a functionally significant polyglutamine repeat. Rare (2.9%) Q₇/Q₈ heterozygous females, but not males, bred significantly later than common (91.5%) Q₇/Q₇ females, consistent with the expectation that ‘long’ alleles cause late breeding, as observed in a resident population of another bird species. Because breeding date depends on arrival date from migration, present results suggest that the association between breeding date and Clock might be mediated by migration phenology. In addition, fecundity selection appears to be operating against Q₇/Q₈ because late migrating/breeding swallows have fewer clutches per season, and late breeding has additional negative selection effects via reduced offspring longevity. Genotype frequencies varied marginally non-significantly with age, as Q₇/Q₈ frequency showed a 4-fold reduction in old individuals. This result suggests negative viability selection against Q₇/Q₈, possibly mediated by costs of late breeding.

Conclusions/Significance

This is the first study of migratory birds showing an association between breeding phenology and Clock genotype and suggesting that negative selection occurs on a phenologically deviant genotype. Low polymorphism at Clock may constrain microevolutionary phenological response to changing climate, and may thus contribute to the decline of barn swallow populations.

Predictive modeling of cardiovascular complications in incident hemodialysis patients

The administration of hemodialysis (HD) treatment leads to the continuous collection of a vast quantity of medical data. Many variables related to the patient health status, to the treatment, and to dialyzer settings can be recorded and stored at each treatment session. In this study a dataset of 42 variables and 1526 patients extracted from the Fresenius Medical Care database EuCliD was used to develop and apply a random forest predictive model for the prediction of cardiovascular events in the first year of HD treatment. A ridge-lasso logistic regression algorithm was then applied to the subset of variables mostly involved in the prediction model to get insights in the mechanisms underlying the incidence of cardiovascular complications in this high risk population of patients.

Role of microvesicles in acute kidney injury

The main function of microvesicles (MVs) is signaling through specific interactions with target cells and transferring gene products. Therefore, they may participate in physiological and pathological processes. Gaining further insights into the molecular specificity of MVs has allowed identifying the cellular source and may provide new diagnostic tools in the future. Indeed, an increasing body of evidence indicates that MVs may offer prognostic information in various diseases such as chronic inflammation, cardiovascular and renal diseases, pathological pregnancy, tumors, and sepsis. The presence of MVs in body fluids makes them readily accessible. Their number, cellular origin, composition and function can be dependent on the state of the disease. In sepsis for example, activated endothelial cells may shed MVs that might trigger leukocyte and monocyte production and release pro-oxidant and inflammatory mediators. MVs from platelets may trigger disseminated intravascular coagulopathy. MVs are no doubt also involved in modulating immunity and future studies will clarify their functional role in negatively modulating the cell response. In addition, the recognition of the signals delivered by MVs may open new therapeutic strategies. The removal of harmful MVs from plasma may be beneficial in pathological conditions where MVs deliver thrombogenic and inflammatory signals. On the other hand, MVs derived from stem cells may reprogram altered functions in target cells suggesting that they could be exploited in regenerative medicine to repair damaged tissues. We will discuss the role of stem cell-derived MVs in the repair of acute kidney injury.

Effects of telmisartan added to Angiotensin-converting enzyme inhibitors on mortality and morbidity in hemodialysis patients with chronic heart failure a double-blind, placebo-controlled trial

OBJECTIVES

the aim of this study was to determine whether telmisartan decreases all-cause and cardiovascular mortality and morbidity in hemodialysis patients with chronic heart failure (CHF) and impaired left ventricular ejection fraction (LVEF) when added to standard therapies with angiotensin-converting enzyme inhibitors.

BACKGROUND

in hemodialysis patients, CHF is responsible for a high mortality rate, but presently very few data are available with regard to this population.

METHODS

A 3-year randomized, double-blind, placebo-controlled, multicenter trial was performed involving 30 Italian clinics. Hemodialysis patients with CHF (New York Heart Association functional class II to III; LVEF ≤ 40%) were randomized to telmisartan or placebo in addition to angiotensin-converting enzyme inhibitor therapy. A total of 332 patients were enrolled (165 telmisartan, 167 placebo). Drug dosage was titrated to a target dose of telmisartan of 80 mg or placebo. Mean follow-up period was 35.5 ± 8.5 months (median: 36 months; range: 2 to 40 months). Primary outcomes were: 1) all-cause mortality; 2) cardiovascular mortality; and 3) CHF hospital stay.

RESULTS

at 3 years, telmisartan significantly reduced all-cause mortality (35.1% vs. 54.4%; p < 0.001), cardiovascular death (30.3% vs. 43.7%; p < 0.001), and hospital admission for CHF (33.9% vs. 55.1%; p < 0.0001). With Cox proportional hazards analysis, telmisartan was an independent determinant of all-cause mortality (hazard ratio [HR]: 0.51; 95% confidence interval [CI]: 0.32 to 0.82; p < 0.01), cardiovascular mortality (HR: 0.42; 95% CI: 0.38 to 0.61; p < 0.0001), and hospital stay for deterioration of heart failure (HR: 0.38; 95% CI: 0.19 to 0.51; p < 0.0001). Adverse effects, mainly hypotension, occurred in 16.3% of the telmisartan group versus 10.7% in the placebo group.

CONCLUSIONS

addition of telmisartan to standard therapies significantly reduces all-cause mortality, cardiovascular death, and heart failure hospital stays in hemodialysis patients with CHF and LVEF ≤ 40%. (Effects Of Telmisartan Added To Angiotensin Converting Enzyme Inhibitors On Mortality And Morbidity In Haemodialysed Patients With Chronic Heart Failure: A Double-Blind Placebo-Controlled Trial; NCT00490958).

Study of the autonomic response in hemodialysis patients with different fluid overload levels

This work aims at studying the autonomic nervous system (ANS) response to hemodialysis (HD) treatment in a population of end stage renal disease (ESRD) patients. ECG Holter recordings and whole body bioimpedance spectroscopy measurements were performed for each patient. Patients were classified according to the fluid overload (FO) values and the systolic blood pressure (SBP) measured before HD. Time domain and frequency domain indices from heart rate variability (HRV) signals were measured for the first 30 minutes and last 30 minutes of HD, the first hour after HD, and night (12.00 p.m.-4 a.m.). Significant differences were obtained in fluid overloaded but normotensive patients (Group IV) with respect to fluid overloaded and hypertensive patients (Group I) and normohydrated and normotensive patients (Group N+Dx). In particular, SDNN, RMSSD, SDSD, pNN50%, indices resulted significantly higher in Group IV with respect to the other groups. Overhydrated patients with hypertension (Group I) showed a blunted parasympathetic activity, which is supposed to contribute to hypertension.

Replacement therapy with growth hormone and pituitary tumor recurrence: the relevance of the problem

Most cases of adult GH deficiency (AGHD) result from hypothalamic-pituitary tumors or their treatment. Some experimental and clinical observations suggest that GH may possess a mitogenic potential, thus raising the question of whether it is a safe treatment in patients with a previous pituitary tumor. Few study results have been reported on this topic. All of them have inevitable methodological flaws that limit their conclusions. However, all studies report that replacement therapy with GH does not seem to increase the risk of tumor progression or recurrence, when compared to historical or matched controls. Considering the slow-growing nature of most of these benign tumors and the absence of conclusive evidence from the available studies, a continuous imaging surveillance and longer follow-up periods are nevertheless mandatory for a definite statement on the safety of GH treatment in patients with previous pituitary tumors.

Heart rate as independent prognostic factor for mortality in normotensive hemodialysed patients

BACKGROUND

Several epidemiological studies have reported heart rate (HR) as a predictor of cardiovascular and noncardiovascular mortality in the general population. Aim of the present study was to investigate a possible relation between increased HR and mortality in normotensive end-stage renal disease (ERSD) patients.

METHODS

Between 1997 and 2001 we recruited 407 normotensive ESRD patients (mean age 56.6 +/- 3.3 years) without coronary artery disease, left ventricular systolic dysfunction or on antiarrhythmic therapy. Baseline electrocardiography (ECG) at rest, 48-hour ambulatory Holter ECG monitoring and standard echocardiography were performed. After a mean follow-up of 46 months (range 12-60 months), cardiovascular and sudden death were considered as end points.

RESULTS

Echocardiogram showed a normal left ventricular ejection fraction (>55%) in 370 patients (91%) and left ventricular hypertrophy (LVH) in 290 patients (71.2%). Mean HR by 48-hour Holter ECG was 81 +/- 10.6 bpm. During the follow-up, all-cause mortality rate was 12% (49 patients); 40 patients died from cardiac cause (9.8%) of which 20 patients (4.9%) by sudden death. By univariate analysis, age, diabetes, ECG-LVH with signs of left ventricular strain, and increased mean HR by 48-hour Holter ECG were all significantly related to global, cardiovascular and sudden death. ROC curve analysis identified optimal cutoff points for HR >85 bpm and age >65 years associated with increased cardiovascular risk (p<0.001). By Cox regression analysis, only age >65 years (p<0.0001) and mean HR >85 bpm (p<0.0001) were independent predictors of cardiovascular events.

CONCLUSIONS

In normotensive ERSD patients, increased mean HR detected by 48-hour Holter ECG is an independent determinant of global and cardiovascular mortality.

Hypokalemic periodic paralysis in a patient with acquired growth hormone deficiency

CONTEXT

Hypokalemic periodic paralysis (HypoPP) is a rare disorder consisting of sudden episodes of muscle weakness with areflexia involving all four limbs, which spontaneously resolve within several hours or days. Primary HypoPP is genetically determined, while secondary acquired HypoPP has been described in association with thyreotoxycosis, hyperaldosteronism, kidney diseases, diuretics and liquorice abuse, gastrointestinal potassium loss, or cysplatinum therapy.

OBJECTIVE

To report a case of HypoPP associated with GH deficiency.

PATIENT

A 33 yr-old man with hypopituitarism and diabetes insipidus secondary to pituitary stalk-localized sarcoidosis, and documented HypoPP episodes.

CLINICAL PRESENTATION

Neurologic exam outside HypoPP episodes was normal. Needle electromyography was normal without myotonia or other spontaneous electric activity. Muscle biopsy documented a vacuolar myopathy with tubular aggregates. However, genetic analysis ruled out common mutations of the voltage-gated calcium channel observed in primary HypoPP. Common causes of secondary HypoPP were also ruled out. The patient was diagnosed with severe GH deficiency with modest fasting hyperinsulinemia and insulin resistance and started on GH replacement therapy, an alpha-glucosidase inhibitor (acarbose) and a diet low in simple carbohydrates.

CONCLUSIONS

GH replacement therapy, acarbose and a diet low in simple carbohydrates resulted in the complete long-term (>2 yr) remission of HypoPP episodes. This is consistent with the hypothesis that the hyperinsulinemia associated to GH deficiency may trigger HypoPP episodes by increasing Na+/K+ ATPase activity and K+ transport into the intracellular compartment with subsequent hypokalemia.

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.

Solid-pseudopapillary tumor of the pancreas and pancreatoblastoma in children: Experience over 18 years at a Pediatric Oncology Center

9062

Background: Malignant pancreatic tumor is a rare disease in the pediatric population. Guidelines for treatment, including management in advanced disease are controversial until this moment (REBHANDL, 2001). Regarding to the pathology findings, pancreatoblastoma and solid-pseudopapillary tumor of the pancreas (Frantz’s tumor) are the more frequent pancreatic tumors in childhood. The aim of this study was a retrospective analysis of all malignant pancreatic tumors diagnosed at Boldrini’s Children Cancer Center during a period of 18 years. Methods: A retrospective analysis of eight cases of malignant pancreatic tumors, admitted at the Boldrini’s Children Cancer Center, from January 1987 to April 2005. All the data concerning age, sex, clinical presentation, stage, treatment and follow-up were collected. Results: The patient’s population refers to 8 children. There were 4 boys and 4 girls. Age ranged from 1 month to 15 years old (median age of 14 years). Main symptom was the increase of abdominal volume. None of them had metastatic disease at diagnosis. Regarding to pathology, 4 patients had pancreatoblastoma and 4 Frantz’s tumor. One patient (who had pancreatoblastoma) received chemotherapy before surgery, with a partial response, and then had a complete tumor ressection. All others (7 patients) had complete surgery as first treatment. Only one patient with Frantz’s tumor had pancreatic injury during tumor ressection and is now receiving enzimatic reposition. All the patients are alive, free of disease, from 6 months to 11 years (median follow-up of 2 years). Table I . Conclusions: Total pancreatectomy is the most important treatment in pancreatoblastoma and Frantz’s tumor. Neoadjuvant chemotherapy may be an important tool in unresectable tumors, increasing the disease-free survival.

[Table: see text]

No significant financial relationships to disclose.