Blood Aluminum Levels in Patients with Hemodialysis and Peritoneal Dialysis

Data-driven, two-stage machine learning algorithm-based prediction scheme for assessing 1-year and 3-year mortality risk in chronic hemodialysis patients

Life expectancy is likely to be substantially reduced in patients undergoing chronic hemodialysis (CHD). However, machine learning (ML) may predict the risk factors of mortality in patients with CHD by analyzing the serum laboratory data from regular dialysis routine. This study aimed to establish the mortality prediction model of CHD patients by adopting two-stage ML algorithm-based prediction scheme, combined with importance of risk factors identified by different ML methods. This is a retrospective, observational cohort study. We included 800 patients undergoing CHD between December 2006 and December 2012 in Shin-Kong Wu Ho-Su Memorial Hospital. This study analyzed laboratory data including 44 indicators. We used five ML methods, namely, logistic regression (LGR), decision tree (DT), random forest (RF), gradient boosting (GB), and eXtreme gradient boosting (XGB), to develop a two-stage ML algorithm-based prediction scheme and evaluate the important factors that predict CHD mortality. LGR served as a bench method. Regarding the validation and testing datasets from 1- and 3-year mortality prediction model, the RF had better accuracy and area-under-curve results among the five different ML methods. The stepwise RF model, which incorporates the most important factors of CHD mortality risk based on the average rank from DT, RF, GB, and XGB, exhibited superior predictive performance compared to LGR in predicting mortality among CHD patients over both 1-year and 3-year periods. We had developed a two-stage ML algorithm-based prediction scheme by implementing the stepwise RF that demonstrated satisfactory performance in predicting mortality in patients with CHD over 1- and 3-year periods. The findings of this study can offer valuable information to nephrologists, enhancing patient-centered decision-making and increasing awareness about risky laboratory data, particularly for patients with a high short-term mortality risk.

Ratiometric near-infrared upconversion fluorescence sensor for selectively detecting and imaging of Al3

Near-infrared (NIR) fluorescent probes provide extremely sensitive Al³⁺ detection for human health purposes. This research develops novel Al³⁺ response molecules (HCMPA) and NIR upconversion fluorescent nanocarriers (UCNPs), which respond to Al³⁺ through ratio NIR fluorescence. UCNPs improve photobleaching and visible light lack in specific HCMPA probes. Additionally, UCNPs are capable of ratio response, which will further enhance signal accuracy. The NIR ratiometric fluorescence sensing system has been successfully used to detect Al³⁺ within the range 0.1-1000 nM with an accuracy limit of 0.06 nM. Alternatively, a NIR ratiometric fluorescence sensing system integrated with a specific molecule can image Al³⁺ within cells. This study demonstrates that a NIR fluorescent probe is an effective and highly stable method of measuring Al³⁺ in cells.

Hungry bone syndrome after parathyroid surgery

INTRODUCTION

Data on the incidence rates of hungry bone syndrome after parathyroidectomy in patients on dialysis are inconsistent, as the published rates vary from 15.8% to 92.9%.

METHODS

Between 2009 and 2019, 120 hemodialysis patients underwent parathyroidectomy for secondary hyperparathyroidism at the Chang Gung Memorial Hospital. The patients were stratified into two groups based on the presence (n = 100) or absence (n = 20) of hungry bone syndrome after parathyroidectomy.

FINDINGS

Subtotal parathyroidectomy was the most common surgery performed (76.7%), followed by total parathyroidectomy with autoimplantation (23.3%). Pathological examination revealed parathyroid hyperplasia. Hungry bone syndrome developed within 0.3 ± 0.3 months and lasted for 11.1 ± 14.7 months. After surgery, compared with patients without hungry bone syndrome, patients with hungry bone syndrome had lower levels of nadir corrected calcium (P < 0.001), as well as lower nadir (P < 0.001) and peak (P < 0.001) intact parathyroid hormone levels. During 59.3 ± 44.0 months of follow-up, persistence and recurrence of hyperparathyroidism occurred in 25 (20.8%) and 30 (25.0%) patients, respectively. Furthermore, patients with hungry bone syndrome had a lower rate of persistent hyperparathyroidism than those without hungry bone syndrome (P < 0.001). Four patients (3.3%) underwent a second parathyroidectomy. Patients with hungry bone syndrome received fewer second parathyroidectomies than those without hungry bone syndrome (P < 0.001). Finally, a multivariate logistic regression model revealed that the preoperative blood ferritin level was a negative predictor of the development of hungry bone syndrome (P = 0.038).

DISCUSSION

Hungry bone syndrome is common (83.3%) after parathyroidectomy for secondary hyperparathyroidism in patients undergoing hemodialysis, and this complication should be monitored and managed appropriately.