•Narrative review of machine learning methods used on end-stage kidney disease.•Techniques to predict survival in kidney transplantation and dialysis therapy.•Models used to analyze the incidence of ...infection or tacrolimus toxicity.•Algorithms used to analyze other dialysis variables to predict mortality.
In the field of medicine, decision-making has traditionally been carried out based on the best available scientific information and the experience of specialists using data found in analog formats such as radiographies, medical reports, and handwritten notes, among others. In this sense, the Big Data phenomenon is changing the world of medicine since the technologies that have been developed have made available to researchers and clinicians enormous amounts of data in digital formats that can be used to complement or help in complex tasks such as mentioned decision making. A key element in this process is data analysis techniques, since without them it is not possible to exploit the information. Currently the most used techniques are based on algorithms in the area of artificial intelligence and more specifically machine learning. This paper focuses on a specific domain of medicine, renal replacement therapies for end-stage renal disease, where machine learning is beginning to be used as a complementary tool to predict or make decisions. This paper provides a narrative review of the main machine learning methods that are being used to conduct end-stage renal disease treatment analyses.
To analyse the association between donor capnometry data and the short-term evolution of kidney grafts in cases of uncontrolled donation after circulatory death (uDCD).
We used an ambispective ...observational study design, conducted in the Community of Madrid between January and December 2019, inclusive. Patients who suffered out-of-hospital cardiac arrest (CA) with no response to advanced cardiopulmonary resuscitation (CPR) were selected as potential donors. Donor capnometry levels were measured at the start, midpoint and transfer to hospital then compared with indicators of renal graft evolution.
The initial selection included 34 possible donors, of which 12 (35.2%) were viable donors from whom 22 (32.3%) kidneys were recovered. There was a correlation between the highest capnometry values and less need for post-transplant dialysis (≥24 mmHg, p < 0.017), fewer dialysis sessions and fewer days to recover correct renal function (Rho −0.47, p < 0.044). There was a significant inverse correlation between the capnometry values at transfer and 1-month post-transplant creatinine levels (Rho −0.62, p < 0.033). There were no significant differences between the capnometry values at transfer and primary nonfunction (PNF) or warm ischaemia time. One-year patient survival was 100% for patient receiving organ donation, while graft survival was 95%.
Capnometry levels at transfer are a useful predictor of the short-term function and viability of kidney transplants from uncontrolled donations after circulatory death.
The recurrence of primary focal segmental glomerulosclerosis (FSGS) after kidney transplantation (KT) appears in 30 % of the recipients. Sometimes it can cause the loss of the allograft. Although ...many treatments for this condition have been reported, 20 %–40 % of the affected patients are refractory or presents frequents relapses. In this paper we describe the evolution of three recipients treated with long-term plasmapheresis therapy after a recurrence of FSGS with a bad or incomplete response to other treatments. Although our findings require confirmation, long-term plasmapheresis could be a therapeutic option for this condition.
Abstract Objective This study aims to determine the failure rate of transplanted kidney grafts in recipients of organs from non-heart beating donors (NHBDs) who have had mechanical chest compressions ...to maintain a circulation before organ retrieval. Methods A retrospective observational study based on review of the emergency medical service database and case histories of NHBDs, and information periodically sent by transplant units about donors and organs. The following variables were studied: age, sex, transfer hospital, time to arrival on the scene of cardiopulmonary arrest, time to arrival in hospital, number and type of organs retrieved, use of mechanical chest compression devices, and kidney function in graft recipients. The study covered the period between January 2008 and November 2009. During 2008 standard manual chest compressions were used and during 2009 mechanical chest compression devices were used. Results In 39 transplanted kidneys from donors receiving mechanical chest compressions primary failure was documented in recipients on two occasions (5.1%). Kidneys transplanted from donors who had manual chest compressions resulted in three primary failures in recipients (9.1%). The difference between the two groups was not significant ( p = 0.5). Three patients achieved successful return of spontaneous circulation in the mechanical chest compression group after initiation of the NHBD donor protocol. Conclusion We have described our experience and protocol for non-heart beating donation using victims of out-of-hospital cardiac arrest in whom cardiopulmonary resuscitation has been unsuccessful as donors. Primary kidney graft failure rates in organs from non-heart beating donors is similar when manual or mechanical chest compression devices are used during cardiopulmonary resuscitation.
Perigraft fluid collection (PFC) is a common complication after kidney transplant. Its etiology is not clear and not all the causes have been identified. The influence of the type of donor has never ...been evaluated. Our aim was to compare the incidence, severity and management of PFC in recipients of grafts from uncontrolled donors after circulatory death (DCD) with normothermic extracorporeal membrane oxygenation (NECMO) versus recipients of grafts from donors after brain death (DBD).
We conducted a retrospective cohort study of 300 kidney transplants performed in our center between 2007 and 2012. Patients were divided in two groups: 150 recipients of Maastricht II DCD graft and 150 recipients of the DBD graft. Incidence, severity according to Clavien scale and management were analyzed in both groups, and comparison was carried out using Chi-square.
Of the 300 kidney recipients analyzed, 93 (31.4%) suffered PFC, showing no difference between DBD (32.0%) and DCD (30.8%) groups (p = 0.9). Complicated PFC rate (defined as a PFC generating vascular compression, fever or urinary tract obstruction) was 22.9% in the DBD group versus 22.2% in the DCD group (p = 1); most complicated PFC were due to urinary tract obstruction (81%), with no difference between the groups (p = 1). Concerning Clavien scale, 78.5% of the PFC in our series were Clavien I, 19.4% Clavien IIIa and 2.2% Clavien IIIb, with no difference between both groups (p = 1).
PFC is a frequent complication that appears in a third of our patients, showing no difference in the incidence or severity between DBD and uncontrolled DCD graft recipients.
Abstract
BACKGROUND AND AIMS
Persistent chronic hypotension affects 5–10% of dialysis patients, and it is associated with high morbidity and mortality 1. Although data regarding the influence of this ...hemodynamic condition on kidney transplantation are scarce, some studies suggest a negative impact on kidney transplant outcomes 2-3. We decided to analyze the evolution of patients with chronic hypotension in dialysis who undergo kidney transplantation in our center.
METHOD
A retrospective observational study was conducted. We evaluated 2308 consecutive kidney transplants performed at Hospital Doce de Octubre between 2004 and 2020. Sixty-six patients with chronic hypotension (defined as systolic blood pressure ≤ 100 mmHg at the time of transplantation) were identified. A control group of 66 non-hypotensive patients was assigned, matched for the source of organs and age (using as control, whenever possible, the recipient of the other kidney from the same donor). The evolution of both groups was compared in terms of primary graft function, graft thrombosis, delay in graft function, serum creatinine at the end of follow-up and renal graft survival.
RESULTS
Patients with chronic hypotension had higher rates of primary non-function (18.2% versus 6.1%, P = 0.033) mainly due to venous thrombosis of the renal graft (15.2% versus 3%, P = 0.015). Delayed graft function was also more common in patients with chronic hypotension (68.2% versus 50%, P< 0001). Mean graft survival was lower in the group of patients with chronic hypotension (81.1 months) compared to the control group (104.1 months) (P = 0.012). At the end of follow-up, there were 67.7% of functioning grafts within the hypotensive group compared with 86.4% in the control group (P = 0.013). Serum creatinine at the end of follow-up was slightly higher in patients with chronic hypotension (1.76 ± 0,70mg/dL versus 1.50 ± 0.49 mg/dL, P = 0.04). Median follow-up time was 37 months (1–122). In multivariate analysis, chronic hypotension was an independent risk factor for renal graft loss RR = 2.8 (1.3–6.4), P = 0012.
CONCLUSION
Chronic hypotension in dialysis has a negative impact in short and long-term kidney transplant outcomes. It is associated with higher rates of primary non-function due to venous graft thrombosis, higher rates of delayed graft function, a higher serum creatinine at the end of follow-up and a worse renal graft survival.
It seems crucial to identify this subgroup of patients in order to implement measures aimed to ameliorate transplant results.
PDF
