A prescription-based model for predicting post-hemodialysis urea and electrolytes: a real-world validation study in acute hemodialysis

Abstract

Background

Urgent-start hemodialysis often involves severe biochemical abnormalities but is associated with a high risk of complications. Currently, no tool has been validated to predict post-hemodialysis urea and electrolytes in acute settings from prescription data alone.

Methods

In this ambispective, two-phase study, we developed and validated a model to predict post-hemodialysis blood urea nitrogen and electrolytes using only prescription parameters in patients with urgent hemodialysis indications and those at risk for dialysis disequilibrium syndrome, severe hyponatremia, or hyperkalemia. The model was validated through statistical assessment of correlation and agreement. The model was integrated into a free, user-friendly web application (Adequator app®).

Results

The development cohort included 303 treatments in 42 chronic hemodialysis patients. Dialyzer clearance, which was calculated via formal urea kinetic modeling, demonstrated excellent correlation and agreement with the proposed model (r = 0.99; 95% CI: 0.99 to 0.99) and a bias of 0.23 ± 1.2 ml/min (95% CI: −2.17 to 2.63). The validation cohort included 44 urgent hemodialysis sessions in 30 patients with chronic kidney disease or acute kidney injury and severe electrolyte abnormalities or at risk of dialysis disequilibrium syndrome. The predicted post-hemodialysis, blood urea nitrogen, sodium, and potassium levels strongly correlated with the measured values: r = 0.96 (95% CI: 0.93 to 0.98), r = 0.96 (95% CI: 0.94 to 0.98), and r = 0.84 (95% CI: 0.73 to 0.91), respectively. Agreement was also high, with biases of 0.4 ± 9.4 mg/dL (95% CI: −19.0 to 18.1) for blood urea nitrogen, −0.6 ± 1.6 mEq/L (95% CI: −3.8 to 2.5) for sodium, and 0.17 ± 0.4 mEq/L (95% CI: 0.6 to 0.9) for potassium. Among patients at risk of dialysis disequilibrium syndrome, 92% achieved the target blood urea nitrogen reduction ( < 40%) without severe neurological events. The sodium correction remained below 6 mEq/L, and potassium normalized in all the cases.

Conclusions

A prescription-based model can reliably predict post-Hemodialysis urea and electrolyte values in urgent dialysis settings. The Adequator App HD-Predictor (https://adequatorapp.com/hd-predictor) offers a non-invasive, accessible tool to guide individualized prescriptions, enhancing safety and precision in high-risk clinical scenarios.

Background

Hemodialysis (HD) indications characterized by extreme urea and electrolyte values are common, particularly in urgent-start dialysis, and have been associated with worse outcomes [1–4]. The ability to predict urea clearance and post-HD electrolytes on the basis of a prescription alone would be valuable in urgent clinical scenarios, such as managing electrolyte imbalances such as severe hyponatremia or hyperkalemia [5, 6] or preventing dialysis disequilibrium syndrome (DDS) [7]. Therefore, the development of a tool capable of predicting post-HD values using only HD prescription parameters could be relevant and could improve clinical decision-making and patient safety.

Delivered urea clearance is traditionally assessed using kinetic models, with the single-pool urea kinetic model being the most widely utilized. This method requires blood samples to determine urea concentrations at the start and end of each HD treatment [8]. A less invasive and widely available alternative is the estimation of ionic dialysance, which serves as a surrogate for effective urea clearance. While this method can predict urea clearance in real time, it provides only the final clearance value at the end of the treatment [9]. However, none of these methods can predict the final effective urea clearance solely based on the HD prescription prior to treatment.

Depner et al. (2004) proposed a mathematical model to predict dialyzer clearance (Kd) using key parameters such as real blood flow (rQb), dialysate flow (Qd), and the in vivo membrane permeability-area coefficient (KoA), enabling the estimation of accurate Kd values [10].

Building upon this model, we developed a novel approach that uses the estimated Kd, treatment time (T), and anthropometric-based urea volume of distribution (V) to predict post-HD urea and electrolytes. This method was subsequently validated both internally and externally to predict post-HD values relying solely on the computation of the prescription parameters in an automated, user-friendly calculator (Adequator app ®), which was hypothesized to be especially relevant for specific acute HD indications, such as patients with a high risk of developing DDS, severe hyponatremia and severe hyperkalemia.

Methods

Study population

This was an, ambispective, two-phase study that included a retrospective cohort (development cohort) to assess the feasibility and accuracy of the calculations, followed by a prospective real-world cohort (validation cohort) to validate the predicted post-HD values in urgent dialysis indications.

Development cohort

A retrospective analysis was conducted on 42 anuric patients undergoing maintenance HD at the Hemodialysis Unit of the Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán in Mexico City. A total of 303 HD treatments were evaluated using a formal urea kinetic modeling approach implemented via de Solute Solver ® software [11] in conjunction with the newly developed method for predicting post-HD values. All patients were treated with high-flux polysulfone dialyzers (F180, F60; Fresenius Medical Care) and 4008 HD machines (Fresenius Medical Care) 2–3 times per week through tunneled central venous access, arteriovenous fistulas (AVFs), or arteriovenous grafts (AVGs). Qb, Qd, and ultrafiltration rates (Qf) were prescribed based on individual clinical requirements. Prevalent HD patients were selected for the model development due to the consistent, iterative nature of their treatment data, which enabled a robust retrospective analysis, without interfering with clinical care or prescription decisions.

Monthly blood samples were collected before and after treatment by reducing Qb to 100 ml/hr for 20 seconds, stopping the blood pump, and drawing the sample from the arterial line [12]. Patient weight and height were obtained and recorded on the same day as the samples were collected. Kd, single-pool KtV, equilibrated Kt/V, and urea reduction ratio (RR) were computed using the urea kinetic model Solute Solver® [11] and the collected blood samples. Additionally, Kd, Kt/V, RR, and predicted final blood urea nitrogen (BUN) values were calculated on the basis of age, weight, height, and HD prescription (including Qb, Qd, Qf, KoA, and T).

Validation cohort

The validation cohort was prospectively conducted in a real-world setting at the Centro Médico ISSEMYM in Ecatepec, Mexico. The study included 30 adult patients admitted to the emergency department with chronic kidney disease (CKD) or acute kidney injury (AKI) who required urgent HD treatment (either their first or second HD) and were at risk for developing DDS or experiencing severe electrolyte derangements. Patients with CKD, a BUN greater than 150 mg/dL, and at least one neurological symptom of uremia were considered at risk of DDS. A pre-HD sodium concentration of < 120 meq/L was considered severe hyponatremia, and a pre-HD potassium concentration of at least 6.5 meq/L was considered severe hyperkalemia.

HD prescriptions, namely, Qb, Qd, Qf, membrane type, T and dialysate concentrations (Cd), were planned and executed using the proposed method to target RRs less than 40% for patients at risk of DDS, sodium correction of less than 6 meq/L for those with severe hyponatremia, and potassium reduction to safe levels for patients with severe hyperkalemia. Blood samples were taken before and after each treatment using the same method used for the development cohort, single pool KtV and equilibrated KtV were calculated using validated formulas [13, 14]. Anthropometric measurements were obtained from medical records. Patients were treated with high-flux polyarylethersulfone and polyvinylpyrrolidone blend membranes (Revaclear 400 or Polyflux 6 H; Vantive US Healthcare), Gambro AK98 HD machines (Vantive US Healthcare) and a temporal 11.5 fr. central venous catheter (Mahurkar^TM; Mozarc Medical). HD parameters were monitored during treatment to remain as prescribed. Post-HD values for BUN, sodium, and potassium were obtained.

Prediction model

The proposed method for estimating post-HD values (C2) relies on calculations from pre-HD levels (C1), T, estimated V Eqs. (5 and 6, estimated Kd Eqs. (1-4), and the established logarithmic relationship between Kt/V and the urea RR [14] (Eq. 7). For predicting post-HD values, the method applies the principle that ionic dialysance approximates urea clearance, therefore equilibration ratio, equals RR, [9], which was used to solve Eq. (8).

Equations

Membrane performance characteristics, as reported in the dialyzer data sheet or calculated from in vitro clearances, were used to derive the in vivo KoA using the following equations [10, 14]:

For Qd rates of 500 m/min or faster, the KoA in vivo was determined via the following equation:

1

For Qd rates of < 500 ml/min, KoA invivo was determined via the following equation:

2

For Qb rates greater than 200 ml/min, rQb was determined via the following equation [10]:

3

Kd was calculated using the KoA invivo Eqs. (1 and 2), rQb (Eq. 3), the prescribed Qd, and the prescribed Qf with the following equation [10, 14]:

4

V was assumed to be equivalent to the total body water (TBW), estimated using the Watson equation [15]:

5

6

The RR was estimated using Kd, V, and T.

7

Finally, C2 was estimated by using C1, Cd, RR, and solving for C2:

8

Automatization into a web app.

All the above calculations were integrated into a free web app designed for automated and user-friendly calculations (Adequator app®) https://adequatorapp.com/hd-predictor [16]. The Adequator app® predicts Kd, Kt, KtV, and urea RR as output data. These predictions are based on input data that include the dialysis prescription such as dialyzer type, Qb, Qd and Uf, as well as patient characteristics such as gender, age, height and weight. Additionally, if the pre-dialysis concentration of solutes such as urea, sodium and potassium are available and entered, the app predicts the post-dialysis concentration.

Statistical analysis

The distribution of continuous variables was evaluated using the Kolmogorov‒Smirnov test. Descriptive statistics are expressed as numbers (percentages), medians (interquartile ranges), and means ± standard deviations (SDs), as appropriate. In addition, a correlation analysis was performed between the formal urea kinetic model (Solute Solver®) and the newly proposed method for predicting post-HD values (Adequator app®) in the development cohort. Correlation analysis was performed between the measured post-HD values and the values predicted by the Adquator app® in the validation cohort. Bland‒Altman plots were used to assess agreement and bias. All the statistical analyses were performed using Prism Version 10.4.2 (GraphPad Software, Inc.).

Results

Development cohort

A total of 42 patients with CKD and 280 HD treatments were included in the development cohort (Fig. 1). The main characteristics of the patients, and HD treatments of the development cohort are available in Tables 1 and 2, respectively. The measured values were as follows: Kd, 239 (216–260) ml/min; single pool Kt/V, 1.86 (1.65–2.07); equilibrated Kt/V, 1.64 (1.4–1.8); BUN RR, 80 (76–82) %; and post-HD BUN, 14 (10.5–17.4) mg/dl (Table 3).

Fig. 1.

Flow chart of patients and treatments analyzed in the development cohort and the validation cohort. CKD: chronic kidney disease, HD: hemodialysis, DDS; dialysis disequilibrium syndrome, AKI: acute kidney injury, RRT: renal replacement therapy

Table 1.

Baseline characteristics of patients analyzed in the development cohort and the validation cohort

Patients	Development Cohort (n = 42)	Validation Cohort (n = 30)
Age (years)	43 (28.7–60.2)	58 (48.7–72)
Female	24 (63%)	12 (26%)
Weight (kg)	63 ± 17.72	70 (65–80)
Height (cm)	159 ± 10	165 (155–169)
CKD	42 (100%)	23 (76%)
- Total patients at risk of DDS	NA	23 (76%)
- Risk of DDS alone	NA	12 (40%)
- Risk of DDS & Severe hyponatremia	NA	6 (20%)
- Risk of DDS & Severe hyperkalemia	NA	5 (16%)
AKI	NA	7 (23%)
- AKI & Severe hyponatremia	NA	1 (3.3%)
- AKI & Severe hyperkalemia	NA	4 (13%)
- AKI & Other indications of RRT	NA	2 (6.6%)

Mean ± SD, median (IQR) kg: kilograms, cm: centimeters, CKD: chronic kidney disease, AKI: Acute kidney injury, DDS: disequilibrium dialysis syndrome, NA: not applicable

Table 2.

Treatment characteristics and baseline biochemical values of the development cohort and the validation cohort

Treatments	Development Cohort (n = 280)	Validation Cohort (n = 44)
Treatment time (hours)	4 (3.5–4)	1.5 (1.4–2)
QB (ml/min)	350 (300–380)	200 (170–250)
QD (ml/min)	800 (500–800)	350 (300–350)
Total UF (liters)	2 (1.3–2.9)	0.9 (0–2)
Dialysate Na (meq/L)	134 ± 3.5	138 ± 3.9
Dialysate K (meq/L)	2 ± 0	1.8 ± 0.8
Predialysis BUN (mg/dl)	69 (56.4–83.9)	166 (148–195)
Predialysis Na (meq/L)	NA	138 (135–141)
Predialysis K (meq/L)	NA	6.5 (4–7)

Mean ± SD, median (IQR). QB: Blood flow, QD; dialysate flow, UF; ultrafiltration, Na: sodium, K: potassium, BUN; blood urea nitrogen, meq/L; milliequivalents per liter, mg/dl; milligrams per deciliter, NA: not applicable

Table 3.

Predicted and measured values in the development cohort and the validation cohort

Parameter	Development cohort (n = 280)		Validation cohort (n = 44)
	Solute solver®	Adequator app®	Measured values	Adequator app®
Kd (ml/min)	239 (216–260)	239 (214–261)	NA	162 (161–163)
Equilibrated Kt/V	1.64 (1.4–1.8)	NA	0.36 (0.28–0.43)	NA
Single Pool Kt/V	1.86 (1.65–2.07)	1.71 (1.5–1.8)	0.49 (0.38–0.57)	0.41 (0.34–0.50)
BUN RR (%)	80 (76–82)	81.9 (78–84)	34.5 ± 8.4	34.8 ± 6.9
Postdialysis BUN (mg/dl)	14 (10.5–17.4)	12.7 (9.8–17)	105 (87–139)	106 (91–128)
Postdialysis sodium (meq/L)	NA	NA	139 (136–141)	138 (135–140)
Postdialysis potassium (meq/L)	NA	NA	3.8 ± 0.77	3.9 ± 0.76

Mean ± SD, median (IQR). Kd: dialyzer clearance, ml/min: milliliters per minute, V; volume of distribution, T: time, RR: removal rate, BUN: blood urea nitrogen, mg/dl: milligrams per deciliter, meq/L: milliequivalents per liter, NA: not applicable

When the values in the development cohort were compared, a very strong correlation and agreement between the formal urea kinetic model (Solute solver®) and the developed method (Adequator app®) were observed for the Kd values, with an r value of 0.99 (95% CI: 0.99 to 0.99) and a bias of 0.23 ± 1.2 ml/min (95% CI: −2.17 to 2.63), demonstrating the accurate computation of the equations used in the proposed method.

When T and urea V were incorporated to predict Kt/V, RR and post-HD BUN, the correlations were as follows: r = 0.77 (95% CI: 0.72 to 0.81) when compared to equilibrated Kt/V, r = 0.74 (95% CI: 0.68 to 0.79) when compared to single pool Kt/V, r = 0.78 (95% CI: 0.73 to 0.82) for RR, and r = 0.88 (95% CI: 0.85 to 0.90) for post-HD BUN.

In terms of agreement, the bias for equilibrated Kt/V was 0.02 ± 0.19 (95% CI: −0.35 to 0.40), that for single pool Kt/V was −0.20 ± 0.22 (95% CI: −0.64 to 0.23), that for RR was 1.29 ± 3.7% (95% CI: −5.9 to 8.5), and that for post-HD BUN was −0.9 ± 2.6 mg/dL (95% CI: −6.14 to 4.34) (Fig. 2).

Fig. 2.

Development cohort correlation and agreement. panel A: Pearson correlation coefficient and bland- Altman plot for dialyzer urea clearance kd. panel B Pearson correlation coefficient and bland- Altman plot for Kt/V vs equilibrated KtV, panel C Pearson correlation coefficient and bland- Altman plot for Kt/V vs single pool KtV panel D Pearson correlation coefficient and bland- Altman plot for RR, panel E Pearson correlation coefficient and bland- Altman plot for posttherapy BUN. Kd: dialyzer clearance, ml/min: milliliters per minute, RR: reduction ratio

Validation cohort

The validation cohort consisted of 30 patients and 44 HD treatments (Fig. 1). The main characteristics of the patients, and HD treatments of the validation cohort are available in Tables 1 and 2, respectively. Among these patients, 23 patients (76%) had CKD, 7 (23%) had AKI, and all patients with CKD were at risk of DDS 23 (76%). Additionally, a total of 7 patients (23%) also had severe hyponatremia, 9 (30%) had severe hyperkalemia prior to HD, and 2 patients with AKI had other indications for renal replacement therapy (RRT), such as volume overload or metabolic acidosis (Table 1). For the measured values, the BUN RR was 34.5 ± 8.4%, the post-HD BUN was 105 (87–139) mg/dl, the post-HD sodium was 139 (136–141) meq/L, and the post-HD potassium was 3.8 ± 0.77 meq/L. Calculated single pool Kt/V was 0.49 (0.38–0.57), and equilibrated Kt/V was 0.36 (0.28–0.43) (Table 3).

In the validation cohort, excellent correlation and agreement were observed between the measured and predicted values using the Adequator app®. Post-HD BUN (mg/dL) demonstrated a strong correlation, with an r value of 0.96 (95% CI: 0.93 to 0.98) and a bias of 0.4 ± 9.4 mg/dL (95% CI: −19 to 18.1). The post-HD sodium level had an r value of 0.96 (95% CI: 0.94 to 0.98) and a bias of −0.6 mEq/L ±1.6 (95% CI: −3.8 to 2.5). Post-HD potassium showed a slightly lower correlation, with an r value of 0.84 (95% CI: 0.73 to 0.91) and a bias of 0.17 ± 0.4 mEq/L (95% CI: −0.6 to 0.9). Finally, the BUN RR had an r value of 0.77 (95% CI: 0.61 to 0.86) and a bias of 0.18 ± 5.3% (95% CI: −10.3 to 10.7) (Fig. 3).

Fig. 3.

Validation cohort correlation and agreement. panel A: Pearson correlation coefficient and bland- Altman plot for posttherapy BUN, panel B: Pearson correlation coefficient and bland- Altman plot posttherapy sodium panel C: Pearson correlation coefficient and bland- Altman plot for posttherapy potassium, panel D: Pearson correlation coefficient and bland- Altman plot for RR%. RR: reduction ratio, BUN: blood urea nitrogen, mg/dl: milligrams per deciliter, meq/L: milliequivalents per liter

Among the 23 patients and 39 treatments identified as being at risk of DDS, the mean pre-HD BUN was 185 ± 17 mg/dL, and 36 of the 39 treatments (92%) successfully met the RR target established before HD. Notably, severe DDS was not observed in any patients, and only 3 patients (12%) reported mild neurological symptoms such as nausea and dizziness. Additionally, among patients with severe pre-HD hyponatremia, the mean sodium level was 117 ± 3 mEq/L, and all 7 achieved the post-HD goal of a sodium increase of < 6 mEq/L. The 9 patients with severe hyperkalemia had a mean pre-HD potassium of 7.5 ± 1.5, and all of them achieved a normal potassium level post-HD.

Discussion

In this study, we identified a strong correlation between the post-HD measured values of patients undergoing urgent HD and the values predicted by the proposed method (Adequator app®). These findings support the safety and accuracy of the model in acute clinical settings, highlighting its potential utility in guiding dialysis adequacy assessments under urgent treatment conditions.

The concept of ionic dialysance approximating urea clearance to predict solute values has been effectively applied in continuous renal replacement therapies (CRRTs), as urea clearance can be readily determined from the volume of the CRRT effluent [17, 18]. However, in the context of intermittent HD, predicting urea clearance and consequently post-HD solute values can be challenging since urea clearance is influenced by multiple factors, such as Qb, Qd and membrane KoA. Although methods such as those proposed by Wendland & Kaplan (2012) [6] have successfully estimated sodium changes in cases of severe hyponatremia, these approaches address primarily sodium transfer and redistribution at a very low Qb without comprehensively considering overall solute clearance.

We propose estimating urea clearance from the calculated Kd, building on previously developed equations that consider membrane characteristics, Qb, and Qd. These equations originate from the logarithmic transmembrane decline of water-soluble solutes described by Michaels [19] and were later refined to account for factors such as the influence of the blood pump on Qb [20, 21], the impact of Qd on KoA [22], and the estimation of in vivo membrane KoA [10]. Eqs. (1-4).

Although these individual concepts are not new, no previous models have combined them into a unified model to predict post HD solute values, particularly in acute care settings. The integrative stepwise process distinguishes our model from the known individual concepts. Although Daugirdas (2016) proposed a similar method using a web-based calculator (Solute Solver: What If) to predict values, it is specifically designed for chronic HD, incorporating weekly schedules, residual kidney function, and incremental HD planning strategies, with no prediction of electrolytes [23]. In contrast, our approach focuses on predicting urea and electrolytes in acute HD indications and settings.

In this study, the proposed method demonstrated accuracy across both cohorts. In the retrospective development cohort, the primary objective was to validate the accurate computation of the Kd equation and assess the correlation between Kt/V and RR with a formal urea kinetic model [11] by incorporating T and measured anthropometric values. The calculation of Kd and RR strongly agreed with the Kd and RR derived from the formal urea kinetic model. The predicted Kt/V showed better agreement with the equilibrated Kt/V than with the single pool Kt/V. This observation can be primarily attributed to the tendency of anthropometric formulas to overestimate V in dialysis patients [24, 25]. As a result, the solved Kt/V which conceptually aligns with a single pool Kt/V is underestimated, making it more consistent with the equilibrated Kt/V than with the single pool Kt/V derived from formal kinetic modeling. Additionally, the predicted post-HD BUN closely matched the measured post-HD BUN, indicating excellent correlation and agreement.

For the external prospective validation cohort, the goal was to design specific prescriptions for patients with acute HD indications undergoing their first or second session. Most participants had end-stage CKD or AKI and were receiving HD for uremic syndrome, hyperkalemia, fluid overload, or metabolic acidosis. For patients at risk of DDS, the prescription aimed to achieve a BUN RR of 40% or lower. Additionally, patients with severe hyponatremia and urgent dialysis indications were treated with a prescription designed to limit sodium changes to no more than 6 mEq/L, with both strategies being successful. The predicted post-HD BUN and sodium levels demonstrated excellent correlation and agreement with the measured values. However, the correlation for potassium was moderate, likely due to its larger V and rapid redistribution. A study that analyzed the effect of potassium removal during HD on the plasma potassium concentration in 20 samples from 7 anuric patients revealed that the decrease in plasma potassium primarily depends on the pre-HD concentration. The authors concluded that predicting post-HD plasma potassium levels cannot rely on a single compartment model and established two different potassium volumes of distribution (kV) by using potassium levels immediately after HD, obtaining a mean kV of 0.73 liters/kg of body weight and a mean kV of 1.58 ± 0.74 liters/kg of body weight with potassium values measured 5 hours post-HD [26]. Based on these observations, we propose the use of a correction factor of 1.6 liters/kg of body weight to estimate kV and apply it in Eq. (7) to predict 5-hour post-HD potassium levels, accounting for a two-compartment model. This approach was implemented as a 2-pool post-HD potassium output feature in the Adequator app® calculator.

Clinical implications

The assessment of delivered urea clearance through kinetic models has diminished in relevance owing to the emergence of broader approaches to HD adequacy that extend beyond small solute clearance [27]. Nonetheless, estimating urea clearance and predicting post-HD values are relevant for patients with urgent HD needs, where the risk of overcorrection or the need for precise prescriptions is critical. The present study demonstrated that the proposed method provides accurate predictions and ensures the safety of planned prescriptions in acute settings.

Limitations

The primary limitation of the proposed model is that its predictions rely on immediate post-HD measurements, corresponding to a single-pool model. As a result, the accuracy of the presented outcomes is ensured only when samples are collected following the standardized method described by Daugirdas [12]. While this may be perceived as a constraint, the standardization of measurement enhances the calculator’s reliability in a specific context, providing a considerable margin of safety when samples are obtained using this approach. Additionally, the model was developed and validated using mainly high efficiency dialyzers. Its performance with low efficiency dialyzers has not been fully assessed, and although the model incorporates KoA as an input variable, further validation may be required in settings where such dialyzers are used.

Conclusions

This study demonstrated that post-HD urea and electrolyte levels can be accurately predicted using a simplified, prescription-based model. The proposed approach strongly agreed with formal urea kinetic modeling and performed well in acute HD settings, particularly for patients at risk of DDS, severe hyponatremia, or severe hyperkalemia. By enabling rapid and automated predictions, this tool may enhance clinical decision-making in urgent scenarios where precise electrolyte and urea control is critical. Future research should explore its integration into routine practice and its impact on patient outcomes in diverse clinical settings.

Abbreviations

HD

Hemodialysis

DDS

Dialysis disequilibrium syndrome

Kd

Dialyzer clearance

rQb

Real blood flow

Qd

Dialysate flow

KoA

In vivo membrane permeability-area coefficient

T

Time

V

Volume of distribution

AVFs

Arteriovenous fistulas

AVGs

Arteriovenous grafts

Qf

Ultrafiltration rate

BUN

Blood urea nitrogen

CKD

Chronic kidney disease

AKI

Acute kidney injury

Cd

Dialysate concentrations

C2

Post-HD values

C1

Pre-HD levels

RR

Reduction ratio

TBW

Total body water

SDs

Standard deviations

UF

Ultrafiltration

RRT

Renal replacement therapy

CRRTs

Continuous renal replacement therapies

kV

Potassium volume of distribution

Author contributions

P.G., O.V.V. and R.G.H. conceived the design and experiments. P.G. developed the model, programmed and designed the calculator. C.R.B., P.G. and R.G.H. collected the data and analyzed the results. O.V.V., N.T.C. and AJ.M. R critically revised the manuscript. All the authors revised the manuscript and agreed to be published.

Funding

All funding was covered by the Instituto Nacional de Nutrición y Ciencias Médicas Salvador Zubiran and Centro Médico ISSEMYM Ecatepec.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubiran, México (NMM-3034–1919–1) and by the Centro Medico ISSEMYM at Ecatepec, Mexico (PICME- 2024–04). Written informed consent was obtained from all patients from the prospective cohort. For the retrospective cohort, the requirement for informed consent was waived by the ethics committees due to the retrospective nature of the study and the use of anonymized data.

Consent for publication

Not applicable.

Competing interests

This study began before AJ.M. R worked at Vantive Healthcare. He is currently an employee of Vantive Healthcare. P.G., C.R.B., R.G.H., N.T.C., and O.V.V. have no conflicts of interest to report.