Introduction: Spot urinary sodium emerged as a useful parameter for assessing decongestion in patients with congestive heart failure (CHF). Growing evidence endorses the therapeutic role of continuous ambulatory peritoneal dialysis (CAPD) in patients with refractory CHF and kidney disease. We aimed to examine the long-term trajectory of urinary, peritoneal, and total (urinary plus peritoneal) sodium removal in a cohort of patients with refractory CHF enrolled in a CAPD program. Additionally, we explored whether sodium removal was associated with the risk of long-term mortality and episodes of worsening heart failure (WHF). Methods: We included 66 ambulatory patients with refractory CHF enrolled in a CAPD program in a single teaching center. 24-h peritoneal, urinary, and total sodium elimination were analyzed at baseline and after CAPD initiation. Its trajectories over time were calculated using joint modeling of longitudinal and survival data. Within the framework of joint frailty models for recurrent and terminal events, we estimated its prognostic effect on recurrent episodes of WHF. Results: At the time of enrollment, the mean age and estimated glomerular filtration rate were 72.8 ± 8.4 years and 28.5 ± 14.3 mL/min/1.73 m2, respectively. The median urinary sodium at baseline was 2.34 g/day (1.40–3.55). At a median (p25%–p75%) follow-up of 2.93 (1.93–3.72) years, we registered 0.28 deaths and 0.24 episodes of WHF per 1 person-year. Compared to baseline (urinary), CAPD led to increased sodium excretion (urinary plus dialyzed) since the first follow-up visit (p < 0.001). Over the follow-up, repeated measurements of total sodium removal were associated with a lower risk of death and episodes of WHF. Conclusions: CAPD increased sodium removal in patients with refractory CHF. Elevated sodium removal identified those patients with a lower risk of death and episodes of WHF.

Fluid overload is responsible for most symptoms and signs of heart failure (HF), especially during decompensations [1‒3]. Diuretics remain the mainstay of pharmacological depletive armamentarium [2‒4]. However, in some cases, traditional diuretic approaches become ineffective, a fact that overshadows the prognosis [2‒4].

Isolated ultrafiltration therapy emerged as a potentially useful strategy for tackling fluid overload in HF [2, 3]. However, the results in most cases are heterogeneous and sometimes discouraging [5]. In this regard, continuous ambulatory peritoneal dialysis (CAPD) has emerged as a promising alternative for decongestion in patients with refractory congestive HF [6‒11]. However, beyond increasing water elimination, the pathophysiological mechanisms behind the benefit of CAPD in refractory congestive HF remain to be elucidated. The amount of sodium removed seems to be a key issue in evaluating the response to depletive therapy in HF patients [4], as urinary sodium removal after the intensification of diuretic therapy identifies those with better clinical response [4]. In line with this statement, we hypothesize that in patients with congestive refractory HF included in a CAPD program, the greater sodium clearance (dialyzed and urinary) will identify those with a better response in terms of mortality and episodes of worsening heart failure (WHF).

In this work, we aimed to evaluate the long-term trajectory of urinary, peritoneal, and total (urinary plus peritoneal) sodium removal in a cohort of patients with congestive refractory HF enrolled in a CAPD program. Additionally, we explored whether the course of sodium elimination was associated with the risk of long-term mortality and total WHF episodes.

Study Population

This is a retrospective study that included patients with refractory congestive HF enrolled in an ultrafiltration program with CAPD in a single teaching center (Hospital Clínico Universitario de Valencia, Spain) from 11th May 2008 to 8th Oct 2021. The eligibility criteria for considering the inclusion of patients with HF in a CAPD program are published elsewhere [6, 7] and included all the following: ambulatory New York Heart Association (NYHA) class III/IV, at least two HF hospitalizations in the last 6 months, persistent congestion despite diuretic therapy maximization, and estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2. The initial cohort included 97 patients (1,315 visits). Patients included in CAPD program before 2014 and those with less than 3 months in CAPD were excluded due to lack of data on sodium excretion (n = 31 patients). The final sample comprises 66 patients (839 visits), with a median (interquartile range [IQR]) number of visits per patient of 12 (5–17) and a range of 1–36 visits.

All participants provided written informed consent. This study protocol was reviewed and approved by INCLIVA (Instituto de Investigación Sanitaria, Valencia, Spain). The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki (revised in 1983), as reflected by an a priori approval by the institution’s human research committee. Patients were not involved in the design and conduct of this research.

Characteristics of the CAPD and Scheme of Follow-Up

CAPD is a renal replacement therapy consisting of the introduction of an osmotic solution into the abdomen through a catheter. The patient’s peritoneum is used as a membrane across which excess fluid and waste products are filtered from the blood. We started a CAPD program for patients with refractory HF with some degree of renal dysfunction in 2008. Our protocol followed current international guidelines regarding CAPD-related infection [12] and peritoneal access [13]. The CAPD program consisted of 1–4 times/day exchange with dialyzate solution (1.36–2.27% glucose or icodextrin) according to the nephrologist criteria and based on clinical status and kidney function, but not based on the exposures. Changes in the prescription were made, if necessary, to achieve a correct state of hydration and based on residual renal function. The number of dialyzate solution exchanges was increased in case of residual renal function decay. Three types of peritoneal dialysis fluids were used: Balance® (glucose) (134 mmol/L), Physioneal® (glucose) (132 mmol/L), and Extraneal® (icodextrin) (133 mmol/L). A baseline peritoneal equilibrium test (PET) was performed in all patients, and all of them were average transporters (D/P creatinine between 0.5 and 0.81).

The medical treatment was individualized and in line with established guidelines. All patients were instructed to restrict dietary sodium intake. All patients were routinely visited by a specialized cardio-renal team at least every 2 months. Additional visits were performed in case of decompensation or complications. Clinical information was collected from all patients. The initial CAPD regimen (type and volume of dialysis fluid infused) was also recorded.

The 24-h peritoneal sodium removal is the result of subtracting the drained sodium (24-h drained volume multiplied by the sodium concentration per liter) minus the infused sodium (24-h infused volume multiplied by the sodium concentration per liter). For the total urinary sodium balance, we added the peritoneal removal of sodium and the 24-h natriuresis of each patient. We recorded the extracted sodium variables (total, peritoneal, and urinary) in grams/24 h.

Endpoints

Each patient was followed up until the end of the analyzed period (October 2021) or the withdrawal from CAPD treatment. The endpoint was the long-term trajectory of urinary, peritoneal, and total sodium following CAPD onset. Additionally, we explored whether the course of sodium removal was associated with the risk of long-term all-cause mortality and repeated episodes of WHF. WHF was defined as new-onset or worsening (gradual or rapid) signs and symptoms of HF that require urgent therapy and result in hospitalization or administration of parenteral diuretic therapy. Data about survival and episodes of WHF were extracted from electronic medical records and follow-up ambulatory visits. Personnel in charge of endpoint adjudication were blinded to the exposures (extracted sodium).

Statistical Analysis

Continuous variables were expressed as mean ± standard deviation or median (IQR), whenever appropriate. Discrete variables were summarized as percentages.

The sodium trajectories over time were calculated using joint modeling of longitudinal and survival data (JM). Within the framework of joint frailty models for recurrent and terminal events, we estimated the prognostic effect of longitudinal values of urinary sodium, dialyzed sodium, and their combination on recurrent hospital readmissions. Because the mortality rate is extremely high in this population, we accounted for informative censoring by including a mortality submodel simultaneously with the HF-readmission submodel. These two outcome models were linked through a random effect and constraining the coefficient of the recurrent outcome random effect to be one. The baseline hazard function of each outcome was fitted using Weibull distributional parameterization. By design, our longitudinal exposure variables were modeled without transformation (linearly). In a separate regression model, we included a time-dependent effect by including the interaction of each longitudinal exposure with time (years). In doing so, we aimed to determine if there were time-driven variations of the predictive effects of the readmission and mortality outcomes and, thus, to corroborate or reject the proportionality assumption over time. The final model for both outcome processes (WHF and mortality) included age (years), gender, prior history of acute myocardial infarction, cumulative time on CAPD (years), the updated value at each time point of the complementary longitudinal exposure (urinary sodium if the exposure tested as dependent variable was peritoneal sodium, and vice versa), time between initiation of CAPD and first exposure measured, and baseline values of hemoglobin, eGFR, serum sodium, and amino-terminal pro-B-type natriuretic peptide.

A 2-sided p value of <0.05 was the threshold used for significance in all analyses. Stata 17.0 (StataCorp. 2021. Stata Statistical Software: Release 17. College Station, TX: StataCorp LLC) was used for data clean-up and preparation and for analyses. Within Stata, the analyses were implemented with “merlin” [14] and “stjm” packages [15].

Baseline Characteristics

The mean age of the sample was 72.8 ± 8.4 years, and 17 (25.8%) were women. The mean left ventricle ejection fraction was 44.6 ± 14.4%. At CAPD initiation, the mean (standard deviation) eGFR was 28.5 ± 14.3 mL/min/1.73 m2. The median (IQR) of amino-terminal pro-B-type natriuretic peptide was 5,153 pg/mL (2,803–12,876). The median urinary sodium at baseline was 2.34 g/day (1.40–3.55). All patients received oral furosemide, with a mean dose of 120 mg/day (80–120); 47.0% of patients were also on treatment with thiazides, and 12.1% on acetazolamide. Baseline characteristics are presented in Table 1. Regarding the CAPD regimens at the beginning of the CAPD program, 23.0% of the patients started with a single 2-L exchange, and 34.8% used icodextrin (alone or with glucose).

Table 1.

Baseline characteristics

Demographics and medical historyCAPD cohort (N = 66)
Age, years 72.8±8.4 
Gender (female), n (%) 17 (25.8) 
Height, cm 167.5±8.9 
Weight, kg 80.2±12.2 
Body mass index, kg/m2 28.7±4.8 
Obesity at baseline, n (%) 6 (9.1) 
Hypertension, n (%) 54 (81.8) 
Dyslipidemia, n (%) 49 (74.2) 
Diabetes, n (%) 40 (60.6) 
Complicated diabetes, n (%) 10 (15.1) 
Current smoker, n (%) 5 (7.6) 
Past smoker, n (%) 21 (31.8) 
History of acute myocardial infarction, n (%) 26 (39.4) 
History of acute pulmonary edema, n (%) 11 (16.7) 
History of stroke, n (%) 5 (7.6) 
Chronic obstructive coronary disease, n (%) 18 (27.3) 
Chronic renal insufficiency, n (%) 58 (87.9) 
Age-adjusted Charlson Comorbidity Index, points 7.9±2.1 
HF etiology, n (%) 
 Coronary artery disease 26 (39.4) 
 Valve heart disease 11 (17.2) 
 Hypertension 13 (19.7) 
 Cardiac amyloidosis 4 (6.3) 
 Nonischemic dilated cardiomyopathy 8 (12.5) 
 Constrictive pericarditis 4 (6.3) 
Pacemaker, n (%) 11 (16.7) 
Implantable cardioverter defibrillator, n (%) 16 (24.2) 
Cardiac resynchronization therapy, n (%) 9 (13.6) 
Atrial fibrillation/flutter at baseline, n (%) 41 (62.1) 
Bundle branch block, n (%) 26 (39.4) 
NYHA class, n (%)  
 III 47 (71.2) 
 IV 19 (28.8) 
Vital signs 
 Systolic blood pressure, mm Hg 121.8±21.4 
 Diastolic blood pressure, mm Hg 70.6±11.9 
Echocardiography 
 LVEF, % 44.6±14.4 
 LVEF <40%, n (%) 29 (43.9) 
 LV diastolic diameter, mm 57.8±10.0 
 LV systolic diameter, mm 44.4±11.6 
 LV septum wall thickness, mm 11.4±2.4 
 LV posterior wall thickness, mm 11.2±2.3 
 Left atrial dimension, mm 47.6±8.3 
 Systolic pulmonary artery pressure1, mm Hg 50.9 (18.6) 
 TAPSE2, mm* 17.0 (15.0–19.0) 
Tricuspid regurgitation severity, n (%) 
 I 20 (30.3) 
 II 9 (13.6) 
 III 12 (18.2) 
 IV 9 (13.6) 
Laboratory 
 Hemoglobin, g/dL 12.1±1.1 
 Hematocrit, % 35.7±5.3 
 Urea, mg/dL 119.6±55.0 
 Creatinine, mg/dL 2.34±1.07 
 eGFR, mL/min/1.73 m2 at baseline 28.5±14.3 
 Serum sodium, mEq/L 136.9±3.9 
 Serum potassium, mEq/L 4.2±0.7 
 NT-proBNP, pg/mL* 5,153 (2803–12876) 
 CA125, U/mL* 35 (18–85.5) 
 Urinary sodium, g/day* 2.34 (1.40–3.55) 
Treatment 
 Furosemide dose, mg/day* 120 (80–120) 
 Thiazides, n (%) 31 (47.0) 
 Acetazolamide, n (%) 8 (12.1) 
 RAASi, n (%) 24 (36.4) 
 Beta-blockers, n (%) 50 (75.7) 
 MRA, n (%) 30 (45.4) 
 Statins, n (%) 51 (77.3) 
 Antiplatelet, n (%) 28 (42.4) 
 Oral anticoagulants, n (%) 40 (60.6) 
 Ivabradine, n (%) 6 (9.1) 
 Digoxin, n (%) 2 (3.0) 
CAPD regimen at baseline 
 Volume of peritoneal fluid infused, mL/day 4,000 (750) 
Type of dialysis fluid used 
 Only glucose, n (%) 43 (65.1) 
 Only icodextrin, n (%) 14 (21.2) 
 Icodextrin plus glucose, n (%) 9 (13.6) 
Demographics and medical historyCAPD cohort (N = 66)
Age, years 72.8±8.4 
Gender (female), n (%) 17 (25.8) 
Height, cm 167.5±8.9 
Weight, kg 80.2±12.2 
Body mass index, kg/m2 28.7±4.8 
Obesity at baseline, n (%) 6 (9.1) 
Hypertension, n (%) 54 (81.8) 
Dyslipidemia, n (%) 49 (74.2) 
Diabetes, n (%) 40 (60.6) 
Complicated diabetes, n (%) 10 (15.1) 
Current smoker, n (%) 5 (7.6) 
Past smoker, n (%) 21 (31.8) 
History of acute myocardial infarction, n (%) 26 (39.4) 
History of acute pulmonary edema, n (%) 11 (16.7) 
History of stroke, n (%) 5 (7.6) 
Chronic obstructive coronary disease, n (%) 18 (27.3) 
Chronic renal insufficiency, n (%) 58 (87.9) 
Age-adjusted Charlson Comorbidity Index, points 7.9±2.1 
HF etiology, n (%) 
 Coronary artery disease 26 (39.4) 
 Valve heart disease 11 (17.2) 
 Hypertension 13 (19.7) 
 Cardiac amyloidosis 4 (6.3) 
 Nonischemic dilated cardiomyopathy 8 (12.5) 
 Constrictive pericarditis 4 (6.3) 
Pacemaker, n (%) 11 (16.7) 
Implantable cardioverter defibrillator, n (%) 16 (24.2) 
Cardiac resynchronization therapy, n (%) 9 (13.6) 
Atrial fibrillation/flutter at baseline, n (%) 41 (62.1) 
Bundle branch block, n (%) 26 (39.4) 
NYHA class, n (%)  
 III 47 (71.2) 
 IV 19 (28.8) 
Vital signs 
 Systolic blood pressure, mm Hg 121.8±21.4 
 Diastolic blood pressure, mm Hg 70.6±11.9 
Echocardiography 
 LVEF, % 44.6±14.4 
 LVEF <40%, n (%) 29 (43.9) 
 LV diastolic diameter, mm 57.8±10.0 
 LV systolic diameter, mm 44.4±11.6 
 LV septum wall thickness, mm 11.4±2.4 
 LV posterior wall thickness, mm 11.2±2.3 
 Left atrial dimension, mm 47.6±8.3 
 Systolic pulmonary artery pressure1, mm Hg 50.9 (18.6) 
 TAPSE2, mm* 17.0 (15.0–19.0) 
Tricuspid regurgitation severity, n (%) 
 I 20 (30.3) 
 II 9 (13.6) 
 III 12 (18.2) 
 IV 9 (13.6) 
Laboratory 
 Hemoglobin, g/dL 12.1±1.1 
 Hematocrit, % 35.7±5.3 
 Urea, mg/dL 119.6±55.0 
 Creatinine, mg/dL 2.34±1.07 
 eGFR, mL/min/1.73 m2 at baseline 28.5±14.3 
 Serum sodium, mEq/L 136.9±3.9 
 Serum potassium, mEq/L 4.2±0.7 
 NT-proBNP, pg/mL* 5,153 (2803–12876) 
 CA125, U/mL* 35 (18–85.5) 
 Urinary sodium, g/day* 2.34 (1.40–3.55) 
Treatment 
 Furosemide dose, mg/day* 120 (80–120) 
 Thiazides, n (%) 31 (47.0) 
 Acetazolamide, n (%) 8 (12.1) 
 RAASi, n (%) 24 (36.4) 
 Beta-blockers, n (%) 50 (75.7) 
 MRA, n (%) 30 (45.4) 
 Statins, n (%) 51 (77.3) 
 Antiplatelet, n (%) 28 (42.4) 
 Oral anticoagulants, n (%) 40 (60.6) 
 Ivabradine, n (%) 6 (9.1) 
 Digoxin, n (%) 2 (3.0) 
CAPD regimen at baseline 
 Volume of peritoneal fluid infused, mL/day 4,000 (750) 
Type of dialysis fluid used 
 Only glucose, n (%) 43 (65.1) 
 Only icodextrin, n (%) 14 (21.2) 
 Icodextrin plus glucose, n (%) 9 (13.6) 

CAPD, continuous ambulatory peritoneal dialysis; CA125, carbohydrate antigen 125; eGFR, estimated glomerular filtration rate; LV, left ventricle; LVEF, left ventricle ejection fraction; MRA, mineralocorticoid receptor antagonists; NT-proBNP, amino-terminal pro-brain natriuretic peptide; NYHA, New York Heart Association; RAASi, renin-angiotensin-aldosterone system inhibitors; TAPSE, tricuspid annular plane systolic excursion.

Continuous variables are expressed as mean ± standard deviation, unless otherwise specified.

*Variable expressed as median (interquartile interval).

1Data available in 53 patients.

2Data available in 51 patients.

Mortality and Recurrent HF Hospitalizations during Follow-Up

At a median follow-up of 2.93 (1.93–3.72) years, 43 (65.2%) patients died (0.28 deaths per 1 person-year) (95% CI = 0.21–0.38; p < 0.001). Likewise, 16 patients experienced repeated episodes of WHF with a rate of 0.24 events per 1 person-year (95% CI = 0.17–0.34; p < 0.001).

Sodium Removal: Trajectory over Time

Compared to baseline sodium excretion (only urinary sodium) (2.64 ± 1.26 g/day), at 1, 6, 12, and 24 months after CAPD initiation, total excretion of sodium increased by 33.7%, 31.9%, 30.4%, and 27.2%, respectively (p < 0.001). After the initial increase, urinary sodium removal linearly decreased during follow-up (p < 0.001), as shown in Figure 1a. Conversely, peritoneal sodium removal linearly increased during follow-up (p = 0.002) (shown in Fig. 1b). Overall, total sodium removal (urinary plus dialyzed) linearly decreased during follow-up (p = 0.001) (shown in Fig. 1c).

Fig. 1.

Predicted longitudinal trajectories during follow-up. a Urinary sodium. b Dialyzed sodium. c Total sodium.

Fig. 1.

Predicted longitudinal trajectories during follow-up. a Urinary sodium. b Dialyzed sodium. c Total sodium.

Close modal

The type of dialysis solution at baseline (glucose vs. icodextrin) did not have a differential effect on urinary (p value for interaction = 0.905), dialyzed (p value for interaction = 0.600), and total sodium removal (p value for interaction = 0.722) as shown in online Supplementary Figure 1 (for all online suppl. material, see https://doi.org/10.1159/000531631).

Longitudinal Sodium Removal and Adverse Clinical Events

Urinary Sodium

Urinary sodium was inversely and linearly related to the risk of all-cause mortality. Indeed, per increase in 1 g/day, the risk of death decreased by 39% (HR: 0.61, 95% CI = 0.45–0.88; p = 0.008). Similarly, the relationship between urinary sodium and risk of recurrent WHF was inversely related, where increments in 1 g/day were associated with a decrease in WHF risk by 34% (HR: 0.66, 95% CI = 0.45–0.96; p = 0.031). We found a time-dependent effect for mortality (p value for time-dependent interactions = 0.021). Thus, the predictive value of urinary sodium was nonsignificant during the first year. However, it became significant and increased the associated effect at longer follow-up (shown in Fig. 2a). We did not find evidence for a differential effect over time for the WHF endpoint (p value for time-dependent interaction = 0.883), as shown in Figure 2b.

Fig. 2.

Predictive value of urinary sodium for mortality and WHF. a All-cause mortality. b Worsening renal failure.

Fig. 2.

Predictive value of urinary sodium for mortality and WHF. a All-cause mortality. b Worsening renal failure.

Close modal

Dialyzed Sodium

Overall, the amount of extracted sodium by the peritoneal dialysis was inversely related to the risk of mortality (HR: 0.59, 95% CI = 0.37–0.95; p = 0.032). The greater the extraction of sodium, the greater the reduction in the risk (per increase in 1 g/day, the risk of mortality was reduced by 61%). Dialyzed sodium was not significantly related to the risk of recurrent episodes of WHF (HR: 0.83, 95% CI = 0.64–1.07; p = 0.157). We did not find evidence for a differential effect over time for both endpoints: mortality (p value for time-dependent interaction = 0.179) and WHF hospitalization (p value for time-dependent interaction = 0.812), as shown in Figure 3. Thus, we may infer that the protective effect of dialyzed sodium was found during the entire follow-up.

Fig. 3.

Predictive value of dialyzed sodium for mortality and WHF. a All-cause mortality. b Worsening renal failure.

Fig. 3.

Predictive value of dialyzed sodium for mortality and WHF. a All-cause mortality. b Worsening renal failure.

Close modal

Total Sodium

Overall, total extracted sodium (dialyzed plus urinary) was significantly associated with a lower risk of all-cause mortality (HR: 0.61; CI 95% = 0.43–0.87; p = 0.006) and recurrent WHF episodes (HR: 0.78; CI 95% = 0.64–0.96; p = 0.017). This relationship was linear and not time-dependent (p value for interaction with time: 0.450 and 0.654 for mortality and WHF, respectively) (shown in Fig. 4).

Fig. 4.

Predictive value of total sodium (urinary plus dialyzed) for mortality and WHF. a All-cause mortality. b Worsening renal failure.

Fig. 4.

Predictive value of total sodium (urinary plus dialyzed) for mortality and WHF. a All-cause mortality. b Worsening renal failure.

Close modal
Fig. 5.

Value of peritoneal and urinary sodium levels for monitoring the course of the disease in patients with refractory congestive heart failure included in an ambulatory peritoneal dialysis program.

Fig. 5.

Value of peritoneal and urinary sodium levels for monitoring the course of the disease in patients with refractory congestive heart failure included in an ambulatory peritoneal dialysis program.

Close modal

In this retrospective analysis that included 66 patients with congestive HF enrolled in a CAPD program because of diuretic-refractory congestion, we found a significant increase in 24-h sodium removal (urinary plus dialyzed sodium) compared to baseline (only urinary sodium). Greater sodium removal identified a patient with a lower risk of long-term mortality and recurrent WHF (Central illustration). To the best of our knowledge, this is the first study showing the utility of 24-h urinary, dialyzed, and total sodium removal for monitoring the response to CAPD in patients with refractory congestive HF. Current findings endorse the clinical utility of sodium excretion assessment in patients with refractory congestive HF enrolled in a CAPD program.

CAPD in Refractory Congestive HF

Fluid overload explains most of the symptoms and signs of HF and is responsible for most of the decompensations [1‒3]. Additionally, there is some evidence linking congestion itself with the progression of the disease [3, 16]. Diuretics are typically the first line of treatment for fluid overload [2, 4]. However, as the disease advances, the odds of diuretic resistance increase [2‒4]. In this latter scenario, ultrafiltration therapies constitute a therapeutic alternative [2, 4, 5, 17]. Along this line, CAPD emerged as a useful ultrafiltration alternative. CAPD offers advantages over other ultrafiltration therapies, such as ambulatory and long-term treatment, preservation of residual kidney function, slow ultrafiltration targeting intravascular and tissue congestion avoiding excessive intravascular depletion, and activation of the renin-angiotensin-aldosterone system [6‒11]. In patients with congestive HF, CAPD has been associated with an acceptable safety profile, lower risk of readmissions, improved quality of life, and functional capacity [6‒10]. Compared to a control group after a propensity matching adjustment, patients with advanced HF and renal dysfunction enrolled in a CAPD program showed a meaningful reduction in the risk of the composite of death or HF hospitalization (HR = 0.32; 95% CI = 0.17–0.61; p = 0.001) [7]. Additionally, in this context, Sanchez et al. [18] reported that CAPD led to a four-fold decrease in annual costs in comparison with the standard therapy. The mechanisms behind the effectiveness of this treatment remain not fully elucidated but appear multiple. Among them, the removal of water, sodium, and other substances such as cytokines, uremic toxins, and even light-chain amyloid protein seem highly attractive [8, 10, 11, 19, 20].

Sodium Excretion following CAPD Initiation in Congestive HF

Urinary sodium excretion has emerged as a useful biomarker in acute HF [4]. Low urinary sodium relates to inappropriate diuretic response and a higher risk of morbimortality [4, 21‒23]. Likewise, in congestive HF, serial assessment of urinary sodium has been shown to be useful for predicting HF decompensations [24]. In addition, current ESC HF guidelines and expert documents recommend tailoring the intensity of diuretic therapy according to the kinetic of spot urinary sodium in the first hours of HF decompensation [2, 4]. Peritoneal dialysis is generally considered an efficient method for removing sodium, and the transport of solutes and water in CAPD is dynamic and can vary within each treatment cycle [11].

In the current study, we expanded the clinical utility of sodium removal for monitoring response to CAPD in patients with congestive HF. Interestingly, after CAPD initiation, we found an early increase in sodium removal mainly because we added a new way to excrete sodium (dialyzed sodium). As expected, urinary sodium decreased over time, probably due to HF and chronic kidney disease progression. On the contrary, peritoneal sodium excretion increased over the first years, probably explained by modifications of CAPD ultrafiltration regimens based on clinical and kidney function status. At the opposite of traditional diuretics that mainly increase the water removal (hypotonic urine) [11], peritoneal ultrafiltrate yields a greater sodium removal (usually isotonic compared to plasma) [14]. Indeed, while loop diuretics generate hypotonic urine (about 60 mmol/L of sodium), the concentration of sodium in peritoneal ultrafiltrate has been reported to be as high as 126–134 mmol/L [24, 25]. This increased and maintained predominant sodium over water removal is an attractive property that may be behind this method’s differential and beneficial properties over other depletive alternatives [11]. Interestingly, in the current work, we found a progressive increase of dialyzed sodium during the first years. Thus, we speculate CAPD may lead to a mid-long-term reduction in tissue sodium, which would translate into a reduction in tissue water avidity and organ dysfunction, commonly found in patients with congestive HF.

Gaps in the Evidence and Further Lines of Research

First, the evidence supporting the role of CAPD in advanced congestive HF does not come from well-controlled dedicated studies. Thus, controlled and, ideally, randomized studies are warranted. Second, the timing and appropriate criteria for selection of the appropriate patient remain unclear. Third, the mechanisms behind the benefits require to be better described and quantified. Finally, the optimal peritoneal dialysis regimen for patients with congestive HF remains elusive. Some authors have proposed customized regimens for these patients including the use of icodextrin among others [11].

First, this is a retrospective analysis of a single teaching center in which several confounders have not been accounted for. Second, this is a small sample size with limited power. It is especially relevant when exploring subgroup analyses in which the limited power increases the risk of type II error. Third, we did not register the longitudinal changes in CAPD regimen, pharmacological therapy, and other surrogates of decongestion such as natriuretic peptides. Thus, we could not examine their contribution in sodium removal and adverse clinical events. Likewise, for this same reason, we could not accurately explore the contribution of the peritoneal equilibrium test to predict natriuresis. Finally, a lack of a control group is a major limitation.

CAPD increased sodium removal in patients with advanced HF. Serial total elevated sodium excretion (peritoneal plus urinary) identified those patients with a lower risk of death and WHF. Further prospective and controlled studies are warranted.

All participants provided written informed consent. This study protocol was reviewed and approved by INCLIVA (Instituto de Investigación Sanitaria, Valencia, Spain). The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki (revised in 1983), as reflected by an a priori approval by the institution’s human research committee. Patients were not involved in the design and conduct of this research.

Gema Miñana has received speaker fees from Abbott Vascular; Rafael de la Espriella reports personal fees from Astra Zeneca, Novartis, Boehringer Ingelheim, and Novo Nordisk; Jose Luis Górriz reports fees for participating in advisory boards and educational activities from Astra Zeneca, Boehringer Ingelheim, Novo Nordisk, Bayer, and Novartis; and Julio Núñez reports personal fees from Astra Zeneca, Novartis, Boehringer Ingelheim, Eli Lilly, ROVI, Novo Nordisk, and Vifor Pharma. The rest of authors have nothing to declare.

This work was supported in part by CIBER Cardiovascular (16/11/00420).

Gema Miñana and Miguel Gonzalez-Rico: conception or design of the work, data acquisition, analysis and data interpretation, drafting, and the final approval of the version to be published. Rafael de la Espriella, Daniel González-Sánchez, and Marco Montomoli: data acquisition, analysis and data interpretation, revision of the work for important intellectual content, and the final approval of the version to be published. Eduardo Núñez: analysis and data interpretation, drafting, revision of the work for important intellectual content, and the final approval of the version to be published. Agustín Fernández-Cisnal: data acquisition, drafting, revision of the work for important intellectual content, and the final approval of the version to be published. Sandra Villar: data acquisition, revision of the work for important intellectual content, and the final approval of the version to be published. Jose Luis Górriz: conception or design of the work, data acquisition, analysis and data interpretation, revision of the work for important intellectual content, and the final approval of the version to be published. Julio Núñez: conception or design of the work, analysis and data interpretation, drafting, and the final approval of the version to be published.

Additional Information

Gema Miñana and Miguel González-Rico contributed equally to this work.

All data generated or analyzed during this study are included in this article and its supplementary material files. Further inquiries can be directed to the corresponding author.

1.
Gheorghiade
M
,
Filippatos
G
,
De Luca
L
,
Burnett
J
.
Congestion in acute heart failure syndromes: an essential target of evaluation and treatment
.
Am J Med
.
2006 Dec
119
12 Suppl 1
S3
10
.
2.
McDonagh
TA
,
Metra
M
,
Adamo
M
,
Gardner
RS
,
Baumbach
A
,
Böhm
M
.
2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure
.
Eur Heart J
.
2021 Sep
42
36
3599
726
.
3.
de la Espriella
R
,
Cobo
M
,
Santas
E
,
Verbrugge
FH
,
Fudim
M
,
Girerd
N
.
Assessment of filling pressures and fluid overload in heart failure: an updated perspective
.
Rev Esp Cardiol
.
2023 Jan
76
1
47
57
.
4.
Mullens
W
,
Damman
K
,
Harjola
VP
,
Mebazaa
A
,
Brunner-La Rocca
HP
,
Martens
P
.
The use of diuretics in heart failure with congestion: a position statement from the Heart Failure Association of the European Society of Cardiology
.
Eur J Heart Fail
.
2019 Feb
21
2
137
55
.
5.
Wang
MJ
,
Zheng
YM
,
Jin
HX
.
Ultrafiltration for patients with acute decompensated heart failure: a systematic review and meta-analysis
.
Medicine
.
2021 Dec 17
100
50
e28029
.
6.
Núñez
J
,
González
M
,
Miñana
G
,
Garcia-Ramón
R
,
Sanchis
J
,
Bodí
V
.
Continuous ambulatory peritoneal dialysis as a therapeutic alternative in patients with advanced congestive heart failure
.
Eur J Heart Fail
.
2012 May
14
5
540
8
.
7.
Núñez
J
,
González
M
,
Miñana
G
,
Garcia-Ramón
R
,
Sanchis
J
,
Bodí
V
.
Continuous ambulatory peritoneal dialysis and clinical outcomes in patients with refractory congestive heart failure
.
Rev Esp Cardiol
.
2012 Nov
65
11
986
95
.
8.
Viglino
G
,
Neri
L
,
Feola
M
.
Peritoneal ultrafiltration in congestive heart failure-findings reported from its application in clinical practice: a systematic review
.
J Nephrol
.
2015 Feb
28
1
29
38
.
9.
Grossekettler
L
,
Schmack
B
,
Meyer
K
,
Brockmann
C
,
Wanninger
R
,
Kreusser
MM
.
Peritoneal dialysis as therapeutic option in heart failure patients
.
ESC Heart Fail
.
2019 Apr
6
2
271
9
.
10.
Pernias
V
,
González
M
,
Miñana
G
,
Górriz
JL
,
Juan
I
,
Chorro
FJ
.
Refractory congestive heart failure: when the solution is outside the heart
.
ESC Heart Fail
.
2020 Feb
7
1
311
4
.
11.
Kazory
A
,
Koratala
A
,
Ronco
C
.
Customization of peritoneal dialysis in cardiorenal syndrome by optimization of sodium extraction
.
Cardiorenal Med
.
2019
;
9
(
2
):
117
24
.
12.
Li
PK
,
Szeto
CC
,
Piraino
B
,
Bernardini
J
,
Figueiredo
AE
,
Gupta
A
.
Peritoneal dialysis-related infections recommendations: 2010 update
.
Perit Dial Int
.
2010 Jul-Aug
30
4
393
423
.
13.
Figueiredo
A
,
Goh
BL
,
Jenkins
S
,
Johnson
DW
,
Mactier
R
,
Ramalakshmi
S
.
Clinical practice guidelines for peritoneal access
.
Perit Dial Int
.
2010 Jul-Aug
30
4
424
9
.
14.
Crowther
MJ
.
Merlin: a unified modeling framework for data analysis and methods development in Stata
.
Stata J
.
2020
;
20
(
4
):
763
84
.
15.
Crowther
MJ
,
Abrams
KR
,
Lambert
PC
.
Joint modeling of longitudinal and survival data
.
Stata J
.
2013
;
13
(
1
):
165
84
.
16.
Colombo
PC
,
Doran
AC
,
Onat
D
,
Wong
KY
,
Ahmad
M
,
Sabbah
HN
.
Venous congestion, endotelial and neurohormonal activation in acute decompensated heart failure: cause or effect
.
Curr Heart Fail Rep
.
2015 Jun
12
3
215
22
.
17.
Kazory
A
.
Peritoneal dialysis for chronic cardiorenal syndrome: lessons learned from ultrafiltration trials
.
World J Cardiol
.
2015 Jul
7
7
392
6
.
18.
Sánchez
JE
,
Ortega
T
,
Rodríguez
C
,
Díaz-Molina
B
,
Martín
M
,
García-Cueto
C
.
Efficacy of peritoneal ultrafiltration in the treatment of refractory congestive heart failure
.
Nephrol Dial Transpl
.
2010 Feb
25
2
605
10
.
19.
Kunin
M
,
Carmon
V
,
Arad
M
,
Levin-Iaina
N
,
Freimark
D
,
Holtzman
EJ
.
Inflammatory biomarkers in refractory congestive heart failure patients treated with peritoneal dialysis
.
BioMed Res Int
.
2015
;
2015
:
590851
.
20.
Núñez
J
,
Teruel
A
,
Quiñones-Torrelo
C
,
García-Blas
S
,
Carratalá
A
,
Miñana
G
.
Continuous ambulatory peritoneal dialysis as a promising therapy for light chain amyloidosis with congestive heart failure
.
Int J Cardiol
.
2016 Nov
223
807
9
.
21.
Miñana
G
,
Llàcer
P
,
Sanchis
I
,
García-Blas
S
,
Bonanad
C
,
Ventura
S
.
Early spot urinary sodium and diuretic efficiency in acute heart failure and concomitant renal dysfunction
.
Cardiorenal Med
.
2020
;
10
(
5
):
362
72
.
22.
Tersalvi
G
,
Dauw
J
,
Gasperetti
A
,
Winterton
D
,
Cioffi
GM
,
Scopigni
F
.
The value of urinary sodium assessment in acute heart failure
.
Eur Heart J Acute Cardiovasc Care
.
2021 Apr
10
2
216
23
.
23.
de la Espriella
R
,
Núñez
E
,
Llàcer
P
,
García-Blas
S
,
Ventura
S
,
Núñez
JM
.
Early urinary sodium trajectory and risk of adverse outcomes in acute heart failure and renal dysfunction
.
Rev Esp Cardiol
.
2021 Jul
74
7
616
23
.
24.
Ali
SS
,
Olinger
CC
,
Sobotka
PA
,
Dahle
TG
,
Bunte
MC
,
Blake
D
.
Loop diuretics can cause clinical natriuretic failure: a prescription for volume expansion
.
Congest Heart Fail
.
2009 Jan-Feb
15
1
1
4
.
25.
Rodríguez-Carmona
A
,
Fontán
MP
.
Sodium removal in patients undergoing CAPD and automated peritoneal dialysis
.
Perit Dial Int
.
2002 Nov-Dec
22
6
705
13
.