Asymptomatic Fluid Volume Imbalance and Peridialysis Blood Pressure Independently Predict Cardiovascular and All-Cause Mortality in Patients Undergoing Hemodialysis
Chronic fluid overload (FO) is a significant contributor to morbidity and mortality in patients undergoing hemodialysis (HD). While hypertension is commonly linked to fluid overload, the prognostic value of blood pressure (BP) in HD patients remains complex due to variations in measurement timing (pre- or post-dialysis) and the interplay between volume status and BP. This study investigates the relationship between fluid volume parameters, peridialysis BP, and their independent effects on cardiovascular and all-cause mortality in HD patients without preexisting coronary artery disease (CAD).
Study Design and Population
This retrospective analysis included 101 adults aged 18–90 years receiving HD for at least three months at a single center. Exclusion criteria encompassed comorbidities such as cancer, systemic vasculitis, congenital heart disease, CAD confirmed by imaging or symptomatic angina, and unstable vital signs during HD. Participants underwent bio-impedance analysis (BIA) within 30 minutes post-HD to measure extracellular fluid (ECF), intracellular fluid (ICF), total body fluid (TBF), and dry weight. Fluid overload (FO) was calculated as the difference between post-dialysis weight and dry weight. Key demographic and clinical parameters, including dialysis vintage, BP measurements, and laboratory values, were extracted from electronic records.
The median age of participants was 48 years, with 63.4% male and 21.8% having diabetes. Primary causes of end-stage renal disease (ESRD) included glomerulonephritis (38.6%), hypertensive nephropathy (25.7%), and diabetic nephropathy (17.8%). Median dialysis vintage was 10.0 months, with 52.5% using autogenous arteriovenous fistulas and 47.5% relying on tunneled dialysis catheters.
Key Measurements and Outcomes
Post-dialysis ECF/ICF ratios averaged 0.66 ± 0.04, with 39.6% of patients exhibiting post-dialysis ECF/TBF ratios ≥0.4, indicating persistent fluid overload despite ultrafiltration. Pre-dialysis systolic BP (SBP) and diastolic BP (DBP) averaged 135.9/79.5 mmHg, while post-dialysis values were 134.3/78.1 mmHg. Median absolute BP changes during HD were 7 mmHg for SBP and 9 mmHg for DBP.
The primary outcomes were cardiovascular death (CVD) and all-cause mortality, with CVD defined as death from heart failure, myocardial infarction, arrhythmia, stroke, or hypovolemic shock. Over a median follow-up of 26.4 months, 12 deaths occurred (11.9%), including nine cardiovascular events. Other causes included severe pneumonia (2) and esophageal cancer (1).
Fluid Volume Parameters and Mortality
Receiver operating characteristic (ROC) analysis identified an ECF/ICF ratio cutoff of 0.65 (AUC = 0.76, P = 0.011) for discriminating cardiovascular mortality. Patients with ECF/ICF ≥0.65 exhibited significantly higher cardiovascular (15.1% vs. 2.1%, P = 0.022) and all-cause mortality (20.8% vs. 2.1%, P = 0.004) compared to those below the threshold. Cox regression confirmed ECF/ICF ratio as an independent predictor of CVD (HR = 1.65 per 0.01 increase, 95% CI: 1.02–2.68, P = 0.043) and all-cause death (HR = 1.27, 95% CI: 1.02–1.57, P = 0.033).
The ECF/ICF ratio correlated inversely with serum albumin (r = -0.47), hemoglobin (r = -0.31), and potassium (r = -0.33), suggesting its role as a composite marker reflecting fluid overload, malnutrition, and inflammation. Elevated ECF/ICF may stem from inflammation-driven fluid shifts (from intracellular to extracellular compartments) and reduced ICF due to muscle wasting.
Blood Pressure and Prognostic Divergence
Peridialysis BP metrics displayed divergent associations with outcomes. Pre-dialysis SBP and post-dialysis DBP were risk factors for mortality, while pre-dialysis DBP and post-dialysis SBP showed protective effects. In multivariable models, post-dialysis DBP independently predicted both CVD (HR = 1.10, P = 0.09) and all-cause death (HR = 1.12, P = 0.09), though statistical significance was marginal. Post-dialysis SBP demonstrated a protective effect against all-cause mortality (HR = 0.86 per 1 mmHg, 95% CI: 0.78–0.95, P < 0.01).
Notably, pre-dialysis SBP trended toward increased CVD risk (HR = 1.12, P = 0.11), while pre-dialysis DBP was inversely associated with all-cause death (HR = 0.86, P = 0.02). These findings highlight the complexity of BP interpretation in HD patients, where post-dialysis measurements may better reflect true cardiovascular function after antihypertensive drug clearance during dialysis.
Clinical Implications and Mechanistic Insights
The study underscores the limitations of relying solely on clinical assessment for fluid management, as 39.6% of patients remained fluid-overloaded post-HD despite ultrafiltration. BIA-guided volume evaluation could improve outcomes by identifying subclinical fluid imbalance. The ECF/ICF ratio’s prognostic value likely integrates multiple pathological pathways:
- Chronic Fluid Overload: Directly contributes to cardiac strain and endothelial dysfunction.
- Malnutrition-Inflammation Complex: Hypoalbuminemia and anemia exacerbate fluid redistribution and reduce osmotic pressure, perpetuating extracellular expansion.
- Electrolyte Imbalance: Lower serum potassium in high ECF/ICF patients (P < 0.05) may reflect diuretic use or dietary restrictions, further complicating cardiovascular risk.
Limitations and Generalizability
The study’s retrospective design introduces potential selection bias, particularly excluding patients with contraindications to BIA (e.g., amputees). The small sample size limited exploration of interaction effects between fluid parameters and BP. Additionally, single-timepoint BIA measurements may not capture dynamic fluid changes, though baseline FO has been linked to long-term outcomes in prior studies.
The cohort’s homogeneity—predominantly Chinese patients from a tertiary center—limits extrapolation to diverse populations. Serial BIA measurements and time-averaged volume data could enhance prognostic accuracy in future research.
Conclusions
In HD patients without CAD, asymptomatic fluid volume imbalance (reflected by ECF/ICF ratio) and peridialysis BP parameters independently predict cardiovascular and all-cause mortality. The lack of correlation between fluid indices and BP underscores the need for multimodal assessment in this population. Clinicians should prioritize objective volume monitoring via BIA and recognize the prognostic nuances of BP measurements taken at different dialysis phases.
doi.org/10.1097/CM9.0000000000001337
Was this helpful?
0 / 0