Early Cardiopulmonary Adaptation Following Arteriovenous Fistula Creation in Pre- Dialysis Chronic Kidney Disease Patients: A Prospective Echocardiographic Study

Technical Summary: Early Cardiopulmonary Adaptation Following Arteriovenous Fistula Creation in Pre-Dialysis CKD Patients

Problem Statement

Arteriovenous fistula (AVF) creation is the preferred vascular access for maintenance hemodialysis due to superior patency and lower infection rates compared to central venous catheters. However, the procedure induces significant systemic hemodynamic alterations, including increased venous return, reduced systemic vascular resistance, and elevated cardiac output. While cardiovascular remodeling after AVF creation has been documented in established hemodialysis populations, prospective echocardiographic data regarding these changes in pre-dialysis chronic kidney disease (CKD) patients remain limited. Understanding the early cardiovascular response in this specific cohort is critical, as these patients often possess pre-existing cardiac abnormalities and may be at higher risk for maladaptive responses, such as high-output cardiac failure or pulmonary hypertension, before dialysis initiation.

Methodology

This was a single-center, prospective observational study conducted at the University of Malaya Medical Centre between March 2021 and December 2022.

• Population: 34 adult pre-dialysis CKD patients (mean age 65.2 ± 8.9 years; 52.9% male) undergoing elective AVF creation. Exclusion criteria included atrial fibrillation, significant valvular disease, NYHA class III/IV symptoms, baseline LVEF <40%, and urgent dialysis initiation.
• Design: Serial transthoracic echocardiography was performed at three timepoints:
  1. Baseline: Before AVF creation.
  2. Early Follow-up: 4–12 weeks post-creation.
  3. Late Follow-up: ≥6 months post-creation.
• Parameters Assessed: Left ventricular (LV) volumes (LVEDV, LVESV), systolic function (LVEF, Stroke Volume, Cardiac Index), LV mass (LVM, LVMI), diastolic function (E/A ratio, IVRT, Deceleration Time), pulmonary hemodynamics (Pulmonary Artery Systolic Pressure [PASP]), and right ventricular (RV) function (TAPSE, RV S′ velocity).
• Subgroup Analysis: AVFs were classified as proximal (at or above the antecubital fossa) or distal (below). Access flow was categorized as low (<600 mL/min) or high (≥600 mL/min).
• Statistical Approach: Paired t-tests or Wilcoxon signed-rank tests were used for paired analyses. Mean differences with 95% confidence intervals were calculated. No formal correction for multiple comparisons was applied due to the exploratory nature of the study.

Key Results

The study enrolled 34 patients for baseline assessment, with 27 completing early follow-up and 14 completing late follow-up.

1. Left Ventricular Volume and Output:

• Volume Loading: Significant increases in LV volumes were observed at early follow-up. LVEDV increased from 79.8 ± 21.8 mL to 104.8 ± 32.0 mL (+25.0 mL, p=0.001). LVESV increased from 29.4 ± 10.5 mL to 37.5 ± 15.5 mL (+8.1 mL, p=0.012). These volume increases persisted at late follow-up.
• Cardiac Output: Stroke Volume (SV) increased significantly from 50.5 ± 13.5 mL to 66.2 ± 21.0 mL (+15.8 mL, p=0.002). Cardiac Index (CI) rose from 2.18 ± 0.65 to 3.05 ± 1.15 L/min/m² (+0.87, p=0.007).
• Systolic Function: LVEF remained preserved throughout the follow-up period, showing no statistically significant change (64.6% to 63.6%, p=0.463).

2. Diastolic and Structural Changes:

• Diastolic Function: Isometric relaxation time (IVRT) increased significantly at early follow-up (0.083s to 0.099s, p=0.023). The E/A ratio showed a non-significant trend toward increase.
• Remodeling: No statistically significant changes in Left Ventricular Mass (LVM) or Left Ventricular Mass Index (LVMI) were observed at either follow-up timepoint.

3. Pulmonary Hemodynamics and Right Ventricular Function:

• PASP: Pulmonary Artery Systolic Pressure increased significantly at early follow-up from 28.0 ± 8.6 mmHg to 34.2 ± 9.9 mmHg (+6.2 mmHg, p=0.032).
• RV Function: Tricuspid Annular Plane Systolic Excursion (TAPSE) increased significantly (21.9 mm to 24.5 mm, p=0.005), while RV S′ velocity remained unchanged.

4. Subgroup Findings (AVF Location):

• Among patients with paired PASP data (n=14), those with proximal AVFs demonstrated a significantly greater increase in PASP compared to those with distal AVFs (+11.4 ± 9.6 mmHg vs. +1.0 ± 6.8 mmHg, p=0.030).
• A significant negative correlation was found between AVF proximity and the magnitude of PASP change (Spearman ρ = −0.621, p=0.018), indicating proximal locations are associated with larger pressure increases.
• No significant correlations were found between AVF flow rate and changes in LV volumes or cardiac index in the available paired data.

Key Contributions and Significance

This study provides prospective, serial echocardiographic evidence that AVF creation in pre-dialysis CKD patients induces early cardiopulmonary adaptation characterized by:

1. Increased Volume Loading: A substantial rise in LV end-diastolic and end-systolic volumes.
2. Augmented Cardiac Output: Significant increases in stroke volume and cardiac index, driven by the Frank-Starling mechanism, while preserving systolic function (LVEF).
3. Pulmonary Hemodynamic Burden: A measurable increase in pulmonary artery systolic pressure, suggesting an increased afterload on the right ventricle.
4. Location-Dependent Risk: The finding that proximal AVFs confer a greater pulmonary hemodynamic burden (higher PASP increases) than distal AVFs supports the hypothesis that shunt volume and location are critical determinants of cardiopulmonary stress.

The authors conclude that while the observed changes represent an adaptive high-output physiological state rather than immediate systolic decompensation, the significant increase in pulmonary pressures warrants attention. The study suggests that selective cardiovascular surveillance may be beneficial for higher-risk patients, particularly those with pre-existing cardiopulmonary disease or those receiving proximal, high-flow AVFs.

Modest Claims and Limitations

The authors explicitly frame these findings as hypothesis-generating due to several limitations:

• Sample Size: The cohort was relatively small, and the late follow-up subgroup was particularly limited (n=14), reducing statistical power and generalizability.
• Attrition Bias: Incomplete follow-up led to variable sample sizes across parameters, and patients completing follow-up may have differed systematically from those lost.
• Observational Design: The study design precludes causal inference.
• Exploratory Nature: Formal correction for multiple comparisons was not performed to avoid Type II errors; thus, findings should be interpreted with caution.
• Lack of Long-term Data: The study does not define the long-term cardiovascular implications of these hemodynamic changes, necessitating further multicenter prospective studies.