Study cohort

Between January 2006 and December 2020, a total of 1047 patients began long-term dialysis treatment in Taipei Veterans General Hospital, with 383 receiving PD and 664 receiving HD. After excluding 190 patients who did not receive an echocardiogram within 3 months of dialysis initiation, 857 patients met the study inclusion criteria. The mean age of the study population was 62 ± 12 years, with 445 (52%) patients being male, 275 (32%) receiving PD, and 131 (15%) having a history of heart failure. The most common causes of kidney failure were diabetes mellitus (39%), hypertension (20%), and glomerulonephritis (14%).

Among the study population, 659 patients (77%) had one or more echocardiographic abnormalities. Patients with diabetes, a history of heart failure, and higher body weight were more likely to have abnormal echocardiograms (Table 1).

Table 1 Baseline Characteristics of the Study PopulationsEchocardiographic findings in kidney failure patients

Most patients with an abnormal echocardiogram have more than one abnormal parameter. The most frequently observed abnormalities were left ventricular hypertrophy (56%), diastolic dysfunction (49%), and left atrial enlargement (45%). Mitral regurgitation was the most common valvular abnormality, present in 9% of patients. Right ventricular systolic dysfunction was observed in 5% of patients (Table 2). The average age was higher among the HD patients compared to PD patients (65.68 ± 16.07 vs 53.42 ± 14.67, p < 0.001). In addition, a greater prevalence of left atrial enlargement, diastolic dysfunction, and pulmonary hypertension among HD patients was also identified in the present study (Fig. 1). The echocardiographic findings relative to the type of dialysis vascular access are presented in Supplementary Table 1. Notably, left atrial size was significantly larger in subjects with arteriovenous fistulas (AVFs) compared to those using central venous catheters (CVCs). Apart from this distinction, other parameters including left ventricular mass index, left ventricular volume index, left ventricular ejection fraction, diastolic function, pulmonary artery systolic pressure (PASP), and right ventricular function showed no significant differences between the two groups.

Table 2 Baseline echocardiographic parameters of the study patients (n = 857)Fig. 1

Prevalence of echocardiographic abnormalities in dialysis patients. HD hemodialysis, PD peritoneal dialysis. *p < 0.05

Echocardiographic abnormalities and patient outcomes

During a mean follow-up period of 4.28 ± 2.61 years, a total of 183 patients developed major adverse cardiovascular events while 185 patients died. The cumulative survival rate at 1 year, 3 years, and 5 years was 93%, 84%, and 79%, respectively, which is consistent with the mortality rate of dialysis patients in Taiwan [16].

Patients who exhibited abnormalities on echocardiography had significantly higher rates of both major adverse cardiovascular events (log-rank p = 0.002) and total mortality (log-rank p = 0.011) compared to those without abnormalities. The Kaplan–Meier figures demonstrated that all echocardiographic abnormalities, except for left ventricular hypertrophy and left atrial enlargement, were associated with an increased risk of 5-year all-cause mortality (Fig. 2). Additionally, most echocardiographic abnormalities, except for left ventricular hypertrophy, were found to be associated with an increased risk of major adverse cardiovascular events (Fig. 3). To assess the association between echocardiographic abnormalities and outcomes, we performed a Cox proportional hazard model analysis, adjusting for age, sex, comorbidities, and medications. Our findings showed that left ventricular systolic dysfunction (HR, 95% CI 2.31, 1.56–3.41, p < 0.001), left ventricular volume index (HR, 95% CI 1.93, 1.30–2.90, p = 0.001), right ventricular dysfunction (HR, 95% CI 1.86, 1.09–3.19, p = 0.024), valvular heart disease (HR, 95% CI 1.75, 1.22–2.51, p = 0.002), and pulmonary hypertension (HR, 95% CI 1.72, 1.25–2.34, p < 0.001) were significantly associated with a higher mortality rate (Table 3). Similarly, right ventricular systolic dysfunction (HR, 95% CI 2.36, 1.44–3.86, p = 0.001), left ventricular systolic dysfunction (HR, 95% CI 2.20, 1.47–3.30, p < 0.001), left ventricular volume index (HR, 95% CI 1.98, 1.34–2.94, p < 0.001), and valvular heart disease (HR, 95% CI 1.81, 1.25–2.61, p = 0.002) were identified as the independent risk factors for major adverse cardiovascular events. (Table 4).

Fig. 2

Survival analysis of abnormal echocardiographic parameters in the dialysis population. Kaplan–Meier survival curve analysis of mortality over a 5-year follow-up period in relation to specific echocardiographic abnormalities including right ventricular systolic dysfunction, left ventricular systolic dysfunction, left ventricular volume index, and moderate to severe valvular heart disease

Fig. 3

Impact of specific echocardiographic abnormalities on major adverse cardiovascular events (MACE) in dialysis patients. Kaplan–Meier survival curve analysis of mortality over a 5-year follow-up period in relation to specific echocardiographic abnormalities including right ventricular systolic dysfunction, left ventricular systolic dysfunction, left ventricular volume index, and moderate to severe valvular heart disease

Table 3 Risk of MACE in each echocardiographic abnormalityTable 4 Risk of 5-year all-cause mortality in each echocardiographic abnormalitySubgroup analysis of the interaction between dialysis modalities and specific echocardiographic abnormalities

The 5-year survival rate was 76.5% and 82.5% in HD and PD subjects, respectively. Hemodialysis can cause fluctuations in blood pressure and volume, as well as changes in electrolyte levels. In contrast, PD does not cause as many rapid shifts in fluid, but both HD and PD can have negative effects on the heart over time [17, 18]. To evaluate the interaction between dialysis modality and each echocardiographic abnormality on mortality and major adverse cardiovascular events, we performed subgroup analysis of interaction between dialysis modality and echocardiogram abnormalities on patient outcomes. The findings demonstrated that in PD patients, left ventricular systolic dysfunction (p for interaction = 0.019) and elevated left ventricular volume index (p for interaction < 0.001) were more strongly associated with increased mortality risk compared to HD patients (Fig. 4A). Conversely, there was no notable interaction between dialysis modality and echocardiographic abnormalities in predicting major adverse cardiovascular events (Fig. 4B).

Fig. 4

Association between distinct echocardiographic abnormalities and the risk of mortality and major adverse cardiovascular events (MACE) among individuals undergoing hemodialysis and peritoneal dialysis. Adjusted for age, gender, comorbidities, and medications. HD hemodialysis, PD peritoneal dialysis