Echocardiographic Changes In Patients With Chronic Kidney Failure (Stage 5) On Hemodialysis: Prospective Cohort Study from tertiary care hospital in karnataka

Cardiovascular diseases (CVD) results in 50% of all deaths in patients undergoing haemodialysis and account for one third of the hospital admissions among CKD patients1. Structural changes in addition to functional cardiovascular changes are detected by echocardiography; which are important for characterization of cardiovascular risk among CKD patients2.

LVMI (left ventricular mass index) is a predictor of cardiovascular morbidity in CKF (chronic kidney failure) patients undergoing hemodialysis3. CKD patients have a huge burden of cardiovascular disorders, ESRD on MHD patients are at the highest risk for development of cardiovascular events besides death 4. Amongst ESRD patients, idiosyncrasies in left ventricular (LV) structure as well as function are predictors of adverse clinical outcomes5.

We conducted a study on cardiovascular end results in terms of morbidity & mortality of end stage kidney disease patients who were initiated on haemodialysis and hence study the structural and functional cardiovascular changes with echocardiographic parameters, attributed to effects of haemodialysis in stage 5 CKD patients6.

Available data of LV structure as well as function are restricted by cross sectional studies, data of clinical trial participants7, which is generally less representative compared with enrolees of observational research data, hence there was a need to conduct a prospective follow up study8.

The Objective of our study was to evaluate the changes in the echocardiographic parameters in patients with chronic kidney failure (stage 5) before and after initiating on maintenance hemodialysis and to study the significance of echocardiographic parameters in determining cardiovascuscular outcomes in CKD stage 5D before and after initiating hemodialysis.

We conducted a Prospective cohort study over a duration of 2 years (September 2019 to September 2021) carried out in patients and out patients of hospitals- a tertiary care, a tertiary care teaching medical centre with alignment with department of nephrology. Institutional ethical committee clearance was obtained.

The calculated Sample Size was 68 patients. After explaining the study procedure and objectives to all the patients, a written informed consent was taken. Patients aged 18 years and above. Who are newly diagnosed as CKD stage 5 and is being initiated on regular haemodialysis (3 times a week) were included in our study.

Patients who had recent history of acute myocardial infarction (less than six months) Recent percutaneous or surgical revascularization, unstable angina or cerebrovascular accident, decompensated congestive heart failure, valvular heart disease. blood pressure >160/110 mmHg, uncontrolled atrial fibrillation or complex ventricular arrhythmia, anaemia with Hb less than 10g% and pregnant women were excluded from the study.

Detailed history, including age, sex, duration of dialysis and predisposing factors of the cardiovascular diseases were assessed and documented in proforma. Clinical examination with assessment of vitals, general and systemic examination were done, a 12-lead surface electrocardiography was recorded. A Doppler echocardiography was performed initially and repeated after six months, by single medical professional, trained and skilled in echocardiography. eGFR (estimated Glomerular Filtration Rate) was calculated using MDRD (Modification of Diet in Renal disease Study) formula. Staging of CKD was done as per by KDIGO (Kidney Disease Improving Global Outcomes). The following echo parameters were assessed:

o              End diastolic septal thickness (EDST),

o              End diastolic posterior wall thickness (EDPWT),

o              Left ventricular end diastolic diameter (LVEDD)

o              Left ventricular end systolic diameter (LVESD).

o              Left ventricular mass (LVM)

o              Left ventricular mass index LVMI. (LVMI = LVM / BSA)

o              The relative wall thickness (RWT)

o              E/A ratio.

o              E/E’ ratio.

A Complete blood picture (CBC) and Serum. Creatinine was tested and documented. Data was entered into Microsoft excel data sheet and was analysed using SPSS 22 version software. Categorical data was represented using the Frequencies and proportions. Continuous data was represented as mean and standard deviation. Independent t test was used as test of significance to identify the mean difference between two quantitative variables. ROC curve of 0.5. Area under ROC curve above 0.8 indicated fairly good prediction. Graphical representation of data: MS Excel and MS word was used to obtain various types of graphs. P value of 0.05 was considered as statistically significant after assuming all the rules of statistical tests.

Our study included 74 patients, after 6 months 3 patients were excluded (2 patients underwent renal transplantation,1 patient died due to non-cardiovascular Cause and 71 patients were studied.

TABLE 1:  Sex distribution

Out of 71 patients, 48 were males, 23 were females.

TABLE 2:   Shows age group distribution.

Maximum patients were in the age group of 50 to 60 years.

TABLE – 3: CAUSES OF CKD

Common cause of CKD among the study population was Type 2DM.  Other causes are as represented in the Table 3

TABLE – 4 Describes the basic parameters of study patients.

TABLE- 5:  Showing echocardiographic parameters with mean and standard deviation at the time of initiating after 6 months of haemodialysis.

TABLE- 6:  Showing echocardiographic parameters with mean standard deviation with student t test and p value.

TABLE – 7: Mortality

TABLE showing number of patients who developed cardiovascular comorbidities during 6 months of follow up period.

TABLE – 8: Showing number of cardiovascular outcomes with s mean and standard deviation of LVM.

TABLE- 9: Table showing number of patients who developed cardiovascular comorbidities during 6 months of follow up with mean and standard deviation of LVM.

Table 10: - Comparison of mean age between two groups

There was no statistically significant difference found between groups with respect to age

Table 11: - Comparison of mean echocardiographic parameters among the patients who developed cardiovascular outcomes and those without cardiovascular outcomes.

There was a statistically significant difference found between groups with respect to LVM and LVMI

FIGURE: 3 SHOWING ECHOCARDIOGRAPHIC PARAMETERS AT THE TIME OF INITIATING HD AND 6 MONTHS AFTER INITIATION OF HD.

Figure: - ROC for LVMI in predicting outcome

TABLE- 12:

The LVMI for patients with cardiovascular outcomes had sensitivity of 66.67% and specificity of 73.08%.  with cut-off of 289.

Figure: - ROC for LVM in predicting outcome

TABLE- 13:

The LVM for patients with cardiovascular outcomes had sensitivity of 90.48% and specificity of 43.08%.  with cut-off of 448.

Figure: - ROC for LVM in predicting outcome

Cardiovascular complications are the main cause of mortality chronic kidney disease patients (CKD) undergoing haemodialysis therapy.9 The diagnosis of LV abnormalities by Doppler echocardiography is essential for the characterization of individuals with higher cardiovascular risk beside estimating the prevalence of primary heart disease to study its predisposing factors, prognostic impact and the effect of therapeutic interventions.10

The Doppler echocardiogram is a complementary, non-invasive examination, broadly used in the assessment of heart structure and function, bringing several ultra-sound techniques together in a single examination. 11

In this study our aim was to evaluate the predictive value of echocardiographic parameters for analysing cardiovascular outcomes in patients with chronic renal failure stage 5 undergoing maintenance haemodialysis.61 patients initiating on maintenance haemodialysis were included in this study. They were subjected to clinical and echocardiographic examination and within 6 months of follow up 16% of patients developed cardiovascular mortality and morbidity, we divided them in two groups – patients with cardiovascular outcomes, and those without cardiovascular outcomes.

Common cause of CKD in our study was type 2 dm.  Mean LVM was 226.8 and LVMI was 126.12 Mean LVM among patients with no cardiovascular outcomes was 461, and in patients with CV outcomes was 635. Mean LVMI in patients with cv outcomes was 345.52, and in those with no cardiovascular outcomes was 254.50 The LVMI for patients with cardiovascular outcomes had sensitivity of 66.67% and specificity of 73.08%.  with cut-off of 289.

The LVM for patients with cardiovascular outcomes had sensitivity of 90.48% and specificity of 43.08%.  with cut-off of 448. Mean LVESD in patients with CVS outcomes was 4.25 compared to those with no CVS outcomes was 2.9327. Mean LVEDD in patients with CVS outcomes was 4.66, compared to patients with no CVS outcomes which was 4.47. EDPWT in patients with CVS outcomes was 1.271, compared to patients with no CVS outcomes which was 1.3869. EDST in patients with CVS outcomes was 1.1171, compare to patients with no cardiovascular outcomes which was 1.1419.

In concordance to our study Zoccali et al in a prospective study in 161 haemodialysis patients tested the prognostic importance of changes in LVM on survival and incident cardiovascular events. After 4 years of follow up they found that the rate of increase of LVMI was significantly (P = 0.029) more amongst patients with incident cardiovascular events than in those without such events and concluded that changes in LVMI have an independent prognostic value for cardio-vascular events and provide scientific support to the use of repeated echocardiographic studies for monitoring cardiovascular risk in dialysis patients.13

In another study Chen et al, who studied 505 patients with stage 3–5 chronic kidney disease with 3 years follows up period found that increase in the left ventricular mass index (LVMI) independently associated with adverse CV outcomes in CKD patients with (P = 0.003). 14 Also, Mostovaya et al, who studied 327 patients on haemodialysis with mean follow up period of 2 years confirmed the relation between LVM and all-cause mortality. Furthermore, it demonstrated a markedly increase in the peril of unexpected death in individuals with a high LVM.15

In concordance with our study, Hillege et al. found that there was a significant correlation between the deterioration of congestive CCF and the progression of kidney failure16. Our study found that, with the decrease of renal function, there was a significant trend for a stepwise increase in LVMI and the presence of LVH and LVEF < 55% and a stepwise decrease in LVEF in diabetic nephropathy patients, which is consistent with the previous study findings.

We found that age and LVMI were the only predictors that can predict 6 months mortality in haemodialytic patients with significant statistical differences LVM (p value = 0.001) and LVMI (p value = 0.001). In another study Chen et al17 who studied 505 patients with stage 3–5 chronic kidney disease with 3 years follow up period found that elevated left ventricular mass index (LVMI) independently associated with adverse CV outcomes in CKD patients with (P = 0.001).

In disconcordance to our results, Siqueira et al18 who studied 60 patients with CKD undergoing haemodialysis for 6 months found that moderate to severe diastolic dysfunction was a non-dependent risk factor for cardiovascular events but in this study diastolic dysfunction was not associated with cardiovascular mortality or morbidity possible explanation may be short follow up period.

Also, in disconcordance to our results, Dogan et al19 who studied 45 chronic haemodialysis patients with complete standard and tissue Doppler imaging echocardiography with follow up period (26month) found that E/e [ratio of mitral E to peak early diastolic mitral annular velocity (e)] were significantly higher in patients who had major events. In Cox proportional hazard analysis only E/e ratio predicted combined endpoint of all-cause mortality and nonfatal cardiovascular events (p = 0.018).in this study no significant statistical differences between two groups (p value = 0.550) possible explanation might be short follow up period.

In our study, Of the total sample size of 68 patients, 26 (38.23%) patients developed cardiovascular comorbidities. 30.8% of the patients had hospital admissions secondary to heart failure, 19.2% had arrhythmias, 19.2% had coronary artery disease.  LVMI, LVM, EDST, LVEDD were significantly elevated in these subgroups of patients with p value of <0.001.

In concordance to our study, study done by Chen, Szu-Chia et al. 20showed  a significant stepwise increase in LVMI (P < 0.001), LVH (P < 0.001), and LVEF < 55% (P = 0.013) and a stepwise decrease in LVEF (P = 0.038) corresponding to advance in CKD stages. Taken together, the interaction of functional and structural derangements in the heart and kidney amplifies the progression of CKD and have unfavourable cardiovascular outcomes.21

Echocardiographic parameters have prognostic significance in determining cardiovascular morbidity and mortality in CKD patients on haemodialysis. LVM and LVMI are significantly elevated in patient who developed cardiovascular outcomes compared to patient who did not develop cardiovascular outcomes.