Simple Laboratory and Clinical Parameters as Predictor for Electrocardiographic Abnormalities Among Hospitalized Patients with Chronic Kidney Disease: A Cross-Sectional Study

Rerdin Julario, Ricardo Adrian Nugraha, Bagus Putra Dharma Khrisna, Tony Santoso Putra, Eka Prasetya Budi Mulia, Ryan Enast Intan, Firas Farisi Alkaff
doi: https://doi.org/10.1101/2021.01.15.21249916

Abstract

Background In developing countries, even electrocardiography (ECG) hasn’t been used widely in most health-care centers. The ability of physicians to refer to chronic kidney disease (CKD) patients for ECG, often collide with several barriers and costs. Therefore, we need to formulate the simplest and most efficient model to predict when CKD patients need to be referred due to potential ECG abnormalities.

Objective The aim of this study was to develop several clinical and laboratory parameters as a predictor of any ECG abnormalities.

Materials and Methods A retrospective cross-sectional study design held at Dr. Soetomo General Academic Hospital, Surabaya, Indonesia. Subjects were hospitalized patients with CKD between 1 January to 31 December 2019. 198 CKD patients (101 males) were enrolled for the study. All patients had demographic information, detailed clinical profile, resting 12-lead ECG recording, complete blood count, serum electrolyte and renal function test profile during admission and results were interpreted blindly by two cardiologists. Statistical analysis was done by SPSS 17.0.

Results A total of 198 patients were included in this study. Mean ages were 52.2±11.8 years old and fifty-one percent were males. Eighty-eight percent of patients from 198 patients had ECG abnormality. AUC of hemoglobin level to discriminate poor R wave progression, pathological Q wave, non-spesific ST-T changes, and frontal axis deviation were 0.532, 0.641, 0.556 and 0.693, respectively. In multivariate logistic regression analysis, only higher systolic blood pressure was determined as an independent predictor of abnormal ECG finding in CKD patients, as systolic blood pressure increase by one unit, the odds of having abnormal ECG is increased 1.02 times (95% CI: 1.00 – 1.02, p=0.042).

Conclusion The ECG abnormalities can be found in hospitalized CKD patients. Fragmented QRS and long QTc were the highest prevalent ECG abnormalities in our study. Serum creatinine and hemoglobin could predict peaked T wave and prolonged QTc among hospitalized CKD patients. Systolic blood pressure could predict prolonged QTc and fragmented QRS in CKD patients.

INTRODUCTION

Chronic kidney disease (CKD) is associated with various kinds of complications that lead to poor health outcomes. The risk of cardiovascular disease (CVD) has increased in patients with CKD even before reaching End-Stage Renal Disease (ESRD). The mortality rate in dialysis patients due to cardiovascular complications is high. The prognosis in developing countries is poor due to inadequate diagnostic and late presentation[1].

KDIGO 2012 recommends active intervention by paying attention to the conditions of CVD and related comorbidities to reduce hospitalization and mortality for people with CKD[2]. Based on the guideline principles, CKD patients need to be closely monitored for early signs and symptoms of CVD. Even annual ECG monitorings are mandatory to detect any sign of CVD at an early stage. Unfortunately, many clinicians, particularly in developing countries, face a big problem to adopt these recommendations [3]. Such as electrocardiography (ECG) hasn’t been used widely in most health-care centers. Therefore, patients need to be referred to bigger health-care centers or hospitals. On the other hand, the ability of physicians to refer chronic kidney disease (CKD) patients for ECG examination or comprehensive specialist treatment, often collide with the geographical barrier, accessibility, patients’ belief, policy, cost, and insurance[3]. Patients with CKD in developing countries usually seek medical advice after falling at an advanced stage due to a lack of awareness, education, financial constraints, and medical facilities[4, 5].

Due to several barriers, the decision of whether to refer CKD patients to comprehensive specialist treatment or even for ECG measurement is challenging and confusing for many clinicians. In addition, only a few studies provide information on the frequency of ECG changes in the local CKD population which can make the physician less aware of the complications of CVD in CKD[6]. The frequency and predictor of ECG abnormalities in CKD patients in Indonesia may also be different from the literature that has been reported[7]. Therefore, we need to formulate the simplest and most efficient model from the clinical and laboratory parameters to predict the time when CKD patients need to be referred due to potential ECG abnormalities. Since there is a lack of Indonesian-based research about predictor model for ECG abnormalities on CKD patients, we develop our model from several clinical and simple laboratory parameters to predict any ECG abnormalities in CKD patients treated in a referral hospital in Indonesia.

MATERIALS AND METHODS

Ethical clearance

Institutional ethics and research committee approved the study. Dr. Soetomo General Academic Hospital conferred ethical clearance for this study. Compliance with ethical standards, this research proposal was approved by the ethical committees of Dr. Soetomo General Academic Hospital in collaboration with Universitas Airlangga College of Medicine Research Ethics Council (Ref: 1811/KEPK/II/2020) under the name of Rerdin Julario as the Principal Investigator.

Study design and study setting

This study was an analytic observational study using a retrospective cross-sectional study design. This study held at Dr. Soetomo General Academic Hospital, Surabaya, Indonesia. Dr. Soetomo General Academic Hospital is the biggest hospital in East Java and also the referral hospitals for East Indonesia region with an academic affiliation with Faculty of Medicine, Universitas Airlangga. All patients between the ages of 15–90 years with known or newly diagnosed CKD admitted at the internal medicine ward were included. The study was conducted over 12 months period from 1 January to 31 December 2019.

Population and sample

All patients in different stages of CKD were included in this study. Patients with unstable hemodynamic, history of cardiac arrest, previously consumed antiarrhythmic drugs and lack of complete medical record data were excluded. This study used consecutive sampling.

Outcomes and variables

Potential cases of CKD were identified by an eGFR 60 ml/min per 1.73 m2. To predict the risk of CV in previous epidemiological studies, we can use an eGFR threshold of 60 ml / min per 1.73 m2 [2]. Estimated GFR (eGFR) was calculated using the CKD-EPI equation [8]. Medical records, Patient’s history, and laboratory information were reviewed to gain data on patient’s age, gender, history of hyperglycemia, hypertension, resting heart rate, systolic and diastolic blood pressure, serum creatinine, blood ureum nitrogen, eGFR, electrolytes, and complete blood count. The patient was classified to have ECG abnormalities if one of the ECG parameters didn’t fulfill the reference range.

Data collection

All patients had resting 12-lead ECG examination. All ECGs were reviewed with the following parameters: heart rhythm, heart rate, frontal axis, P wave, PR interval, QRS duration, pathological Q wave, fragmented QRS morphology, QTc interval, ST-segment deviation, and T wave morphology; by two independent consultant physicians. The aforementioned parameters were studied and measured manually and compared with published normal values. The normal limit and definition used for this study were shown in table 1 and table 2. The abnormal findings were defined if were away from minimal or maximal limits or normal morphology. Precordial lead subjective assessments of the peaking of the T wave were also noted. Subsequent ECGs if done were not analyzed for study purposes. Descriptive results were presented in frequency and percentage. Analytic results were presented in ROC curve.

View this table:
Table 1.

Normal limit of ECG parameters

View this table:
Table 2.

Definition used for ECG abnormality

Statistical analysis

For bivariate analysis, χ2-test was used for analyzing categorical data, and independent t-test or Mann-Whitney U test was used for continuous data as appropriate. For univariate and multivariate analysis, the logistic regression model was performed with backward selection. Model calibration was tested with the Hosmer-Lemeshow test. The accuracy of the test was evaluated based on ROC curve analysis. The results on the level of p < 0.05 were assumed to be statistically significant. All statistical analyses were performed using SPSS 26.0 (IBM SPSS Statistics for Windows, IBM Corporation, Armonk, NY).

RESULTS

Baseline characteristic of the study

A total of 198 patients were included in this study. Characteristics of the study population are summarized in (Table 3). The mean age of all patients was 52.2±11.8 years old. A total of 101 patients (51%) were males. Mean serum creatinine was 10.5±8.0 mg/dL and mean eGFR was 10.6±14.4 mL/min/1.73 m2. Most patients were in stage 5 of CKD, which was 164 patients (82.8%). 56.1% (111 patients from 198 patient) patients had hypertension, 37.9% had diabetes mellitus, and 24.7% had a known history of CVD. The mean systolic and diastolic blood pressure was 142.1±28.2 mmHg and 83.7±16.3 mmHg respectively. 143 (72.2%) had anemia with mean hemoglobin level was 8.1±2.3 g/dL. As the result, the ECG abnormality can be found in 176 or almost 88.9%. Another abundant comorbidity was metabolic acidosis (31.8%). Most patients had high (45.5%) or normal (44.9%) serum potassium, and only 9.1% of patients had low serum potassium, with mean serum potassium was 4.8±1.06 mEq/L.

View this table:
Table 3.

Baseline characteristic of study population

Abnormal ECG findings

Various ECG changes are observed in Table 4

View this table:
Table 4.

ECG Abnormalities

Cut-off value for serum creatinine and other biomarkers to discriminate long QTc interval

The cut-off value of serum creatinine to discriminate long QTc can be seen in ROC Curve in figure 1. For bivariate analysis, serum creatinine is positively discriminate long QTc based on ROC curve (Figure 1) with area under curve (AUC) 0.544 (P<0.001) with the best cut off serum creatinine level at 7.58 (sensitivity 62% and specificity 50%) (Table 5).

View this table:
Table 5.

Area under the curve (AUC) of clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of long QT interval by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Figure 1.
Figure 1.

Receiver operator characteristic analysis of the clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of long QT interval by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Cut-off value for serum creatinine and other biomarkers to discriminate fragmented QRS

The cut-off value of serum creatinine to discriminate fragmented QRS can be seen in ROC Curve (Figure 2). For bivariate analysis, serum creatinine is a negative predictor for fragmented QRS based on ROC curve with area under curve (AUC) 0.496 (Table 6).

View this table:
Table 6.

Area under the curve (AUC) of clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of fragmented QRS by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Figure 2.
Figure 2.

Receiver operator characteristic analysis of the clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of fragmented QRS by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Cut-off value for serum creatinine and other biomarkers to discriminate left ventricle hypertrophy

The cut-off value of serum creatinine to discriminate left ventricle hypertrophy (LVH) can be seen in ROC Curve (Figure 3). For bivariate analysis, serum creatinine is positively discriminate LVH based on ROC curve with area under curve (AUC) 0.635 (P<0.001) with the best cut off serum creatinine level at 7.045 (sensitivity 81% and specificity 46%) (Table 7).

View this table:
Table 7.

Area under the curve (AUC) of clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of LVH by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Figure 3.
Figure 3.

Receiver operator characteristic analysis of the clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of LVH by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Cut-off value for serum creatinine and other biomarkers to discriminate peaked T wave

The cut-off value of serum creatinine to discriminate peaked T wave can be seen in ROC Curve (Figure 4). For bivariate analysis, serum creatinine is positively discriminate peaked T wave based on ROC curve with area under curve (AUC) 0.708 (P<0.001) with the best cut off serum creatinine level at 9.03 (sensitivity 77% and specificity 61%) (Table 8).

View this table:
Table 8.

Area under the curve (AUC) of clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of peaked T wave by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Figure 4.
Figure 4.

Receiver operator characteristic analysis of the clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of peaked T wave by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Cut-off value for hemoglobin concentration and other biomarkers to discriminate poor precordial R wave progression

For bivariate analysis, serum creatinine was a negative predictor for poor precordial R wave progression based on ROC curve (Figure 5) with AUC 0.464 (Table 9).

View this table:
Table 9.

Area under the curve (AUC) of clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of poor precordial R wave progression by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Figure 5.
Figure 5.

Receiver operator characteristic analysis of the clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of poor precordial R wave progression by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Cut-off value for hemoglobin concentration and other biomarkers to discriminate pathological Q wave

For bivariate analysis, serum creatinine was a negative predictor for pathological Q wave based on ROC curve (Figure 6) with AUC 0.397 (Table 10).

View this table:
Table 10.

Area under the curve (AUC) of clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of pathological Q wave by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Figure 6.
Figure 6.

Receiver operator characteristic analysis of the clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of pathological Q wave by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Cut-off value for hemoglobin concentration and other biomarkers to discriminate non-specific ST-T changes

For bivariate analysis, serum creatinine was a negative predictor for non-specific ST-T changes based on ROC curve (Figure 7) with AUC 0.406 (Table 11).

View this table:
Table 11.

Area under the curve (AUC) of clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of non-specific ST-T changes by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Figure 7.
Figure 7.

Receiver operator characteristic analysis of the clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of non-specific ST-T changes by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Cut-off value for hemoglobin concentration and other biomarkers to discriminate frontal axis deviation

For bivariate analysis, serum creatinine was a negative predictor for frontal axis deviation based on ROC curve (Figure 8) with AUC 0.462 (Table 12).

View this table:
Table 12.

Area under the curve (AUC) of clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of frontal axis deviation by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Figure 8.
Figure 8.

Receiver operator characteristic analysis of the clinical prediction model for age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. Prediction of frontal axis deviation by age, BUN, serum creatinine, eGFR, systolic blood pressure, diastolic blood pressure, hemoglobin level and serum potassium. P < 0.001 for all ROC curves compared with 0.5 curves.

Univariate and multivariate analysis for ECG abnormalities

A multivariate analysis was used to see the most important factor for ECG abnormalities in CKD patients. Prolong QTc interval and fragmented QRS was used as the marker of ECG abnormality, because of their importance as predictors of long-term mortality and also as the most common abnormal ECG findings in our population. There were 110 (55.55%) patients who have QTc interval prolongation or fragmented QRS, or both. The variables whose P-value <0.25 in univariate analysis from baseline characteristics and laboratory findings were specified as potential risk markers and included in the full model (Table 13). In the univariate analysis, blood urea nitrogen (BUN), systolic and diastolic blood pressure, and potassium level were identified as potential risk predictors. However, in multivariate logistic regression analysis, only higher systolic blood pressure was determined as an independent predictor of abnormal ECG finding in CKD patients, as systolic blood pressure increase by one unit, the odds of having abnormal ECG are increased 1.02 times (95% CI: 1.00 – 1.02, p=0.042). Further analysis was done using ROC curve to determine systolic blood pressure cut off, and the best optimal cut off was 140 mmHg (sensitivity 60%, specificity 59%, AUC 0.57).

View this table:
Table 13.

Univariate and multivariate logistic regression for prediction of ECG abnormalities among chronic kidney disease patients

DISCUSSION

The long-term CKD influences the pathogenesis of cardiorenal syndromes, cardiovascular diseases, cardiovascular risk factors, and substantial cardiovascular mortality, there is a need to screen Indonesian CKD patients who are at risk of getting earlier complications [9]. Renal dysfunction is an independent and significant contributor to worse cardiovascular outcomes[10]. Serum creatinine levels are frequently used as a screening test to assess renal dysfunction[11]. This study also explored the potential role of serum creatinine as a parameter for hospitalized CKD patients for getting cardiovascular diseases and/or cardiac arrhythmias. Even though serum creatinine is conventional and old-fashioned diagnostic tools, however, it is cost-effective, rapidly, and highly applicable to larger cohorts of CKD patients compared with the more expensive, more sophisticated, more laborious, time-consuming serum human neutrophil gelatinase-associated interleukin-18 (IL-18), cystatin-C (cys-C), lipocalin (NGAL), kidney injury molecule-1 (KIM-1), fibroblast growth factor-23 (FGF-23), carboxy-terminal fragment of insulin-like growth factor binding protein-4 (CT-IGFBP-4) and liver-type fatty acid-binding proteins (L-FABP) required for the identification of early cardiovascular diseases in acute kidney injury patients[12, 13].

In our model, serum creatinine had the highest AUC value to predict left ventricular hypertrophy (LVH). Our results might be supported by Leoncini et al (2004), which concluded that creatinine clearance proved to be a very sensitive marker of clinical and subclinical cardiovascular damage[14]. LVH is a well-established sign of cardiovascular damage and also an independent predictor of future cardiovascular morbidity in CKD patients[14]. Although conventional ECG has been thought to be a less specific method than echocardiography for detecting LVH, however, its specificity is even greater than echocardiography. Thus, the more severe the LVH is on echocardiography, the more likely it would be diagnosed on ECG[15]. Serum creatinine also had higher AUC to predict peaked T wave and long QTc interval, compared to serum potassium. This fact might be interesting because serum potassium had long been used as a strong predictor for peaked T wave. Activation of the potassium channels can stimulate speeding up membrane repolarisation, cause inactivates sodium channels can make membrane become depolarized[16]. in hyperkalemia, typical ECG findings progress from shortened QTc interval, and tall, peaked T waves [17]. Hyperkalemia stimulates activation of potassium channel and inactivation of sodium channels which causes shortening QTc interval, sluggish cardiac conduction, widening of the QRS complex and smaller P waves [16]. Moreover, in our model, serum creatinine showed a stronger predictor for peaked T wave compared to serum potassium in hospitalized CKD patients. This fact might be explained by an absence of hydro-electrolytic alterations in many CKD patients. However, calcium also plays a role in T wave amplitude. [18]. Another literature had published that hyperkalemia doesn’t always include changes in the abnormality ECG, especially in the amplitude of T wave, as a progressive elevation of serum potassium doesn’t always make the auricular muscle unresponsive[19, 20].

The phenomenon of higher serum creatinine correlate with long QTc interval could be explained by numerous factors influencing the QT interval: 1) Patients with CKD often take many medications that can cause QT prolongation; 2) Secondary hyperparathyroidism, a common complication of CKD, could lead to hypocalcemia and hypomagnesemia; 3) Renal anemia, macrocytosis and anisocytosis are also related to prolonged QT intervals due to subsequent hypoxia, autonomic dysfunction and decreased myocardial oxygen supply; 4) Impairment of delayed rectifier potassium channels; 5) Hypertension can significantly reduce potassium current densities (Ipeak, Ito, IKur, Iss, and Ik1) and increase the L-type calcium channel; and 6) Reactive oxygen species and uremic toxin leading to inflammatory condition and oxidative stress itself may also predispose to cardiac electrical remodeling[21, 22]. Therefore, the prolongation of the QTc interval is common in hospitalized patients with CKD.

Hemoglobin concentration often reflects duration and chronicity in individuals with CKD, although its level fluctuates frequently [23]. Recent studies suggest that a lower hemoglobin concentration is related with subsequent cardiovascular mortality and increased risk of coronary artery disease in hospitalized patients with CKD, such as CONFIRM-HF Trial [24], IRON OUT HF Trial [25], TIME Trial [26], GUSTO IIb Trial [27], PURSUIT Trial [28], PARAGON B Trial [29], ANCHOR Study [30], and ARIC Study [31]. On the other hand, our study explained that higher hemoglobin concentration showed a strong predictor for pathological Q wave, non-specific ST-T changes, and frontal axis deviation. We believed that the effect of higher hemoglobin concentrations on ECG abnormalities varies across various subtypes of cardiovascular diseases. If the state of anemia continues chronically, the hemodynamic changes found in anemia can contribute to left ventricular hypertrophy and progressive arterial walls [32]. In addition, the hemodynamic changes found in the higher hemoglobin concentration might be primarily determined by the viscosity of blood. Greater hematocrit concentrations would thus significantly slowing its flow rate throughout the body, thicken the blood, raising the peripheral vascular resistance, and reducing coronary blood flow and perfusion to various tissues including the cardiomyocyte. The linked, mechanisms, pathogenesis and pathophysiology of coronary artery disease (CAD) in CKD are complex. Many shreds of evidence clearly stated a significant association between lower or higher hemoglobin concentration and the development of CAD or poor outcomes in patients with known CAD, Therefore, ECG abnormalities such as pathological Q wave (reflects previous myocardial infarction), non-specific ST/T changes (reflects intrinsic myocardial diseases i.e. ischemia) and frontal axis deviation (sign of hypertrophy or reduced muscle mass or conduction abnormalities) are just suggesting the existence of the relationship between CKD and cardiovascular complication.

Because hypertension in CKD is predominantly systolic [33], we explored the relationship between systolic blood pressure (BP) with ECG abnormalities in a cross-sectional study. In our logistic regression models, only systolic BP is associated with ECG abnormalities (QTc interval prolongation or fragmented QRS, or both). The severity of diastolic dysfunction in CKD usually progressed (from grade I to grade III) with the rise in systolic BP readings. The presence of diastolic dysfunction itself also the risk factor to develop any ECG abnormalities. We did not find an association between diastolic BP with ECG abnormalities, even in bivariate analysis, regardless of BP technique. One possible reason for the lack of association between diastolic BP with ECG abnormalities maybe because the mean diastolic BP in our study population is nearly normal. In perspective, the differences between predictors of systolic BP and diastolic BP might be explainable [34].

Our study suffers from many limitations. Firstly, further risk factors for ECG abnormalities were not taken into account for the analysis. Secondly, this was a cross-sectional study, single-center with small sample sizes in once measurement and therefore we cannot imply causality from association studies. Thirdly, necessary information such as echocardiography or chest x-ray may be unavailable. Fourthly, we did not have data on medications that may have influenced cardiovascular parameters in this study. Fifthly, the interpretation of several laboratories and clinical parameters changes over time. For example, serum creatinine is a breakdown product of muscle, its level is directly associated with muscle mass, which is lower in the elderly, women and whites[35]. The strength of this study was that data already exists, prevalence estimation could be done at one point in time, complete study populations minimizing selection bias and independently collected data.

CONCLUSION

In hospitalized CKD patients, ECG abnormalities are common, where fragmented QRS and prolonged QTc were the most prevalent ECG abnormalities. Serum creatinine and hemoglobin could predict peaked T wave and prolonged QTc among hospitalized CKD patients. Systolic blood pressure could predict prolonged QTc and fragmented QRS in CKD patients. Longitudinal studies are needed to show the causality of each parameter with ECG abnormalities.

Data Availability

The datasets generated during and/or analyzed during the current study are not publicly available due to protecting participant confidentiality but are available from the corresponding author on reasonable request.

Ethical Clearance

Dr. Soetomo General Academic Hospital conferred ethical clearance for this study. Compliance with ethical standards, this research proposal was approved by the ethical committees of Dr. Soetomo General Academic Hospital in collaboration with Universitas Airlangga College of Medicine Research Ethics Council (Ref: 1811/KEPK/II/2020) under the name of Rerdin Julario as the Principal Investigator.

Conflict of Interest

None declared

Funding

None declared

Availability of Data and Material

The datasets generated during and/or analyzed during the current study are not publicly available due to protecting participant confidentiality but are available from the corresponding author on reasonable request.

Data Availability Statements

The datasets generated and analyzed during this study are not publicly available but are available by contacting the corresponding author.

Author Contribution

R.J. conceived the idea. R.A.N. designed the research and initially wrote the manuscript. B.P.D.K. and T.S.P. carried out the observations. E.P.B.M. helped the project administration. R.E.I. analyzed and interpreted the data, developed the model formulation. F.F.A. helped draft and reviewed the manuscript. All authors read and approved the final version of the manuscript.

Provenance and peer review

Not commissioned; externally peer reviewed.

Acknowledgment

The authors would also like to offer special thanks to all staff, fellows, residents, and nurses from the Department of Cardiology and Vascular Medicine & Divison of Nephrology Department of Medicine, Dr. Soetomo Academic General Hospital, Surabaya for their technical contribution.

References

  1. [1].↵
    Shastri S, Sarnak MJ. Cardiovascular Disease and CKD: Core Curriculum 2010. Am J Kidney Dis 2010; 56: 399–417.
    OpenUrlCrossRefPubMedWeb of Science
  2. [2].↵
    KDIGO. KDIGO 2017 Clinical Practice Guideline Update for the Diagnosis, Evaluation, Prevention, and Treatment of Chronic Kidney Disease–Mineral and Bone Disorder (CKD-MBD). Kidney Int Suppl 2017; 7: 1–59.
    OpenUrl
  3. [3].↵
    Lunney M, Alrukhaimi M, Ashuntantang GE, et al. Guidelines, policies, and barriers to kidney care: findings from a global survey. Kidney Int Suppl 2018; 8: 30–40.
    OpenUrl
  4. [4].↵
    Fink JC, Brown J, Hsu VD, et al. CKD as an underrecognized threat to patient safety. Am J Kidney Dis 2009; 53: 681–8.
    OpenUrlCrossRefPubMedWeb of Science
  5. [5].↵
    Saeed ZI, Hussain SA. Chronic kidney disease in Pakistan: an under-recognized public health problem. Kidney Int 2012; 81: 1151; author reply 1151-2.
    OpenUrlPubMed
  6. [6].↵
    Jayanti A, Foden P, Mitra S. Multidisciplinary staff attitudes to home haemodialysis therapy. Clin Kidney J 2017; 10: 269–275.
    OpenUrl
  7. [7].↵
    Shafi S, Saleem M, Anjum R, et al. ECG Abnormalities In Patients With Chronic Kidney Disease. J Ayub Med Coll Abbottabad 2017; 29: 61–64.
    OpenUrl
  8. [8].↵
    Levey AS, Stevens LA. Estimating GFR using the CKD Epidemiology Collaboration (CKD-EPI) creatinine equation: more accurate GFR estimates, lower CKD prevalence estimates, and better risk predictions. Am J Kidney Dis 2010; 55: 622–7.
    OpenUrlCrossRefPubMedWeb of Science
  9. [9].↵
    Hung TW, Huang JY, Jong GP. Long-term outcomes of dialysis in patients with chronic kidney disease and new-onset atrial fibrillation: A population-based cohort study. PLoS One 2019; 14: 1– 10.
    OpenUrlCrossRefPubMed
  10. [10].↵
    Liu M, Li XC, Lu L, et al. Cardiovascular disease and its relationship with chronic kidney disease. Eur Rev Med Pharmacol Sci 2014; 18: 2918–2926.
    OpenUrlPubMed
  11. [11].↵
    Drosos G, Ampatzidou F, Sarafidis P, et al. Serum Creatinine and Chronic Kidney Disease- Epidemiology Estimated Glomerular Filtration Rate: Independent Predictors of Renal Replacement Therapy following Cardiac Surgery. Am J Nephrol 2018; 48: 108–117.
    OpenUrl
  12. [12].↵
    Adeoye AM, Adebayo O, Abiola O, et al. Association between Selected Molecular BiOmarkers and AmbuLatory Blood Pressure Pattern in African Chronic Kidney Disease and Hypertensive Patients compared with normotensIve Controls: The SyMBOLIC Study Protocol (Preprint). JMIR Res Protoc 2019; 9: 1–8.
    OpenUrl
  13. [13].↵
    Chen SC, Huang JC, Su HM, et al. Prognostic cardiovascular markers in chronic kidney disease. Kidney Blood Press Res 2018; 43: 1388–1407.
    OpenUrl
  14. [14].↵
    Leoncini G, Viazzi F, Parodi D, et al. Creatinine clearance and signs of end-organ damage in primary hypertension. J Hum Hypertens 2004; 18: 511–516.
    OpenUrlCrossRefPubMed
  15. [15].↵
    Shirafkan A, Motahari M, Mojerlou M, et al. Association between left ventricular hypertrophy with retinopathy and renal dysfunction in patients with essential hypertension. Singapore Med J 2009; 50: 1177–1183.
    OpenUrlPubMed
  16. [16].↵
    El-Sherif N, Turitto G. Electrolyte disorders and arrhythmogenesis. Cardiol J 2011; 18: 233–245.
    OpenUrlPubMedWeb of Science
  17. [17].↵
    Levis JT. CLINICAL MEDICINE ECG Diagnosis?: Hyperkalemia. Perm Journal 2013; 17: 69.
    OpenUrl
  18. [18].↵
    Cobo Sánchez JL, Alconero Camarero AR, Casaus Pérez M, et al. Hyperkalaemia and haemodialysis patients: Electrocardiographic changes. J Ren Care 2007; 33: 124–129.
    OpenUrlPubMed
  19. [19].↵
    Aslam S, Friedman EA, Ifudu O. EKG is Unreliable in Detecting Potentially Lethal Hyperkalemia in Hemodialysis (HD) Patients. ASAIO J 2001; 47: 158.
    OpenUrl
  20. [20].↵
    Sharma S, Gupta H, Ghosh M, et al. Severe hyperkalemia with normal electrocardiogram. Indian J Crit Care Med 2007; 11: 215–217.
    OpenUrl
  21. [21].↵
    Liu P, Han D, Sun X, et al. Prevalence and risk factors of acquired long QT syndrome in hospitalized patients with chronic kidney disease. J Investig Med 2019; 67: 289–294.
    OpenUrlAbstract/FREE Full Text
  22. [22].↵
    Bodi I, Grünert SC, Becker N, et al. Mechanisms of acquired long QT syndrome in patients with propionic academia. Hear Rhythm 2016; 13: 1335–1345.
    OpenUrl
  23. [23].↵
    Kalantar-Zadeh K, Aronoff GR. Hemoglobin variability in anemia of chronic kidney disease. J Am Soc Nephrol 2009; 20: 479–487.
    OpenUrlAbstract/FREE Full Text
  24. [24].↵
    Ponikowski P, Van Veldhuisen DJ, Comin-Colet J, et al. Beneficial effects of long-term intravenous iron therapy with ferric carboxymaltose in patients with symptomatic heart failure and iron deficiency. Eur Heart J 2015; 36: 657–668.
    OpenUrlCrossRefPubMedWeb of Science
  25. [25].↵
    Lewis GD, Malhotra R, Hernandez AF, et al. Effect of oral iron repletion on exercise capacity in patients with heart failure with reduced ejection fraction and iron deficiency the IRONOUT HF randomized clinical trial. JAMA - J Am Med Assoc 2017; 317: 1958–1966.
    OpenUrl
  26. [26].↵
    Muzzarelli S, Pfisterer M. Anemia as independent predictor of major events in elderly patients with chronic angina. Am Heart J 2006; 152: 991–996.
    OpenUrlCrossRefPubMedWeb of Science
  27. [27].↵
    Armstrong PW, Fu Y, Chang WC, et al. Acute coronary syndromes in the GUSTO-IIb trial: Prognostic insights and impact of recurrent ischemia. Circulation 1998; 98: 1860–1868.
    OpenUrlAbstract/FREE Full Text
  28. [28].↵
    Harrington RA. PATIENTS WITH ACUTE CORONARY SYNDROMES.
  29. [29].↵
    Harrington RA. Randomized, placebo-controlled trial of titrated intravenous lamifiban for acute coronary syndromes. Circulation 2002; 105: 316–321.
    OpenUrlAbstract/FREE Full Text
  30. [30].↵
    Go AS, Yang J, Ackerson LM, et al. Hemoglobin level, chronic kidney disease, and the risks of death and hospitalization in adults with chronic heart failure - The anemia in chronic heart failure: Outcomes and Resource Utilization (ANCHOR) Study. Circulation 2006; 113: 2713–2723.
    OpenUrlAbstract/FREE Full Text
  31. [31].↵
    Sarnak MJ, Tighiouart H, Manjunath G, et al. Anemia as a risk factor for cardiovascular disease in the Atherosclerosis Risk in Communities (ARIC) study. J Am Coll Cardiol 2002; 40: 27–33.
    OpenUrlFREE Full Text
  32. [32].↵
    Rymer JA, Rao S V. Anemia and coronary artery disease: Pathophysiology, prognosis, and treatment. Coron Artery Dis 2018; 29: 161–167.
    OpenUrl
  33. [33].↵
    Andersen MJ, Khawandi W, Agarwal R. Home blood pressure monitoring in CKD. Am J Kidney Dis 2005; 45: 994–1001.
    OpenUrlCrossRefPubMedWeb of Science
  34. [34].↵
    Agarwal R, Andersen MJ. Correlates of systolic hypertension in patients with chronic kidney disease. Hypertension 2005; 46: 514–520.
    OpenUrlCrossRefPubMed
  35. [35].↵
    Bargnoux A-S, Kuster N, Cavalier E, et al. Serum creatinine: advantages and pitfalls. J Lab Precis Med 2018; 3: 71–71.
    OpenUrl
  36. [36].
    Lilly LS. Pathophysiology of Heart Disease. 6th ed. Philadelphia: Wolters Kluwer -- Medknow Publications, 2016.
  37. [37].
    Take Y, Morita H. Fragmented QRS: What Is The Meaning? Indian Pacing Electrophysiol J 2012; 12: 213.
    OpenUrlCrossRefPubMed
  38. [38].
    Surawicz B, Childers R, Deal BJ, et al. AHA/ACCF/HRS Recommendations for the Standardization and Interpretation of the Electrocardiogram: Part III: Intraventricular Conduction Disturbances A Scientific Statement From the American Heart Association Electrocardiography and Arrhythmias Committee,. J Am Coll Cardiol 2009; 53: 976–981.
    OpenUrlFREE Full Text
  39. [39].
    Kim S-H, Kwak MH, Kim HJ, et al. Prevalence and positive predictive value of poor R-wave progression and impact of the cardiothoracic ratio. Korean Circ J 2009; 39: 418–22.
    OpenUrlPubMed
  40. [40].
    Tsao CW, Josephson ME, Hauser TH, et al. Accuracy of electrocardiographic criteria for atrial enlargement: validation with cardiovascular magnetic resonance. J Cardiovasc Magn Reson 2008; 10: 7.
  41. [41].
    Sklyar E, Ginelli P, Barton A, et al. Validity of electrocardiographic criteria for increased left ventricular mass in young patients in the general population. World J Cardiol 2017; 9: 248–254.
    OpenUrl
  42. [42].
    Whitman IR, Patel V V, Soliman EZ, et al. Validity of the surface electrocardiogram criteria for right ventricular hypertrophy: the MESA-RV Study (Multi-Ethnic Study of Atherosclerosis-Right Ventricle). J Am Coll Cardiol 2014; 63: 672–681.
    OpenUrlFREE Full Text
  43. [43].
    Ibanez B, James S, Agewall S, et al. 2017 ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur Heart J 2018; 39: 119–177.
    OpenUrlPubMed
  44. [44].
    Hanna EB, Glancy DL ST.-segment depression and T-wave inversion: Classification, differential diagnosis, and caveats | Cleveland Clinic Journal of Medicine. Cleve Clin J Med 2011; 78: 404–414.
    OpenUrlAbstract/FREE Full Text
  45. [45].
    Bao H, Cai H, Zhao Y, et al. Nonspecific ST-T changes associated with unsatisfactory blood pressure control among adults with hypertension in China: Evidence from the CSPPT study. Medicine (Baltimore) 2017; 96: e6423.
    OpenUrl
  46. [46].
    Arjmand A, Eshraghi A, Sani ZA, et al. Value of pathologic Q wave in surface electrocardiography in the prediction of myocardial nonviability: A cardiac magnetic resonance imaging-based study. J Adv Pharm Technol Res 2018; 9: 162.
    OpenUrl
  47. [47].
    Pappal R, Espina I, Rafique Z. Peaked T-Waves and Potassium: The Reliability of Peaked T-Waves as Indicators of Hyperkalemia in End-Stage Renal Patients Undergoing Hemodialysis. Am J Undergrad Res; 13. Epub ahead of print August 2016. DOI: 10.33697/ajur.2016.027.
  48. [48].
    Andrianto A, Dewi PR. Clinical Implications of Low-voltage Electrocardiogram. Indones J Cardiol 2017; 206–12.
Back to top
Posted January 20, 2021.
Download PDF
Data/Code
Email
Simple Laboratory and Clinical Parameters as Predictor for Electrocardiographic Abnormalities Among Hospitalized Patients with Chronic Kidney Disease: A Cross-Sectional Study
Rerdin Julario, Ricardo Adrian Nugraha, Bagus Putra Dharma Khrisna, Tony Santoso Putra, Eka Prasetya Budi Mulia, Ryan Enast Intan, Firas Farisi Alkaff
medRxiv 2021.01.15.21249916; doi: https://doi.org/10.1101/2021.01.15.21249916
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area