Fluid overload is an important factor of morbidity and mortality in hemodialysis patients, and providing normovolemia is an important component of the prescription and sufficiency of hemodialysis. Also known as “dry weight”, normovolemia is attempted to be estimated with several methods such as clinical evaluation, echocardiography, the diameter of the inferior vena cava, lung ultrasonography, and bioimpedance, and through measurement of biochemical parameters such as brain natriuretic peptide (BNP) or N-terminal-pro brain natriuretic peptide (NT-proBNP). Because overestimated dry weight may lead to serious morbidities such as fluid load, pulmonary oedema, cardiac failure, and hypertension, while underestimated dry weight may cause muscle cramps, intradialytic hypotension and associated ischemic attacks, resulting in increased mortality in hemodialysis patients.1–5 Today dry weight which is attempted to be predicted by clinic evaluation and combination of these methods remains an important issue of hemodialysis practice. Within this context, it is subjected to new method searching. Tissue Doppler Imaging (TDI) technique analyzes myocardial velocities, providing an evaluation of the cardiac functions. In clinical practice, it allows evaluating systolic and diastolic functions of the ventricles regionally or as a whole. Positive or negative Doppler waves are obtained according to the movement direction of myocardium at systole and diastole (Fig. 1). Since the obtained data belong only to the region of sampling, systolic and diastolic functions of the myocardium can be evaluated individually for each segment. TDI can detect diastolic and systolic dysfunctions that are resulted from ageing, hypertension, ischemic heart disease, aortic stenosis, hypertrophic cardiomyopathy and various diseases showing myocardial involvement. Parameters obtained with TDI can be used not only for diagnosis but also for the prediction of prognosis and evaluation of the effects of treatment. Early diastolic myocardial peak velocity (é) has been shown to have a linear association with the decrease rate of left ventricular pressure at early diastole.6 Regional systolic dysfunction can be detected with tissue Doppler imaging in ischemic heart diseases. Ventricular systolic myocardial peak velocity (S′) was found lower in the infarction areas with TDI.7 Although heart catheterization is the gold standard method in measurement of the ventricular filling pressure which is an important marker of the volume status, its use in the practice is limited since this method is expensive and not reproducible. With the TDI technique, left ventricular end-diastole pressure can be calculated by evaluating the transmitral flow. The relationship between ventricular filling pressure changes and Doppler velocity changes has been shown with simultaneously made invasive left ventricular pressure measurements and Doppler measurements.8 The ratio of mitral flow early diastolic velocity (E) to early diastolic myocardial peak velocity obtained from the mitral annulus with TDI (é), ie (E/é) is correlated with invasively measured left ventricular end-diastole pressure. When E/é>10, left ventricular end diastole pressure was found to be>15mmHg with 85% sensitivity, and 77% specificity. Again using these data, estimated pulmonary capillary wedge pressure can be calculated using the formula of ePCWP=1.9+1.24 (E/é).9

Fig. 1.

Positive or negative waves in pulsed wave doppler and tissue doppler imaging. Abbreviations: A; mitral late diastolic peak velocity, A′; late diastolic myocardial peak velocity, DT; deceleration time, E; early mitral diastolic peak velocity, é; early diastolic myocardial peak velocity, IVC; isovolumetric contraction, IVRT; isovolumetric relaxation time, S′; systolic myocardial peak velocity.

(0.25MB).

Studies evaluating cardiac function with this method in chronic kidney patients are limited and quite new. It is possible to obtain important data for the evaluation of volume load in the cases of end-stage renal disease (ESRD) with TDI because of the detailed information about ventricular filling pressure provided by this method. In this study, we aimed to evaluate the relationship between ePCWP calculated with TDI and the other parameters that are related to volume load.

No significant difference was found between the groups in terms of demographics and biochemical analyses (Tables 1 and 2).

When the groups were compared before dialysis; ePCWP (26.5±4.8 vs 14.1±2.6, p=0.0001), systolic blood pressure (161±35 vs 126±27mmHg, p=0.008), mean arterial pressure (113±20 vs 94±19mmHg, p=0.026) and pulse pressure (73±28 vs 48±17mmHg, p=0.011) values were statistically significantly higher in the second group. Comparing the echocardiographic data; parameters related to volume load including left atrial diameter (4.6±0.3 vs 4.1±0.7cm, p=0.021), left atrial volume (9.5±2.4 vs 7.6±2cm3, p<0.034), E/é ratio (18.9±3.4 vs 10.1±2.6, p=0.0001) and E/Vp (2.8±0.5 vs 1.9±0.6, p=0.0001) ratio was statistically significantly higher in Group 2 (Table 3).

On the other hand, a strong correlation was found between predialysis ePCWP and volume related hemodynamic parameters including systolic blood pressure (r=0.420, p=0.006), mean arterial pressure (r=0.364, p=0.019) and pulse pressure (r=0.430, p=0.005), biochemical parameters such as NT-ProBNP (r=0.472, p=0.002), and echocardiographic parameters such as left atrial diameter (r=0.421, p=0.006), E/é ratio (r=0.952, p=0.0001) and E/Vp ratio (r=0.358, p=0.022) (Table 4).

When the groups were compared in the post-dialysis period; ePCWP (20.5±2.9 vs 12.7±3.4, p=0.0001), and pulse pressure (65±22 vs 44±19, p=0.023) were significantly higher, and pulse (73±11 vs 87±17, p=0.019) was significantly lower in the second group.

No significant difference was found between the groups in terms of the other hemodynamic parameters evaluated. Comparing echocardiographic data in the post-dialysis period; left atrial diameter (4.2±0.4 vs 3.6±0.7, p=0.011), left atrial volume (8.9±1.6 vs 7±1.8, p=0.012), left atrial volume index (5.6±1.7 vs 4.4±1.1, p=0.021), stroke volume (92.8±18.6 vs 77.9±29.6, p=0.043), E/é ratio 15.6±3.1 vs 8.7±2.5, p=0.0001), and E/Vp ratio (2.3±1.1 vs 1.4±0.6, p=0.040) were significantly higher in Group 2 (Table 5).

On the other hand, it was found that the correlation between ePCWP and volume status related biochemical, hemodynamic, and echocardiographic parameters such as systolic blood pressure (r=0.313, p=0.046), pulse pressure (r=0.319, p=0.042), pulse (r=−0.340, p=0.030), NT-ProBNP (r=0.365, p=0.019), left ventricular end-systolic diameter (r=0.317, p=0.043), left atrial diameter (r=0.46521, p=0.002), left atrial volume (r=0.385, p=0.013), left atrial volume index (r=0.329, r=0.036) E/é ratio (r=0.960, p=0.0001) and E/Vp ratio (r=0.478, p=0.002) was continued in the post-dialysis period, similar to the pre-dialysis period (Table 6).

A new method based on estimated PCWP by tissue doppler has been presented which provides an objective prediction of optimal volume status or dry weight in this paper. The annual rate of mortality from cardiovascular diseases in end-stage renal failure is 9%, accounting for 50% of deaths.11 Cardiovascular problems largely begin in the predialysis period and continues with an increase in size during dialysis. A high portion such as 25–50% of patients who newly starts to dialysis had a cardiac failure, and this causes a dramatic decrease in survival rates. Hypervolemia induced increase in the left ventricular pressure and volume is an important factor which significantly increases mortality and morbidity.12–14 Hypertension (HT) which is a very important cardiac risk factor varies according to the cause and stage of CKD (chronic kidney disease), although it is seen between 60% and 100% with the most important cause being increased intravascular volume.15,16 On the other hand, increased volume load increases ventricular diastolic pressure, decreasing coronary perfusion pressure.

Over fluid load is one of the most common, and reversible risk factors associated with a high mortality rate in HD patients.17 With this regard, the determination of dry weight correctly is important. On the other hand, the correct estimate of dry weight is an important and challenging issue of hemodialysis practice, and remain a great problem for nephrologists. It is attempted to be predicted with a combination of clinical evaluation and the data obtained with many old and new methods. None of the physical and other methods gives definite results.18 Today there is not gold standard to evaluate volume status in ESRD patients, and also there is no a uniform standard for the definition of dry weight. Alternative methods used for prediction of dry weight have significant limitations. This brings new seeking together.

Brain natriuretic peptide (BNP) levels are widely used to estimate fluid overload in hemodialysis patients.19,20 Similarly to BNP, NT proBNP levels are strongly associated with fluid overload, and thus functions as guidelines for fluid load in HD patients.21–24 However, BNP levels also increase with the other causes of myocardial stress.25 BMP levels were persistently high in patients with chronic heart failure despite a basal dry weight.26 These factors limit the reliability of BNP in the prediction of dry weight.

Bioimpedance spectroscopy (BIS) is a promising application which has been come into prominence recently the correct determination of dry weight. It has been reported that heart rate variability and blood pressure parameters can be used together with BIS in the determination of dry weight.27–29 However, skin changes, inadequate adhesion to the electrodes, electrical parasites and obesity limit the use of the BIS method for determination of dry weight.28,30 Lung ultrasonography (LUS) is a new and simple method which appropriately evaluates extravascular lung water (EVLW).31 In the presence of interstitial–alveolar fluid transition, vertical artefacts resulted from the pleura appear. These are called B-lines. The number of B-lines depends on the degree of pulmonary ventilation loss. Because a positive linear correlation has been shown between the amount of B-lines and EVLW.32–34 As EVLW increases, air content decreases and the number of B-lines increases in the lungs.35 These observations indicate that early-onset pulmonary congestion can be detected with LUS. It has been shown that B-lines significantly decrease as the result of decreased pulmonary congestion with the treatment of acute cardiac failure.36 It has been reported that LUS is a reliable method in the detection of the rapid changes in extravascular lung fluid in hemodialysis patients.37,38 On the other hand, it has been stated that LUS may be a more suitable method together with BIS in the determination of dry weight.39 Within this context, LUS comes into prominence with being a non-invasive method which can be performed at the bedside, rapidly and in every patient position, and its use is increasingly becoming widespread.

However, being person dependent, obesity, presence of emphysema and rib artefacts limit its suitability. The fluid load is an important cause of increased blood pressure in hemodialysis patients. The volume increase is significantly correlated with 24-h arterial pressure and randomized pre-dialysis blood pressure measurements, and blood pressure values are decreased with normalization of the hydration status of the patient.40 From this point of view, it has been proposed that blood pressure parameters, and especially pre-dialysis pulse pressure, post-dialysis pulse pressure, and systolic blood pressure can be used as a reference in predicting dry weight.41 However, besides volume status many factors change hemodynamic parameters including hemodialysis procedure itself,42 limiting blood pressure parameters to be used as appropriate criteria.

Inferior vena cava diameter (IVCD) and collapse index have been shown to reflect individual fluid status.43 Although there are numerous studies showing the validity of IVCD in the determination of dry weight, optimal timing for a rebalancing of interstitial and intravascular compartments, and post-HD after some time of HD is not clear, measurements show personal variability and are influenced by the right heart valve pathologies, limiting its reliability.44 Again in a study with BIS, no correlation could be demonstrated between IVCD and BIS volume parameters.18,45,46

Blood volume monitoring (BVM) is a commonly used technique in guiding ultrafiltration (UF) in HD. However, percent change in blood volume (BV) during HD sessions depends not only on the amount of extracellular fluids but also on several factors such as oncotic pressure and more importantly rate of UF. Therefore, BV is a weak marker of volume load in HD patients.47 Besides a correct prediction, an important factor in these methods is practicability and reproducibility.48

In summary, many methods have been developed in the evaluation of dry weight, and are currently being developed. In the present study, a method with high practicability and reproducibility which could be an alternative to the above-mentioned methods was presented. It has been reported that acute volume changes in ESRD patients is associated with é velocity determined by TDI, and E/é ratio is an important parameter in evaluating the left ventricular (LV) functions.49 Also, a high E/é ratio has been shown to better predict cardiovascular events compared to the other echocardiographic parameters in hemodialysis patients.50 Within this context, it has been suggested that only conventional evaluation of LV systolic and diastolic functions would not be sufficient, and they should also be further evaluated with TDI. In their TDI study with HD patients, Drighil et al. reported a significant decrease in post-HD LV é velocity, which was associated with the amount of UF. At the end of that study, the authors reported that LV and RV systolic and diastolic myocardial velocities are preload dependent.51 In another study, a direct linear correlation has been shown between E/Vp and E/é ratios and left ventricular filling pressure52. In their TDI study with 220 HD patients, Wang et al. reported that E/é ratio was increased in 62% of the cases, this increase showed substantial correlations with LV volume index, residual GFR loss, advanced age, and poor EF, and that the increase in E/é ratio during a median follow-up of 48 months was an independent predictor of cardiovascular mortality and all-cause mortality.53 Recent studies have shown that E/é ratio is associated with high levels of NT-pro BNP and troponin T, and correlated with BIS data.54,55

In the present study, we found statistically significantly higher volume load related parameters such as left atrial diameter (p=0.021), left atrial volume (p=0.034), E/é (p=0.0001), E/Vp (p=0.0001), systolic blood pressure (p=0.008), mean arterial pressure (p=0.026) and pulse pressure (p=0.011) in the group with high ePCWP value in the pre-dialysis period. In addition, we found as statistically significant correlation between ePCWP and volume load related parameters such as systolic blood pressure (r=0.420, p=0.006), mean arterial pressure (r=0.364, p=0.019), pulse pressure (r=0.430, p=0.005), NT-ProBNP (r=0.472, p=0.002), left atrial diameter (r=0.421, p=0.006), E/é ratio (r=0.952, p=0.0001) and E/é ratio (r=0.358, p=0.022). It was found that, similar to pre-dialysis period, the correlation between ePCWP and systolic blood pressure (r=0.313, p=0.046), pulse pressure (r=0.319, p=0.042), pulse (r=−0.340, p=0.030), NT-ProBNP (r=0.365, p=0.019), left ventricular end-systolic diameter (r=0.317, p=0.043), left atrial diameter (r=0.46521, p=0.002), left atrial volume (r=0.385, p=0.013), left atrial volume index (r=0.329, r=0.036) E/é ratio (r=0.960, p=0.0001) and E/Vp ratio (r=0.478, p=0.002) parameters continued also in the post-dialysis period. Even in the post-dialysis period pulse was significantly lower in the group with high ePCWP. This was evaluated as a reflection of improved myocardial stress depending on the decrease in volume load. From this aspect, ePCWP can be used as a novel method in determination of dry weight.

In the light of our results, it can be predicted that TDI method could provide a significant contribution in prediction of dry weight by giving information about ventricular filling pressures, and its ability to estimate PCWP, which is the gold standard of volume load.

The major limitations of this study is being a single-centre study with a relatively small number of patients. Nevertheless, these efforts and results are statistically significant.

Prediction and maintenance of dry weight are important in hemodialysis patients because of its short term effects on morbidity and mortality. From this aspect, the combined use of inferior vena cava diameter, BIS, blood volume monitoring, and methods such as ePCWP which is estimated by PUS or TDI together with physical examination findings would be appropriate for a prediction of more correct dry weight. On the other hand, further large clinical studies are needed to determine the effectiveness and benefits of TDI method and whether it is superior over the other methods.

The authors declare no conflicts of interest.