Assessment of volemia status using ultrasound examination of the inferior vena cava and spectroscopic bioimpendance in hemodialysis patients

Background/Aim. Hypervolemia is an important risk factor for the development of cardiovascular morbidity and mortality in patients treated with regular hemodialysis. There is still no reliable method for assessing the status of volemia in these patients. The aim of the study was to assess the status of volemia in patients treated with regular hemodialysis by measuring the parameters of the inferior vena cava (IVC) and bioimpedance. Methods. The effect of hemodialysis treatment on ultrasound parameters of the IVC, as well as on the parameters measured by bioimpedance, was examined before and after hemodialysis. The values of the N-terminal prohormone of brain natriuretic peptide (NT-proBNP) were measured both before and after hemodialysis. Forty-five patients were involved in this non-interventional cross-section study, including the patients treated with standard bicarbonate dialysis. According to the interdialytic yield, the patients were divided into three groups: I (up to 2,000 mL), II (2,000?3,000 mL), and III (over 3,000 mL). Results. The values of the IVC parameters and the parameters measured with bioimpedance were significantly lower after treatment with hemodialysis (p < 0.005). The third group of patients had a significantly higher total fluid volume in the body com-pared to the group I, as well as a significantly greater volume of extracellular fluid (p < 0.005). The significantly lower values of NT-proBNP in all groups (p < 0.005) were detected after hemodialysis. After treatment with hemodialysis, a positive correlation was observed between the concentration of NT-proBNP in the serum and the extracellular/intracellular water ratio. However, the correlation between NT-proBNP concentration and total fluid measured by bioimpedance spectroscopy did not reach statistical significance. Conclusion. Measurement of the IVC ultrasound parameters and volemia parameters using bioimpedance significantly contributes to the assessment of the status of volemia. Nevertheless, it cannot be used as a separate parameter, only in combination with all other methods.


Introduction
Hypervolemia is an important risk factor for the development of cardiovascular morbidity and mortality, as well as adverse outcome in patients treated with regular hemodialysis 1 . The assessment of volemia status in these patients is performed by clinical examination, lung radiography, ultrasound examination of the lungs, ultrasound examination of the inferior vena cava, ultrasound of the heart, measurement of volemia using bioimpedance spectroscopy (BIS), and by monitoring of N-terminal prohormone of brain natriuretic peptide (NT-proBNP) in the serum 1 .
The goals of hemodialysis are to eliminate excess fluid, achieve adequate dry weight and depuration of the organism from uremic toxins and regulation of electrolyte imbalance.
Dry weight is the weight obtained at the end of a regular dialysis session, below which the patients will most likely develop symptomatic hypotension 2 .
Symptomatic hypotension is associated with the rate of ultrafiltration during hemodialysis treatment and the rate at which the removal is performed. Inadequate dry weight assessment leads to chronic volume overload. Removing too much excess fluid (below the dry weight) causes hypovolemia and vertigo, headache, pain and cramps in the muscles, and a decrease in perfusion of vital organs occurs. In cases where the elimination of fluid is not sufficient (weight above the dry weight), a chronic hypervolemic condition and complications occur, such as: hypertension, left ventricular hypertrophy, diastolic dysfunction, congestive heart failure and oedema of the lungs 3 .
Radiography of the heart and lungs is highly specific but with low sensitivity for the conditions of hypervolemia. Disadvantage of this diagnostic method is that it sometimes takes several hours for the radiographic changes of hypervolemia to occur. In addition, in 20 to 40% of patients, radiographic changes of hypervolemia are absent 4 . What is important when assessing the status of volemia is echocardiographic examination of the heart to assess the structure and function of the left ventricle, as well as the disorder of left ventricular systolic and diastolic function. This examination may indicate the condition of hypervolemia, but it does not give us information that the patient is in normovolemia 5,6,7 .
Ultrasound of the lungs is used to estimate the volume of fluid in the extravascular lung section and the severity of the degree of hypervolemia 8 . This method is useful in detecting pulmonary congestion even before the manifested clinical picture, and it correlates well with systolic cardiac function. The disadvantage of this method is that it may not always correlate with hypervolemia 9 .
The results of numerous studies indicate that hyper-or hypovolemia can be determined in a dialysis patient using ultrasound measurement of the diameter of the inferior vena cava (IVC) 10,11 . However, the recommended values of the IVC diameter (IVCd) that would correlate with optimal body weight are not generally accepted because of individual variations, as well as due to significant subjectivity in the measurement. In their study, Jose Muniz Pazeli, et al concluded that ultrasound of the IVC could be performed by nephrologists who have little experience in ultrasound and the findings were potentially useful for dry weight assessment in patients on dialysis 12 . The IVCd, IVC index (IVCi), and IVC collapsibilty index (IVCci) are measured using ultrasound examination of IVC. The normal diameter of the IVC is 15-20 mm, and it varies depending on the breathing cycle. In the inspirium, the diameter decreases and amounts to 0-15 mm, and in the expirium it increases and amounts to 15-20 mm. The anterior-posterior diameter of the IVC was measured subxiphoidally, at a distance of 2-3 cm from the right atrium during spontaneous breathing, forced inspiration and forced expiration (Figure 1). The IVCi is the ratio of the IVCd and the body surface (BS) of the patient IVCd/BS (mm/m 2 ). The normal IVCi is 8.0-11.5 mm/m 2 (euvolemia). The IVCi below 8 mm/m 2 indicates hypovolemia, and over 11.5 mm/m 2 is considered as hypervolemia. In order to calculate the IVCci, it is necessary to measure its diameter in inspirium (IVC insp) and expirium (IVC exp). The VCIci is calculated using the formula: IVCci = [(IVC exp-IVC insp)/ IVC exp] × 100%. If the IVCci is 50 to 75%, we are talking about euvolemia, if it is below 50% about hypervolemia, and over 75% about hypovolemia 13 .
Additionally, in assessing the status of volemia in hemodialysis patients was used to concentration of NT-proBNP in plasma, which has a significant prognostic value for cardiovascular mortality in these patients 14,15 . NT-proBNP is often increased in patients with chronic renal failure 16,17 . When examining patients with asymptomatic chronic renal failure who have not yet started dialysis treatments, an increased level of NT-proBNP was observed in more than half of the patients 17 . However, there is no clear cut-off in the literature for the concentration of NT-proBNP that would distinguish cardiac from renal failure (cut-off values range from 5000 to 7000 pg/ml). NT-proBNP is not only specific for hypervolemia but also for nutritional status, systolic and diastolic left ventricular dysfunction 18 .
Precise assessment of the status of volemia in patients on dialysis also includes the analysis of bioimpedance spectroscopy. Bioimpedance represents the total resistance of tissue and fluid to the flow of micro-amperage alternating current through the body 19 .
There are two methods that work on the principle of bioimpedance spectroscopy (BIS).
One is the so-called "whole body BIS", where the electrodes are placed on the wrist and ankle on the same side of the body, and using appropriate mathematical and physiological tissue models, we obtain data on the total amount of water in the body (TBW) and on the amount of water in tissue cells 19,20 . The second method of bioimpedance spectroscopy is a segmental BIS, which separately measures the amount of water in the extremities and the trunk. It can be done with eight electrodes, it is very precise, but the examination lasts for longer time and in practice it has not shown an advantage in relation to the whole body BIS 21,22 .
A shift in the application of bioimpedance spectroscopy to determine the condition of volemia in hemodialysis patients was provided by two studies published in 2009. These studies have shown that: 1) hypervolemia prior to dialysis treatment greater than 15% of normal extracellular fluid (approximately 2.5 L in an average person of 70 kg) correlated with a twice higher risk of fatal outcome during the 3.5 year follow-up period, compared to those patients in whom this pre-dialysis hypervolemia was lower 23 ; 2) Reduction of this "critical" hypervolemia can improve arterial blood pressure values and reduce therapy in hemodialysis patients 24 .
The aim of this study was to investigate the degree of association between the status of

Influence of the rate of ultrafiltration on the examined parameters used to assess volemia
By comparing the mean values of ECF between the three groups, it was found that they were not equal. Namely, the third group (ultrafiltration greater than 3000 ml) had significantly higher ECF values than the first group (p=0.017) and compared to the second group (p=0.025). Statistically significant difference in comparison of TBW was recorded only between the first and the third group of patients (p=0.046) in Table 2.

Mean values of extracellular fluid in examined patient groups
Statistically significant differences were not detected between the groups when the values of NT-proBNP, IVCd, IVCi, IVC insp, IVC exp, IVCci and the ECF/ICF ratio were compared. The mean values of ECF are shown in Graph 2.

Comparison of the examined parameters before and after hemodialysis
The values of laboratory findings, ultrasound (IVCd, IVCi, IVCci) and volemia parameters used to assess the clinical condition of patients were statistically significantly lower after hemodialysis, except in the measurement of IVC values (p=0.990). The values of IVCci 30 minutes after hemodialysis were statistically significantly higher compared to the values of this parameter before hemodialysis (48.15±11.38 vs. 55.45±15.07; p=0.008). This result suggests that the patients switched from the state of hypervolemia to the state of euvolemia after hemodialysis. This was corroborated by the result of weight changes before and after hemodialysis. The results of these comparisons are given in Table 3.

The degree of correlation of serum NT-proBNP values with other before-hemodialysis parameters
Positive correlation was detected of serum NT-proBNP with ultrasound parameters VCI d, VCI i, VCI ins, VCI exp, VCI ci and changes in body weight. NT-proBNP is also positively correlated with the ECF/ICF index and weight change. Persons with higher values of NT-proBNP lose more body fluid. NT-proBNP is in a negative correlation with the VCI collapsibility index, which means that higher NT-proBNP values are in correlation with lower VCI collapsibility index. That is, the patients in hypervolemia have high NT-proBNP and low VCI collapsibility index in Table 4.

The degree of correlation of serum NT-proBNP values with other parameters after hemodialysis
Correlations of NT-proBNP with ultrasound parameters and volemia condition parameters after dialysis are shown in Table 5.
It can be seen that after hemodialysis, NT-proBNP is positively correlated with volemia, body weight, and ECF/ICF index. However, a negative correlation was recorded between NT-proBNP and IVCci, which was on the border of statistical significance (p=0.060).

Discussion
The results of many studies suggest that hyper-or hypovolemia in patients on hemodialysis can be determined by ultrasound measurement of the diameter of the inferior vena cava (IVC d) 26,27 . In terms of the significance of the IVCd and the correlation of this parameter with hypervolemia, the results of previous studies are contradictory. Sonmez, et al 28 , based on the results of their study, recommend that the values of normal hydration be defined as IVCd values greater then ≥11.6 mm/m² in the expirium or above 9.8 mm/m² in the inspirium. Contrary to these results, it is usually stated in the literature that the IVCd indexed to body surface (IVCi) greater than 11.5 mm/m 2 correlates with mean pressure in the right atrium greater than 7 mmHg, or with significant hypervolemia, while this index smaller than 8 mm/m 2 correlates with significant circulatory hypovolemia. However, this index is not sufficiently accurate, as other factors such as: heart rate, arterial pressure and antihypertensive drug therapy have an effect on it.
The results of our study show that there was no statistically significant difference in the measurement of the IVCd during normal breathing, inspirium and expirium, between the three groups of patients with different degrees of ultrafiltration. A statistically significant difference was observed by comparing the IVCd before and 30 minutes after hemodialysis. After hemodialysis, the IVCd significantly decreases.
In the available literature, it is stated that the values of NT-proBNP positively correlate with the risk of mortality in patients on dialysis 29 . However, one measurement of the NT-proBNP value can't be a reliable parameter of the assessment of the condition of volemia, because it depends on both the volemia and the degree of myocardial damage.
However, serial measurement of this marker may be a good indicator of the condition of volemia measured by bioimpedance 29 . In our study we measured the values of NT-proBNP before and after hemodialysis. Our results show that the measured values of NT-proBNP before hemodialysis positively correlate with the inferior vena cava diameter. Also, the higher NT-pBNP values before hemodialysis were positively correlated with ECF/ICF and body fluid changes. However, only positive correlation between the value of this marker and the condition of volemia was recorded by measuring the value of NT-pBNP after hemodialysis. The results of our study are similar to the results of a study in which the cutoff for the value of NT-proBNP was determined (from 5000 to 7700 pg/ml) as an indicator of hypervolemia 30 . In their cross section study, Velasco, et al described an excellent correlation between the average volemia assessed by BCM before hemodialysis and the levels of NT-pBNP measured before hemodialysis in patients aged up to 72 years 30 . In the study by Paunic et al. normovolemic patients (also estimated by BIS measurement) has NT-proBNP up to 4700 pg/ml and hypervolemic has this value above 5800 pg/ml 31 .
A recent study by David, et al showed a significant correlation between serum NT-proBNP and the ratio of extracellular water and body weight only in hemodialysis patients with systolic left ventricular dysfunction, but not in those without systolic dysfunction 32 .
Although the level of NT-proBNP can be increased with the increase in volume load, the summary of currently available evidence gives the impression that this marker has a limited role in assessing the state of hydration in the dialysis population. The use of this cardiac biomarker in assessing the condition of volemia in patients on hemodialysis has not yet entered the standard procedure due to the high cost and poor specificity of the tests, as well as due to the lack of a clear criterion for the normal range of NT-pBNP values in hemodialysis patients 32 . Consistent with the above study results, the results of our study cannot provide a clear cut-off value of NT-proBNP (range 5188 to 7536 pg/ml).
Our results show that hemodialysis treatment has a statistically significant effect on the values of parameters for assessing the state of volemia measured by BIS. The values of the parameters measured by BIS after hemodialysis treatment are highly statistically significantly lower compared to the values before hemodialysis. Our results correlate with the results of the study they conducted by Chamney, et al. 33 .
Also, our results show that the parameters obtained by BIS provide objective, qualitative and useful data on the state of volemia in patients treated with hemodialysis.
The results obtained by this method correlate with the biohumoral cardiac marker NT-proBNP. Similar results were obtained in the study by Velasco, et al. 30 .

Conclusion
Neither the inferior vena cava diameter nor bioimpedance spectroscopy nor NT/proBNP measurement can't be used as a stand alone test for monitoring the condition of hydration of patients on hemodialysis. The results obtained by these methods must be interpreted individually and adapted for each patient. Only in this way, these parameters could add the value to the clinical judgment of an individual hemodialysis patient optimal body weight.

Conflict of interest:
None declared.