Hypertension is a major cause of cardiac disease in patients on hemodialysis (Purcell et al., 2004). The pathogenesis of hypertension is multifactorial and is not completely understood (Kuhlmann et al., 2005), yet hypervolemia has always been considered a major pathogenic factor (Lopez-Gomez et al., 2005). Whereas control of blood pressure in hemodialysis patients using anti-hypertensive medications is feasible, recent work has emphasized that the better control of fluid status would more effectively prevent the damaging effects of fluid overload and hypertension while avoiding the side effects of anti-hypertensive medications (Steuer et al., 1998).
There are several studies that assessed the effect of volume status on blood pressure (Kirchner, 1997; Luik et al., 1997; Jain et al., 2001; Leypoldt et al., 2002; Michael et al., 2004) These studies, however, showed conflicting data. In addition, most of these studies were performed in adults (Kirchner, 1997; Luik et al., 1997; Jain et al., 2001); very few data on children are published (Jain et al., 2001; Michael et al., 2004). The present study aimed to assess the correlation between hypertension and excess volume parameters in pediatric patients with end stage renal disease on chronic hemodialysis, to show whether adequate control of volume status and adequate estimation of dry weight would effectively normalize blood pressure in this population. Pre- and post dialysis plasma proteins were used to calculate the intradialytic change in plasma volume. A previous study used the same approach on adult population (Leypoldt et al., 2002).
MATERIALS AND METHODS
The study was performed on 50 pediatric patients with ESRD undergoing maintenance hemodialysis in the Center of Pediatric Nephrology and Transplantation, Children Hospital, Cairo University. Patients who recently started hemodialysis with unidentified dry weight were excluded from the study.
Hemodialysis machines with volumetric control (Fresenius Medical Care 4008B and 4008S) were used. The standard dialysis bath consisted of sodium 135 mEq L-1, potassium 2 mEq L-1, calcium 3 mEq L-1 and bicarbonate 35 mEq L-1. Dialysis prescription was based on delivering a minimum single-pool Kt v-1 of 1.2. Dialysis session duration ranged from 2.5-3.5 h. Blood pump flow ranged from 80-180 mL min-1. Dialysis sessions were prescribed 3 times/week in most patients except for 4 patients who were dialyzing 4 times/week. The ultrafiltration rate was programmed to reach the patients optimal dry weight which is defined as the post dialysis body weight below which the patients developed symptomatic hypotension or muscle cramps in the absence of edema (Purcell et al., 2004). Dialyzer size was chosen to match each patients body surface area and to limit the circuit extracorporeal volume to <10% of patients total body volume. Filters used were polysulfone (Fresenius USA) with surface areas 0.4, 0.7, 1.0, or 1.3 for F3, F4, F5 and F6, respectively. The dialysate flow rate (Qd) was 500 mL min-1. Heparin was used for anticoagulation. Some patients had heparin free dialysis, for example those having thrombocytopenia or bleeding tendency.
We prospectively studied 150 sessions in 50 patients; the 3 sessions of the week were included for every patient. Data collected from the patients included demographic data, original renal disease, dry weight, antihypertensives given, co-morbid conditions, dialysis prescription and efficiency of dialysis as measured by Kt v-1 calculated in the month of the study.
Pre- and post dialysis body weights were measured by a standardized digital scale, pre- and post dialysis blood pressures were measured using a standardized mercurial sphygmomanometer with the patient in the sitting position. Predialysis blood samples were taken directly from the blood line and post dialysis samples taken from the arterial side after reducing pump speed for 30 sec. Blood samples were centrifuged after collection and the plasma was sent for determination of total protein concentration by electrochemiluminescence. The test is performed on the dimension Dade Behring.
The intradialytic decrease in body weight, mean arterial blood pressure and intradialytic change in blood pressure were calculated.
The intradialytic decrease in plasma volume was calculated from the pre- and
post dialysis total protein concentrations and expressed as a percentage of
the plasma volume at the beginning of hemodialysis. It was calculated using
the following equation:
Where TP denotes the total plasma protein concentration and subscripts pre
and post denote predialysis and postdialysis values, respectively.
Statistical analysis: SPSS for Windows, version 7.5 computer program
was used for statistical analysis. A p value of less than 0.05 was considered
statistically significant. Data are represented as the mean±SD. The t-test
was used to compare between two independent means. Paired-Samples t-test was
used to compare between two dependent means. One-way analysis of variance was
used to compare between more than 2 independent means. Pearson correlation coefficient
r was used to measure the linear relationship between different continuous variables.
Multiple regression analysis was used to provide us with an equation to estimate
value of criterion (dependable) variable using values in predictor variables.
The study included 50 patients with ESRD on maintenance hemodialysis. Their age ranged from 2-16 years (mean age 9.98±3.32 years); their weight ranged from 7.6-45 kg (mean 21.58±7.0 kg). They consisted of 32 males and 18 females with a male to female ratio of 1.8:1.
Most of patients had an unknown etiology of renal disease (46%). The most common known etiology was obstructive uropathy (28%), followed by cystinosis (8%), focal segmental glomerulosclerosis (6%), primary oxalosis (4%), polycystic kidney disease (2%), hemolytic uremic syndrome (2%), medullary cystic disease of the kidney (2%) and mesangioproliferative glomerulonephritis (2%).
Only 8 of our patients (16%) were not receiving anti-hypertensive medications, 19 patients (38%) were receiving 1 drug, 19 patients (38%) were receiving 2 drugs and 4 patients (8%) were using 3 drugs. Nifedipine was the most commonly used anti-hypertensive (38/42, 90% of hypertensive patients) followed by captopril (25/42, 59% of hypertensive patients) and prazocin (6/42 patients, 14.2% of hypertensive patients).
A statistically significant drop of systolic, diastolic and mean blood pressures occurred following the hemodialysis sessions (Table 1).
The intradialytic decrease in body weight ranged from 0.1-2.5 kg with a mean of 0.99±0.54. The plasma protein concentration increased after dialysis; predialysis values ranged from 4.2-7.8 mg dL-1 (mean 6.1±0.63) and post dialysis values ranged from 5.0-11.3 mg dL-1 (mean 7.0±1.02).
The pre- and post dialysis plasma protein concentrations were used to calculate the percent of decrease of plasma volume. It ranged from 15-40% (mean 17±9%).
There was a significant positive correlation between intradialytic decrease
in systolic, diastolic and mean blood pressures with decrease in plasma volume
% (Fig. 1A-C), as well as with decrease in body weight (r
= 0.27, p = 0.001, r = 0.37, p<0.001; r = 0.23, p = 0.005, respectively).
||Mean±SD systolic, diastolic and mean blood pressures
before and after hemodialysis session
||Scatter diagrams and regression lines of Intradialytic decrease
in systolic (A), diastolic (B) and mean blood pressure (C) versus decrease
in plasma volume percent
|| Correlation (r) between predialysis systolic, diastolic and
mean blood pressures versus intradialytic decrease in body weight and plasma
|| Intradialytic decrease in plasma volume % and body weight
% in relation to the order of the session
|*There is a statistically significant difference between Group
1 versus Groups 2 and 3 (p<0.05)
||Scatter diagrams and regression lines of Kt v-1
versus decrease in body weight percent (A) and plasma volume percent (B)
Also significant positive correlations between predialysis systolic, diastolic
and mean blood pressures versus decrease in body weight and plasma volume %
were present (Table 2).
We calculated the decrease of body weight as a percent of total patients body weight instead of being used as an absolute number, in order to normalize this decrease to patients size; and correlated this with the same parameters. Similar results were obtained.
Regression analysis showed that each 1 kg decrease of body weight was associated with 19 mm Hg decrease in systolic blood pressure and 13 mm Hg decrease in diastolic blood pressure and each 10% decrease of plasma volume was associated with 15 mm Hg decrease in systolic blood pressure and 10.5 mm Hg decrease in diastolic blood pressure.
The correlation between the KT V-1 (which is a measure of dialysis efficiency) with the percent of decrease in body weight and decrease of plasma volume is shown in Fig. 2A and B.
Correlation of the intradialytic decrease in plasma volume and body weight percents with the order of the session is shown in Table 3.
Hypertension is common in chronic hemodialysis patients and likely contributes to the excess morbidity and mortality in these patients (Port et al., 1999; Mazzuchi et al., 2000). Several previous studies have examined the effect of interdialytic weight gain or intradialytic reduction in body weight on blood pressure (Charra and Chazot, 2003; Kuhlmann et al., 2005). However, this approach does not assess postdialysis volume status. Moreover, the vast majority of these studies were performed in adults. In our study, we determined the association of the intradialytic decrease in body weight (as an indicator of interdialytic fluid gain) and the intradialytic decrease in plasma volume (as an indicator of postdialysis volume status) with predialysis and the intradialytic drop of blood pressure in pediatric patients.
The percent of hypertensive patients in our study (84%) was higher than in
other studies. In a study by Grekas et al. (1998) which included 113
patients on regular hemodialysis, 46 patients (40%) were not receiving antihypertensives,
26% were receiving 1 drug, 22% were receiving 2 drugs and 10.5% were receiving
3 drugs. In the study by Chen et al. (2002) only 51% of their patients
were hypertensive. Both studies were performed in adults. In the study performed
in children by Sorof et al. (1999) the percent of hypertensive patients
was also lower than in our study where only 5 of 12 patients (41.6%) were hypertensive,
whereas in another study by Michael et al. (2004) 40% of hemodialysed
children required anti-hypertensive medication. This higher percent of hypertensive
patients in our study may be due to the lower age range of our patients, even
when compared to studies performed in children. The low compliance to antihypertensive
drugs and difficulty of dietary regulation may explain this situation.
The intradialytic decrease of plasma volume was calculated as a fraction of the predialysis value by an equation using the change in plasma protein concentration. The percent of decrease of plasma volume ranged from 1.5-40% with a mean of 17±9%. Some studies have proposed alternative methods of evaluating postdialysis volume status such as the diameter of the inferior vena cava (Katzarski et al., 1997), the plasma concentration of atrial natriuretic peptide (Zucchelli et al., 2001; Wallin et al., 2004), extracellular volume assessed by multifrequency bioelectric impedance (Katzarski et al., 1997), or total body water volume assessed by single frequency bioelectric impedance (Dionisio et al., 1997; Lins et al., 1997). In other studies which used plasma proteins for assessment of postdialysis volume status (Leypoldt et al., 2002; Kuhlmann et al., 2005) the mean decrease of plasma volume was approximately 10±9.5% which is lower than that of our study. Their studies, however, were conducted in adults. This shows that the volume overload may play a more important role in pathogenesis of hypertension in children than in adults with ESRD on maintenance hemodialysis.
In the present work, intradialytic decrease of systolic, diastolic and mean blood pressures were significantly correlated with intradialytic decrease of body weight and plasma volume percent, in contrast to Leypoldt et al. (2002) who found positive correlation with the decrease in systolic blood pressure only. They attributed this to the relative difficulty in accurately measuring diastolic blood pressure.
There was a significant positive correlation between the magnitude of decrease of body weight and plasma volume with predialysis systolic, diastolic and mean blood pressures (p<0.001 in all), which indicates that larger interdialytic weight gain leads to higher predialysis blood pressures in agreement with Grekas et al. (1998) and Leypoldt et al. (2002). However, Sherman et al. (1993) in their study reported only limited correlation between interdialytic weight gain and predialysis blood pressure. In a study by Fishbane et al. (1996) they have prolonged the dialysis time to 8 h/session to allow for complete fluid removal. They reported that <2% of their patients continued to require antihypertensive medications and concluded that volume may be the primary cause of hypertension in the vast majority of hemodialysis patients and that dialysis-refractory hypertension may simply be due to inadequate volume removal.
All analyses were performed using the magnitude of the intradialytic decrease in body weight not normalized by patient size, so we had performed similar analyses using this parameter as a percent of total body weight. We correlated the percent of decrease of body weight with the predialysis systolic and diastolic blood pressures and found a significant positive correlation. We then correlated the percent of decrease of body weight with reduction of systolic and diastolic blood pressures (the difference between the pre- and post dialysis blood pressures) and also a positive significant correlation was found. This may indicate that the use of absolute decreases of body weight and the percent of decrease of body weight give nearly identical results. This is in agreement with Leypoldt et al. (2002) who reached a similar conclusion.
We looked for an association between Kt v-1 and the volume status. We found that the more the Kt v-1, the more the decrease in percent of body weight and plasma volume and this association was statistically significant. This may be explained by the fact that calculation of Kt v-1 uses time, ultrafiltration and postdialysis weight as parameters, in addition to pre- and post dialysis blood urea and thus prolonging time of dialysis and more ultrafiltration is expected to lead to a higher Kt v-1. Whether these higher values really indicate a better dialysis efficiency needs to be further investigated. As mentioned before, when dialysis time was prolonged to increase ultrafiltration in the study by Fishbane et al. (1996) this led to more control of blood pressure but they did not correlate this with the Kt v-1.
To further clarify the effect of volume status on blood pressure, correlation was done between intradialytic decrease in body weight and plasma volume with order of the session of the week. It is usual that the first session of the week shows a higher interdialytic weight gain due to the longer interdialytic interval (Harris and Stewart, 1994). Present results showed that the intradialytic decrease in body weight and the intradialytic decrease in plasma volume were highest in the first session of the week, compared to the second and third sessions, but was statistically significant only with the plasma volume reduction.
In conclusion, this study shows that volume status influences both predialysis and postdialysis blood pressure. The intradialytic reduction in body weight and plasma volume percent are helpful in describing volume status and predicting blood pressure in chronic hemodialysis patients. The effect of volume status on blood pressure may be higher in pediatric population than adults.