Intradialytic Hypertension and the Association with Interdialytic Ambulatory Blood Pressure

Abstract

Summary

Background and objectives

Increasing BP during maintenance hemodialysis or intradialytic hypertension is associated with increased morbidity and mortality. In hemodialysis patients, ambulatory BP measurements predict adverse cardiovascular outcomes better than in-center measurements. We hypothesized that patients with intradialytic hypertension have higher interdialytic ambulatory systolic BP than those without intradialytic hypertension.

Design, setting, participants, & measurements

We performed a case-control study in adult hemodialysis patients. Cases consisted of subjects with intradialytic-hypertension (systolic BP increase ≥10 mmHg from pre- to posthemodialysis in at least four of six treatments), and controls were subjects with ≥10 mmHg decreases from pre- to posthemodialysis in at least four of six treatments. The primary outcome was mean interdialytic 44-hour systolic ambulatory BP.

Results

Fifty subjects with a mean age of 54.5 years were enrolled (25 per group) among whom 80% were men, 86% diabetic, 62% Hispanic, and 38% African American. The mean prehemodialysis systolic BP for the intradialytic-hypertension and control groups were 144.0 and 155.5 mmHg, respectively. Mean posthemodialysis systolic BP was 159.0 and 128.1 mmHg, for the intradialytic-hypertension and control groups, respectively. The mean systolic ambulatory BP was 155.4 and 142.4 mmHg for the intradialytic-hypertension and control groups, respectively (P = 0.005). Both daytime and nocturnal systolic BP were higher among those with intradialytic hypertension as compared with controls. There was no difference in interdialytic weight gain between groups.

Conclusions

Time-integrated BP burden as measured by 44-hour ambulatory BP is higher in hemodialysis patients with intradialytic hypertension than those without intradialytic hypertension.

Introduction

Hypertension (HTN) is highly prevalent in end-stage renal disease ESRD patients on maintenance hemodialysis (HD). However, whereas high BP is common, BP targets are difficult to establish because the linear relationship between BP and mortality existing in the general population (1) does not exist in HD patients using HD-unit BP measurements (2,3). Furthermore, individual HD-unit measurements do not necessarily reflect interdialytic BP burden or intradialytic BP changes. Increased interdialytic ambulatory BP (ABP) and intradialytic HTN (defined as increases in BP from pre- to post-HD) are both associated with increased morbidity and mortality in HD patients (4–8). However, it is unknown what the overall interdialytic BP burden is in patients with intradialytic HTN defined by intradialytic BP changes that were associated with adverse outcomes in retrospective studies (5,6).

The prevalence of intradialytic HTN is estimated to be 5 to 15% of the HD population (9,10). Mechanisms proposed to explain intradialytic HTN include (1) volume overload, (2) renin-angiotensin-aldosterone and sympathetic nervous system activation, (3) removal of antihypertensive medications during dialysis, (4) endothelial cell dysfunction, and (5) electrolyte imbalances involving dialysate sodium, calcium, or potassium. Recent reviews discuss these mechanisms further (11–13). Intradialytic HTN, when defined as systolic BP increases ≥10 mmHg from pre- to post-HD, has recently been associated with increased risk for short-term (6 months) and long-term (2 years) morbidity and mortality (5,6). However, these studies were unable to discern what other possible factors associated with intradialytic HTN contributed to the increased risk. The question remains whether intradialytic HTN is a marker for modifiable risk factors already known to affect cardiovascular morbidity and mortality. The purpose of this study was to determine whether the interdialytic BP burden differs between patients with intradialytic HTN, defined in this study as systolic BP increases ≥10 mmHg from pre- to post-HD, and maintenance HD control patients whose BP decreases during HD. We hypothesized that mean 44-hour interdialytic systolic ABP would be higher in intradialytic HTN patients compared with HD controls.

Study Population and Methods

Subjects

ESRD patients on maintenance HD at three University of Texas Southwestern-affiliated units were screened for enrollment using consecutive sampling to identify eligible subjects. Inclusion criteria were (1) age 18 to 80 years, (2) HD vintage >1 month, (3) ability to consent, (4) achievement of estimated dry weight, and (5) pre-HD systolic BP ≥140 mmHg or post-HD systolic BP ≥130 mmHg and either of the following BP patterns: (1) case subjects were defined as subjects with systolic BP increases from pre- to post-HD ≥10 mmHg in at least four out of the six HD treatments before enrollment or (2) control subjects were subjects with systolic BP decreases from pre- to post-HD ≥10 mmHg in at least four out of the six HD treatments before enrollment. The major exclusion criteria were (1) open wound or active infection and (2) active malignancy.

All of the subjects signed written, informed consent before any procedures. The protocol was approved by the Institutional Review Board at the University of Texas Southwestern Medical Center-Dallas. All of the procedures were in accordance with the Declaration of Helsinki. This study was part of a registered clinical trial entitled Mechanisms and Treatment of Intradialytic Hypertension (14).

Design

This was a case-control study where cases were defined by the presence of intradialytic HTN, and the dependent variable was the interdialytic systolic ABP. Case subjects were identified and enrolled, and control subjects were enrolled within strata to ensure equal distribution of the variables age (<40 years, 40 to 60 years, and >60 years), sex, and presence/absence of diabetes between groups.

Study Procedures

Hemodialysis Procedure and Weight Measurement.

Each subject's University of Texas Southwestern nephrologist prescribed the treatments and assessed the estimated dry weight. Pre-HD weight was measured on arrival to the unit on a standing scale and recorded in kilograms. Post-HD weight was measured after HD on the same scale. Pre- and post-HD weights from the treatments just before and after the ABP was measured were analyzed. Interdialytic weight gain was the difference between the pre-HD weight after completion of ABP measurements and the post-HD weight just before the ABP measurements. The percentage of interdialytic weight gain was 100 times interdialytic weight gain divided by the estimated dry weight.

Hemodialysis Blood Pressure Measurements.

BP was measured and recorded by an automated sphygmomanometer. Pre-HD BP was measured in the nonaccess arm after a 5-minute rest while the patient was seated with both feet on the floor before placement of a dialysis needle. Five minutes after HD, BP was measured again in a similar fashion. Data for the six treatments (2 weeks) before the ABP measurement was averaged for the pre- and post-HD systolic and diastolic BP. Similarly, the “mean HD-unit” systolic BP was calculated by averaging all pre- and postsystolic BP measurements during those 2 weeks.

Ambulatory Blood Pressure Measurements.

After a mid-week HD treatment, subjects wore a Spacelabs 90207 ABP monitor (Redmond, WA) on the nonaccess arm placed by the principal investigator or nurses trained by the study principle investigator. Arm size was measured to determine appropriate cuff size. The ABP monitor was turned on, and the subjects left the HD unit after the first cuff inflation. Subjects were instructed to wear the cuff and monitor for the entire 44-hour interdialytic period except for bathing (subjects removed the cuff immediately after a measurement and replaced it within the measurement interval). BP was measured every 30 minutes from 6:00 a.m. until 10:00 p.m. and hourly from 10:00 p.m. to 6:00 a.m. Subjects continued their typical diets and antihypertensive regimens. Subjects returned to the dialysis unit for the following treatment still wearing the monitor. The monitor was turned off before starting HD and picked up by study personnel. In cases where the cuff had been turned off or had taken insufficient recordings, the procedure was repeated after the next mid-week treatment.

Statistical Analyses

Categorical measurements were analyzed using χ² analysis. Continuous measurements were tested for normality. Normally distributed data are reported as the means (SD), and non-normally distributed data are reported as median (25^th to 75^th percentile). Between-group comparisons of normally distributed data were performed with unpaired t tests.

The primary outcome variable was the mean 44-hour ambulatory systolic BP, which was analyzed between groups with a two-sided unpaired t test, using a P value of 0.05 to determine significance. The mean systolic and diastolic ambulatory BP for daytime and nocturnal time periods were analyzed between groups as secondary outcomes. Other secondary outcomes included interdialytic weight gain and percentages of interdialytic weight gain. These variables were analyzed between groups with an unpaired t test and in a mixed linear model as independent variables with demographic features (age, sex, race, and presence of diabetes) and case-control group, using systolic ABP as the dependent variable. Correlations between continuous variables such as ABP, pre-HD systolic BP, post-HD systolic BP, and mean HD-unit systolic BP were analyzed using Pearson product moment correlation.

The mean of the individual SD for the 44-hour, daytime, and nocturnal ambulatory systolic and diastolic BP measurements were used to analyze between-group differences in BP variability for each period.

Results

Subjects

The subjects were predominantly men with a mean age of 54.5 years. There was a large percentage of minorities and subjects with diabetes. Intradialytic HTN subjects took more antihypertensive medications, and only angiotensin-converting enzyme inhibitors (ACEi) were used less frequently by controls (the difference in ACEi or angiotensin receptor blocker use was not significant). Fifteen intradialytic-hypertension and seven control subjects took lisinopril, whereas one intradialytic-hypertension and one control subject took quinapril. Detailed descriptions of the subjects are shown in Table 1. Dialysis prescriptions were similar between groups, and there were no differences in dialysate calcium, potassium, bicarbonate, or in treatment time or blood or dialysate flow rates (data not shown).

Table 1.

Baseline characteristics of study subjects

	Intradialytic Hypertension (n = 25)	Control (n = 25)	P
Age (years)	53.9 (11.1)	55.1 (7.9)	0.7
Male (%)	80	80	0.9
African American (%)	36	40	0.8
Hispanic (%)	64	60	0.8
Diabetic (%)	88	84	0.7
Hypertensive (%)	100	100	0.9
Antihypertensives
angiotensin-converting enzyme inhibitor	64	32	0.02
angiotensin-receptor blocker	8	20	0.4
beta blocker	68	68	0.9
alpha blocker	24	12	0.5
calcium channel blocker	60	44	0.3
clonidine	32	24	0.5
hydralazine	24	12	0.5
takes before hemodialysis	60	48	0.4
Other medications
erythropoiesis-stimulating agents	88	92	0.9
intravenous iron	88	64	0.05
vitamin D analogue	84	76	0.5
calcium-containing phosphorus binder	56	52	0.8
noncalcium-containing phosphorus binder	44	56	0.4
Cardiovascular disease
coronary artery disease (% self-reported)	16	8	0.4
congestive heart failure (% self-reported)	16	12	0.9
cerebrovascular disease (% self-reported)	20	20	1.0
peripheral vascular disease (% self-reported)	4	8	0.6
any cardiovascular disease (% self-reported)	32	36	0.8
Laboratory parameters
hemoglobin (g/dl)	12.1 (1.1)	11.9 (1.3)	0.5
ferritin (ng/ml)	384 (210)	444 (223)	0.3
transferrin saturation (%)	27.6 (14.9)	32.7 (10.0)	0.2
serum calcium (mg/dl)	8.7 (0.6)	8.9 (0.9)	0.4
serum phosphorus (mg/dl)	5.1 (1.2)	5.1 (1.4)	0.9
serum parathyroid hormone (pg/ml)	356 (225.8)	443 (534.5)	0.5
serum albumin (g/dl)	3.8 (0.6)	3.9 (0.2)	0.5
serum potassium (mEq/L)	4.7 (0.7)	4.9 (0.5)	0.3
Dialysis prescription
with dialysate sodium of 140 mEq/L (%)	96	84	0.3
with sodium profiling (%)	4	20	0.08
Predialysis plasma sodium (mEq/L)	135.8 (3.7)	137.3 (2.8)	0.1
Dialysate to plasma sodium gradient	4.4 (3.0)	3.0 (3.2)	0.2

Blood Pressure

On the basis of Kidney Disease Outcomes Quality Initiative recommendations for BP control in HD patients, 100% of the subjects were hypertensive (15). The 2-week averaged pre-HD systolic and diastolic BP were 144.0 (9.7) and 77.5 (9.2) mmHg for intradialytic HTN group and 155.5 (15.5) and 82.5 (9.2) mmHg for controls (P = 0.003 for systolic BP, 0.06 for diastolic BP) (Table 2). Two-week averaged post-HD systolic and diastolic BP were 159.0 (9.3) and 80.9 (7.6) mmHg for intradialytic HTN subjects and 128.1 (10.9) and 71.9 (6.6) mmHg for controls (P < 0.0001 for systolic BP and <0.0001 for diastolic BP). The mean systolic BP changes from pre- to post-HD were +15.0 (9.1) and −27.4 (12.8) mmHg for the intradialytic HTN and control groups (Figure 1). The mean systolic BP (the average of all pre- and post-HD systolic BP measurements from the 2 weeks screened) was 151 and 142 mmHg for the intradialytic-HTN and control groups, respectively.

Table 2.

Ambulatory and hemodialysis unit blood pressure measurements

	Blood Pressure Measurements (in mmHg; mean [SD])		P
	IH (n = 25)	Control (n = 25)
44-hour ambulatory SBP	155.4 (14.2)	142.4 (16.5)	0.005
44-hour ambulatory DBP	82.4 (10.8)	76.9 (8.6)	0.05
Daytime ambulatory SBP	155.7 (14.9)	141.8 (16.4)	0.003
Daytime ambulatory DBP	83.2 (11.8)	77.2 (8.6)	0.05
Nocturnal ambulatory SBP	155.6 (16.4)	143.1 (19.1)	0.02
Nocturnal ambulatory DBP	80.9 (10.6)	76.5 (10.1)	0.14
Mean percent nocturnal dip in Ambulatory SBP	−1.2	0.1	0.68
Two-week averaged predialysis SBP	144.0 (9.7)	155.5 (15.5)	0.003
Two-week averaged predialysis DBP	77.5 (9.2)	82.5 (9.2)	0.06
Two-week averaged postdialysis SBP	159.0 (9.3)	128.1 (10.9)	<0.0001
Two-week averaged postdialysis DBP	80.9 (7.6)	71.9 (6.6)	<0.0001
Two-week averaged change in dialysis unit SBP (post − pre)	+15.0 (9.1)	−27.4 (12.8)	<0.0001

SBP, systolic blood pressure; DBP, diastolic blood pressure.

Figure 1.

Hemodialysis unit 2-week time averaged systolic BP measurements for control and intradialytic hypertension subjects. The 2-week average prehemodialysis and posthemodialysis systolic BP for each subject in the control and intradialytic hypertension group (n = 25 for each group). For the control group, the mean systolic BP was 155.5 mmHg prehemodialysis and 128.1 mmHg posthemodialysis. For the intradialytic hypertension group, the mean systolic BP was 144.0 mmHg prehemodialysis and 159.0 mmHg posthemodialysis. HD, hemodialysis.

The ABP monitor recorded a mean of 60.9 measurements per subject. The systolic ABP for the intradialytic-HTN and control groups were 155.4 (14.2) and 142.4 (16.5) mmHg, respectively (P = 0.005) (Figure 2). Systolic BP was higher in the intradialytic-HTN group compared with controls for the first 24 hours after HD (Figure 3). BP decreased in the intradialytic-HTN group and increased in the control group during this time. From 24 hours after HD until shortly before the next HD treatment, BP increased in both groups. The diastolic ABP for the intradialytic-HTN and control groups were 82.4 (10.8) and 76.9 (8.6) mmHg, respectively (P = 0.05). The daytime and nocturnal systolic and diastolic BP values are shown in Table 2. No control subjects had a normal (defined as a decrease in mean nocturnal systolic BP of 10% from the daytime value) nocturnal decrease in BP. Two subjects with intradialytic HTN had a normal nocturnal decrease in BP. The average ambulatory systolic BP during each daytime and nighttime period is shown in Figure 4.

Figure 2.

44-hour ambulatory blood pressure measurements in control and intradialytic-hypertension subjects. Shown is a box and whisker plot of the mean 44-hour systolic ambulatory BP measurements for the control and intradialytic-hypertension groups. The plot demonstrates the maximum value, third quartile, median, first quartile, and minimum value. The mean systolic BP was 142.4 mmHg for the control group and 155.4 mmHg for the intradialytic-hypertension group.

Figure 3.

Average systolic BP measured during each hour after hemodialysis in intradialytic-hypertension and control subjects. For each hour during the entire interdialytic time period, the average ambulatory systolic BP (measured every 30 minutes during the daytime and every hour at night) is shown for the intradialytic-hypertension and control groups. From the first hour after dialysis until approximately the 24^th hour, systolic BP is higher in the intradialytic-hypertension group than controls. BP decreases during the initial 24 hours after hemodialysis in the intradialytic-hypertension group and then increases until shortly before the next treatment. BP generally increases during the entire interdialytic time period in the control group until shortly before the next treatment. For the final hour, the 2-week average pre-HD systolic BP is added for each group.

Figure 4.

Average systolic BP per postdialysis time period in subjects with intradialytic-hypertension and controls. The entire 44-hour interdialytic time period is separated into five periods reflecting the different daytime and nocturnal periods. Average systolic BP for each group during each time period is shown. During the first nocturnal time period, BP decreases in intradialytic-hypertension subjects and increases in controls. During the second nocturnal time period, BP increases in both groups.

Overall, systolic ABP correlated best with the mean HD-unit systolic BP (r = 0.65, P < 0.0001), followed by post-HD unit systolic BP (r = 0.6, P < 0.0001), pre-HD unit systolic BP (r = 0.28, P = 0.04), and change in systolic BP (r = 0.28, P = 0.05). Post-HD BP was a better correlate with systolic ABP than pre-HD BP in the intradialytic-HTN group (r = 0.521 for post-HD and 0.405 for pre-HD), but pre-HD BP was better than post-HD BP in the control group (r = 0.586 for pre-HD and 0.537 for post-HD). There was no difference in 44-hour, daytime or nocturnal systolic or diastolic BP variability between groups (Table 3).

Table 3.

Ambulatory blood pressure measurements

	ABP (mmHg; SD)		P
	Intradialytic Hypertension (n = 25)	Control (n = 25)
44-hour SBP	155.4 (14.2)	142.4 (16.5)	0.005
Blood pressure variability of 44-h SBP	16.5 (5.5)	17.2 (3.4)	0.57
Blood pressure variability of daytime SBP	15.8 (5.1)	17.0 (3.7)	0.33
Blood pressure variability of nocturnal SBP	14.8 (7.8)	15.6 (4.0)	0.68
44-hour DBP	82.4 (10.8)	76.9 (8.6)	0.05
Blood pressure variability of 44-hour DBP	10.0 (2.5)	10.2 (2.5)	0.80
Blood pressure variability of daytime DBP	9.7 (2.5)	10.1 (2.6)	0.51
Blood pressure variability of nocturnal DBP	8.7 (3.6)	9.8 (2.3)	0.20

Blood pressure variability was defined as the mean of the individual standard deviations of the BP measurement. ABP, ambulatory blood pressure; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Intradialytic and Interdialytic Weight Changes

There were no between-group differences in estimated dry weight, ultrafiltration volume, total, or percentage of interdialytic weight gain during the ABP measurements (Table 4). A mixed linear model including age, sex, race, diabetic status, percentage of interdialytic weight gain, and case versus control group as independent variables and systolic ABP as the dependent variable confirmed the independent association of intradialytic HTN with systolic ABP (β = 12.0, P = 0.007). There was no significant correlation between dry weight, ultrafiltration volume, or interdialytic weight gain and systolic ABP (data not shown).

Table 4.

Predialysis and postdialysis body weights for treatments before and after ambulatory blood pressure measurements and weight gained during ambulatory blood pressure measurements

	Weight (kg)		P
	Intradialytic Hypertension (n = 25)	Control (n = 25)
Estimated dry weight	81.8 (19.9)	82.9 (14.6)	0.8
Predialysis weight treatment 1	85.2 (21.1)	86.4 (15.2)	0.8
Postdialysis weight treatment 1	82.0 (20.3)	83.3 (14.5)	0.8
Ultrafiltration during treatment 1 (% of EDW)	3.9 (1.6)	3.7 (1.4)	0.6
Predialysis weight treatment 2	84.7 (20.8)	85.9 (15.0)	0.8
Postdialysis weight treatment 2	81.7 (20.4)	83.3 (14.9)	0.8
Interdialytic weight gain	2.7 (1.1)	2.7 (1.3)	0.9
Interdialytic weight gain (% of EDW)	3.4 (1.2)	3.2 (1.4)	0.6

EDW, estimated dry weight.

Discussion

The clinically important finding of our study was that the mean interdialytic systolic ABP is higher in patients with intradialytic HTN compared with HD patients whose systolic BP decreases during HD. Systolic ABP was not associated with interdialytic weight gain during ABP measurement or percentage of interdialytic weight gain. Intradialytic-HTN patients also had higher mean HD-unit systolic BP. Our study is the first to investigate ABP specifically in patients whose pre- to post-HD systolic BP increases ≥10 mmHg. These findings offer further insight into the increased morbidity and mortality seen in prior retrospective studies of patients with intradialytic HTN of this magnitude. Furthermore, these findings offer practical information that can be applied to HTN management in ESRD because nephrologists may better assess interdialytic BP burden on the basis of recurrent BP patterns in the HD unit.

Previous studies have compared ABP and HD-unit BP in HD patients (16–18). One study demonstrated that post-HD systolic BP correlated better with interdialytic ABP than pre-HD systolic BP did in 22 subjects (16). An important limitation of that study is the low prevalence of HTN (mean systolic ABP was 129), thus the findings cannot be generalized to interpreting HD-unit BP in the context of HTN management. A recent meta-analysis evaluating ABP and HD-unit BP measurement showed poor overall agreement between ABP and HD-unit BP measurements (19). Generally, in this meta-analysis, post-HD systolic BP underestimated ABP and pre-HD systolic BP overestimated ABP. Pre-HD BP overestimating interdialytic BP has previously been attributed to a possible “white-coat effect” (17). In the meta-analysis, post-HD systolic BP was less biased than pre-HD systolic BP, but neither was considered a reliable ABP predictor. For nearly all studies included in that analysis, pre-HD systolic BP was greater than post-HD systolic BP. Our population differs from those in previous studies in that all subjects had Kidney Disease Outcomes Quality Initiative-defined HTN, with case and control groups diverging on the basis of BP response during HD.

We demonstrated that systolic ABP significantly correlated with pre-HD, post-HD, and mean HD-unit systolic BP. A previous study has shown that in patients with large intradialytic BP decreases, ABP best correlates with pre-HD systolic BP; however, in patients with large intradialytic increases in BP post-HD SBP correlates best with ABP (18). In our study, systolic ABP correlated best with mean HD-unit systolic BP in both groups. However, in intradialytic-HTN subjects, post-HD BP correlated better with ABP than pre-HD SBP did; in controls, ABP correlated better with pre-HD than with post-HD systolic BP. This emphasizes the limitations of the predictive capacity of either pre- or post-HD measurements outside of the context of the intradialytic BP pattern.

Recent studies associating intradialytic HTN with increased morbidity and mortality (5,6) are intriguing and require further investigation to understand potential underlying and modifiable risk factors that may contribute to worse outcomes. In this study, in which we stratified subjects by age, sex, and diabetic status, we detected no between-group difference in baseline left ventricular hypertrophy, heart failure, or coronary artery disease. The finding that the intradialytic-HTN group had higher ABP suggests that these subjects might be at higher risk for these comorbidities over time and therefore might benefit from tighter BP control. There was no significant difference in the number of subjects using erythropoiesis-stimulating agents or the dose of erythropoiesis-stimulating agent used. ACEi were more frequently used by intradialytic-HTN subjects who took more antihypertensives in general. It is uncertain whether this is an indication bias or whether the fact that lisinopril and quinapril are dialyzable contributed to the increased incidence of intradialytic hypertension in the case group. Currently, evidence to support a role for activation of the renin-angiotensin-aldosterone and sympathetic nervous systems is lacking (20). Studies investigating the role of nitric oxide and endothelin-1 in intradialytic HTN suggest that imbalances between these mediators of vascular tone may explain the BP increases during HD (20–22). Further investigation into the role of endothelial-cell dysfunction using measurements other than nitric oxide or endothelin-1 may offer more insight into some of these mechanisms.

Volume overload was an initial theory suggested to explain intradialytic HTN (23,24) and remains under investigation. We did not have bioimpedance or noninvasive monitoring of the hematocrit available to confirm the accuracy of the clinical dry weight assessment. However, we found no between-group differences in interdialytic weight gain. This confirms previous studies of single interdialytic time periods and ABP in HD populations (25,26). Furthermore, increased interdialytic weight gain has been shown to result in greater systolic BP decreases during HD (27). More recently, it has been shown that the slope of intradialytic BP change increases (indicating greater reduction in BP from pre- to post-HD) after dry weight reduction and further ultrafiltration over several weeks (28). This suggests that the frequency of intradialytic HTN may decrease with more aggressive ultrafiltration. Our study was not designed to address this issue because our subjects continued their regular dialysis prescriptions to achieve the currently estimated dry weight. We conclude that the differences in ABP between groups could not be accounted for by weight gain during a single interdialytic period. The possibility exists that both interdialytic ABP and the intradialytic BP change could have been reduced with further ultrafiltration over time. Considering the findings of our study and those of Agarwal, it remains a valid option to probe dry weight slowly over the course of several weeks in patients with intradialytic HTN, particularly with evidence of elevated BP on non-HD days. Our recommendations are to evaluate each patient individually to determine whether pharmacologic or nonpharmacologic therapy via dry weight reduction is appropriate. The analysis of the dry weight reduction study (28) also reported a weak but statistically significant association of change in intradialytic BP slope with change in interdialytic systolic BP. Our findings support this concept that qualitative intradialytic BP changes are related to interdialytic ABP.

Limitations of our study include its observational nature, which does not allow us to assign causality between intradialytic HTN and interdialytic HTN assessed by ABP. Another limitation was that the case and control groups had such different post-HD systolic BP, we were unable to fully assess how post-HD systolic BP contributes to interdialytic BP between groups. A larger study with more overlap in post-HD systolic BP may further clarify this. Additionally, we did not include the average of all intradialytic BP measurements in our analysis, which has been shown to better correlate with ABP than pre-HD, post-HD, or the average of the pre- and post-HD measurements (29). However, our study confirms the significance of using several weeks of BP patterns in predicting ambulatory BP. Finally, there is no standard definition of intradialytic HTN. We used the definition used in retrospective studies by Inrig et al. (5,6), because this definition has been clinically important on the basis of its association with adverse short and long term outcomes. An alternative definition using linear regression analyses of all intradialytic BP measurements to determine the slope of intradialytic BP change has been utilized by Agarwal and Light (28). This definition offers the benefit of analyzing multiple intradialytic BP measurements along a continuous spectrum. As the definition of intradialytic HTN is refined, such as the method proposed by Agarwal, it will be important in future studies to ascertain how such definitions associate with long term outcomes.

Conclusions

In conclusion, in hypertensive HD patients, 44-hour ambulatory interdialytic systolic BP was higher in those with intradialytic hypertension compared with controls. This relationship was not affected by interdialytic weight gain but was correlated with mean HD-unit systolic BP. These results suggest an explanation for the increased morbidity and mortality seen in intradialytic HTN. These findings should alert nephrologists that patients with increases in BP during HD may have an overall higher interdialytic BP burden that warrants consideration of more aggressive lowering of BP. The effects of interventions that target ABP reduction on the frequency of intradialytic HTN and vice versa remain to be determined, and further studies are required to ascertain the exact mechanisms responsible for intradialytic HTN.

Disclosures

None.