Effects of acute intradialytic exercise on cardiovascular responses in hemodialysis patients

Abstract

Background:

In patients with kidney failure requiring hemodialysis (HD) treatment, intradialytic exercise (IDEX) has been advocated for its feasibility and effectiveness in improving important health outcomes. However, IDEX as an adjunct therapeutic strategy is infrequently implemented, in part due to potential risks of IDEX, especially in patients with chronic volume overload. This study was performed to evaluate the safety of IDEX performed at different time points by examining its effect on intradialytic cardiovascular hemodynamics.

Methods:

In a randomized cross-over study (n = 12), intradialytic changes in brachial, aortic, and cardiac hemodynamics and autonomic function were examined during a HD session; (1) without exercise; (2) with 30 min of IDEX performed in the first hour of treatment; or (3) with 30 min of IDEX in the third hour of treatment.

Results:

IDEX during either the first or third hour did not exacerbate hemodynamic instability during treatment regardless of patient’s hydrations status. While there were transient increases in stroke volume, cardiac output, and heart rate during IDEX, intradialytic changes in brachial and aortic blood pressure, cardiac hemodynamics, and autonomic function were similar on days with and without IDEX.

Conclusion:

These results indicate that IDEX does not exacerbate hemodynamic instability during HD, regardless of a patient’s hydration status or the timing of exercise.

INTRODUCTION

In patients with kidney failure undergoing chronic hemodialysis (HD) treatment, intradialytic exercise (IDEX) has been advocated for its feasibility and effectiveness in improving cardiovascular (CV) function, physical function, and quality of life.^1–3 However, IDEX as an adjunct therapeutic strategy is infrequently implemented, in part due to putative concerns clinicians have about risks associated with IDEX.^4,5

A primary concern with IDEX is its potential impact on hemodynamic stability during treatment.^5,6 This is especially true for patients with chronic fluid overload (FO), where the high target ultrafiltration (UF) volume often exceeds the entire plasma volume pool. High UF rates during HD often cause complications such as hypotensive episodes and cramping, especially in the last hour of treatment.⁷ Because exercise provides an acute hemodynamic stress, typically causing transient increases in blood pressure (BP) during exercise, as well as postexercise hypotension (PEH),⁸ some have recommended that IDEX should be contraindicated in the late stages (e.g., third hour) of treatment.^6,9 However, very few studies to date have evaluated hemodynamic concerns with IDEX,⁹ and thus, there is little data to guide IDEX prescription. The purpose of this study was to explore the safety of IDEX in HD patients with different hemodynamic characteristics, as well as evaluate the impact of IDEX early vs. late in the treatment.

METHODS

Twelve HD patients from a dialysis clinic in Central Illinois completed the study. Inclusion criteria were: (1) thrice weekly HD; (2) 30 to 70 years of age; (3) on HD for >3 months; (4) physically able to exercise; (5) no pulmonary disease or decompensated congestive heart failure; (6) medical clearance from a nephrologist; (7) no changes in antihypertensive medicines <1 month before the trial. Participants were asked to refrain from NSAIDs for 7 days prior to the study, and to refrain from caffeine before or during HD. All patients provided informed consent and the trial was approved by Institutional Review Board of University of Illinois and registered at Clinicaltrials.gov (NCT02753868).

Intervention

This study was a cross-over design in which patients were assigned to receive each of three treatments: 1) a standard HD session (CON); 2) HD with 30-minutes of intradialytic cycling starting 30 minutes into treatment (first hour IDEX); and 3) HD with 30-minutes of cycling starting 3 hours into treatment (third hour IDEX). The CON and first hour IDEX days were completed in random order, while the third hour IDEX day was always performed last to ensure that first hour IDEX is adequately and safely performed by each patient. All interventions were conducted during a mid-week HD session, 1 week apart. During IDEX sessions, a stationary cycle (Monark Rehab Trainer 881E, Sweden) was placed in front of the dialysis chair and patients were asked to cycle at a self-selected pace and resistance that coincides with a rating of perceived exertion rate between 11 and 13 on the Borg 20-point scale.

Pre- and post-HD measurements

Patients were asked to lay in a supine position for up to 30 min before and immediately after each HD treatment to collect CV measures and hydration status. Transthoracic echocardiography was performed (ProSound α−7, Aloka, Japan) according to the recommendations of the American Society of Echocardiography.¹⁰ Arterial function including aortic augmentation index corrected for heart rate (HR) of 75 bpm (Aix75) and aortic pulse wave velocity (PWV) were measured using an automated BP cuff (Mobil-O-Graph, IEM, Germany) that was placed on the nonaccess arm. Whole-body fluid status, including extracellular water (ECW), intracellular water (ICW), and total body water (TBW), was measured using multifrequency bioelectrical impedance spectroscopy (SFB7, Impedimed, Inc., USA). Hydration status, represented as FO was defined as FO% = Absolute FO (L)/ECW) * 100,¹¹ where absolute FO (L) = 1.136*ECW(L) – 0.430*ICW(L) – 0.114*Body weight (kg).¹²

Intra-HD measurements

Brachial BP was measured on the nonaccess arm every 15 min throughout the HD session using an automated BP cuff integrated into the dialysis unit. Aortic variables were also obtained every hour using an automated BP cuff (Mobil-O-Graph). Beat-to-beat recordings of stroke volume (SV), HR, and cardiac output (CO) by transthoracic impedance and peripheral BP and total peripheral resistance (TPR) by finger plethysmography were collected throughout a HD session (Task Force Monitor, CNSystems, Austria). The variability of the intervals between successive heart beats was analyzed by heart rate variability (HRV) analysis to examine autonomic function. Frequency domain analysis generates low-frequency (LF, 0.04–0.15 Hz) and high-frequency (HF, 0.15–0.40 Hz) components, from which the LF/HF ratio is calculated. Low-frequency band suggests a mixed modulation of sympathetic and parasympathetic activities. The HF component reflects vagal modulation of the sinoatrial node and is thus used a surrogate marker of parasympathetic modulation.¹³ Normalized units of LF and HF to the total power were also calculated (LFnu and HFnu, respectively). Baroreflex sensitivity (BRS) was estimated using the spontaneous sequence method.¹⁴ The beat-to-beat values were averaged every 30 min for analysis.

Intradialytic CV-adverse events and symptom assessment

Intradialytic hypotension (IDH) was defined as (1) systolic blood pressure (SBP) <90 mmHg, and/or (2) a fall of SBP of more than 25% from the start of HD associated with symptoms including dizziness, vomiting, nausea, and muscle cramps, and/or modifying dialysate temperature, stopping UF, or saline solution boluses.¹⁵ Intradialytic hypertension (IDHTN) was an increase in mean arterial pressure (MAP) over 15 mmHg during or after an HD session.¹⁶

Self-reported intradialytic symptoms were assessed by a questionnaire including symptoms of nausea, dizziness, and cramps, using a 5-point Likert scale, where 0 was no symptoms and 4 was severe.¹⁷ Post-HD symptoms were collected using the same questionnaire and the fatigue severity scale questionnaire by contacting patients over the phone 5 h after the preceding HD treatment.¹⁸

Statistical analysis

Variables collected at pre- and post-HD were compared by paired t-tests and tested for difference between three intervention days by one-way analysis of variance (ANOVA). The changes in hemodynamic variables during HD were analyzed by Mixed Model Analysis with Repeated Measures with fixed effects of Group and Time and a random effect of patients to control for their associated intra-class correlation. This analysis was conducted in all participating patients (n = 12) and in the subset of patients who completed all three intervention sessions (n = 8). This model also tolerates an unequal number of response variables. To examine the influence of hydration status on intradialytic hemodynamic changes, markers of hydration status (UF goal, interdialytic weight gain (IDWG), and FO) were also entered into the same models. Between-group comparisons were conducted to examine the difference between interventions at each measured time point (i.e.,15-min interval for BP and 30-min interval for cardiac and autonomic data) throughout HD. For example, for SBP, both the absolute (${\text{SBP}}_{\text{i}-\text{min}}$, the level of SBP at i-minutes into HD) and the difference ($\text{Δ}$ values between the beginning and i-minutes of HD (${\text{ΔSBP}}_{\text{i}-0-\min }={\text{SBP}}_{\text{i}-\text{min}}-{\text{SBP}}_{0-\min }$) were compared between three intervention days by one-way ANOVA. LSD post hoc analysis was used. Low-frequency and HF measures in HRV data were not normally distributed by Shapiro–Wilk tests, and thus the log transformed variables of LF and HF were used for analysis.

RESULTS

Descriptive characteristics of study participants are presented in Table 1. Three patients did not complete third hour IDEX day because they no longer wanted to participate, and one patient finished only the CON day due to an ankle injury not related to the study.

Table 1.

Patient characteristics

Characteristics (n = 12)	Mean ± SD
Age (year)	55.9 ± 8.6
Height (cm)	171.5 ± 10.0
Body weight at pre-HD (kg)	83.3 ± 20.2
BMI (kg/m2)	28.07 ± 9.13
Gender (male/female)	7/5
Race
African American (n, %)	10 (83.3)
Caucasian (n, %)	2 (16.7)
Diabetes (n, %)	4 (33.3)
Hypertension (n, %)	11 (91.7)
Smoking (n, %)	8 (66.7)
Albumin (g/dL)	3.9 ± 0.3
Hemoglobin (g/dL)	11.3 ± 0.8
iPTH (pg/mL)	583.6 ± 394.9
Neutrophil:lymphocyte ratio	2.8 ± 0.9
HD characteristics
Blood flow rate (mL/min)	415 ± 48
HD flow rate (mL/min)	638 ± 80
UF rate	774 ± 270
HD duration (h)	3.86 ± 0.25

BMI = body mass index; iPTH = intact parathyroid hormone.

Comparison of parameters at pre- and post-HD

A comparison of CV parameters is presented in Table 2. Due to patient’s time constraints before or after treatment, this data were only collected on 8, 6, and 3 patients on the CON, first hour IDEX, and third hour IDEX days, respectively. There was no difference between intervention days in any parameters measured at pre- and post-HD, except for Aix75. Aix75 at post-HD was significantly lower in third hour IDEX than in first hour IDEX (P < 0.05). When all intervention days were combined, variables known to be influenced by patient’s hydration status, including body weight, TBW, ECW, and cardiac $\text{E}/{\text{E}}^{\prime }$, a measure of diastolic function, were significantly lower post-HD compared to pre-HD values.

Table 2.

Parameters at pre- and post-HD treatment

Variables	Total	CON	First hour EX	Third hour EX
Hydration status parameters		(n = 12)	(n = 12)	(n = 8)
IDWG (kg)	2.3 ± 1.1	2.2 ± 1.1	2.4 ± 1.1	2.3 ± 1.1
Ultrafiltration goal (L)	2.83 ± 1.09	2.94 ± 0.84	2.74 ± 1.14	2.77 ± 1.44
Body Weight at pre-HD (kg)	83.3 ± 20.2	84.0 ± 20.8	85.0 ± 22.9	80.0 ± 17.4
Body Weight at post-HD (kg)	82.1 ± 21.3*	83.1 ± 22.2	84.6 ± 24.1	77.5 ± 18.4
TBW at pre-HD (%)	61.0 ± 11.5	61.3 ± 14.3	60.7 ± 10.1	61.0 ± 10.0
TBW at post-HD (%)	58.6 ± 11.6*	58.4 ± 11.8	57.7 ± 12.9	60.1 ± 11.2
ECW at pre-HD (%)	44.3 ± 3.5	44.3 ± 4.3	44.2 ± 3.4	44.6 ± 2.7
ECW at post-HD (%)	42.3 ± 4.3*	41.6 ± 5.3	43.4 ± 3.4	43.0 ± 3.9
FO at pre-HD (%)	14.9 ± 13.2	13.9 ± 14.6	16.1 ± 14.1	14.9 ± 10.7
FO at post-HD (%)	8.9 ± 13.1*	7.3 ± 11.8	11.0 ± 14.1	8.3 ± 15.1
Physical symptomsa
Nausea during HD (0–4)	0.5 ± 1.0	0.3 ± 0.7	0.8 ± 1.5	0.4 ± 0.7
Dizziness during HD (0–4)	0.4 ± 1.0	0.3 ± 0.7	0.6 ± 1.2	0.4 ± 1.1
Cramping during HD (0–4)	1.0 ± 1.1	0.9 ± 1.0	1.1 ± 1.4	1.0 ± 1.2
Nausea at 5 h post-HD (0–4)	0.5 ± 1.3	0.0 ± 0.0	0.6 ± 1.5	0.8 ± 1.6
Dizziness at 5 h post-HD (0–4)	0.5 ± 1.3	0.0 ± 0.0	0.4 ± 1.1	1.0 ± 2.0
Cramping at 5 h post-HD (0–4)	0.9 ± 1.4	0.8 ± 1.8	0.7 ± 1.3	1.3 ± 1.5
Fatigue at 5 h post-HD (0–10)	5.2 ± 2.9	4.8 ± 2.6	5.3 ± 3.4	5.7 ± 3.8
IDH and IDHTNa
IDH (HD session, n)	8	2	2	4
IDHTN(HD session, n)	3	1	1	1
Cardiac parameters		(n = 8)	(n = 6)	(n = 3)
SV at pre-HD (mL)	73.2 ± 23.4	74.5 ± 24.9	74.2 ± 24.5	67.7 ± 25.7
SV at post-HD (mL)	67.5 ± 30.3	81.2 ± 41.6	58.7 ± 15.8	53.0 ± 2.8
CO at pre-HD (L)	5.4 ± 1.7	5.7 ± 1.6	5.5 ± 1.9	4.4 ± 1.6
CO at post-HD (L)	5.1 ± 2.7	6.4 ± 3.5	4.5 ± 1.8	3.4 ± 0.6
EF at pre-HD (%)	57.8 ± 18.7	63.0 ± 18.5	55.2 ± 20.0	49.2 ± 19.0
EF at post-HD (%)	59.7 ± 17.6	64.2 ± 8.2	57.1 ± 22.4	54.1 ± 30.8
E/A at pre-HD	1.1 ± 0.2	1.1 ± 0.2	1.2 ± 0.2	1.0 ± 0.0
E/A at post-HD	0.9 ± 0.2	0.9 ± 0.3	0.9 ± 0.1	1.0 ± 0.2
${\text{E}}^{\prime }$ at pre-HD	10.5 ± 3.4	9.8 ± 3.0	11.4 ± 4.8	10.4 ± 1.6
${\text{E}}^{\prime }$ at post-HD	9.2 ± 2.3	9.1 ± 1.7	9.8 ± 2.8	8.3 ± 2.9
$\text{E}/{\text{E}}^{\prime }$ at pre-HD	9.6 ± 5.9	10.8 ± 6.4	9.4 ± 7.6	7.5 ± 0.6
$\text{E}/{\text{E}}^{\prime }$ at post-HD	7.5 ± 3.8*	7.5 ± 4.1	7.7 ± 4.7	7.2 ± 2.8
Arterial parameters		(n = 7)	(n = 6)	(n = 4)
bSBP at pre-HD (mmHg)	126.2 ± 19.3	127.6 ± 20.7	130.7 ± 23.8	117.2 ± 6.1
bSBP at post-HD (mmHg)	132.1 ± 20.6	130.0 ± 17.7	142.7 ± 23.9	116.0 ± 5.0
bDBP at pre-HD (mmHg)	76.5 ± 17.1	78.4 ± 17.0	79.0 ± 21.5	69.2 ± 11.2
bDBP at post-HD (mmHg)	77.5 ± 13.5	75.0 ± 9.8	82.8 ± 17.5	71.7 ± 10.7
aSBP at pre-HD (mmHg)	117.3 ± 19.0	117.3 ± 20.8	122.2 ± 23.0	110.0 ± 7.7
aSBP at post-HD (mmHg)	120.6 ± 19.6	118.8 ± 17.4	129.5 ± 22.4	106.3 ± 11.2
aDBP at pre-HD (mmHg)	77.5 ± 17.0	79.7 ± 17.2	80.2 ± 20.9	69.8 ± 11.0
aDBP at post-HD (mmHg)	78.9 ± 13.5	76.2 ± 10.1	84.7 ± 17.0	72.7 ± 10.7
Aix75 at pre-HD (mmHg)	18.3 ± 13.6	17.6 ± 13.1	21.7 ± 17.7	14.5 ± 8.7
Aix75 at post-HD (mmHg)	16.1 ± 12.5	13.5 ± 9.7	24.8 ± 12.4	4.0 ± 3.5b
PWV at pre-HD (m/sec)	8.2 ± 0.7	8.2 ± 0.8	8.1 ± 0.6	8.2 ± 0.6
PWV at post-HD (m/sec)	8.3 ± 0.9	8.3 ± 1.0	8.7 ± 0.5	7.8 ± 0.9

^aRepresents an analysis of data from patients who completed all the three interventions (n = 8).

^bIndicates a significant difference between three intervention days by ANOVA and a significant difference from first hour EX by LSD post hoc analysis.

${\text{A}}^{\prime }$ = peak late diastolic annular velocity; aDBP = aortic diastolic blood pressure; Aix75 = augmentation pressure at heart rate 75; aSBP = aortic systolic blood pressure; AugP = augmentation pressure; bDBP = brachial diastolic blood pressure; bSBP = brachial systolic blood pressure; BUN = blood urea nitrogen; CO = cardiac output; DecT = deceleration time of ${\text{E}}^{\prime }$; ${\text{E}}^{\prime }$ = peak early diastolic annular velocity; E/A = the ratio of early to late diastolic filling pressures; ECW = extracellular water; $\text{E}/{\text{E}}^{\prime }$ = the ratio of early diastolic filling pressure to tissue velocity; EF = ejection fraction; FO = fluid overload; HD = hemodialysis; ICW = intracellular water; IDH = intradialytic hypotension; IDHPT = intradialytic hypertension; IDWG = interdialytic weight gain; K = potassium; LVM = left ventricular mass; Na = natrium; PWV= pulse wave velocity; ${\text{S}}^{\prime }$ = peak systolic annular velocity; SV = stroke volume; TBW = total body water.

^*Indicates a significant difference between pre- and post-HD levels by paired t-test (P < 0.05).

CV hemodynamic changes during HD

There were no significant group × time interactions for any of arterial hemodynamics including brachial and aortic SBP, diastolic blood pressure (DBP), pulse pressure (PP), MAP, Aix75, and PWV, and cardiac hemodynamic including SV, CO, and TPR and autonomic parameters including LF, HF, LFnu, HFnu, and LF/HF by HRV and BRS (P > 0.05 for all) (Figure 2). This indicates that changes of brachial, aortic, and cardiac hemodynamics and autonomic activity during HD were similar regardless of whether or when patients exercised. Similar trends were seen when the influence of fluid status (IDWG, UF goal, and FO) were added as covariates in the model or when only completers (n = 8) were included in the analysis. However, there was a significant main effect of time in the overall population for several outcome measures. Specifically, brachial SBP, MAP, and PP and aortic SBP and PWV decreased over time to a similar degree during each treatment (P < 0.001 for all). While LFnu by HRV (F_1,28 = 6.85, P = 0.014) and LF/HF ratio by HRV (F_1,28 = 3.92, P = 0.057) increased, HF by HRV (F_1,28 = 4.58, P = 0.041) and HFnu by HRV (F_1,28 = 6.88, P = 0.013) decreased over time.

Figure 2.

Changes in cardiovascular hemodynamics during a hemodialysis treatment with and without exercise. HRV = heart rate variability. #, indicates a significant effect of Time in the overall group (P < 0.05); $\mathit{\Phi }$, indicates a significant difference between first hour IDEX and CON and third hour IDEX; *, indicates a significant difference between third hour IDEX and CON; δ, indicates a significant difference between third hour IDEX and first hour IDEX.

Between-group comparisons at each time point demonstrated that SBP in third hour IDEX day was lower than CON day at 90, 105, and 135 min, and lower than first hour IDEX day at 90 and 165 min into HD (P <0.05 for all). SBP during the first hour IDEX day was significantly higher than on the other 2 days in the later hours of HD (at 210 and 225 min; P < 0.05 for all). Regarding cardiac measures, the first hour IDEX day had higher CO and HR and faster left ventricular ejection time compared to the other treatment days at 30- and 60-min into HD when exercise was performed (P < 0.05 for all). On the third hour IDEX day, there was a higher SV and CO compared to the other intervention days at 150- and 180-min into HD, when cycling was performed (P < 0.05 for all). Using the cardiac $\text{Δ}$ values, similar trends for exercise-induced increases in SV and CO were seen during the time intervals when IDEX was performed: increased ${\text{ΔSV}}_{60-30}$ and ${\text{ΔCO}}_{60-30}$ in first hour IDEX day; and increased ${\text{ΔSV}}_{120-30}$ and ${\text{ΔCO}}_{120-30}$ in third hour IDEX day (P < 0.05 for all). There was a numerical decrease in TPR after the third hour IDEX, but this was not significant (${\text{ΔTPR}}_{240-180}=-222.87$; CI-627.79, 178.03; P = 0.174) (Figure 2).

DISCUSSION

This study examined changes in brachial, aortic, and cardiac hemodynamics and autonomic function during a standard HD session without exercise, or when 30-min of cycling exercise was performed during the first or third hour into HD. The primary finding was that IDEX performed during either the first or third hour does not appear to exacerbate hemodynamic instability during HD. While there were transient increases in SV, CO, and HR during IDEX, the intradialytic changes in BP parameters, cardiac hemodynamics, and autonomic function were similar on days with and without IDEX. This null effect of IDEX on hemodynamic parameters during HD was demonstrated regardless of the timing of exercise or patient’s underlying CV characteristics or hydration status.

Brachial hemodynamic response during HD with and without IDEX

Exercise during HD has been advocated for its convenience as a time-efficient strategy to increase physical activity during a forced sedentary period.³ However, concerns regarding its potential impact on hemodynamic instability have been a cause for concern for clinicians, despite a lack of evidence for these effects. Hemodialysis imparts a significant CV stress mainly due to the progressive loss of blood volume during treatment, and in theory, exercise may exacerbate these effects. In particular, PEH may increase the risk of ischemic adverse events, particularly during the latter phase of HD when the total blood volume is already low.¹⁹ Despite this hypothetical concern, the present study demonstrated that intradialytic changes in BP parameters were not different between HD sessions with and without IDEX. This finding is consistent with previous studies in which IDEX was well tolerated and did not elicit abnormal hemodynamic responses,^9,20–22 though much of these data came from small pilot studies lacking a control condition. The only previous study that systematically investigated the CV safety of IDEX used a short submaximal exercise bout (5-min of cycling at 60% of VO₂peak) that was repeated every hour during a HD treatment. Stable hemodynamic responses to IDEX were demonstrated through the second hour of HD, but five out of eight patients were unable to perform cycling during the third hour due to IDH that was accompanied by reductions in HR and CO. One potential explanation for this is that the patients who were unable to exercise in the third hour due to hypotensive symptoms had a higher UF volume compared to the patients who were able to exercise in the third hour (4.6 ± 1.3 vs. 3.2 ± 1.1 L, respectively).⁹ This suggests that exercise may indeed be well tolerated in the third hour of treatment in the absence of an excessive UF rate, so may not be contraindicated in all patients. The average intradialytic SBP drop in this study was 18 mmHg, which is comparable to other published data.²³ Both CO and TPR were numerically reduced over the course of HD, but these changes were not statistically significant.

Interestingly, brachial BP levels were elevated in the last hour of HD after performing first hour IDEX. Although speculative, when total blood volume is critically low and the expected compensatory responses are suboptimal, the acute stimulation of the CV system by exercise may help maintain CV homeostasis. Indeed, recent evidence indicated that IDEX may have the potential to promote plasma refilling during the postexercise period by increased intravascular colloid osmotic pressure²⁴ or by increased capillary membrane pore size.^25,26

Cardiac hemodynamic and vascular resistance response during HD with and without IDEX

During exercise, peripheral vasodilation occurs to meet the increased metabolic demands in skeletal muscle. At the same time, local signals integrated from baroreflex, chemoreflex, and skeletal muscle receptors help to augment sympathetic outflow in the heart, the adrenal gland, and the splanchnic vasculatures, helping to maintain or increase systemic BP levels during exercise.⁸ In this study, increases in SV, CO, and HR observed during the IDEX period indicate that most patients appeared to have a normal CV response to the moderate intensity exercise bout.

However, we did not observe a reduction in BP following the cessation of exercise. This lack of PEH may have been due to autonomic dysfunction, though it could also have been due to the relatively modest intensity of the exercise bouts. It is also possible that the removal of fluid by UF prevents the signals that are normally responsible for PEH. Postexercise reduction in TPR also influences the magnitude of PEH. The magnitude of local vasodilation in active tissues, despite an increased systemic sympathetic outflow, normally becomes greater as exercise intensity increases.^27,28 Vasoconstriction is an important compensatory mechanism to the progressive hypovolemia and thus presumably occurred to some extents in our subjects. Thus, it can be speculated that the effect of mild to moderate intensity IDEX on the CV system might be offset by the profound impact of HD, resulting in the absence of PEH. Although beyond the scope of the present study, autonomic dysfunction, impaired vasoreactive capacity, and HD-driven increases in systemic inflammation cannot be ruled out as factors responsible for the blunted postexercise responses in HD patients.

Autonomic response during HD with and without IDEX

HD patients are characterized by suppressed HRV, enhanced sympatho-excitation^29,30 and decreased BRS.³¹ In this study, there was a shift of sympatho-vagal balance toward a sympathetic predominance during HD. Our data are consistent with previous findings demonstrating increased sympathetic activity during HD.^32,33

No previous study has investigated the acute effect of IDEX on autonomic function, though several studies have demonstrated compromised exercise-induced autonomic regulation, particularly in relation to abnormal BP responses in patients on non-HD days.^30,34,35 In this study, there was no effect of IDEX on the intradialytic change in autonomic activity. The explanation for this is likely multifactorial, though HD is known to create disturbances in autonomic modulation. A patient’s hydration status,^36,37 the composition³⁸ and temperature of the dialysate,³⁹ and PTH levels⁴⁰ can also influence HRV. Despite the methodological challenges, further investigation of autonomic regulation during IDEX is warranted.

Arterial compliance response during HD with and without IDEX

There are conflicting data regarding the acute effect of HD treatment on arterial compliance. Decreased Aix^41,42 and aortic PWV⁴³ and increased aortic and brachial PWV^42,44 have been shown after HD, though other studies indicated no effect of HD on aortic PWV^45,46 and markers of large artery compliance.^47,48 These inconsistent results might be attributed to methodological and population discrepancies between studies, as well as the acute effects of HD on the CV system. In theory, volume overload increases arterial distension and consequently augments arterial stiffness. However, with decreasing blood volume as a result of UF, the renin angiotensin aldosterone system is activated and angiotensin II levels are increased, resulting in vasoconstriction and arterial stiffness. Furthermore, HD not only creates a significant volume fluctuation, but also changes levels of some vasoactive molecules such as endothelin, nitric oxide, and angiotensin II.⁴⁹ The timing and amplitude of the reflected waves from the peripheral vasculature to the aorta are altered following submaximal exercise, resulting in decreased aortic pulse amplification, aortic PP, and PWV.⁵⁰ While the present study showed no significant effect of IDEX on aortic hemodynamics, there was a significant effect of time, with decreasing trends in aortic BP and PWV over the course of an HD treatment. The intradialytic change in PWV was likely to be driven by the intradialytic BP change, as the time effect of PWV was no longer significant when divided by MAP.

Adverse events

Of the 32 HD sessions examined, 32.3% were associated with IDH and 16.1% with IDHTN. While the frequency of adverse events was not influenced by IDEX, this data should be interpreted with caution due to the small total number of events. Two patients developed severe hypotensive symptoms (cramping and nausea) accompanied by a rapid drop in SBP immediately after performing the third hour exercise. However, those patients had signs of underhydration; one patient had a post-HD weight 1 kg below the prescribed dry weight and the other patient had a negative pre- and post-HD FO% (−3% and −14%, respectively) on the third hour exercise day. Thus, it is not clear whether the late-hour exercise, excessive UF, or both, contributed to the development of the hypotensive symptoms. It also should be recognized that brachial and aortic BP were lower throughout the third hour exercise day compared to other intervention days during the mid-hours of HD. Although the reason for this is unclear, the lower BP before performing the third hour exercise indicates that the last hour exercise was unlikely to be the reason for the hypotensive events. For the sake of maximal safety, it should be advised to examine potential factors that may increase the risk of HD-intolerance such as abnormal hydration status at pre-HD, as well as recent illness, when considering IDEX in the third hour of treatment.

Limitations

This study has several limitations. First, exercise intensity and workload were not objectively measured. The blunted CV response, as indicted by no significant increases in HR, CO, and BP, during the third-hour exercise may also be attributed to the subjectivity of the exercise intensity estimates. Patients may be less motivated to exercise during the last hour than the earlier hours during HD sessions due to HD-mediated physical and mental exhaustion. Second, the total sample size was relatively limited, and only a subset of patients was available for cardiac and arterial measures at pre- and post-HD. Thus, the impact of CV comorbidities on the intradialytic hemodynamic responses was not adequately powered to examine as confounding factors. Nevertheless, our data indicate that the combination of comprehensive hydration status measures at pre-HD, including extracellular volume and BP, were able to explain the two 3rd-hr IDEX-related adverse events. Therefore, hydration assessments at pre-HD may help the management of IDEX-related intradialytic hemodynamic stability. Third, patient’s IDWG and the subsequent UF volume were not controlled between intervention sessions, although patients were advised to maintain the typical eating patterns during the study period. Environmental factors as well as HD prescriptions should be considered as potential confounding factors for hemodynamic changes during HD in future studies.

Conclusion

Our data suggest that IDEX does not exacerbate hemodynamic instability during HD regardless of the timing of the exercise and hydration status. Despite the gradual decreases in brachial and aortic BP throughout HD, IDEX did not cause further hemodynamic instability. These results help improve our understanding regarding the CV safety of IDEX and provide more guidance on how it should be prescribed.

Figure 1.

Study protocol. Aix = augmentation index; BP = blood pressure; CO = cardiac output; EX = exercise; HD = hemodialysis; HRV = heart rate variability; PWV = pulse wave velocity; SV = stroke volume; TPR = total peripheral resistance.