http://www.kidney-international.org
clinical investigation
& 2014 International Society of Nephrology
Increased postdialysis systolic blood pressure is associated with extracellular overhydration in hemodialysis outpatients Arkom Nongnuch1, Neil Campbell1, Edward Stern1, Sally El-Kateb1, Laura Fuentes1 and Andrew Davenport1 1
UCL Center for Nephrology, Royal Free Hospital, University College London Medical School, London, UK
Recently, intradialytic hypertension was reported to be associated with increased mortality for hemodialysis patients. To determine whether volume status plays a role in dialysisassociated hypertension, we prospectively audited 531 patients that had volume assessments measured by multiplefrequency bioelectrical impedance during their midweek dialysis session. Mean pre- and postdialysis weights were 73.2 vs 71.7 kg, and systolic blood pressures (SBPs) 140.5 vs. 130.3 mm Hg, respectively. Patients were divided into groups based on a fall in SBP of 20 mm Hg or more (32%), an increased SBP of 10 mm Hg or more (18%), and a stable group (50%). There were no differences in patient demographics, dialysis prescriptions, predialysis weight, total body (TBW), and extracellular (ECW) and intracellular water (ICW). However, the change in weight was significantly less in the increased blood pressure group (1.01 kg vs. stable 1.65, and 1.7 hypotensive). The ratio of ECW to TBW was significantly higher in the increased blood pressure group, particularly post dialysis (39.1 vs. stable 38.7% and fall in blood pressure group 38.7%). ECW overhydration was significantly greater in the increased blood pressure group post dialysis (0.7 (0.17 to 1.1) vs. stable 0.39 ( 0.2 to 0.95) and fall in blood pressure group 0.38 ( 0.19 to 0.86) liter). We found that patients who had increased blood pressure post dialysis had greater hydration status, particularly ECW. Thus, patients who increase their blood pressure post dialysis should have review of target weight, consideration of lowering the post-dialysis weight, and may benefit from increasing dialysis session time or frequency. Kidney International advance online publication, 30 July 2014; doi:10.1038/ki.2014.276 KEYWORDS: extracellular fluid; hypotension; hemodialysis; hypertension; multifrequency bioimpedance
Correspondence: Andrew Davenport, UCL Centre for Nephrology, Royal Free Hospital, University College London Medical School, Rowland Hill Street, London NW3 2PF, UK. E-mail: [email protected] Received 27 December 2013; revised 16 June 2014; accepted 19 June 2014 Kidney International
Although a fall in blood pressure remains the commonest complication associated with routine outpatient hemodialysis treatments,1,2 a minority of patients become hypertensive either toward or at the end of dialysis or shortly after completion of the dialysis session. Although a recognized complication of hemodialysis,3,4 it is only relatively recently that intradialytic hypertension has been recognized as an independent risk factor for both more frequent hospital admissions and also decreased patient survival.5,6 The definition of intradialytic hypertension has varied over the years,3–7 but most clinical studies now report intradialytic hypertension as an increase in systolic blood pressure (SBP) during dialysis of greater than 10 mm Hg above the predialysis blood pressure measurement.5,6 Intradialytic hypertension has been reported to occur in about 21% of hemodialysis sessions, affecting more than 15% of patients,8 and it is more frequently reported in older patients, those prescribed more antihypertensive medications, and those with lower serum creatinine.9 Although intravascular volume is typically reduced during dialysis, changes in relative blood volume do not necessarily closely mirror changes in blood pressure,10 and this has led to the hypothesis that intradialytic hypertension is more likely to be due to changes in vascular tone.11 The cause of intradialytic hypertension remains to be determined, with different authors speculating on increased renin and angiotensin release and sympathetic nervous system activation in response to ultrafiltration, elimination of antihypertensive medications in patients with established hypertension, increased viscosity and hematocrit following ultrafiltration, and high dialysate calcium concentrations.3,4 However, as there is a link between sodium balance and hypertension in hemodialysis patients,12,13 we speculated that patients with intradialytic hypertension could also potentially have an expanded extracellular volume to account for the rise in blood pressure during dialysis. We therefore reviewed the dialysis records of 531 patients who had volume assessments measured by multiple-frequency bioelectrical impedance to determine whether volume status had a role in dialysis-associated hypertension. 1
clinical investigation
Multifrequency bioelectrical impedance assessments were made on 531 stable adult patients attending for their midweek hemodialysis session; the mean age was 60.3±16.5 years, 62.0% were male, 37.3% were diabetic, 42.2% were Caucasoid, and the median dialysis vintage was 42.0 (23.7–48.5) months. Predialysis weight was 73.2±18.0 kg, and postdialysis weight was 71.7±17.7 kg. Predialysis SBP was 140.5±26.5 mm Hg with a diastolic blood pressure of 71.7±17.7 mm Hg, and postdialysis systolic pressure was 130.3±26.3 mm Hg with a diastolic blood pressure of 71.8± 15.2 mm Hg. Predialysis hemoglobin was 112.6±15.2 g/l, median glucose was 6.2 (5.2–8.3) mmol/l, serum albumin was 38.6±6.2 g/l, and median C-reactive protein was 5 (2–12) mg/l. A total of 67 patients were not included in the audit as bioimpedance was not performed, because 14 patients had pacemakers/defibrillators, 15 were unable to stand owing to amputations or active foot ulceration, 12 were recent starters, 6 had recent hospital in-patient stays, and the remaining 20 patients had a variety of metallic prosthetic implants—hips and knees—or were unable to stand. As such bioimpedance measurements were made on 89% of patients attending for their midweek dialysis session. Patients were divided into three groups, those in whom the blood pressure decreased during dialysis by X20 mm Hg,14 n ¼ 171 (32.2%), those in whom the SBP increased by X10 mm Hg,5 n ¼ 96 (18.1%), and 264 patients (49.7%) in whom the SBP did not meet the definitions of intradialytic hypotension or hypertension, from here on termed the ‘stable’ group (Supplementary Figure online). These three groups did not differ in demographics, or cardiac history and medications (Supplementary Table S1 online), although those with a fall in blood pressure with dialysis were of an older dialysis vintage (46 (15–74) months) compared with both those with ‘stable’ (27 (9–64) months) and those with an increase in SBP post dialysis (23 (5–63.6) months), Po0.05 (Supplementary Table S1 online); in addition, the ‘stable’ blood pressure group had fewer diabetics (30.7%) compared with the fall in SBP (45%) group and with the increased blood pressure group (41.7%). There were no differences in dialysis prescription (Supplementary Table S2 online). A total of 17 nursing interventions were recorded during the dialysis sessions; 10 patients had their ultrafiltration rate reduced or stopped (six in the fall in blood pressure group and four in the stable group), and seven patients were given a bolus of intravenous fluid (two in the fall in blood pressure group and four in the stable group); the mean SBP for these patients was 111.2±10.4 mm Hg. Predialysis hemoglobin and serum biochemistries did not differ between the groups, nor did dialysate composition or dialysis modality (Supplementary Table S2 online). Similarly, there was no difference in the dialysate to serum sodium gradient, and also when corrected for glucose effect, or between the preserum and postserum sodium (Supplementary Table S2 online). Urea reduction ratios were marginally, but 2
statistically significantly, greater in the stable cohort compared with those who increased their blood pressure post dialysis, 74.8±7.2 vs. 71.4±8.5% (P ¼ 0.032), but there was no difference between the group with a fall in SBP with dialysis, 74.4±7.1%. Postdialysis SBP differed between the groups (Figure 1). However, the predialysis SBP was higher in those patients who were documented to have a fall in SBP of X20 mm Hg, and it was lower in those who had a rise in SBP of X10 mm Hg (Figure 1). Predialysis diastolic blood pressures showed a similar pattern to predialysis SBP, but the difference after dialysis was limited to those groups that had decreased and increased blood pressures post dialysis, respectively (Figure 2). Neither weight, total body water (TBW), nor extracellular water (ECW) and intracellular water (ICW) differed between the groups before or after dialysis (Table 1). However, the change in absolute weight was less for the group in which blood pressure increased post dialysis (1.01±1.2 vs. stable 1.65±1.4, and 1.7±1.0 fall in blood pressure group, Po0.05), as was the percentage weight loss of 1.47±1.5 for the group in which blood pressure increased post dialysis (2.3±1.8 in the stable group and 2.3±1.5 in the fall in blood pressure group, respectively, Po0.05). The relative ratio of ECW to TBW was greater in those patients who had an increase in SBP post dialysis of X10 mm Hg (Table 1). Estimation of ECW overhydration was similarly greater for this group (Figure 3). A logistic regression model comparing patients in whom SBP increased post dialysis compared with patients with stable blood pressure or a fall in blood pressure showed that only ECW hydration status remained significant (b 0.382, s.e. 0.12, Wald 10.07 P ¼ 0.002). This finding was supported by a sensitivity analysis (Supplementary Table S3 online). An additional multivariable multilevel regression random intercept model was analyzed with the outcome SBP, clustered within subjects, examining the effect of volume status in tertiles and the dialysis procedure adjusted for age, sex and diabetes. Patients in the highest ECW/TBW tertile Systolic blood pressure (mm Hg)
RESULTS
A Nongnuch et al.: Changes in blood pressure with dialysis
250 200
Hypotensive Stable
*** ***
150
***
Hypertensive
100 50 0 Predialysis
Postdialysis
Figure 1 | Predialysis and postdialysis systolic blood pressure according to groupings. Patients with a fall in systolic blood pressure X20 mm Hg between predialysis and postdialysis systolic blood pressure recordings (hypotensive), patients with an increase in systolic blood pressure of X10 mm Hg (hypertensive), and those patients with systolic blood pressure change of o10 mm Hg to 19 mm Hg (stable). Values expressed as mean±s.d. ***Po0.001 vs. hypertensive group. Kidney International
clinical investigation
A Nongnuch et al.: Changes in blood pressure with dialysis
had the smallest change in SBP with hemodialysis compared with those in the lowest tertile, when adjusted for the mean preSBP (Figure 4). DISCUSSION
100
**
Hypotensive
**
80
Stable Hypertensive
60 40 20 0
ECW overhydration I
Diastolic blood pressure (mm Hg)
A fall in SBP remains the most common complication of routine outpatient hemodialysis treatments,2 and we observed a
fall of 20 mm Hg or greater in 32.2% of our patients attending their midweek dialysis session. On the other hand, an increase in SBP of 10 mm Hg or greater was documented in 18.1% of dialysis sessions, in keeping with more recent previous reports.8,9 Earlier reports on intradialytic hypertension suggested that it was more common in older patients and in those prescribed more antihypertensive medications.9 We did not find any differences in age or the number of
Hypotensive
1.5 1.0
Stable
*
0.5
Postdialysis
Figure 2 | Predialysis and postdialysis diastolic blood pressure according to groupings. Patients with a fall in systolic blood pressure X20 mm Hg between predialysis and postdialysis systolic blood pressure recordings (hypotensive), patients with an increase in systolic blood pressure of X10 mm Hg (hypertensive), and those patients with systolic blood pressure change o10 mm Hg to 19 mm Hg. Values expressed as mean±s.d., **Po0.01 vs. hypertensive group.
Hypertensive
*
*
0.0 –0.5
Predialysis
*
Predialysis
Postdialysis
Figure 3 | Predialysis and postdialysis relative extracellular water excess according to groupings. Patients with a fall in systolic blood pressure X20 mm Hg between predialysis and postdialysis systolic blood pressure recordings (hypotensive), patients with an increase in systolic blood pressure of X10 mm Hg (hypertensive), and those patients with systolic blood pressure change o10 mm Hg to 19 mm Hg. Values expressed as median (interquartile range), *Po0.05 vs. hypertensive group.
Table 1 | Patients were divided into three groups according to the change in postdialysis systolic blood pressure from the predialysis systolic blood pressure
Age, years Male (%) Caucasoid (%) Medical history (%) Hypertension (%) IHD (%) Valvular disease (%) BP meds (%) Hemoglobin, g/l Albumin, g/l CRP, mg/l Sodium, mmol/l Glucose, mmol/l Weight before dialysis, kg Weight after dialysis, kg SBP before dialysis, mm Hg SBP after dialysis, mm Hg TBW before dialysis, l ECW before dialysis, l ICW before dialysis, l ECW/TBW before dialysis TBW after dialysis, l ECW after dialysis, l ICW after dialysis, l ECW/TBW after dialysis
Hypotensive
Stable
Hypertensive
P-value
61.5±15.1 59.1 43.9
59.4±17.4 61.4 39.4
60.9±16.6 64.6 46.9
40.05 40.05 40.05
80.7 15.8 9.4 52.6 114.0±14.9 39.4±7.4 5.0 (2.0–11.0) 139.2±3.2 6.6 (5.2–8.8) 74.1±18.2 72.5±18.2 152.6±25.1 119.3±22.1 36.6±8.6 14.4±3.2 22.1±5.2 39.5±1.3 34.9±7.7 13.5±2.9 21.5±4.9 38.7±1.5
80.7 22.3 9.5 47.7 113.0±15.8 38.6±8.2 5.0 (2.0–14.0) 138.7±8.4 5.9 (5.0–7.7) 73.1±17.5 71.4±17.1 136.1±25.1 130±24.7 37.8±9.4 14.8±3.6 23.0±6.0 39.2±2.3 36.1±10.1 13.8±3.3 22.1±5.8 38.6±2.2
80.2 19.8 8.3 57.3 110±14.2 37.2±9.2 4.0 (2.0–14.0) 139.2±3.5 6.8 (5.4–9.1) 72.1±18.1 71.3±18.5 130.7±25.1 151.1±24.5 37.8±8.5 15.2±3.6 22.8±5.2 39.9±1.4 35.9±7.8 14.1±2.9 22.1±4.8 39.1±1.4
40.05 40.05 40.05 40.05 40.05 40.05 40.05 40.05 40.05 40.05 40.05 o0.001 o0.001 40.05 40.05 40.05 0.006 40.05 40.05 40.05 0.027
Abbreviations: BP, blood pressure; CRP, C-reactive protein; ECW, extracellular water; ICW, intracellular water; IHD, ischemic heart disease; SBP, systolic blood pressure; TBW, total body water. Patients with a fall in systolic blood pressure X20 mm Hg between predialysis and postdialysis systolic blood pressure recordings (hypotensive), patients with an increase in systolic blood pressure of X10 mm Hg (hypertensive), and those patients with systolic blood pressure change of o10 mm Hg to 19 mm Hg (stable). Ischemic heart disease (myocardial infarction, coronary artery bypass surgery, or coronary artery stenting), valvular heart disease (valvular disease), peripheral vascular disease (PVD), cerebrovascular disease (CVD), and percentage of patients prescribed antihypertensive drugs (BP meds). Multifrequency bioelectrical impedance assessments. TBW, ECW, and ICW, all reported in liters. Values are expressed as mean ±s.d., median (interquartile range), or percentage. P-values between hypertensive and hypotensive groups.
Kidney International
3
A Nongnuch et al.: Changes in blood pressure with dialysis
140
***
130 120 110
H ig he st EC W /T BW
M id dl e
EC W /T BW
100 Lo w es tE C W /T BW
Postdialysis systolic BP (mm Hg)
clinical investigation
Figure 4 | Postdialysis systolic blood pressure according to predialysis tertile of extracellular water (ECW) to total body water (TBW), analyzed by multivariable multilevel modeling, adjusted for predialysis systolic blood pressure of 129.5 mm Hg and for sex, age, and diabetes. ***Po0.001 vs. lowest tertile.
patients prescribed antihypertensive medications and the number of different antihypertensive drugs prescribed to patients. It has been suggested that, as activation of the renin–angiotensin system may have a role in intradialytic hypertension, angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) may help prevent this condition.3,4 However, we found a similar number of patients prescribed ACEIs among the groups, and if anything more patients with increased blood pressure post dialysis were prescribed ARBs or ACEIs and ARBs than the stable blood pressure group. This would suggest that in our patient cohort, prescription of ACEIs/ARBs did not significantly affect intradialytic hypertension. Activation of the sympathetic nervous system has also been proposed as a cause of intradialytic hypertension,4,9 but we did not observe any differences between the groups in terms of the prescription of beta-blockers or central sympathetic nervous system blockers, such as moxonidine. Other reports have raised the possibility that high concentrations of calcium in dialysates could be associated with intradialytic hypertension.3,9 However, we found no association with either dialysate electrolyte composition or temperature, or dialysis modality14 and whether patients had an increase in blood pressure post dialysis. An increased dialysate to serum sodium concentration gradient has also been reported to lead to hypertension, but again we could find no difference in the dialysate to serum sodium gradient,15 or differences in the predialysis to postdialysis serum sodium between the groups. In keeping with previous reports, we noted that those patients who had a fall in SBP of X20 mm Hg started dialysis with higher SBPs, and those who had a X10 mm Hg increase in postdialysis SBP had lower initial blood pressures.8,9 However, other studies have reported that although patients with intradialytic hypertension had lower predialysis blood pressures, they had similar 24-h blood pressures compared 4
with those with either intradialytic hypotension or stable intradialytic blood pressure.16 This highlights the difficulties of assuming that predialysis blood pressure can be used as a guide to intradialytic blood pressure control, and developing clinical guidelines based on predialysis blood pressure measurements.17 Although TBW and both ICW and ECW volumes were similar between the groups, this was probably because of the large variation in patient size18 and different ethnicities.19 As water content of tissues, in particular fat and muscle, differs, we chose to compare patients using the ratio of ECW/TBW.20 Those patients who had a rise in blood pressure post dialysis had higher ECW/TBW before dialysis, had lower weight loss during dialysis, and higher ECW/TBW post dialysis, suggesting that these patients were more volume overloaded than patients who had either a fall in SBP or stable blood pressure during dialysis. As a raised ECW/TBW ratio can also be caused by inflammation and a loss in ICW, we looked at predialysis albumin and CRP concentrations, but found no differences. We estimated the relative ECW to ICW as recommended by the European Society of Parenteral Nutrition, and this showed that those patients who increased their postdialysis blood pressure had ECW overhydration both before and after dialysis compared with the other groups. As such our study would suggest that patients who had increased blood pressure post dialysis are most likely to be volume overloaded, and downward revision of their target weight should be considered. Our study does not exclude changes in the vascular response to volume overload.11 We only analyzed predialysis and postdialysis blood pressures rather than changes in intradialytic blood pressure. However, the number of patients requiring nursing interventions was modest at 3.2%, with only 1.3% of patients requiring intravenous fluids for treating symptomatically low blood pressure during dialysis. As such the risk of misclassifying patients between the groups was small. We chose to measure bioimpedance before and then after the midweek dialysis session, and, as patient weight gains and volume status changes during the course of the dialysis week, it would be interesting to serially follow up patients to determine whether they change the group according to differences in volume status. Our study has a number of limitations. First, we only included a single hemodialysis session for each patient, and a more robust approach would be to classify changes in blood pressure with hemodialysis over time. However, we did not have the serial bioimpedance data to perform this analysis. Blood pressure was recorded in the dialysis centers using integral blood pressure and stand-alone blood pressure machines. Although the stand-alone blood pressure machines are regularly serviced and calibrated, the integral blood pressure devices were calibrated by the dialysis machine manufacturer. We have internal audit data looking at the reliability of both integral and stand-alone blood pressure devices in our dialysis centers, and this does not show any systematic bias between the various devices. As our bioimpedance device passes electrical currents through the Kidney International
clinical investigation
A Nongnuch et al.: Changes in blood pressure with dialysis
body, we were unable to perform bioimpedance measurements in all patients, excluding those with pacemakers, metallic implants, and those with amputations and active foot ulceration. Even so, with more than 80% of the dialysis population evaluated, we feel that our data are robust. Our study does support other reports that, after intradialytic hypotension, intradialytic hypertension is a relatively common complication of routine outpatient dialysis treatments. Patients with intradialytic hypertension have been reported to have an increased mortality risk,5,6 and our study would suggest that this increased risk of death is associated with persistent volume overload. As such patients with postdialysis hypertension should have their target weight reviewed and may benefit from longer or more frequent dialysis sessions. MATERIALS AND METHODS The bioimpedance records of 531 prevalent adult hemodialysis outpatients who had attended their routine midweek dialysis session were audited. Multifrequency bioelectrical impedance analysis measurements were made before and then approximately 20 min after dialysis (InBody 720 Body Composition Analysis, Biospace, Seoul, South Korea)21 using a direct segmental multifrequency bioelectrical impedance analysis method with tetrapolar 8-point tactile electrodes. A total of 30 impedance measurements were taken by using six frequencies (1–1000 kHz) at each of five segments (right arm, left arm, trunk, right leg, and left leg), and reactance by 15 impedance measurements using three frequencies (5–250 kHz) at each of the five segments.22 ECW volume excess or ECW overhydration was estimated according to the method recommended by the European Society for Parenteral and Enteral Nutrition.23 Patients with cardiac pacemakers, implantable defibrillators, amputees, and those unable to stand on the bioimpedance machine were excluded from study. Similarly 18 patients who had either recently started on dialysis who had not achieved their clinically determined target weight or established hemodialysis patients recently discharged from hospital due to recent illness requiring emergency admission who had not re-established a stable target weight were excluded. All patients were dialyzed using Fresenius F4000H or 5000H dialysis machines (Fresenius, Bad Homburg, Germany), Braun Dialog þ (B. Braun, Melsungen, Germany) with integrated blood pressure monitoring, and polysulfone high-flux dialyzers (Nipro, Osaka, Japan),24 with ultrapure quality dialysis water at a modal temperature of 35 1C and anticoagulated with a tinzaparin bolus (Leo Laboratories, Princes Risborough, UK).25 Delivered dialysate sodium was regularly checked by both flame photometry and ion electrophoresis methods.26 Predialysis and postdialysis blood samples were measured using a standard laboratory autoanalyzer (Roche Integra, Roche Diagnostics, Lewes, UK), with an indirect ion-selective electrode technique for sodium, and predialysis serum sodium was also corrected for glucose interference.27 Serum albumin was determined by the bromcresol green method and hemoglobin by autoanalyzer (XE-2100 Sysmex, Kobe, Japan). Blood pressure was taken in a standardized manner both immediately before starting dialysis and after dialysis in the nonfistula arm while in the sitting position using the hemodialysis machine integrated electronic blood pressure monitor. In cases of dialysis machines without functioning integral blood pressure measuring devices, blood pressure was measured using a Dinamap (Dinamap Pro100, Critikon, Tampa, FL). Blood pressure devices Kidney International
were regularly serviced and calibrated by the medical physics department at the Royal Free Hospital. All patients received dietetic advice to reduce dietary sodium intake. Patients did not receive intravenous iron or erythropoietin-stimulating agents during their midweek dialysis session. Ethics approval for this retrospective audit fulfilled the UK National Health Service audit and clinical service development guidelines. Statistical analysis Results are expressed as mean±s.d., or median and interquartile range, or percentage. Statistical analysis was by w2-analysis, corrected for small numbers by Yates’ correction, Students’ t-test for parametric data, and the Mann–Whitney U-test for nonparametric data, with Bonferroni correction for multiple analyses where appropriate, and by analysis of variance or Kruskall–Wallis test with Tukey’s or Dunn’s post hoc correction, as appropriate, as well as by backward logistic regression analysis, excluding variables that were not statistically relevant unless they improved model fit. In addition, to analyze the effect of predialysis and postdialysis SBP, multivariable multilevel regression random intercept analysis was undertaken adjusted for age, sex, and diabetes. Statistical analysis used Graph Pad Prism version 6.0 (Graph Pad, San Diego, CA), SPSS version 20 (University Chicago, IL), and Stata version 10a (StataCorp, College Station, TX). Statistical significance was taken at or below the 5% level. DISCLOSURE
All the authors declared no competing interests. ACKNOWLEDGMENTS
We thank D Nitsch and B Caplin for statistical advice and statistical analysis. This work was funded by Royal Free Hospital. SUPPLEMENTARY MATERIAL Figure S1. Scatter plot of pre and post dialysis systolic blood pressures for the 3 groups, patients with a fall in systolic blood pressure X20 mmHg between pre and post dialysis systolic blood pressure recordings (hypotensive), patients with an increase in systolic blood pressure of X10 mmHg (Hypertensive) and those patients with systolic blood pressure change of o10 mmHg to 19 mmHg (Stable). Table S1. Patients divided into 3 groups: intradialytic hypotension (Hypotensive) fall in systolic blood pressure 420 mmHg, intradialytic hypertension (Hypertensive) systolic blood pressure rose by 410 mmHg, and those patients with no hypertension or hypotension (Stable). Table S2. Patients divided into 3 groups: intradialytic hypotension (Hypotensive) fall in systolic blood pressure 420 mmHg, intradialytic hypertension (Hypertensive) systolic blood pressure rose by 410 mmHg, and those patients with no hypertension or hypotension (Stable). Table S3. Sensitivity analysis comparing those patients who had an increase of systolic blood pressure of 10 mmHg or greater compared to those patients who had an increase in systolic blood pressure of up to 9 mmHg or a fall in systolic blood pressure post dialysis. Supplementary material is linked to the online version of the paper at http://www.nature.com/ki REFERENCES 1. 2. 3.
Caplin B, Kumar S, Davenport A. Patients’ perspective of haemodialysisassociated symptoms. Nephrol Dial Transplant 2011; 26: 2656–2663. Davenport A. Intradialytic complications during haemodialysis. Hemodial Int 2006; 10: 162–167. Levin NW. Intradialytic hypertension: I. Semin Dial 1993; 6: 370–371.
5
clinical investigation
4. 5.
6.
7. 8.
9. 10.
11.
12.
13.
14. 15.
16.
6
Fellner SK. Intradialytic hypertension: II. Semin Dial 1993; 6: 371–373. Inrig JK, Oddone EZ, Hasselblad V et al. Association of intradialytic blood pressure changes with hospitalization and mortality rates in prevalent ESRD patients. Kidney Int 2007; 71: 454–461. Inrig JK, Patel UD, Toto R et al. Association of blood pressure increases during haemodialysis with 2-year mortality in incident haemodialysis patients: a secondary analysis of the Dialysis Morbidity and Mortality Wave 2 Study. Am J Kid Dis 2009; 54: 881–890. Mees D. Rise in blood pressure during hemodialysis-ultrafiltration: a "paradoxical" phenomenon? Int J Artif Organs 1996; 19: 569–570. Van Buren PN, Kim C, Toto RD et al. The prevalence of persistent intradialytic hypertension in a hemodialysis population with extended follow-up. Int J Artif Organs 2012; 35: 1031–1038. Inrigg JK. Intradialytic hypertension: a less-recognized cardiovascular complication of haemodialysis. Am J Kidney Dis 2010; 55: 580–589. Booth J, Pinney J, Davenport A. Do changes in relative blood volume monitoring correlate to hemodialysis-associated hypotension? Nephron Clin Pract 2011; 117: c179–c183. Inrig JK, Van Buren P, Kim C et al. Probing the mechanisms of intradialytic hypertension: a pilot study targeting endothelial cell dysfunction. Clin J Am Soc Nephrol 2012; 7: 1300–1309. Shah A, Davenport A. Does a reduction in dialysate sodium improve blood pressure control in haemodialysis patients? Nephrology (Carlton) 2012; 17: 358–363. Davenport A. Audit of the effect of dialysate sodium concentration on inter-dialytic weight gains and blood pressure control in chronic haemodialysis patients. Nephron Clin Pract 2006; 104: c120–c125. Kooman J, Basci A, Pizzarelli F et al. EBPG guideline on haemodynamic instability. Nephrol Dial Transplant 2007; 22(Suppl 2): ii22–ii44. Pinney JH, Oates T, Davenport A. Haemodiafiltration does not reduce the frequency of intradialytic hypotensive episodes when compared to cooled high-flux haemodialysis. Nephron Clin Pract 2011; 119: c138–c144. Van Buren PN, Toto R, Inrig JK. Interdialytic ambulatory blood pressure in patients with intradialytic hypertension. Curr Opin Nephrol Hypertens 2012; 21: 15–23.
A Nongnuch et al.: Changes in blood pressure with dialysis
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
Davenport A, Cox C, Thuraisingham R. Achieving blood pressure targets during dialysis improves control but increases intradialytic hypotension. Kidney Int 2008; 73: 759–764. Davenport A. Differences in prescribed Kt/V and delivered haemodialysis dose—why obesity makes a difference to survival for haemodialysis patients when using a ’one size fits all’ Kt/V target. Nephrol Dial Transplant 2013; 28(Suppl 4): iv219–iv223. Kumar S, Khosravi M, Maasart A et al. The effects of racial differences in body composition and total body water measured by multifrequency bioelectrical impedance analysis affect delivered Kt/V dialysis dosing. Nephron Clin Pract 2014; 124: 60–66. Van der Kerkhof J, Hermans M, Beerenhout C et al. Reference values for multifrequency bioimpedance analysis in dialysis patients. Blood Purif 2004; 22: 301–306. Davenport A. Negative dialysate to sodium gradient does not lead to intracellular volume expansion post haemodialysis. Int J Artif Organs 2010; 33: 700–705. Fu¨rstenberg A, Davenport A. Comparison of multifrequency bioelectrical impedance analysis and dual-energy X-ray absorptiometry assessments in outpatient haemodialysis patients. Am J Kidney Dis 2010; 57: 123–129. Kyle UG, Bosaeus I, De Lorenzo AD et al. Composition of the ESPEN Working Group. Bioelectrical impedance analysis—part I: review of principles and methods. Clin Nutr 2004; 23: 1226–1243. Vernon K, Peasegood J, Riddell A et al. Dialyzers designed to increase internal filtration do not result in significantly increased platelet activation and thrombin generation. Nephron Clin Pract 2011; 117: c403–c408. Davenport A. Low-molecular-weight heparin as an alternative anticoagulant to unfractionated heparin for routine outpatient haemodialysis treatments. Nephrology (Carlton) 2009; 14: 455–461. Sandhu E, Crawford C, Davenport A. Weight gains and increased blood pressure in outpatient hemodialysis patients due to change in acid dialysate concentrate supplier. Int J Artif Organs 2012; 35: 642–647. Davenport A. Interdialytic weight gain in diabetic haemodialysis patients and diabetic control as assessed by glycated haemoglobin. Nephron Clin Pract 2009; 113: c33–c37.
Kidney International
clinical investigation
& 2014 International Society of Nephrology
Increased postdialysis systolic blood pressure is associated with extracellular overhydration in hemodialysis outpatients Arkom Nongnuch1, Neil Campbell1, Edward Stern1, Sally El-Kateb1, Laura Fuentes1 and Andrew Davenport1 1
UCL Center for Nephrology, Royal Free Hospital, University College London Medical School, London, UK
Recently, intradialytic hypertension was reported to be associated with increased mortality for hemodialysis patients. To determine whether volume status plays a role in dialysisassociated hypertension, we prospectively audited 531 patients that had volume assessments measured by multiplefrequency bioelectrical impedance during their midweek dialysis session. Mean pre- and postdialysis weights were 73.2 vs 71.7 kg, and systolic blood pressures (SBPs) 140.5 vs. 130.3 mm Hg, respectively. Patients were divided into groups based on a fall in SBP of 20 mm Hg or more (32%), an increased SBP of 10 mm Hg or more (18%), and a stable group (50%). There were no differences in patient demographics, dialysis prescriptions, predialysis weight, total body (TBW), and extracellular (ECW) and intracellular water (ICW). However, the change in weight was significantly less in the increased blood pressure group (1.01 kg vs. stable 1.65, and 1.7 hypotensive). The ratio of ECW to TBW was significantly higher in the increased blood pressure group, particularly post dialysis (39.1 vs. stable 38.7% and fall in blood pressure group 38.7%). ECW overhydration was significantly greater in the increased blood pressure group post dialysis (0.7 (0.17 to 1.1) vs. stable 0.39 ( 0.2 to 0.95) and fall in blood pressure group 0.38 ( 0.19 to 0.86) liter). We found that patients who had increased blood pressure post dialysis had greater hydration status, particularly ECW. Thus, patients who increase their blood pressure post dialysis should have review of target weight, consideration of lowering the post-dialysis weight, and may benefit from increasing dialysis session time or frequency. Kidney International advance online publication, 30 July 2014; doi:10.1038/ki.2014.276 KEYWORDS: extracellular fluid; hypotension; hemodialysis; hypertension; multifrequency bioimpedance
Correspondence: Andrew Davenport, UCL Centre for Nephrology, Royal Free Hospital, University College London Medical School, Rowland Hill Street, London NW3 2PF, UK. E-mail: [email protected] Received 27 December 2013; revised 16 June 2014; accepted 19 June 2014 Kidney International
Although a fall in blood pressure remains the commonest complication associated with routine outpatient hemodialysis treatments,1,2 a minority of patients become hypertensive either toward or at the end of dialysis or shortly after completion of the dialysis session. Although a recognized complication of hemodialysis,3,4 it is only relatively recently that intradialytic hypertension has been recognized as an independent risk factor for both more frequent hospital admissions and also decreased patient survival.5,6 The definition of intradialytic hypertension has varied over the years,3–7 but most clinical studies now report intradialytic hypertension as an increase in systolic blood pressure (SBP) during dialysis of greater than 10 mm Hg above the predialysis blood pressure measurement.5,6 Intradialytic hypertension has been reported to occur in about 21% of hemodialysis sessions, affecting more than 15% of patients,8 and it is more frequently reported in older patients, those prescribed more antihypertensive medications, and those with lower serum creatinine.9 Although intravascular volume is typically reduced during dialysis, changes in relative blood volume do not necessarily closely mirror changes in blood pressure,10 and this has led to the hypothesis that intradialytic hypertension is more likely to be due to changes in vascular tone.11 The cause of intradialytic hypertension remains to be determined, with different authors speculating on increased renin and angiotensin release and sympathetic nervous system activation in response to ultrafiltration, elimination of antihypertensive medications in patients with established hypertension, increased viscosity and hematocrit following ultrafiltration, and high dialysate calcium concentrations.3,4 However, as there is a link between sodium balance and hypertension in hemodialysis patients,12,13 we speculated that patients with intradialytic hypertension could also potentially have an expanded extracellular volume to account for the rise in blood pressure during dialysis. We therefore reviewed the dialysis records of 531 patients who had volume assessments measured by multiple-frequency bioelectrical impedance to determine whether volume status had a role in dialysis-associated hypertension. 1
clinical investigation
Multifrequency bioelectrical impedance assessments were made on 531 stable adult patients attending for their midweek hemodialysis session; the mean age was 60.3±16.5 years, 62.0% were male, 37.3% were diabetic, 42.2% were Caucasoid, and the median dialysis vintage was 42.0 (23.7–48.5) months. Predialysis weight was 73.2±18.0 kg, and postdialysis weight was 71.7±17.7 kg. Predialysis SBP was 140.5±26.5 mm Hg with a diastolic blood pressure of 71.7±17.7 mm Hg, and postdialysis systolic pressure was 130.3±26.3 mm Hg with a diastolic blood pressure of 71.8± 15.2 mm Hg. Predialysis hemoglobin was 112.6±15.2 g/l, median glucose was 6.2 (5.2–8.3) mmol/l, serum albumin was 38.6±6.2 g/l, and median C-reactive protein was 5 (2–12) mg/l. A total of 67 patients were not included in the audit as bioimpedance was not performed, because 14 patients had pacemakers/defibrillators, 15 were unable to stand owing to amputations or active foot ulceration, 12 were recent starters, 6 had recent hospital in-patient stays, and the remaining 20 patients had a variety of metallic prosthetic implants—hips and knees—or were unable to stand. As such bioimpedance measurements were made on 89% of patients attending for their midweek dialysis session. Patients were divided into three groups, those in whom the blood pressure decreased during dialysis by X20 mm Hg,14 n ¼ 171 (32.2%), those in whom the SBP increased by X10 mm Hg,5 n ¼ 96 (18.1%), and 264 patients (49.7%) in whom the SBP did not meet the definitions of intradialytic hypotension or hypertension, from here on termed the ‘stable’ group (Supplementary Figure online). These three groups did not differ in demographics, or cardiac history and medications (Supplementary Table S1 online), although those with a fall in blood pressure with dialysis were of an older dialysis vintage (46 (15–74) months) compared with both those with ‘stable’ (27 (9–64) months) and those with an increase in SBP post dialysis (23 (5–63.6) months), Po0.05 (Supplementary Table S1 online); in addition, the ‘stable’ blood pressure group had fewer diabetics (30.7%) compared with the fall in SBP (45%) group and with the increased blood pressure group (41.7%). There were no differences in dialysis prescription (Supplementary Table S2 online). A total of 17 nursing interventions were recorded during the dialysis sessions; 10 patients had their ultrafiltration rate reduced or stopped (six in the fall in blood pressure group and four in the stable group), and seven patients were given a bolus of intravenous fluid (two in the fall in blood pressure group and four in the stable group); the mean SBP for these patients was 111.2±10.4 mm Hg. Predialysis hemoglobin and serum biochemistries did not differ between the groups, nor did dialysate composition or dialysis modality (Supplementary Table S2 online). Similarly, there was no difference in the dialysate to serum sodium gradient, and also when corrected for glucose effect, or between the preserum and postserum sodium (Supplementary Table S2 online). Urea reduction ratios were marginally, but 2
statistically significantly, greater in the stable cohort compared with those who increased their blood pressure post dialysis, 74.8±7.2 vs. 71.4±8.5% (P ¼ 0.032), but there was no difference between the group with a fall in SBP with dialysis, 74.4±7.1%. Postdialysis SBP differed between the groups (Figure 1). However, the predialysis SBP was higher in those patients who were documented to have a fall in SBP of X20 mm Hg, and it was lower in those who had a rise in SBP of X10 mm Hg (Figure 1). Predialysis diastolic blood pressures showed a similar pattern to predialysis SBP, but the difference after dialysis was limited to those groups that had decreased and increased blood pressures post dialysis, respectively (Figure 2). Neither weight, total body water (TBW), nor extracellular water (ECW) and intracellular water (ICW) differed between the groups before or after dialysis (Table 1). However, the change in absolute weight was less for the group in which blood pressure increased post dialysis (1.01±1.2 vs. stable 1.65±1.4, and 1.7±1.0 fall in blood pressure group, Po0.05), as was the percentage weight loss of 1.47±1.5 for the group in which blood pressure increased post dialysis (2.3±1.8 in the stable group and 2.3±1.5 in the fall in blood pressure group, respectively, Po0.05). The relative ratio of ECW to TBW was greater in those patients who had an increase in SBP post dialysis of X10 mm Hg (Table 1). Estimation of ECW overhydration was similarly greater for this group (Figure 3). A logistic regression model comparing patients in whom SBP increased post dialysis compared with patients with stable blood pressure or a fall in blood pressure showed that only ECW hydration status remained significant (b 0.382, s.e. 0.12, Wald 10.07 P ¼ 0.002). This finding was supported by a sensitivity analysis (Supplementary Table S3 online). An additional multivariable multilevel regression random intercept model was analyzed with the outcome SBP, clustered within subjects, examining the effect of volume status in tertiles and the dialysis procedure adjusted for age, sex and diabetes. Patients in the highest ECW/TBW tertile Systolic blood pressure (mm Hg)
RESULTS
A Nongnuch et al.: Changes in blood pressure with dialysis
250 200
Hypotensive Stable
*** ***
150
***
Hypertensive
100 50 0 Predialysis
Postdialysis
Figure 1 | Predialysis and postdialysis systolic blood pressure according to groupings. Patients with a fall in systolic blood pressure X20 mm Hg between predialysis and postdialysis systolic blood pressure recordings (hypotensive), patients with an increase in systolic blood pressure of X10 mm Hg (hypertensive), and those patients with systolic blood pressure change of o10 mm Hg to 19 mm Hg (stable). Values expressed as mean±s.d. ***Po0.001 vs. hypertensive group. Kidney International
clinical investigation
A Nongnuch et al.: Changes in blood pressure with dialysis
had the smallest change in SBP with hemodialysis compared with those in the lowest tertile, when adjusted for the mean preSBP (Figure 4). DISCUSSION
100
**
Hypotensive
**
80
Stable Hypertensive
60 40 20 0
ECW overhydration I
Diastolic blood pressure (mm Hg)
A fall in SBP remains the most common complication of routine outpatient hemodialysis treatments,2 and we observed a
fall of 20 mm Hg or greater in 32.2% of our patients attending their midweek dialysis session. On the other hand, an increase in SBP of 10 mm Hg or greater was documented in 18.1% of dialysis sessions, in keeping with more recent previous reports.8,9 Earlier reports on intradialytic hypertension suggested that it was more common in older patients and in those prescribed more antihypertensive medications.9 We did not find any differences in age or the number of
Hypotensive
1.5 1.0
Stable
*
0.5
Postdialysis
Figure 2 | Predialysis and postdialysis diastolic blood pressure according to groupings. Patients with a fall in systolic blood pressure X20 mm Hg between predialysis and postdialysis systolic blood pressure recordings (hypotensive), patients with an increase in systolic blood pressure of X10 mm Hg (hypertensive), and those patients with systolic blood pressure change o10 mm Hg to 19 mm Hg. Values expressed as mean±s.d., **Po0.01 vs. hypertensive group.
Hypertensive
*
*
0.0 –0.5
Predialysis
*
Predialysis
Postdialysis
Figure 3 | Predialysis and postdialysis relative extracellular water excess according to groupings. Patients with a fall in systolic blood pressure X20 mm Hg between predialysis and postdialysis systolic blood pressure recordings (hypotensive), patients with an increase in systolic blood pressure of X10 mm Hg (hypertensive), and those patients with systolic blood pressure change o10 mm Hg to 19 mm Hg. Values expressed as median (interquartile range), *Po0.05 vs. hypertensive group.
Table 1 | Patients were divided into three groups according to the change in postdialysis systolic blood pressure from the predialysis systolic blood pressure
Age, years Male (%) Caucasoid (%) Medical history (%) Hypertension (%) IHD (%) Valvular disease (%) BP meds (%) Hemoglobin, g/l Albumin, g/l CRP, mg/l Sodium, mmol/l Glucose, mmol/l Weight before dialysis, kg Weight after dialysis, kg SBP before dialysis, mm Hg SBP after dialysis, mm Hg TBW before dialysis, l ECW before dialysis, l ICW before dialysis, l ECW/TBW before dialysis TBW after dialysis, l ECW after dialysis, l ICW after dialysis, l ECW/TBW after dialysis
Hypotensive
Stable
Hypertensive
P-value
61.5±15.1 59.1 43.9
59.4±17.4 61.4 39.4
60.9±16.6 64.6 46.9
40.05 40.05 40.05
80.7 15.8 9.4 52.6 114.0±14.9 39.4±7.4 5.0 (2.0–11.0) 139.2±3.2 6.6 (5.2–8.8) 74.1±18.2 72.5±18.2 152.6±25.1 119.3±22.1 36.6±8.6 14.4±3.2 22.1±5.2 39.5±1.3 34.9±7.7 13.5±2.9 21.5±4.9 38.7±1.5
80.7 22.3 9.5 47.7 113.0±15.8 38.6±8.2 5.0 (2.0–14.0) 138.7±8.4 5.9 (5.0–7.7) 73.1±17.5 71.4±17.1 136.1±25.1 130±24.7 37.8±9.4 14.8±3.6 23.0±6.0 39.2±2.3 36.1±10.1 13.8±3.3 22.1±5.8 38.6±2.2
80.2 19.8 8.3 57.3 110±14.2 37.2±9.2 4.0 (2.0–14.0) 139.2±3.5 6.8 (5.4–9.1) 72.1±18.1 71.3±18.5 130.7±25.1 151.1±24.5 37.8±8.5 15.2±3.6 22.8±5.2 39.9±1.4 35.9±7.8 14.1±2.9 22.1±4.8 39.1±1.4
40.05 40.05 40.05 40.05 40.05 40.05 40.05 40.05 40.05 40.05 40.05 o0.001 o0.001 40.05 40.05 40.05 0.006 40.05 40.05 40.05 0.027
Abbreviations: BP, blood pressure; CRP, C-reactive protein; ECW, extracellular water; ICW, intracellular water; IHD, ischemic heart disease; SBP, systolic blood pressure; TBW, total body water. Patients with a fall in systolic blood pressure X20 mm Hg between predialysis and postdialysis systolic blood pressure recordings (hypotensive), patients with an increase in systolic blood pressure of X10 mm Hg (hypertensive), and those patients with systolic blood pressure change of o10 mm Hg to 19 mm Hg (stable). Ischemic heart disease (myocardial infarction, coronary artery bypass surgery, or coronary artery stenting), valvular heart disease (valvular disease), peripheral vascular disease (PVD), cerebrovascular disease (CVD), and percentage of patients prescribed antihypertensive drugs (BP meds). Multifrequency bioelectrical impedance assessments. TBW, ECW, and ICW, all reported in liters. Values are expressed as mean ±s.d., median (interquartile range), or percentage. P-values between hypertensive and hypotensive groups.
Kidney International
3
A Nongnuch et al.: Changes in blood pressure with dialysis
140
***
130 120 110
H ig he st EC W /T BW
M id dl e
EC W /T BW
100 Lo w es tE C W /T BW
Postdialysis systolic BP (mm Hg)
clinical investigation
Figure 4 | Postdialysis systolic blood pressure according to predialysis tertile of extracellular water (ECW) to total body water (TBW), analyzed by multivariable multilevel modeling, adjusted for predialysis systolic blood pressure of 129.5 mm Hg and for sex, age, and diabetes. ***Po0.001 vs. lowest tertile.
patients prescribed antihypertensive medications and the number of different antihypertensive drugs prescribed to patients. It has been suggested that, as activation of the renin–angiotensin system may have a role in intradialytic hypertension, angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) may help prevent this condition.3,4 However, we found a similar number of patients prescribed ACEIs among the groups, and if anything more patients with increased blood pressure post dialysis were prescribed ARBs or ACEIs and ARBs than the stable blood pressure group. This would suggest that in our patient cohort, prescription of ACEIs/ARBs did not significantly affect intradialytic hypertension. Activation of the sympathetic nervous system has also been proposed as a cause of intradialytic hypertension,4,9 but we did not observe any differences between the groups in terms of the prescription of beta-blockers or central sympathetic nervous system blockers, such as moxonidine. Other reports have raised the possibility that high concentrations of calcium in dialysates could be associated with intradialytic hypertension.3,9 However, we found no association with either dialysate electrolyte composition or temperature, or dialysis modality14 and whether patients had an increase in blood pressure post dialysis. An increased dialysate to serum sodium concentration gradient has also been reported to lead to hypertension, but again we could find no difference in the dialysate to serum sodium gradient,15 or differences in the predialysis to postdialysis serum sodium between the groups. In keeping with previous reports, we noted that those patients who had a fall in SBP of X20 mm Hg started dialysis with higher SBPs, and those who had a X10 mm Hg increase in postdialysis SBP had lower initial blood pressures.8,9 However, other studies have reported that although patients with intradialytic hypertension had lower predialysis blood pressures, they had similar 24-h blood pressures compared 4
with those with either intradialytic hypotension or stable intradialytic blood pressure.16 This highlights the difficulties of assuming that predialysis blood pressure can be used as a guide to intradialytic blood pressure control, and developing clinical guidelines based on predialysis blood pressure measurements.17 Although TBW and both ICW and ECW volumes were similar between the groups, this was probably because of the large variation in patient size18 and different ethnicities.19 As water content of tissues, in particular fat and muscle, differs, we chose to compare patients using the ratio of ECW/TBW.20 Those patients who had a rise in blood pressure post dialysis had higher ECW/TBW before dialysis, had lower weight loss during dialysis, and higher ECW/TBW post dialysis, suggesting that these patients were more volume overloaded than patients who had either a fall in SBP or stable blood pressure during dialysis. As a raised ECW/TBW ratio can also be caused by inflammation and a loss in ICW, we looked at predialysis albumin and CRP concentrations, but found no differences. We estimated the relative ECW to ICW as recommended by the European Society of Parenteral Nutrition, and this showed that those patients who increased their postdialysis blood pressure had ECW overhydration both before and after dialysis compared with the other groups. As such our study would suggest that patients who had increased blood pressure post dialysis are most likely to be volume overloaded, and downward revision of their target weight should be considered. Our study does not exclude changes in the vascular response to volume overload.11 We only analyzed predialysis and postdialysis blood pressures rather than changes in intradialytic blood pressure. However, the number of patients requiring nursing interventions was modest at 3.2%, with only 1.3% of patients requiring intravenous fluids for treating symptomatically low blood pressure during dialysis. As such the risk of misclassifying patients between the groups was small. We chose to measure bioimpedance before and then after the midweek dialysis session, and, as patient weight gains and volume status changes during the course of the dialysis week, it would be interesting to serially follow up patients to determine whether they change the group according to differences in volume status. Our study has a number of limitations. First, we only included a single hemodialysis session for each patient, and a more robust approach would be to classify changes in blood pressure with hemodialysis over time. However, we did not have the serial bioimpedance data to perform this analysis. Blood pressure was recorded in the dialysis centers using integral blood pressure and stand-alone blood pressure machines. Although the stand-alone blood pressure machines are regularly serviced and calibrated, the integral blood pressure devices were calibrated by the dialysis machine manufacturer. We have internal audit data looking at the reliability of both integral and stand-alone blood pressure devices in our dialysis centers, and this does not show any systematic bias between the various devices. As our bioimpedance device passes electrical currents through the Kidney International
clinical investigation
A Nongnuch et al.: Changes in blood pressure with dialysis
body, we were unable to perform bioimpedance measurements in all patients, excluding those with pacemakers, metallic implants, and those with amputations and active foot ulceration. Even so, with more than 80% of the dialysis population evaluated, we feel that our data are robust. Our study does support other reports that, after intradialytic hypotension, intradialytic hypertension is a relatively common complication of routine outpatient dialysis treatments. Patients with intradialytic hypertension have been reported to have an increased mortality risk,5,6 and our study would suggest that this increased risk of death is associated with persistent volume overload. As such patients with postdialysis hypertension should have their target weight reviewed and may benefit from longer or more frequent dialysis sessions. MATERIALS AND METHODS The bioimpedance records of 531 prevalent adult hemodialysis outpatients who had attended their routine midweek dialysis session were audited. Multifrequency bioelectrical impedance analysis measurements were made before and then approximately 20 min after dialysis (InBody 720 Body Composition Analysis, Biospace, Seoul, South Korea)21 using a direct segmental multifrequency bioelectrical impedance analysis method with tetrapolar 8-point tactile electrodes. A total of 30 impedance measurements were taken by using six frequencies (1–1000 kHz) at each of five segments (right arm, left arm, trunk, right leg, and left leg), and reactance by 15 impedance measurements using three frequencies (5–250 kHz) at each of the five segments.22 ECW volume excess or ECW overhydration was estimated according to the method recommended by the European Society for Parenteral and Enteral Nutrition.23 Patients with cardiac pacemakers, implantable defibrillators, amputees, and those unable to stand on the bioimpedance machine were excluded from study. Similarly 18 patients who had either recently started on dialysis who had not achieved their clinically determined target weight or established hemodialysis patients recently discharged from hospital due to recent illness requiring emergency admission who had not re-established a stable target weight were excluded. All patients were dialyzed using Fresenius F4000H or 5000H dialysis machines (Fresenius, Bad Homburg, Germany), Braun Dialog þ (B. Braun, Melsungen, Germany) with integrated blood pressure monitoring, and polysulfone high-flux dialyzers (Nipro, Osaka, Japan),24 with ultrapure quality dialysis water at a modal temperature of 35 1C and anticoagulated with a tinzaparin bolus (Leo Laboratories, Princes Risborough, UK).25 Delivered dialysate sodium was regularly checked by both flame photometry and ion electrophoresis methods.26 Predialysis and postdialysis blood samples were measured using a standard laboratory autoanalyzer (Roche Integra, Roche Diagnostics, Lewes, UK), with an indirect ion-selective electrode technique for sodium, and predialysis serum sodium was also corrected for glucose interference.27 Serum albumin was determined by the bromcresol green method and hemoglobin by autoanalyzer (XE-2100 Sysmex, Kobe, Japan). Blood pressure was taken in a standardized manner both immediately before starting dialysis and after dialysis in the nonfistula arm while in the sitting position using the hemodialysis machine integrated electronic blood pressure monitor. In cases of dialysis machines without functioning integral blood pressure measuring devices, blood pressure was measured using a Dinamap (Dinamap Pro100, Critikon, Tampa, FL). Blood pressure devices Kidney International
were regularly serviced and calibrated by the medical physics department at the Royal Free Hospital. All patients received dietetic advice to reduce dietary sodium intake. Patients did not receive intravenous iron or erythropoietin-stimulating agents during their midweek dialysis session. Ethics approval for this retrospective audit fulfilled the UK National Health Service audit and clinical service development guidelines. Statistical analysis Results are expressed as mean±s.d., or median and interquartile range, or percentage. Statistical analysis was by w2-analysis, corrected for small numbers by Yates’ correction, Students’ t-test for parametric data, and the Mann–Whitney U-test for nonparametric data, with Bonferroni correction for multiple analyses where appropriate, and by analysis of variance or Kruskall–Wallis test with Tukey’s or Dunn’s post hoc correction, as appropriate, as well as by backward logistic regression analysis, excluding variables that were not statistically relevant unless they improved model fit. In addition, to analyze the effect of predialysis and postdialysis SBP, multivariable multilevel regression random intercept analysis was undertaken adjusted for age, sex, and diabetes. Statistical analysis used Graph Pad Prism version 6.0 (Graph Pad, San Diego, CA), SPSS version 20 (University Chicago, IL), and Stata version 10a (StataCorp, College Station, TX). Statistical significance was taken at or below the 5% level. DISCLOSURE
All the authors declared no competing interests. ACKNOWLEDGMENTS
We thank D Nitsch and B Caplin for statistical advice and statistical analysis. This work was funded by Royal Free Hospital. SUPPLEMENTARY MATERIAL Figure S1. Scatter plot of pre and post dialysis systolic blood pressures for the 3 groups, patients with a fall in systolic blood pressure X20 mmHg between pre and post dialysis systolic blood pressure recordings (hypotensive), patients with an increase in systolic blood pressure of X10 mmHg (Hypertensive) and those patients with systolic blood pressure change of o10 mmHg to 19 mmHg (Stable). Table S1. Patients divided into 3 groups: intradialytic hypotension (Hypotensive) fall in systolic blood pressure 420 mmHg, intradialytic hypertension (Hypertensive) systolic blood pressure rose by 410 mmHg, and those patients with no hypertension or hypotension (Stable). Table S2. Patients divided into 3 groups: intradialytic hypotension (Hypotensive) fall in systolic blood pressure 420 mmHg, intradialytic hypertension (Hypertensive) systolic blood pressure rose by 410 mmHg, and those patients with no hypertension or hypotension (Stable). Table S3. Sensitivity analysis comparing those patients who had an increase of systolic blood pressure of 10 mmHg or greater compared to those patients who had an increase in systolic blood pressure of up to 9 mmHg or a fall in systolic blood pressure post dialysis. Supplementary material is linked to the online version of the paper at http://www.nature.com/ki REFERENCES 1. 2. 3.
Caplin B, Kumar S, Davenport A. Patients’ perspective of haemodialysisassociated symptoms. Nephrol Dial Transplant 2011; 26: 2656–2663. Davenport A. Intradialytic complications during haemodialysis. Hemodial Int 2006; 10: 162–167. Levin NW. Intradialytic hypertension: I. Semin Dial 1993; 6: 370–371.
5
clinical investigation
4. 5.
6.
7. 8.
9. 10.
11.
12.
13.
14. 15.
16.
6
Fellner SK. Intradialytic hypertension: II. Semin Dial 1993; 6: 371–373. Inrig JK, Oddone EZ, Hasselblad V et al. Association of intradialytic blood pressure changes with hospitalization and mortality rates in prevalent ESRD patients. Kidney Int 2007; 71: 454–461. Inrig JK, Patel UD, Toto R et al. Association of blood pressure increases during haemodialysis with 2-year mortality in incident haemodialysis patients: a secondary analysis of the Dialysis Morbidity and Mortality Wave 2 Study. Am J Kid Dis 2009; 54: 881–890. Mees D. Rise in blood pressure during hemodialysis-ultrafiltration: a "paradoxical" phenomenon? Int J Artif Organs 1996; 19: 569–570. Van Buren PN, Kim C, Toto RD et al. The prevalence of persistent intradialytic hypertension in a hemodialysis population with extended follow-up. Int J Artif Organs 2012; 35: 1031–1038. Inrigg JK. Intradialytic hypertension: a less-recognized cardiovascular complication of haemodialysis. Am J Kidney Dis 2010; 55: 580–589. Booth J, Pinney J, Davenport A. Do changes in relative blood volume monitoring correlate to hemodialysis-associated hypotension? Nephron Clin Pract 2011; 117: c179–c183. Inrig JK, Van Buren P, Kim C et al. Probing the mechanisms of intradialytic hypertension: a pilot study targeting endothelial cell dysfunction. Clin J Am Soc Nephrol 2012; 7: 1300–1309. Shah A, Davenport A. Does a reduction in dialysate sodium improve blood pressure control in haemodialysis patients? Nephrology (Carlton) 2012; 17: 358–363. Davenport A. Audit of the effect of dialysate sodium concentration on inter-dialytic weight gains and blood pressure control in chronic haemodialysis patients. Nephron Clin Pract 2006; 104: c120–c125. Kooman J, Basci A, Pizzarelli F et al. EBPG guideline on haemodynamic instability. Nephrol Dial Transplant 2007; 22(Suppl 2): ii22–ii44. Pinney JH, Oates T, Davenport A. Haemodiafiltration does not reduce the frequency of intradialytic hypotensive episodes when compared to cooled high-flux haemodialysis. Nephron Clin Pract 2011; 119: c138–c144. Van Buren PN, Toto R, Inrig JK. Interdialytic ambulatory blood pressure in patients with intradialytic hypertension. Curr Opin Nephrol Hypertens 2012; 21: 15–23.
A Nongnuch et al.: Changes in blood pressure with dialysis
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
Davenport A, Cox C, Thuraisingham R. Achieving blood pressure targets during dialysis improves control but increases intradialytic hypotension. Kidney Int 2008; 73: 759–764. Davenport A. Differences in prescribed Kt/V and delivered haemodialysis dose—why obesity makes a difference to survival for haemodialysis patients when using a ’one size fits all’ Kt/V target. Nephrol Dial Transplant 2013; 28(Suppl 4): iv219–iv223. Kumar S, Khosravi M, Maasart A et al. The effects of racial differences in body composition and total body water measured by multifrequency bioelectrical impedance analysis affect delivered Kt/V dialysis dosing. Nephron Clin Pract 2014; 124: 60–66. Van der Kerkhof J, Hermans M, Beerenhout C et al. Reference values for multifrequency bioimpedance analysis in dialysis patients. Blood Purif 2004; 22: 301–306. Davenport A. Negative dialysate to sodium gradient does not lead to intracellular volume expansion post haemodialysis. Int J Artif Organs 2010; 33: 700–705. Fu¨rstenberg A, Davenport A. Comparison of multifrequency bioelectrical impedance analysis and dual-energy X-ray absorptiometry assessments in outpatient haemodialysis patients. Am J Kidney Dis 2010; 57: 123–129. Kyle UG, Bosaeus I, De Lorenzo AD et al. Composition of the ESPEN Working Group. Bioelectrical impedance analysis—part I: review of principles and methods. Clin Nutr 2004; 23: 1226–1243. Vernon K, Peasegood J, Riddell A et al. Dialyzers designed to increase internal filtration do not result in significantly increased platelet activation and thrombin generation. Nephron Clin Pract 2011; 117: c403–c408. Davenport A. Low-molecular-weight heparin as an alternative anticoagulant to unfractionated heparin for routine outpatient haemodialysis treatments. Nephrology (Carlton) 2009; 14: 455–461. Sandhu E, Crawford C, Davenport A. Weight gains and increased blood pressure in outpatient hemodialysis patients due to change in acid dialysate concentrate supplier. Int J Artif Organs 2012; 35: 642–647. Davenport A. Interdialytic weight gain in diabetic haemodialysis patients and diabetic control as assessed by glycated haemoglobin. Nephron Clin Pract 2009; 113: c33–c37.
Kidney International
