Hemodynamic response to fluid removal during hemodialysis: categorization of causes of intradialytic hypotension

Hemodynamic response to fluid removal during hemodialysis: categorization of causes of... Abstract Background Intradialytic hypotension is a clinically significant problem, however, the hemodynamics that underlie ultrafiltration and consequent hypotensive episodes has not been studied comprehensively. Methods Intradialytic cardiac output, cardiac power and peripheral resistance changes from pretreatment measurements were evaluated using a novel regional impedance cardiographic device (NICaS, NI Medical, Peta Tikva, Israel) in 263 hemodialysis sessions in 54 patients in dialysis units in the USA and Brazil with the goal of determining the various hemodynamic trends as blood pressure decreases. Results Hypotensive episodes occurred in 99 (13.5%) of 736 intra- and postdialytic evaluations. The hemodynamic profiles of the episodes were categorized: (i) The cardiac power index significantly decreased in 35% of episodes by 36%, from 0.66 [95% confidence interval (CI) 0.60–0.72] to 0.43 (95% CI 0.37–0.48) [w/m2] with a small reduction in the total peripheral resistance index. (ii) The total peripheral resistance index significantly decreased in 37.4% of episodes by 33%, from 3342 (95% CI 2824–3859) to 2251 (95% CI 1900–2602) [dyn × s/cm5 × m2] with a small reduction in the cardiac power index. (iii) Both the cardiac power index and total peripheral resistance index significantly decreased in 27.3% of episodes, the cardiac power index by 25% from 0.63 (95% CI 0.57–0.70) to 0.48 (95% CI 0.42–0.53) [w/m2] and the total peripheral resistance index by 23% from 2964 (95% CI 2428–3501) to 2266 (95% CI 1891–2642). Conclusions The hemodynamic profiles clearly define specific hemodynamic mechanisms of cardiac power reduction and/or vasodilatation as underlying intradialytic hypotensive episodes. A reduction in cardiac power (reduction of both blood pressure and cardiac output) could be the result of preload reduction due to a high ultrafiltration rate with not enough refilling or low target weight. A reduction in peripheral resistance (reduction in blood pressure and increase in cardiac output) could be the result of relative vasodilatation as arteries do not contract to compensate for volume reduction due to autonomous dysfunction. As both phenomena are independent, they may appear at the same time. Based on these results, a reduction of ultrafiltration rate and an increase in target weight to improve preload or immediate therapeutic actions to increase peripheral resistance are rational measures that could be taken to maintain blood pressure and prevent hypotensive ischemic complications in dialysis patients. cardiovascular, hemodialysis, intradialytic hypotensive episodes, online intradialytic hemodynamics, ultrafiltration INTRODUCTION Hypotension occurring during chronic hemodialysis is a major clinical problem with consequent ischemia of the heart, brain and gut [1]. Reduced cardiac preload due to hypovolemia may be further diminished in the presence of diastolic dysfunction [2]. An increase in sympathetic tone may add functional problems [3]. These factors are associated with an increase in mortality in hemodialysis patients [4]. The hemodynamics underlying ultrafiltration and consequent hypotensive episodes have not been studied comprehensively. Zucchelli and Santoro in 1993 [5] suggested that intradialytic hypotensive episodes may be the result of hypovolemia, left ventricular diastolic dysfunction or a ‘breakdown’ in peripheral resistance, but they did not provide data showing intradialytic hemodynamic trends. The objective of this study is to report the hemodynamic changes during hemodialysis, specifically focusing on those responsible for intradialytic hypotensive episodes. All hemodynamic measurements were made using a regional impedance cardiography device (NICaS, NI Medical, Peta Tikva, Israel), which have been shown to correlate well with pulmonary artery catheterization thermodilution and echocardiographic data [6–8] and are more accurate than thoracic impedance cardiographic measurements [9]. The accuracy of the device during dialysis has been recently validated, demonstrating a strong correlation of stroke volume (SV) with echocardiographic measurements, with insignificant bias [10]. MATERIALS AND METHODS Patients Fifty-four established chronic hemodialysis patients at the Queens Artificial Kidney Center, New York (n = 26) and at the Biocor Hospital, Belo Horizonte, Brazil (n = 28) were studied. Patients were randomly selected. There were no inclusion criteria other than dialysis vintage of at least 3 months. All gave written informed consent. The study was approved by the New England Institutional Review Board (IRB) and by the Biocor Hospital de Doencas Cardiovasculares, Belo Horizonte IRB. Study protocol Hemodynamic trends were assessed using the device in randomly selected dialysis sessions. Sitting blood pressures were measured immediately prior to each hemodynamic measurement. Gender, age, height, weight, electrode location and blood pressure data were entered into the device. The device measures and calculates hemodynamic parameters on each heart beat during 60 s and provides the averaged parameters. Each patient’s hemodynamics were measured over a range of dialysis treatments with an average of 4.9 ± 2.3 treatments per patient. Measurements were made at baseline (pretreatment), at the middle, just before the end and 10 min after the end of the treatment. The dialysis staff was blinded to all measurements derived from the device. Technology used for hemodynamic measurements The device (NICaS, NI Medical) is a noninvasive regional bioimpedance cardiac measurement and analysis system (FDA 510k clearance no. K080941, 12 June 2009). The US Food and Drug Administration indication for use of the device states ‘NICaS is intended to monitor and display hemodynamic parameters in males and females with known or suspected cardiac disorders needing cardiac assessment’. SV is measured by applying an alternating electrical current of 1.4 mA at 30 kHz frequency through the patient’s body via two pairs of tetrapolar sensors, one pair placed on the wrist of the nonaccess arm above the radial pulse and the other pair on the contralateral ankle above the posterior tibial pulse (Figure 1). SV is calculated by Frinerman’s formula: SV = (dR/R) × ρ × (L2/Ri) × (α + β)/β × KW × HF [2–4], where dR is the impedance change in the arterial system as a result of intraarterial expansion during systole, R is basal resistance, ρ is blood electrical resistance, L is the patient’s height, Ri is basal resistance corrected for gender and age, KW is the correction of weight according to ideal values, HF is a hydration factor that takes into account the ratio between R and body mass index (BMI), which is correlated to body water volume, α + β is the electrocardiogram (ECG) R-R wave interval and β is the diastolic time interval. SV is automatically calculated every 20 s and is the average of three measurements obtained consecutively during 60 s of monitoring. The SV index is calculated as SV/body surface area using the Du Bois formula [11]. Heart rate is calculated from a one channel ECG and cardiac (output) index = SV index × heart rate/1000. FIGURE 1 View largeDownload slide ECG waveform and impedance waveform (left). Sensor locations (right). An alternating electrical current of 1.4 mA at 30 kHz frequency is applied through the patient’s body via two pairs of tetrapolar sensors. The impedance waveform provides the changes in electrical resistivity of the arterial system with each heartbeat. SV is calculated based on these changes. Heart rate (HR) is measured by the ECG that is sensed through the same sensors. Cardiac output (CO) is calculated as CO = SV×HR. FIGURE 1 View largeDownload slide ECG waveform and impedance waveform (left). Sensor locations (right). An alternating electrical current of 1.4 mA at 30 kHz frequency is applied through the patient’s body via two pairs of tetrapolar sensors. The impedance waveform provides the changes in electrical resistivity of the arterial system with each heartbeat. SV is calculated based on these changes. Heart rate (HR) is measured by the ECG that is sensed through the same sensors. Cardiac output (CO) is calculated as CO = SV×HR. Using an oscillometric method, sitting systolic and diastolic blood pressure measurements were made automatically by the dialysis machine. Mean arterial pressure [2 × (diastolic + systolic)/3], cardiac power index [CPI; mean arterial pressure (MAP) × cardiac index × 0.0022 w/m2; normal range 0.45–0.85 w/m2] [12, 13] and total peripheral resistance (MAP/cardiac index × 80 dyn × s/cm5 × m2; normal range 1600–3000 dyn × s/cm5 × m2) [13] were calculated. As the device measures pulsatile flow and is blinded to constant flow, fluid removal during dialysis has no impact on measurement accuracy. This was recently validated by correlating SV to ECG measurements during hemodialysis treatments. Good correlation was maintained during treatment. Further, NICaS performance immunity to fluid reduction was demonstrated by the maintenance of correlation to ECG results throughout dialysis treatments [10]. The results are drawn on hemodynamic graphs showing the MAP (y-axis) as a function of cardiac index (x-axis); curves of total peripheral resistance index (TPRI) and CPI are displayed. Ranges for the normal population are depicted by the dotted octagon (Figure 2). FIGURE 2 View largeDownload slide Hemodynamic trends of each individual evaluation in the three IDH subgroups using three graphs of hemodynamic trends of MAP versus CI. Representative lines of both CPI and TPRI are also shown. Hemodynamic ranges for the normal population are illustrated by a dashed octagon. Each arrow provides the MAP and CI changes from pretreatment of all intra- and postdialytic evaluations: substantial reductions of CPI and CI for subgroup 1, substantial reduction of TPRI and increase of CI in subgroup 2 and substantial reductions of both CPI and TPRI with no change of CI in subgroup 3 are shown. FIGURE 2 View largeDownload slide Hemodynamic trends of each individual evaluation in the three IDH subgroups using three graphs of hemodynamic trends of MAP versus CI. Representative lines of both CPI and TPRI are also shown. Hemodynamic ranges for the normal population are illustrated by a dashed octagon. Each arrow provides the MAP and CI changes from pretreatment of all intra- and postdialytic evaluations: substantial reductions of CPI and CI for subgroup 1, substantial reduction of TPRI and increase of CI in subgroup 2 and substantial reductions of both CPI and TPRI with no change of CI in subgroup 3 are shown. Statistical analysis The Kolmogorov–Smirnov test for normally distributed data was used. Descriptive statistics are presented as number (percentages) and mean ± SD. The comparisons for each parameter between the pretreatment and the intra- and postdialytic measurements were made according to the occurrence of intradialytic hypotension (IDH) episodes and by IDH subgroups related to specific hemodynamic changes. Since each evaluation was not only influenced by individual cardiovascular functional responses but also by sodium intake between dialysis and dialysis treatment variables, such as target weight, fluid removal rates and dialysate temperatures, each patient was measured during a few dialysis treatments and analysis of all comparisons was performed using a mixed model for repeated measurements [14] employing a mixed procedure (SAS/STAT software, version 9.4; SAS Institute, Cary, NC, USA) and adjusted for multiple comparisons using the Tukey–Kramer method [15]. Results are presented as mean and 95% confidence intervals (CIs). P-values ≤ 0.05 were considered to be significant. RESULTS: NON-IDH AND THREE IDH SUBGROUPS Fifty-four patients from two sites in the USA and Brazil were studied. The composition of the study group included 29 males (54%), mean age 67 ± 10 years, mean weight 74 ± 17 kg, mean BMI 27.6 ± 5.2 and 30 diabetic patients (56%). Adjusted pretreatment mean systolic blood pressure (SBP) was 139 (95% CI 133–144) mmHg, MAP was 95 (95% CI 92–98) mmHg, mean cardiac output index (CI) was 2.8 (95% CI 2.6–3.0) L/min/m2, mean CPI was 0.59 (95% CI 0.54–0.63) w/m2 and mean TPRI was 3147 (95% CI 2706–3589) dyn × s/cm5 × m2. Criteria for IDH were defined as an SBP reduction ≥20 mmHg OR MAP reduction ≥10 mmHg AND intradialytic nadir SBP to <100 mmHg or MAP <70 mmHg. Patients with at least one IDH episode in at least 30% of dialysis treatments were defined as having IDH. In a total of 263 treatments, 736 intra- and postdialytic measurements were evaluated for IDH episodes by comparing them with the relevant predialytic measurements. A total of 99 (13.5%) evaluations met the criteria for IDH. Fourteen (26%) patients were categorized as having IDH. The demographics of the IDH and non-IDH patients did not differ significantly (Table 1). IDH episodes were recorded in 61 (23.2%) of 263 treatments. In this latter group, pretreatment hemodynamics, fluid removal rates, total fluid removed and duration of treatments did not significantly differ from non-IDH treatments (Table 2). Table 1 Demographics of all patients, patients with no IDH episodes and patients with at least one IDH episode in 30% of treatments Parameter All (n = 54) Non-IDH [n = 40 (74.1%)] IDH [n = 14 (25.9%)] IDH–non-IDH P-value Male, n (%) 29 (54) 20 (50) 9 (64) 0.35 Age (years), mean ± SD 67 ± 10 67 ± 11 66 ± 9 0.87 Weight (kg), mean ± SD 74 ± 17 74 ± 14 75 ± 20 0.80 BMI, mean ± SD 27.6 ± 5.2 26.8 ± 4.50 28.3 ± 5.8 0.30 Diabetes, n (%) 30 (56) 22 (55) 8 (57) 0.81 Parameter All (n = 54) Non-IDH [n = 40 (74.1%)] IDH [n = 14 (25.9%)] IDH–non-IDH P-value Male, n (%) 29 (54) 20 (50) 9 (64) 0.35 Age (years), mean ± SD 67 ± 10 67 ± 11 66 ± 9 0.87 Weight (kg), mean ± SD 74 ± 17 74 ± 14 75 ± 20 0.80 BMI, mean ± SD 27.6 ± 5.2 26.8 ± 4.50 28.3 ± 5.8 0.30 Diabetes, n (%) 30 (56) 22 (55) 8 (57) 0.81 Table 1 Demographics of all patients, patients with no IDH episodes and patients with at least one IDH episode in 30% of treatments Parameter All (n = 54) Non-IDH [n = 40 (74.1%)] IDH [n = 14 (25.9%)] IDH–non-IDH P-value Male, n (%) 29 (54) 20 (50) 9 (64) 0.35 Age (years), mean ± SD 67 ± 10 67 ± 11 66 ± 9 0.87 Weight (kg), mean ± SD 74 ± 17 74 ± 14 75 ± 20 0.80 BMI, mean ± SD 27.6 ± 5.2 26.8 ± 4.50 28.3 ± 5.8 0.30 Diabetes, n (%) 30 (56) 22 (55) 8 (57) 0.81 Parameter All (n = 54) Non-IDH [n = 40 (74.1%)] IDH [n = 14 (25.9%)] IDH–non-IDH P-value Male, n (%) 29 (54) 20 (50) 9 (64) 0.35 Age (years), mean ± SD 67 ± 10 67 ± 11 66 ± 9 0.87 Weight (kg), mean ± SD 74 ± 17 74 ± 14 75 ± 20 0.80 BMI, mean ± SD 27.6 ± 5.2 26.8 ± 4.50 28.3 ± 5.8 0.30 Diabetes, n (%) 30 (56) 22 (55) 8 (57) 0.81 Table 2 Pretreatment hemodynamics and fluid removal data in all treatments in those with no IDH episodes and those with at least one IDH episode Parameter All treatments (n = 263) Non-IDH treatments [n= 202 (76.8%)] IDH treatments [n = 61(23.2%)] IDH versus non-IDH adjusted P-valuea Pretreatment hemodynamics, adjusted meana (95% CIa) SBP (mmHg) 139 (133–144) 138 (133–144) 139 (133–145) 0.66 MAP (mmHg) 95 (92–98) 95 (92–98) 96 (93–100) 0.28 CI (L/min/m2) 2.8 (2.6–3.0) 2.8 (2.6–3.0) 2.9 (2.6–3.1) 0.21 CPI (w/m2) 0.59 (0.54–0.63) 0.58 (0.54–0.63) 0.61 (0.56–0.66) 0.09 TPRI (dyn × s/cm5 × m2) 3147 (2706–3589) 3159 (2695–3622) 3142 (2653–3632) 0.85 Fluid removal data, adjusted meana 95% CIa TFR (mL) 2332 (2136–2530) 2328 (2131–2525) 2393 (2200–2586) 0.47 Treatment duration (h:min) 3:37 (3:26–3:47) 3:37 (3:24–3:49) 3:36 (3:23–3:50) 0.87 UFR (mL/h) 662 (601–723) 661 (594–729) 678 (595–762) 0.53 Parameter All treatments (n = 263) Non-IDH treatments [n= 202 (76.8%)] IDH treatments [n = 61(23.2%)] IDH versus non-IDH adjusted P-valuea Pretreatment hemodynamics, adjusted meana (95% CIa) SBP (mmHg) 139 (133–144) 138 (133–144) 139 (133–145) 0.66 MAP (mmHg) 95 (92–98) 95 (92–98) 96 (93–100) 0.28 CI (L/min/m2) 2.8 (2.6–3.0) 2.8 (2.6–3.0) 2.9 (2.6–3.1) 0.21 CPI (w/m2) 0.59 (0.54–0.63) 0.58 (0.54–0.63) 0.61 (0.56–0.66) 0.09 TPRI (dyn × s/cm5 × m2) 3147 (2706–3589) 3159 (2695–3622) 3142 (2653–3632) 0.85 Fluid removal data, adjusted meana 95% CIa TFR (mL) 2332 (2136–2530) 2328 (2131–2525) 2393 (2200–2586) 0.47 Treatment duration (h:min) 3:37 (3:26–3:47) 3:37 (3:24–3:49) 3:36 (3:23–3:50) 0.87 UFR (mL/h) 662 (601–723) 661 (594–729) 678 (595–762) 0.53 a Calculated by using a mixed model for repeated measurements and adjusted for multiple comparisons using the Tukey–Kramer method. TFR, total fluid removed; UFR, ultrafiltration rate; CI, cardiac index; TPRI, total peripheral resistance index (dyn × s/cm5 × m2) method. Table 2 Pretreatment hemodynamics and fluid removal data in all treatments in those with no IDH episodes and those with at least one IDH episode Parameter All treatments (n = 263) Non-IDH treatments [n= 202 (76.8%)] IDH treatments [n = 61(23.2%)] IDH versus non-IDH adjusted P-valuea Pretreatment hemodynamics, adjusted meana (95% CIa) SBP (mmHg) 139 (133–144) 138 (133–144) 139 (133–145) 0.66 MAP (mmHg) 95 (92–98) 95 (92–98) 96 (93–100) 0.28 CI (L/min/m2) 2.8 (2.6–3.0) 2.8 (2.6–3.0) 2.9 (2.6–3.1) 0.21 CPI (w/m2) 0.59 (0.54–0.63) 0.58 (0.54–0.63) 0.61 (0.56–0.66) 0.09 TPRI (dyn × s/cm5 × m2) 3147 (2706–3589) 3159 (2695–3622) 3142 (2653–3632) 0.85 Fluid removal data, adjusted meana 95% CIa TFR (mL) 2332 (2136–2530) 2328 (2131–2525) 2393 (2200–2586) 0.47 Treatment duration (h:min) 3:37 (3:26–3:47) 3:37 (3:24–3:49) 3:36 (3:23–3:50) 0.87 UFR (mL/h) 662 (601–723) 661 (594–729) 678 (595–762) 0.53 Parameter All treatments (n = 263) Non-IDH treatments [n= 202 (76.8%)] IDH treatments [n = 61(23.2%)] IDH versus non-IDH adjusted P-valuea Pretreatment hemodynamics, adjusted meana (95% CIa) SBP (mmHg) 139 (133–144) 138 (133–144) 139 (133–145) 0.66 MAP (mmHg) 95 (92–98) 95 (92–98) 96 (93–100) 0.28 CI (L/min/m2) 2.8 (2.6–3.0) 2.8 (2.6–3.0) 2.9 (2.6–3.1) 0.21 CPI (w/m2) 0.59 (0.54–0.63) 0.58 (0.54–0.63) 0.61 (0.56–0.66) 0.09 TPRI (dyn × s/cm5 × m2) 3147 (2706–3589) 3159 (2695–3622) 3142 (2653–3632) 0.85 Fluid removal data, adjusted meana 95% CIa TFR (mL) 2332 (2136–2530) 2328 (2131–2525) 2393 (2200–2586) 0.47 Treatment duration (h:min) 3:37 (3:26–3:47) 3:37 (3:24–3:49) 3:36 (3:23–3:50) 0.87 UFR (mL/h) 662 (601–723) 661 (594–729) 678 (595–762) 0.53 a Calculated by using a mixed model for repeated measurements and adjusted for multiple comparisons using the Tukey–Kramer method. TFR, total fluid removed; UFR, ultrafiltration rate; CI, cardiac index; TPRI, total peripheral resistance index (dyn × s/cm5 × m2) method. In the non-IDH evaluations [637 of 736 (86.5%)], hemodynamics did not significantly change during treatments. CPI decreased by 3% (95% CI −5% to −2%) from a pretreatment level of 0.59 (95% CI 0.54–0.63) w/m2, P = 0.28; CI did not change (95% CI −1–1%) from a pretreatment level of 2.8 (95% CI 2.6–3.0) L/min/m2, P = 0.94 and TPRI decreased by 6% (95% CI −10% to −1%) from a pretreatment level of 3147 (95% CI 2706–3589) dyn × s/cm5 × m2, P = 0.37. SBP decreased by 4 (95% CI −1 to −8) from a pretreatment level of 138 (95% CI 133–144) mmHg, P = 0.71 and MAP decreased by 3 (95% CI −1 to −5) from a pretreatment of level 95 (95% CI 92–98) mmHg, P = 0.43 (Table 3). Table 3 Hemodynamics trends from pretreatment to intra- and postdialytic nadir measurements (n = 736) of the non-IDH and each of the three IDH subgroups Parameter Non-IDH [n = 637 (86.5%)] IDH, n = 99 (13.5%) Adj. P-valuea (1) CPI decreased [n = 35 (35.3%)] (2) TPRI decreased (vasodilatation) [n = 37 (37.4%)] (3) Combined CPI and TPRI decrease [n = 27 (27.3%)] (1) CPI decrease versus non-IDH (2) TPRI decrease versus non-IDH (1) CPI versus (2) TPRI decrease SBP (mmHg) Pre 138 (133–144) 145 (137–152) 137 (130–144) 135 (128–143) 0.14 0.97 0.21 Nadir 134 (129–139) 109 (101–116) 107 (100–114) 102 (95–110) <0.001 <0.001 0.95 Change −4 (−1 to −8) −35 (−28 to −43) −30 (−23 to −37) −33 (−25 to −41) <0.001 <0.001 0.67 Adj. P-valuea 0.71 <0.001 <0.001 <0.001 MAP (mmHg) Pre 95 (92–98) 98 (94–103) 96 (91–100) 95 (90–99) 0.27 0.98 0.68 Nadir 92 (89–95) 75 (70–80) 74 (69–78) 73 (68–78) <0.001 <0.001 0.80 Change −3 (−1 to −5) −23 (−18 to −28) −22 (−18 to −27) −22 (−17 to −27) <0.001 <0.001 0.99 Adj. P valuea 0.43 <0.001 <0.001 <0.001 CI (mL/m2) Pre 2.8 (2.6–3.0) 3.1 (2.8–3.3) 2.6 (2.3–2.8) 3.0 (2.8–3.3) 0.022 0.062 0.002 Nadir 2.8 (2.6–2.9) 2.6 (2.4–2.9) 3.2 (2.9–3.4) 3.1 (2.8–3.3) 0.07 <0.001 <0.001 Change (%) 0% (−1 to 1) −15% (−14 to −16) 23% (21 to 26) 1% (1 to 2) 0.001 <0.001 <0.001 Adj. P valuea 0.94 <0.001 <0.001 0.73 CPI (w/m2) Pre 0.59 (0.54–0.63) 0.66 (0.60–0.72) 0.55 (0.49–0.60) 0.63 (0.57–0.70) 0.01 0.23 0.002 Nadir 0.56 (0.53–0.60) 0.43 (0.37–0.48) 0.49 (0.44–0.55) 0.48 (0.42–0.53) <0.001 0.01 0.06 Change −3% (−5 to −2) −36% (−33 to −38) −9% (−8 to −10) −25% (−24 to −26) <0.001 0.576 <0.001 Adj. P- valuea 0.28 <0.001 <0.05 <0.001 TPRI Pre 3147 (2706–3589) 3069 (2539–3601) 3342 (2824–3859) 2964 (2428–3501) 0.92 0.53 0.49 Nadir 2959 (2690–3228) 2596 (2229–2963) 2251 (1900–2602) 2266 (1891–2642) 0.04 <0.001 0.08 Change −6% (−10 to −1) −14% (−12 to −18) −33% (−32 to −33) −23% (−22 to −25) 0.23 <0.001 0.002 Adj. P-valuea 0.37 <0.01 <0.001 <0.001 Parameter Non-IDH [n = 637 (86.5%)] IDH, n = 99 (13.5%) Adj. P-valuea (1) CPI decreased [n = 35 (35.3%)] (2) TPRI decreased (vasodilatation) [n = 37 (37.4%)] (3) Combined CPI and TPRI decrease [n = 27 (27.3%)] (1) CPI decrease versus non-IDH (2) TPRI decrease versus non-IDH (1) CPI versus (2) TPRI decrease SBP (mmHg) Pre 138 (133–144) 145 (137–152) 137 (130–144) 135 (128–143) 0.14 0.97 0.21 Nadir 134 (129–139) 109 (101–116) 107 (100–114) 102 (95–110) <0.001 <0.001 0.95 Change −4 (−1 to −8) −35 (−28 to −43) −30 (−23 to −37) −33 (−25 to −41) <0.001 <0.001 0.67 Adj. P-valuea 0.71 <0.001 <0.001 <0.001 MAP (mmHg) Pre 95 (92–98) 98 (94–103) 96 (91–100) 95 (90–99) 0.27 0.98 0.68 Nadir 92 (89–95) 75 (70–80) 74 (69–78) 73 (68–78) <0.001 <0.001 0.80 Change −3 (−1 to −5) −23 (−18 to −28) −22 (−18 to −27) −22 (−17 to −27) <0.001 <0.001 0.99 Adj. P valuea 0.43 <0.001 <0.001 <0.001 CI (mL/m2) Pre 2.8 (2.6–3.0) 3.1 (2.8–3.3) 2.6 (2.3–2.8) 3.0 (2.8–3.3) 0.022 0.062 0.002 Nadir 2.8 (2.6–2.9) 2.6 (2.4–2.9) 3.2 (2.9–3.4) 3.1 (2.8–3.3) 0.07 <0.001 <0.001 Change (%) 0% (−1 to 1) −15% (−14 to −16) 23% (21 to 26) 1% (1 to 2) 0.001 <0.001 <0.001 Adj. P valuea 0.94 <0.001 <0.001 0.73 CPI (w/m2) Pre 0.59 (0.54–0.63) 0.66 (0.60–0.72) 0.55 (0.49–0.60) 0.63 (0.57–0.70) 0.01 0.23 0.002 Nadir 0.56 (0.53–0.60) 0.43 (0.37–0.48) 0.49 (0.44–0.55) 0.48 (0.42–0.53) <0.001 0.01 0.06 Change −3% (−5 to −2) −36% (−33 to −38) −9% (−8 to −10) −25% (−24 to −26) <0.001 0.576 <0.001 Adj. P- valuea 0.28 <0.001 <0.05 <0.001 TPRI Pre 3147 (2706–3589) 3069 (2539–3601) 3342 (2824–3859) 2964 (2428–3501) 0.92 0.53 0.49 Nadir 2959 (2690–3228) 2596 (2229–2963) 2251 (1900–2602) 2266 (1891–2642) 0.04 <0.001 0.08 Change −6% (−10 to −1) −14% (−12 to −18) −33% (−32 to −33) −23% (−22 to −25) 0.23 <0.001 0.002 Adj. P-valuea 0.37 <0.01 <0.001 <0.001 Hemodynamic data are shown as adjusted mean (95% CI). a All hemodynamic data and P-values were calculated using a mixed model for repeated measurements and adjusted for multiple comparisons using the Tukey–Kramer method. Adj., adjusted; Pre, pretreatment; TPRI, total peripheral resistance index (dyn × s/cm5 × m2). Table 3 Hemodynamics trends from pretreatment to intra- and postdialytic nadir measurements (n = 736) of the non-IDH and each of the three IDH subgroups Parameter Non-IDH [n = 637 (86.5%)] IDH, n = 99 (13.5%) Adj. P-valuea (1) CPI decreased [n = 35 (35.3%)] (2) TPRI decreased (vasodilatation) [n = 37 (37.4%)] (3) Combined CPI and TPRI decrease [n = 27 (27.3%)] (1) CPI decrease versus non-IDH (2) TPRI decrease versus non-IDH (1) CPI versus (2) TPRI decrease SBP (mmHg) Pre 138 (133–144) 145 (137–152) 137 (130–144) 135 (128–143) 0.14 0.97 0.21 Nadir 134 (129–139) 109 (101–116) 107 (100–114) 102 (95–110) <0.001 <0.001 0.95 Change −4 (−1 to −8) −35 (−28 to −43) −30 (−23 to −37) −33 (−25 to −41) <0.001 <0.001 0.67 Adj. P-valuea 0.71 <0.001 <0.001 <0.001 MAP (mmHg) Pre 95 (92–98) 98 (94–103) 96 (91–100) 95 (90–99) 0.27 0.98 0.68 Nadir 92 (89–95) 75 (70–80) 74 (69–78) 73 (68–78) <0.001 <0.001 0.80 Change −3 (−1 to −5) −23 (−18 to −28) −22 (−18 to −27) −22 (−17 to −27) <0.001 <0.001 0.99 Adj. P valuea 0.43 <0.001 <0.001 <0.001 CI (mL/m2) Pre 2.8 (2.6–3.0) 3.1 (2.8–3.3) 2.6 (2.3–2.8) 3.0 (2.8–3.3) 0.022 0.062 0.002 Nadir 2.8 (2.6–2.9) 2.6 (2.4–2.9) 3.2 (2.9–3.4) 3.1 (2.8–3.3) 0.07 <0.001 <0.001 Change (%) 0% (−1 to 1) −15% (−14 to −16) 23% (21 to 26) 1% (1 to 2) 0.001 <0.001 <0.001 Adj. P valuea 0.94 <0.001 <0.001 0.73 CPI (w/m2) Pre 0.59 (0.54–0.63) 0.66 (0.60–0.72) 0.55 (0.49–0.60) 0.63 (0.57–0.70) 0.01 0.23 0.002 Nadir 0.56 (0.53–0.60) 0.43 (0.37–0.48) 0.49 (0.44–0.55) 0.48 (0.42–0.53) <0.001 0.01 0.06 Change −3% (−5 to −2) −36% (−33 to −38) −9% (−8 to −10) −25% (−24 to −26) <0.001 0.576 <0.001 Adj. P- valuea 0.28 <0.001 <0.05 <0.001 TPRI Pre 3147 (2706–3589) 3069 (2539–3601) 3342 (2824–3859) 2964 (2428–3501) 0.92 0.53 0.49 Nadir 2959 (2690–3228) 2596 (2229–2963) 2251 (1900–2602) 2266 (1891–2642) 0.04 <0.001 0.08 Change −6% (−10 to −1) −14% (−12 to −18) −33% (−32 to −33) −23% (−22 to −25) 0.23 <0.001 0.002 Adj. P-valuea 0.37 <0.01 <0.001 <0.001 Parameter Non-IDH [n = 637 (86.5%)] IDH, n = 99 (13.5%) Adj. P-valuea (1) CPI decreased [n = 35 (35.3%)] (2) TPRI decreased (vasodilatation) [n = 37 (37.4%)] (3) Combined CPI and TPRI decrease [n = 27 (27.3%)] (1) CPI decrease versus non-IDH (2) TPRI decrease versus non-IDH (1) CPI versus (2) TPRI decrease SBP (mmHg) Pre 138 (133–144) 145 (137–152) 137 (130–144) 135 (128–143) 0.14 0.97 0.21 Nadir 134 (129–139) 109 (101–116) 107 (100–114) 102 (95–110) <0.001 <0.001 0.95 Change −4 (−1 to −8) −35 (−28 to −43) −30 (−23 to −37) −33 (−25 to −41) <0.001 <0.001 0.67 Adj. P-valuea 0.71 <0.001 <0.001 <0.001 MAP (mmHg) Pre 95 (92–98) 98 (94–103) 96 (91–100) 95 (90–99) 0.27 0.98 0.68 Nadir 92 (89–95) 75 (70–80) 74 (69–78) 73 (68–78) <0.001 <0.001 0.80 Change −3 (−1 to −5) −23 (−18 to −28) −22 (−18 to −27) −22 (−17 to −27) <0.001 <0.001 0.99 Adj. P valuea 0.43 <0.001 <0.001 <0.001 CI (mL/m2) Pre 2.8 (2.6–3.0) 3.1 (2.8–3.3) 2.6 (2.3–2.8) 3.0 (2.8–3.3) 0.022 0.062 0.002 Nadir 2.8 (2.6–2.9) 2.6 (2.4–2.9) 3.2 (2.9–3.4) 3.1 (2.8–3.3) 0.07 <0.001 <0.001 Change (%) 0% (−1 to 1) −15% (−14 to −16) 23% (21 to 26) 1% (1 to 2) 0.001 <0.001 <0.001 Adj. P valuea 0.94 <0.001 <0.001 0.73 CPI (w/m2) Pre 0.59 (0.54–0.63) 0.66 (0.60–0.72) 0.55 (0.49–0.60) 0.63 (0.57–0.70) 0.01 0.23 0.002 Nadir 0.56 (0.53–0.60) 0.43 (0.37–0.48) 0.49 (0.44–0.55) 0.48 (0.42–0.53) <0.001 0.01 0.06 Change −3% (−5 to −2) −36% (−33 to −38) −9% (−8 to −10) −25% (−24 to −26) <0.001 0.576 <0.001 Adj. P- valuea 0.28 <0.001 <0.05 <0.001 TPRI Pre 3147 (2706–3589) 3069 (2539–3601) 3342 (2824–3859) 2964 (2428–3501) 0.92 0.53 0.49 Nadir 2959 (2690–3228) 2596 (2229–2963) 2251 (1900–2602) 2266 (1891–2642) 0.04 <0.001 0.08 Change −6% (−10 to −1) −14% (−12 to −18) −33% (−32 to −33) −23% (−22 to −25) 0.23 <0.001 0.002 Adj. P-valuea 0.37 <0.01 <0.001 <0.001 Hemodynamic data are shown as adjusted mean (95% CI). a All hemodynamic data and P-values were calculated using a mixed model for repeated measurements and adjusted for multiple comparisons using the Tukey–Kramer method. Adj., adjusted; Pre, pretreatment; TPRI, total peripheral resistance index (dyn × s/cm5 × m2). Since the primary objective of the study was to categorize IDH episodes in terms of the immediate hemodynamic cause(s) of hypotension, and not a comparison of individual patients, the episodes (n = 99) were divided into three subgroups according to the observed functional changes per episode. The three subgroups were initially defined according to changes in cardiac output (decrease, increase or no change) as blood pressure decreased. As cardiac power is the product of blood pressure and cardiac output, a reduction in either or both will reduce cardiac power. Since peripheral resistance is the ratio between blood pressure and cardiac output, cardiac output will decrease as a result of a decrease in peripheral resistance. As a result, the three subgroups where named as follows: Predominantly due to CPI decrease: defined as a CPI decrease >15% and a CI decrease >5%. Thirty-five (35.3%) of IDH episodes were in this subgroup, with CPI decreased by 36% (95% CI −33 to −38) from a pretreatment level of 0.66 (95% CI 0.60–0.72) w/m2, P < 0.001; CI decreased by 15% (95% CI −14 to −16) from 3.1 (95% CI 2.8–3.3) L/min/m2, P < 0.001 and TPRI decreased by 14% (95% CI −12 to −18) from 3069 (95% CI 2539–3601) dyn × s/cm5 × m2, P < 0.01. SBP decreased by 35 (95% CI −28 to −43) from 145 (95% CI 137–152) mmHg, P < 0.001 and MAP decreased by 23 (95% CI −18 to −28) from 98 (95% CI 94–103) mmHg, P < 0.001. Predominantly due to TPRI decrease (vasodilatation): defined as a TPRI decrease >15% and a CI increase >5%. Thirty-seven 37 (37.4%) IDH episodes were in this group with TPRI decreased by 33% (95% CI −32 to −33) from a pretreatment level of 3342 (95% CI 2824–3859) dyn × s/cm5 × m2, P < 0.001; CI increased by 23% (95% CI 21–26) from a pretreatment level of 2.6 (95% CI 2.3–2.8) L/min/m2, P < 0.001 and CPI decreased by 9% (95% CI −8 to −10) from a pretreatment level of 0.55 (95% CI 0.49–0.60) w/m2, P < 0.05. SBP decreased by 30 (95% CI −23 to −37) from a pretreatment level of 137 (95% CI 130–144) mmHg, P < 0.001 and MAP decreased by 22 (95% CI −18 to −27) from a pretreatment level of 96 (95% CI 91–100) mmHg, P < 0.001. Due to a combination of CPI and TPRI decrease: defined as CPI and TPRI decreases >15% with a CI change of not >5%. Twenty-seven (27.3%) IDH episodes were in this subgroup, with CPI decreased by 25% (95% CI −24 to −26) from a pretreatment level of 0.63 (95% CI 0.57–0.70) w/m2, P < 0.001; TPRI decreased by 23% (95% CI −22 to −25) from a pretreatment level of 2964 (95% CI 2428–3501) dyn × s/cm5 × m2, P < 0.001 and CI was not significantly changed: 1.3% (95% CI 1.2–1.7) from a pretreatment level of 3.0 (95% CI 2.8–3.3) L/min/m2, P = 0.730. SBP decreased by 30 (95% CI −23 to −37) from a pretreatment level of 135 (95% CI 128–143) mmHg, P < 0.001 and MAP decreased by 22 (95% CI −18 to −27) from a pretreatment level of 95 (95% CI 90–99) mmHg, P < 0.001. Differences in changes in CI, CPI and TPRI between IDH subgroups 1 and 2 were significant: P < 0.001 for changes in CI and CPI and P = 0.002 for changes in TPRI. Differences in changes in SBP and MAP between IDH subgroups 1 and 2 were not significant (P = 0.67 and 0.99, respectively). Table 3 shows the hemodynamic trends of the non-IDH and the three IDH subgroups. Figure 2 illustrates the hemodynamic trends of each individual evaluation in the three IDH subgroups using three hemodynamic graphs of MAP versus CI. Representative lines of both CPI and TPRI are also shown. Hemodynamic ranges for a normal population are illustrated by a dashed octagon. Each arrow provides the MAP and CI changes from pretreatment of all intra- and postdialytic evaluations, which are substantial reductions of CPI and CI for subgroup 1, a substantial reduction of TPRI and an increase of CI in subgroup 2 and substantial reductions of both CPI and TPRI with no change of CI in subgroup 3. DISCUSSION The clinically vital relationship between intradialytic hypotension and consequent morbidity and mortality is unquestioned [16, 17]. The primary objective of this article was to report changes of cardiac output, peripheral resistance and cardiac power during incidents of IDH. The stringent definition of IDH used here includes a nadir SBP <100 mmHg or a nadir MAP <70 mmHg and was used to compensate for the lack of information about intradialytic symptoms, which is included in some definitions of IDH as additional subjective criteria. This study utilized regional impedance cardiographic technology, which has been validated for measurements during chronic hemodialysis [10]. It measures cardiac output, and with the additional information of blood pressure, calculates cardiac power and total peripheral resistance. All parameters were indexed to body surface area. Cardiac power reduction may occur as a result of a preload reduction due to high ultrafiltration rates with inadequate vascular refilling or a low target weight, together resulting in hypovolemia and a consequent reduction in blood pressure and cardiac output. The presence of diastolic dysfunction may increase cardiac sensitivity to preload reduction [18]. Low cardiac power has been shown to be an independent predictor of inhospital mortality in a broad spectrum of patients with primary cardiac disease [12]. As total peripheral resistance is the ratio between blood pressure and cardiac output, a reduction of blood pressure and an increase in cardiac output are the result of a reduction in total peripheral resistance. Such a reduction could be the result of volume reduction without compensatory vasoconstriction, which creates a situation of relative vasodilatation. Autonomic dysfunction, often associated with diabetes, could be the underlying reason for failure of vasoconstriction. Low peripheral resistance has been shown to be an independent predictor of in-hospital mortality in both sepsis and nonseptic patients [19] As hypovolemia (preload reduction) and autonomous dysfunction are independent parameters, both may occur in the same patient during the same dialysis, resulting in unchanged cardiac output. This means that both cardiac power and peripheral resistance may decrease at the same time. The new ability to categorize the underlying hemodynamic abnormalities leading to IDH in each patient and each dialysis session provides significant potential to improve the management of hypotension. A reduction of the ultrafiltration rate or an increase in the prescribed target weight in response to a progressive decrease in cardiac power or by the use of short-acting alpha adrenergic agonists, such as midodrine, in the prophylaxis or treatment of peripheral resistance decrease [20] are rational interventions that could reduce intradialytic episodes and prevent organ ischemia and consequent damage. The unexpected relatively high frequency of vasodilatation as a direct cause of hypotensive episodes either alone or together with decreased cardiac power (collectively 64.7% of all IDH episodes) suggests that more emphasis might be placed on autonomic dysfunctional disorders, particularly in patients with diabetes. The similarity in pretreatment hemodynamics in the IDH and non-IDH groups increases the value of intradialytic characterization of IDH mechanisms for appropriate interventions. Limitations The number of patients studied was small. However, the emphasis of this article is the analysis of all hemodynamic responses occurring during each dialysis, not a comparison of individual patients. Measurements were not made continuously and therefore IDH episodes occurring earlier in treatments would have been missed; however, the hemodynamics of every recorded IDH episode were analyzed. Comorbid and cardiovascular status data and information on inter- and intradialytic drug use were not collected since the main purpose of this study was to characterize IDH episodes in the course of routine dialysis treatments. An increase in the accuracy of blood pressure measurements would improve interpretation of small changes in hemodynamics. In conclusion, intradialytic noninvasive cardiac output measurement has enabled the assessment of changes in cardiac power and peripheral resistance. These have been demonstrated to be distinctly different causes of hypotensive episodes in routine dialysis patients. This information can potentially improve the care of dialysis patients prone to IDH, by interventions based on specific cause. ACKNOWLEDGEMENTS The authors would like to thank Loretta Andersen, RN, Clinical Manager, for facilitating the quality improvement program on which this work was based. FUNDING Partial financial support for the performance of the research was provided by NI Medical. This support consisted solely of the lending of devices and sensors for performance of the research and the financing of a study coordinator. AUTHORS’ CONTRIBUTIONS N.W.L. confirms that he has full access to the data in the study and final responsibility for the decision to submit for publication. Contributing authors were involved in designing the work, acquiring data and interpreting the results. N.W.L. conceived and partially designed the work, acquired data and played an important role in interpreting the results and drafting and revised the manuscript. M.H.F.G. d.A., L.E.B., H.A.T.F. acquired data and approved the final version. R.S. assisted in the design of the work, acquired data and assisted in data coordination. S.G. and T.H. assisted in the design of the work and acquired data. S.L. played a major role in the analysis of the data. C.W. assisted in the design and analysis of the work. CONFLICT OF INTEREST STATEMENT None declared. REFERENCES 1 McIntyre CW , Burton JO , Selby NM et al. Hemodialysis-induced cardiac dysfunction is associated with an acute reduction in global and segmental myocardial blood flow . Clin J Am Nephrol 2008 ; 3 : 19 – 26 Google Scholar CrossRef Search ADS 2 Ritz E , Rambausek M , Mall G et al. Cardiac changes in uremia and their possible relation to cardiovascular instability on dialysis . Contrib Nephrol 1990 ; 78 : 221 – 229 Google Scholar CrossRef Search ADS PubMed 3 Converse RL Jr , Jacobsen TN , Jost CM et al. Paradoxical withdrawal of reflex vasoconstriction as a cause of hemodialysis-induced hypotension . J Clin Invest 1992 ; 90 : 1657 – 1665 Google Scholar CrossRef Search ADS PubMed 4 Shoji T , Tsubakihara Y , Fujii M et al. Hemodialysis-associated hypotension as an independent risk factor for two-year mortality in hemodialysis patients . Kidney Int 2004 ; 66 : 1212 – 1220 Google Scholar CrossRef Search ADS PubMed 5 Zucchelli P , Santoro A. Dialysis-induced hypotension: a fresh look at pathophysiology . Blood Purif 1993 ; 11 : 85 – 98 Google Scholar CrossRef Search ADS PubMed 6 Paredes OL , Shite J , Shinke T et al. Impedance cardiography for cardiac output estimation—reliability of wrist-to-ankle electrode configuration . Circ J 2006 ; 70 : 1164 – 1168 Google Scholar CrossRef Search ADS PubMed 7 Cotter G , Moshkovitz Y , Kaluski E et al. Accurate, non-invasive, continuous monitoring of cardiac output by whole body electrical bio-impedance . Chest 2004 ; 125 : 1431 – 1440 Google Scholar CrossRef Search ADS PubMed 8 Leitman L , Sucher E , Kaluski E et al. Non-invasive measurement of cardiac output by whole-body bio-impedance during dobutamine stress echocardiography: clinical implications in patients with left ventricular dysfunction and ischaemia . Eur J Heart Fail 2006 ; 8 : 136 – 140 Google Scholar CrossRef Search ADS PubMed 9 Cotter G , Schachner A , Sasson L et al. Impedance cardiography revisited . Physiol Meas 2006 ; 27 : 817 – 827 Google Scholar CrossRef Search ADS PubMed 10 Germain M , Nathanson BH , Chait Y et al. Comparison of stroke volume measurements during hemodialysis using bioimpedance cardiography and echocardiography. Hemodial Int 2017 ; doi: 10.1111/hdi.12589 11 Du Bois D , Du Bois EF. A formula to estimate the approximate surface area if height and weight be known . Arch Intern Med (Chic) 1916 ; XVII : 863 – 871 Google Scholar CrossRef Search ADS 12 Mendoza D , Cooper HA , Panza JA. Cardiac power output predicts mortality across a broad spectrum of patients with acute cardiac disease . Am Heart J 2007 ; 153 : 366 – 370 Google Scholar CrossRef Search ADS PubMed 13 Cotter G , Moshkovitz Y , Kaluski E et al. The role of cardiac power and systemic vascular resistance in the pathophysiology and diagnosis of patients with acute congestive heart failure . Eur J Heart Fail 2003 ; 5 : 443 – 451 Google Scholar CrossRef Search ADS PubMed 14 Verbeke G , Molenberghs G. Linear mixed models for longitudinal data . New York : Springer , 2000 15 Tukey JW. Computing individual means in the analysis of variance . Biometrics 1949 ; 5 : 99 – 114 Google Scholar CrossRef Search ADS PubMed 16 Bradshaw W. The importance of mean arterial pressure as a patient assessment tool: in haemodialysis and acute care . Aust Nurs J 2012 ; 20 : 26 – 29 Google Scholar PubMed 17 Flythe JE , Xue H , Lynch KE et al. Association of mortality risk with various definitions of intradialytic hypotension . J Am Soc Nephrol 2015 ; 26 : 724 – 734 Google Scholar CrossRef Search ADS PubMed 18 Acharya P , Ranabhat K , Trikhatri Y et al. Effect of preload reduction by hemodialysis on Doppler indices of diastolic function in patients with end-stage renal disease . Kathmandu Univ Med J 2008 ; 6 : 98 – 101 19 Melo J , Peters J . Low systemic vascular resistance: differential diagnosis and outcome . Crit Care 1999 ; 3 : 71 – 77 Google Scholar CrossRef Search ADS PubMed 20 Prakash S , Garg AX , Heidenheim AP et al. Midodrine appears to be safe and effective for dialysis-induced hypotension: a systematic review . Nephrol Dial Transplant 2004 ; 19 : 2553 – 2558 Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nephrology Dialysis Transplantation Oxford University Press

Hemodynamic response to fluid removal during hemodialysis: categorization of causes of intradialytic hypotension

Loading next page...
 
/lp/ou_press/hemodynamic-response-to-fluid-removal-during-hemodialysis-dbVdsPqTQr
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
ISSN
0931-0509
eISSN
1460-2385
D.O.I.
10.1093/ndt/gfy048
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Intradialytic hypotension is a clinically significant problem, however, the hemodynamics that underlie ultrafiltration and consequent hypotensive episodes has not been studied comprehensively. Methods Intradialytic cardiac output, cardiac power and peripheral resistance changes from pretreatment measurements were evaluated using a novel regional impedance cardiographic device (NICaS, NI Medical, Peta Tikva, Israel) in 263 hemodialysis sessions in 54 patients in dialysis units in the USA and Brazil with the goal of determining the various hemodynamic trends as blood pressure decreases. Results Hypotensive episodes occurred in 99 (13.5%) of 736 intra- and postdialytic evaluations. The hemodynamic profiles of the episodes were categorized: (i) The cardiac power index significantly decreased in 35% of episodes by 36%, from 0.66 [95% confidence interval (CI) 0.60–0.72] to 0.43 (95% CI 0.37–0.48) [w/m2] with a small reduction in the total peripheral resistance index. (ii) The total peripheral resistance index significantly decreased in 37.4% of episodes by 33%, from 3342 (95% CI 2824–3859) to 2251 (95% CI 1900–2602) [dyn × s/cm5 × m2] with a small reduction in the cardiac power index. (iii) Both the cardiac power index and total peripheral resistance index significantly decreased in 27.3% of episodes, the cardiac power index by 25% from 0.63 (95% CI 0.57–0.70) to 0.48 (95% CI 0.42–0.53) [w/m2] and the total peripheral resistance index by 23% from 2964 (95% CI 2428–3501) to 2266 (95% CI 1891–2642). Conclusions The hemodynamic profiles clearly define specific hemodynamic mechanisms of cardiac power reduction and/or vasodilatation as underlying intradialytic hypotensive episodes. A reduction in cardiac power (reduction of both blood pressure and cardiac output) could be the result of preload reduction due to a high ultrafiltration rate with not enough refilling or low target weight. A reduction in peripheral resistance (reduction in blood pressure and increase in cardiac output) could be the result of relative vasodilatation as arteries do not contract to compensate for volume reduction due to autonomous dysfunction. As both phenomena are independent, they may appear at the same time. Based on these results, a reduction of ultrafiltration rate and an increase in target weight to improve preload or immediate therapeutic actions to increase peripheral resistance are rational measures that could be taken to maintain blood pressure and prevent hypotensive ischemic complications in dialysis patients. cardiovascular, hemodialysis, intradialytic hypotensive episodes, online intradialytic hemodynamics, ultrafiltration INTRODUCTION Hypotension occurring during chronic hemodialysis is a major clinical problem with consequent ischemia of the heart, brain and gut [1]. Reduced cardiac preload due to hypovolemia may be further diminished in the presence of diastolic dysfunction [2]. An increase in sympathetic tone may add functional problems [3]. These factors are associated with an increase in mortality in hemodialysis patients [4]. The hemodynamics underlying ultrafiltration and consequent hypotensive episodes have not been studied comprehensively. Zucchelli and Santoro in 1993 [5] suggested that intradialytic hypotensive episodes may be the result of hypovolemia, left ventricular diastolic dysfunction or a ‘breakdown’ in peripheral resistance, but they did not provide data showing intradialytic hemodynamic trends. The objective of this study is to report the hemodynamic changes during hemodialysis, specifically focusing on those responsible for intradialytic hypotensive episodes. All hemodynamic measurements were made using a regional impedance cardiography device (NICaS, NI Medical, Peta Tikva, Israel), which have been shown to correlate well with pulmonary artery catheterization thermodilution and echocardiographic data [6–8] and are more accurate than thoracic impedance cardiographic measurements [9]. The accuracy of the device during dialysis has been recently validated, demonstrating a strong correlation of stroke volume (SV) with echocardiographic measurements, with insignificant bias [10]. MATERIALS AND METHODS Patients Fifty-four established chronic hemodialysis patients at the Queens Artificial Kidney Center, New York (n = 26) and at the Biocor Hospital, Belo Horizonte, Brazil (n = 28) were studied. Patients were randomly selected. There were no inclusion criteria other than dialysis vintage of at least 3 months. All gave written informed consent. The study was approved by the New England Institutional Review Board (IRB) and by the Biocor Hospital de Doencas Cardiovasculares, Belo Horizonte IRB. Study protocol Hemodynamic trends were assessed using the device in randomly selected dialysis sessions. Sitting blood pressures were measured immediately prior to each hemodynamic measurement. Gender, age, height, weight, electrode location and blood pressure data were entered into the device. The device measures and calculates hemodynamic parameters on each heart beat during 60 s and provides the averaged parameters. Each patient’s hemodynamics were measured over a range of dialysis treatments with an average of 4.9 ± 2.3 treatments per patient. Measurements were made at baseline (pretreatment), at the middle, just before the end and 10 min after the end of the treatment. The dialysis staff was blinded to all measurements derived from the device. Technology used for hemodynamic measurements The device (NICaS, NI Medical) is a noninvasive regional bioimpedance cardiac measurement and analysis system (FDA 510k clearance no. K080941, 12 June 2009). The US Food and Drug Administration indication for use of the device states ‘NICaS is intended to monitor and display hemodynamic parameters in males and females with known or suspected cardiac disorders needing cardiac assessment’. SV is measured by applying an alternating electrical current of 1.4 mA at 30 kHz frequency through the patient’s body via two pairs of tetrapolar sensors, one pair placed on the wrist of the nonaccess arm above the radial pulse and the other pair on the contralateral ankle above the posterior tibial pulse (Figure 1). SV is calculated by Frinerman’s formula: SV = (dR/R) × ρ × (L2/Ri) × (α + β)/β × KW × HF [2–4], where dR is the impedance change in the arterial system as a result of intraarterial expansion during systole, R is basal resistance, ρ is blood electrical resistance, L is the patient’s height, Ri is basal resistance corrected for gender and age, KW is the correction of weight according to ideal values, HF is a hydration factor that takes into account the ratio between R and body mass index (BMI), which is correlated to body water volume, α + β is the electrocardiogram (ECG) R-R wave interval and β is the diastolic time interval. SV is automatically calculated every 20 s and is the average of three measurements obtained consecutively during 60 s of monitoring. The SV index is calculated as SV/body surface area using the Du Bois formula [11]. Heart rate is calculated from a one channel ECG and cardiac (output) index = SV index × heart rate/1000. FIGURE 1 View largeDownload slide ECG waveform and impedance waveform (left). Sensor locations (right). An alternating electrical current of 1.4 mA at 30 kHz frequency is applied through the patient’s body via two pairs of tetrapolar sensors. The impedance waveform provides the changes in electrical resistivity of the arterial system with each heartbeat. SV is calculated based on these changes. Heart rate (HR) is measured by the ECG that is sensed through the same sensors. Cardiac output (CO) is calculated as CO = SV×HR. FIGURE 1 View largeDownload slide ECG waveform and impedance waveform (left). Sensor locations (right). An alternating electrical current of 1.4 mA at 30 kHz frequency is applied through the patient’s body via two pairs of tetrapolar sensors. The impedance waveform provides the changes in electrical resistivity of the arterial system with each heartbeat. SV is calculated based on these changes. Heart rate (HR) is measured by the ECG that is sensed through the same sensors. Cardiac output (CO) is calculated as CO = SV×HR. Using an oscillometric method, sitting systolic and diastolic blood pressure measurements were made automatically by the dialysis machine. Mean arterial pressure [2 × (diastolic + systolic)/3], cardiac power index [CPI; mean arterial pressure (MAP) × cardiac index × 0.0022 w/m2; normal range 0.45–0.85 w/m2] [12, 13] and total peripheral resistance (MAP/cardiac index × 80 dyn × s/cm5 × m2; normal range 1600–3000 dyn × s/cm5 × m2) [13] were calculated. As the device measures pulsatile flow and is blinded to constant flow, fluid removal during dialysis has no impact on measurement accuracy. This was recently validated by correlating SV to ECG measurements during hemodialysis treatments. Good correlation was maintained during treatment. Further, NICaS performance immunity to fluid reduction was demonstrated by the maintenance of correlation to ECG results throughout dialysis treatments [10]. The results are drawn on hemodynamic graphs showing the MAP (y-axis) as a function of cardiac index (x-axis); curves of total peripheral resistance index (TPRI) and CPI are displayed. Ranges for the normal population are depicted by the dotted octagon (Figure 2). FIGURE 2 View largeDownload slide Hemodynamic trends of each individual evaluation in the three IDH subgroups using three graphs of hemodynamic trends of MAP versus CI. Representative lines of both CPI and TPRI are also shown. Hemodynamic ranges for the normal population are illustrated by a dashed octagon. Each arrow provides the MAP and CI changes from pretreatment of all intra- and postdialytic evaluations: substantial reductions of CPI and CI for subgroup 1, substantial reduction of TPRI and increase of CI in subgroup 2 and substantial reductions of both CPI and TPRI with no change of CI in subgroup 3 are shown. FIGURE 2 View largeDownload slide Hemodynamic trends of each individual evaluation in the three IDH subgroups using three graphs of hemodynamic trends of MAP versus CI. Representative lines of both CPI and TPRI are also shown. Hemodynamic ranges for the normal population are illustrated by a dashed octagon. Each arrow provides the MAP and CI changes from pretreatment of all intra- and postdialytic evaluations: substantial reductions of CPI and CI for subgroup 1, substantial reduction of TPRI and increase of CI in subgroup 2 and substantial reductions of both CPI and TPRI with no change of CI in subgroup 3 are shown. Statistical analysis The Kolmogorov–Smirnov test for normally distributed data was used. Descriptive statistics are presented as number (percentages) and mean ± SD. The comparisons for each parameter between the pretreatment and the intra- and postdialytic measurements were made according to the occurrence of intradialytic hypotension (IDH) episodes and by IDH subgroups related to specific hemodynamic changes. Since each evaluation was not only influenced by individual cardiovascular functional responses but also by sodium intake between dialysis and dialysis treatment variables, such as target weight, fluid removal rates and dialysate temperatures, each patient was measured during a few dialysis treatments and analysis of all comparisons was performed using a mixed model for repeated measurements [14] employing a mixed procedure (SAS/STAT software, version 9.4; SAS Institute, Cary, NC, USA) and adjusted for multiple comparisons using the Tukey–Kramer method [15]. Results are presented as mean and 95% confidence intervals (CIs). P-values ≤ 0.05 were considered to be significant. RESULTS: NON-IDH AND THREE IDH SUBGROUPS Fifty-four patients from two sites in the USA and Brazil were studied. The composition of the study group included 29 males (54%), mean age 67 ± 10 years, mean weight 74 ± 17 kg, mean BMI 27.6 ± 5.2 and 30 diabetic patients (56%). Adjusted pretreatment mean systolic blood pressure (SBP) was 139 (95% CI 133–144) mmHg, MAP was 95 (95% CI 92–98) mmHg, mean cardiac output index (CI) was 2.8 (95% CI 2.6–3.0) L/min/m2, mean CPI was 0.59 (95% CI 0.54–0.63) w/m2 and mean TPRI was 3147 (95% CI 2706–3589) dyn × s/cm5 × m2. Criteria for IDH were defined as an SBP reduction ≥20 mmHg OR MAP reduction ≥10 mmHg AND intradialytic nadir SBP to <100 mmHg or MAP <70 mmHg. Patients with at least one IDH episode in at least 30% of dialysis treatments were defined as having IDH. In a total of 263 treatments, 736 intra- and postdialytic measurements were evaluated for IDH episodes by comparing them with the relevant predialytic measurements. A total of 99 (13.5%) evaluations met the criteria for IDH. Fourteen (26%) patients were categorized as having IDH. The demographics of the IDH and non-IDH patients did not differ significantly (Table 1). IDH episodes were recorded in 61 (23.2%) of 263 treatments. In this latter group, pretreatment hemodynamics, fluid removal rates, total fluid removed and duration of treatments did not significantly differ from non-IDH treatments (Table 2). Table 1 Demographics of all patients, patients with no IDH episodes and patients with at least one IDH episode in 30% of treatments Parameter All (n = 54) Non-IDH [n = 40 (74.1%)] IDH [n = 14 (25.9%)] IDH–non-IDH P-value Male, n (%) 29 (54) 20 (50) 9 (64) 0.35 Age (years), mean ± SD 67 ± 10 67 ± 11 66 ± 9 0.87 Weight (kg), mean ± SD 74 ± 17 74 ± 14 75 ± 20 0.80 BMI, mean ± SD 27.6 ± 5.2 26.8 ± 4.50 28.3 ± 5.8 0.30 Diabetes, n (%) 30 (56) 22 (55) 8 (57) 0.81 Parameter All (n = 54) Non-IDH [n = 40 (74.1%)] IDH [n = 14 (25.9%)] IDH–non-IDH P-value Male, n (%) 29 (54) 20 (50) 9 (64) 0.35 Age (years), mean ± SD 67 ± 10 67 ± 11 66 ± 9 0.87 Weight (kg), mean ± SD 74 ± 17 74 ± 14 75 ± 20 0.80 BMI, mean ± SD 27.6 ± 5.2 26.8 ± 4.50 28.3 ± 5.8 0.30 Diabetes, n (%) 30 (56) 22 (55) 8 (57) 0.81 Table 1 Demographics of all patients, patients with no IDH episodes and patients with at least one IDH episode in 30% of treatments Parameter All (n = 54) Non-IDH [n = 40 (74.1%)] IDH [n = 14 (25.9%)] IDH–non-IDH P-value Male, n (%) 29 (54) 20 (50) 9 (64) 0.35 Age (years), mean ± SD 67 ± 10 67 ± 11 66 ± 9 0.87 Weight (kg), mean ± SD 74 ± 17 74 ± 14 75 ± 20 0.80 BMI, mean ± SD 27.6 ± 5.2 26.8 ± 4.50 28.3 ± 5.8 0.30 Diabetes, n (%) 30 (56) 22 (55) 8 (57) 0.81 Parameter All (n = 54) Non-IDH [n = 40 (74.1%)] IDH [n = 14 (25.9%)] IDH–non-IDH P-value Male, n (%) 29 (54) 20 (50) 9 (64) 0.35 Age (years), mean ± SD 67 ± 10 67 ± 11 66 ± 9 0.87 Weight (kg), mean ± SD 74 ± 17 74 ± 14 75 ± 20 0.80 BMI, mean ± SD 27.6 ± 5.2 26.8 ± 4.50 28.3 ± 5.8 0.30 Diabetes, n (%) 30 (56) 22 (55) 8 (57) 0.81 Table 2 Pretreatment hemodynamics and fluid removal data in all treatments in those with no IDH episodes and those with at least one IDH episode Parameter All treatments (n = 263) Non-IDH treatments [n= 202 (76.8%)] IDH treatments [n = 61(23.2%)] IDH versus non-IDH adjusted P-valuea Pretreatment hemodynamics, adjusted meana (95% CIa) SBP (mmHg) 139 (133–144) 138 (133–144) 139 (133–145) 0.66 MAP (mmHg) 95 (92–98) 95 (92–98) 96 (93–100) 0.28 CI (L/min/m2) 2.8 (2.6–3.0) 2.8 (2.6–3.0) 2.9 (2.6–3.1) 0.21 CPI (w/m2) 0.59 (0.54–0.63) 0.58 (0.54–0.63) 0.61 (0.56–0.66) 0.09 TPRI (dyn × s/cm5 × m2) 3147 (2706–3589) 3159 (2695–3622) 3142 (2653–3632) 0.85 Fluid removal data, adjusted meana 95% CIa TFR (mL) 2332 (2136–2530) 2328 (2131–2525) 2393 (2200–2586) 0.47 Treatment duration (h:min) 3:37 (3:26–3:47) 3:37 (3:24–3:49) 3:36 (3:23–3:50) 0.87 UFR (mL/h) 662 (601–723) 661 (594–729) 678 (595–762) 0.53 Parameter All treatments (n = 263) Non-IDH treatments [n= 202 (76.8%)] IDH treatments [n = 61(23.2%)] IDH versus non-IDH adjusted P-valuea Pretreatment hemodynamics, adjusted meana (95% CIa) SBP (mmHg) 139 (133–144) 138 (133–144) 139 (133–145) 0.66 MAP (mmHg) 95 (92–98) 95 (92–98) 96 (93–100) 0.28 CI (L/min/m2) 2.8 (2.6–3.0) 2.8 (2.6–3.0) 2.9 (2.6–3.1) 0.21 CPI (w/m2) 0.59 (0.54–0.63) 0.58 (0.54–0.63) 0.61 (0.56–0.66) 0.09 TPRI (dyn × s/cm5 × m2) 3147 (2706–3589) 3159 (2695–3622) 3142 (2653–3632) 0.85 Fluid removal data, adjusted meana 95% CIa TFR (mL) 2332 (2136–2530) 2328 (2131–2525) 2393 (2200–2586) 0.47 Treatment duration (h:min) 3:37 (3:26–3:47) 3:37 (3:24–3:49) 3:36 (3:23–3:50) 0.87 UFR (mL/h) 662 (601–723) 661 (594–729) 678 (595–762) 0.53 a Calculated by using a mixed model for repeated measurements and adjusted for multiple comparisons using the Tukey–Kramer method. TFR, total fluid removed; UFR, ultrafiltration rate; CI, cardiac index; TPRI, total peripheral resistance index (dyn × s/cm5 × m2) method. Table 2 Pretreatment hemodynamics and fluid removal data in all treatments in those with no IDH episodes and those with at least one IDH episode Parameter All treatments (n = 263) Non-IDH treatments [n= 202 (76.8%)] IDH treatments [n = 61(23.2%)] IDH versus non-IDH adjusted P-valuea Pretreatment hemodynamics, adjusted meana (95% CIa) SBP (mmHg) 139 (133–144) 138 (133–144) 139 (133–145) 0.66 MAP (mmHg) 95 (92–98) 95 (92–98) 96 (93–100) 0.28 CI (L/min/m2) 2.8 (2.6–3.0) 2.8 (2.6–3.0) 2.9 (2.6–3.1) 0.21 CPI (w/m2) 0.59 (0.54–0.63) 0.58 (0.54–0.63) 0.61 (0.56–0.66) 0.09 TPRI (dyn × s/cm5 × m2) 3147 (2706–3589) 3159 (2695–3622) 3142 (2653–3632) 0.85 Fluid removal data, adjusted meana 95% CIa TFR (mL) 2332 (2136–2530) 2328 (2131–2525) 2393 (2200–2586) 0.47 Treatment duration (h:min) 3:37 (3:26–3:47) 3:37 (3:24–3:49) 3:36 (3:23–3:50) 0.87 UFR (mL/h) 662 (601–723) 661 (594–729) 678 (595–762) 0.53 Parameter All treatments (n = 263) Non-IDH treatments [n= 202 (76.8%)] IDH treatments [n = 61(23.2%)] IDH versus non-IDH adjusted P-valuea Pretreatment hemodynamics, adjusted meana (95% CIa) SBP (mmHg) 139 (133–144) 138 (133–144) 139 (133–145) 0.66 MAP (mmHg) 95 (92–98) 95 (92–98) 96 (93–100) 0.28 CI (L/min/m2) 2.8 (2.6–3.0) 2.8 (2.6–3.0) 2.9 (2.6–3.1) 0.21 CPI (w/m2) 0.59 (0.54–0.63) 0.58 (0.54–0.63) 0.61 (0.56–0.66) 0.09 TPRI (dyn × s/cm5 × m2) 3147 (2706–3589) 3159 (2695–3622) 3142 (2653–3632) 0.85 Fluid removal data, adjusted meana 95% CIa TFR (mL) 2332 (2136–2530) 2328 (2131–2525) 2393 (2200–2586) 0.47 Treatment duration (h:min) 3:37 (3:26–3:47) 3:37 (3:24–3:49) 3:36 (3:23–3:50) 0.87 UFR (mL/h) 662 (601–723) 661 (594–729) 678 (595–762) 0.53 a Calculated by using a mixed model for repeated measurements and adjusted for multiple comparisons using the Tukey–Kramer method. TFR, total fluid removed; UFR, ultrafiltration rate; CI, cardiac index; TPRI, total peripheral resistance index (dyn × s/cm5 × m2) method. In the non-IDH evaluations [637 of 736 (86.5%)], hemodynamics did not significantly change during treatments. CPI decreased by 3% (95% CI −5% to −2%) from a pretreatment level of 0.59 (95% CI 0.54–0.63) w/m2, P = 0.28; CI did not change (95% CI −1–1%) from a pretreatment level of 2.8 (95% CI 2.6–3.0) L/min/m2, P = 0.94 and TPRI decreased by 6% (95% CI −10% to −1%) from a pretreatment level of 3147 (95% CI 2706–3589) dyn × s/cm5 × m2, P = 0.37. SBP decreased by 4 (95% CI −1 to −8) from a pretreatment level of 138 (95% CI 133–144) mmHg, P = 0.71 and MAP decreased by 3 (95% CI −1 to −5) from a pretreatment of level 95 (95% CI 92–98) mmHg, P = 0.43 (Table 3). Table 3 Hemodynamics trends from pretreatment to intra- and postdialytic nadir measurements (n = 736) of the non-IDH and each of the three IDH subgroups Parameter Non-IDH [n = 637 (86.5%)] IDH, n = 99 (13.5%) Adj. P-valuea (1) CPI decreased [n = 35 (35.3%)] (2) TPRI decreased (vasodilatation) [n = 37 (37.4%)] (3) Combined CPI and TPRI decrease [n = 27 (27.3%)] (1) CPI decrease versus non-IDH (2) TPRI decrease versus non-IDH (1) CPI versus (2) TPRI decrease SBP (mmHg) Pre 138 (133–144) 145 (137–152) 137 (130–144) 135 (128–143) 0.14 0.97 0.21 Nadir 134 (129–139) 109 (101–116) 107 (100–114) 102 (95–110) <0.001 <0.001 0.95 Change −4 (−1 to −8) −35 (−28 to −43) −30 (−23 to −37) −33 (−25 to −41) <0.001 <0.001 0.67 Adj. P-valuea 0.71 <0.001 <0.001 <0.001 MAP (mmHg) Pre 95 (92–98) 98 (94–103) 96 (91–100) 95 (90–99) 0.27 0.98 0.68 Nadir 92 (89–95) 75 (70–80) 74 (69–78) 73 (68–78) <0.001 <0.001 0.80 Change −3 (−1 to −5) −23 (−18 to −28) −22 (−18 to −27) −22 (−17 to −27) <0.001 <0.001 0.99 Adj. P valuea 0.43 <0.001 <0.001 <0.001 CI (mL/m2) Pre 2.8 (2.6–3.0) 3.1 (2.8–3.3) 2.6 (2.3–2.8) 3.0 (2.8–3.3) 0.022 0.062 0.002 Nadir 2.8 (2.6–2.9) 2.6 (2.4–2.9) 3.2 (2.9–3.4) 3.1 (2.8–3.3) 0.07 <0.001 <0.001 Change (%) 0% (−1 to 1) −15% (−14 to −16) 23% (21 to 26) 1% (1 to 2) 0.001 <0.001 <0.001 Adj. P valuea 0.94 <0.001 <0.001 0.73 CPI (w/m2) Pre 0.59 (0.54–0.63) 0.66 (0.60–0.72) 0.55 (0.49–0.60) 0.63 (0.57–0.70) 0.01 0.23 0.002 Nadir 0.56 (0.53–0.60) 0.43 (0.37–0.48) 0.49 (0.44–0.55) 0.48 (0.42–0.53) <0.001 0.01 0.06 Change −3% (−5 to −2) −36% (−33 to −38) −9% (−8 to −10) −25% (−24 to −26) <0.001 0.576 <0.001 Adj. P- valuea 0.28 <0.001 <0.05 <0.001 TPRI Pre 3147 (2706–3589) 3069 (2539–3601) 3342 (2824–3859) 2964 (2428–3501) 0.92 0.53 0.49 Nadir 2959 (2690–3228) 2596 (2229–2963) 2251 (1900–2602) 2266 (1891–2642) 0.04 <0.001 0.08 Change −6% (−10 to −1) −14% (−12 to −18) −33% (−32 to −33) −23% (−22 to −25) 0.23 <0.001 0.002 Adj. P-valuea 0.37 <0.01 <0.001 <0.001 Parameter Non-IDH [n = 637 (86.5%)] IDH, n = 99 (13.5%) Adj. P-valuea (1) CPI decreased [n = 35 (35.3%)] (2) TPRI decreased (vasodilatation) [n = 37 (37.4%)] (3) Combined CPI and TPRI decrease [n = 27 (27.3%)] (1) CPI decrease versus non-IDH (2) TPRI decrease versus non-IDH (1) CPI versus (2) TPRI decrease SBP (mmHg) Pre 138 (133–144) 145 (137–152) 137 (130–144) 135 (128–143) 0.14 0.97 0.21 Nadir 134 (129–139) 109 (101–116) 107 (100–114) 102 (95–110) <0.001 <0.001 0.95 Change −4 (−1 to −8) −35 (−28 to −43) −30 (−23 to −37) −33 (−25 to −41) <0.001 <0.001 0.67 Adj. P-valuea 0.71 <0.001 <0.001 <0.001 MAP (mmHg) Pre 95 (92–98) 98 (94–103) 96 (91–100) 95 (90–99) 0.27 0.98 0.68 Nadir 92 (89–95) 75 (70–80) 74 (69–78) 73 (68–78) <0.001 <0.001 0.80 Change −3 (−1 to −5) −23 (−18 to −28) −22 (−18 to −27) −22 (−17 to −27) <0.001 <0.001 0.99 Adj. P valuea 0.43 <0.001 <0.001 <0.001 CI (mL/m2) Pre 2.8 (2.6–3.0) 3.1 (2.8–3.3) 2.6 (2.3–2.8) 3.0 (2.8–3.3) 0.022 0.062 0.002 Nadir 2.8 (2.6–2.9) 2.6 (2.4–2.9) 3.2 (2.9–3.4) 3.1 (2.8–3.3) 0.07 <0.001 <0.001 Change (%) 0% (−1 to 1) −15% (−14 to −16) 23% (21 to 26) 1% (1 to 2) 0.001 <0.001 <0.001 Adj. P valuea 0.94 <0.001 <0.001 0.73 CPI (w/m2) Pre 0.59 (0.54–0.63) 0.66 (0.60–0.72) 0.55 (0.49–0.60) 0.63 (0.57–0.70) 0.01 0.23 0.002 Nadir 0.56 (0.53–0.60) 0.43 (0.37–0.48) 0.49 (0.44–0.55) 0.48 (0.42–0.53) <0.001 0.01 0.06 Change −3% (−5 to −2) −36% (−33 to −38) −9% (−8 to −10) −25% (−24 to −26) <0.001 0.576 <0.001 Adj. P- valuea 0.28 <0.001 <0.05 <0.001 TPRI Pre 3147 (2706–3589) 3069 (2539–3601) 3342 (2824–3859) 2964 (2428–3501) 0.92 0.53 0.49 Nadir 2959 (2690–3228) 2596 (2229–2963) 2251 (1900–2602) 2266 (1891–2642) 0.04 <0.001 0.08 Change −6% (−10 to −1) −14% (−12 to −18) −33% (−32 to −33) −23% (−22 to −25) 0.23 <0.001 0.002 Adj. P-valuea 0.37 <0.01 <0.001 <0.001 Hemodynamic data are shown as adjusted mean (95% CI). a All hemodynamic data and P-values were calculated using a mixed model for repeated measurements and adjusted for multiple comparisons using the Tukey–Kramer method. Adj., adjusted; Pre, pretreatment; TPRI, total peripheral resistance index (dyn × s/cm5 × m2). Table 3 Hemodynamics trends from pretreatment to intra- and postdialytic nadir measurements (n = 736) of the non-IDH and each of the three IDH subgroups Parameter Non-IDH [n = 637 (86.5%)] IDH, n = 99 (13.5%) Adj. P-valuea (1) CPI decreased [n = 35 (35.3%)] (2) TPRI decreased (vasodilatation) [n = 37 (37.4%)] (3) Combined CPI and TPRI decrease [n = 27 (27.3%)] (1) CPI decrease versus non-IDH (2) TPRI decrease versus non-IDH (1) CPI versus (2) TPRI decrease SBP (mmHg) Pre 138 (133–144) 145 (137–152) 137 (130–144) 135 (128–143) 0.14 0.97 0.21 Nadir 134 (129–139) 109 (101–116) 107 (100–114) 102 (95–110) <0.001 <0.001 0.95 Change −4 (−1 to −8) −35 (−28 to −43) −30 (−23 to −37) −33 (−25 to −41) <0.001 <0.001 0.67 Adj. P-valuea 0.71 <0.001 <0.001 <0.001 MAP (mmHg) Pre 95 (92–98) 98 (94–103) 96 (91–100) 95 (90–99) 0.27 0.98 0.68 Nadir 92 (89–95) 75 (70–80) 74 (69–78) 73 (68–78) <0.001 <0.001 0.80 Change −3 (−1 to −5) −23 (−18 to −28) −22 (−18 to −27) −22 (−17 to −27) <0.001 <0.001 0.99 Adj. P valuea 0.43 <0.001 <0.001 <0.001 CI (mL/m2) Pre 2.8 (2.6–3.0) 3.1 (2.8–3.3) 2.6 (2.3–2.8) 3.0 (2.8–3.3) 0.022 0.062 0.002 Nadir 2.8 (2.6–2.9) 2.6 (2.4–2.9) 3.2 (2.9–3.4) 3.1 (2.8–3.3) 0.07 <0.001 <0.001 Change (%) 0% (−1 to 1) −15% (−14 to −16) 23% (21 to 26) 1% (1 to 2) 0.001 <0.001 <0.001 Adj. P valuea 0.94 <0.001 <0.001 0.73 CPI (w/m2) Pre 0.59 (0.54–0.63) 0.66 (0.60–0.72) 0.55 (0.49–0.60) 0.63 (0.57–0.70) 0.01 0.23 0.002 Nadir 0.56 (0.53–0.60) 0.43 (0.37–0.48) 0.49 (0.44–0.55) 0.48 (0.42–0.53) <0.001 0.01 0.06 Change −3% (−5 to −2) −36% (−33 to −38) −9% (−8 to −10) −25% (−24 to −26) <0.001 0.576 <0.001 Adj. P- valuea 0.28 <0.001 <0.05 <0.001 TPRI Pre 3147 (2706–3589) 3069 (2539–3601) 3342 (2824–3859) 2964 (2428–3501) 0.92 0.53 0.49 Nadir 2959 (2690–3228) 2596 (2229–2963) 2251 (1900–2602) 2266 (1891–2642) 0.04 <0.001 0.08 Change −6% (−10 to −1) −14% (−12 to −18) −33% (−32 to −33) −23% (−22 to −25) 0.23 <0.001 0.002 Adj. P-valuea 0.37 <0.01 <0.001 <0.001 Parameter Non-IDH [n = 637 (86.5%)] IDH, n = 99 (13.5%) Adj. P-valuea (1) CPI decreased [n = 35 (35.3%)] (2) TPRI decreased (vasodilatation) [n = 37 (37.4%)] (3) Combined CPI and TPRI decrease [n = 27 (27.3%)] (1) CPI decrease versus non-IDH (2) TPRI decrease versus non-IDH (1) CPI versus (2) TPRI decrease SBP (mmHg) Pre 138 (133–144) 145 (137–152) 137 (130–144) 135 (128–143) 0.14 0.97 0.21 Nadir 134 (129–139) 109 (101–116) 107 (100–114) 102 (95–110) <0.001 <0.001 0.95 Change −4 (−1 to −8) −35 (−28 to −43) −30 (−23 to −37) −33 (−25 to −41) <0.001 <0.001 0.67 Adj. P-valuea 0.71 <0.001 <0.001 <0.001 MAP (mmHg) Pre 95 (92–98) 98 (94–103) 96 (91–100) 95 (90–99) 0.27 0.98 0.68 Nadir 92 (89–95) 75 (70–80) 74 (69–78) 73 (68–78) <0.001 <0.001 0.80 Change −3 (−1 to −5) −23 (−18 to −28) −22 (−18 to −27) −22 (−17 to −27) <0.001 <0.001 0.99 Adj. P valuea 0.43 <0.001 <0.001 <0.001 CI (mL/m2) Pre 2.8 (2.6–3.0) 3.1 (2.8–3.3) 2.6 (2.3–2.8) 3.0 (2.8–3.3) 0.022 0.062 0.002 Nadir 2.8 (2.6–2.9) 2.6 (2.4–2.9) 3.2 (2.9–3.4) 3.1 (2.8–3.3) 0.07 <0.001 <0.001 Change (%) 0% (−1 to 1) −15% (−14 to −16) 23% (21 to 26) 1% (1 to 2) 0.001 <0.001 <0.001 Adj. P valuea 0.94 <0.001 <0.001 0.73 CPI (w/m2) Pre 0.59 (0.54–0.63) 0.66 (0.60–0.72) 0.55 (0.49–0.60) 0.63 (0.57–0.70) 0.01 0.23 0.002 Nadir 0.56 (0.53–0.60) 0.43 (0.37–0.48) 0.49 (0.44–0.55) 0.48 (0.42–0.53) <0.001 0.01 0.06 Change −3% (−5 to −2) −36% (−33 to −38) −9% (−8 to −10) −25% (−24 to −26) <0.001 0.576 <0.001 Adj. P- valuea 0.28 <0.001 <0.05 <0.001 TPRI Pre 3147 (2706–3589) 3069 (2539–3601) 3342 (2824–3859) 2964 (2428–3501) 0.92 0.53 0.49 Nadir 2959 (2690–3228) 2596 (2229–2963) 2251 (1900–2602) 2266 (1891–2642) 0.04 <0.001 0.08 Change −6% (−10 to −1) −14% (−12 to −18) −33% (−32 to −33) −23% (−22 to −25) 0.23 <0.001 0.002 Adj. P-valuea 0.37 <0.01 <0.001 <0.001 Hemodynamic data are shown as adjusted mean (95% CI). a All hemodynamic data and P-values were calculated using a mixed model for repeated measurements and adjusted for multiple comparisons using the Tukey–Kramer method. Adj., adjusted; Pre, pretreatment; TPRI, total peripheral resistance index (dyn × s/cm5 × m2). Since the primary objective of the study was to categorize IDH episodes in terms of the immediate hemodynamic cause(s) of hypotension, and not a comparison of individual patients, the episodes (n = 99) were divided into three subgroups according to the observed functional changes per episode. The three subgroups were initially defined according to changes in cardiac output (decrease, increase or no change) as blood pressure decreased. As cardiac power is the product of blood pressure and cardiac output, a reduction in either or both will reduce cardiac power. Since peripheral resistance is the ratio between blood pressure and cardiac output, cardiac output will decrease as a result of a decrease in peripheral resistance. As a result, the three subgroups where named as follows: Predominantly due to CPI decrease: defined as a CPI decrease >15% and a CI decrease >5%. Thirty-five (35.3%) of IDH episodes were in this subgroup, with CPI decreased by 36% (95% CI −33 to −38) from a pretreatment level of 0.66 (95% CI 0.60–0.72) w/m2, P < 0.001; CI decreased by 15% (95% CI −14 to −16) from 3.1 (95% CI 2.8–3.3) L/min/m2, P < 0.001 and TPRI decreased by 14% (95% CI −12 to −18) from 3069 (95% CI 2539–3601) dyn × s/cm5 × m2, P < 0.01. SBP decreased by 35 (95% CI −28 to −43) from 145 (95% CI 137–152) mmHg, P < 0.001 and MAP decreased by 23 (95% CI −18 to −28) from 98 (95% CI 94–103) mmHg, P < 0.001. Predominantly due to TPRI decrease (vasodilatation): defined as a TPRI decrease >15% and a CI increase >5%. Thirty-seven 37 (37.4%) IDH episodes were in this group with TPRI decreased by 33% (95% CI −32 to −33) from a pretreatment level of 3342 (95% CI 2824–3859) dyn × s/cm5 × m2, P < 0.001; CI increased by 23% (95% CI 21–26) from a pretreatment level of 2.6 (95% CI 2.3–2.8) L/min/m2, P < 0.001 and CPI decreased by 9% (95% CI −8 to −10) from a pretreatment level of 0.55 (95% CI 0.49–0.60) w/m2, P < 0.05. SBP decreased by 30 (95% CI −23 to −37) from a pretreatment level of 137 (95% CI 130–144) mmHg, P < 0.001 and MAP decreased by 22 (95% CI −18 to −27) from a pretreatment level of 96 (95% CI 91–100) mmHg, P < 0.001. Due to a combination of CPI and TPRI decrease: defined as CPI and TPRI decreases >15% with a CI change of not >5%. Twenty-seven (27.3%) IDH episodes were in this subgroup, with CPI decreased by 25% (95% CI −24 to −26) from a pretreatment level of 0.63 (95% CI 0.57–0.70) w/m2, P < 0.001; TPRI decreased by 23% (95% CI −22 to −25) from a pretreatment level of 2964 (95% CI 2428–3501) dyn × s/cm5 × m2, P < 0.001 and CI was not significantly changed: 1.3% (95% CI 1.2–1.7) from a pretreatment level of 3.0 (95% CI 2.8–3.3) L/min/m2, P = 0.730. SBP decreased by 30 (95% CI −23 to −37) from a pretreatment level of 135 (95% CI 128–143) mmHg, P < 0.001 and MAP decreased by 22 (95% CI −18 to −27) from a pretreatment level of 95 (95% CI 90–99) mmHg, P < 0.001. Differences in changes in CI, CPI and TPRI between IDH subgroups 1 and 2 were significant: P < 0.001 for changes in CI and CPI and P = 0.002 for changes in TPRI. Differences in changes in SBP and MAP between IDH subgroups 1 and 2 were not significant (P = 0.67 and 0.99, respectively). Table 3 shows the hemodynamic trends of the non-IDH and the three IDH subgroups. Figure 2 illustrates the hemodynamic trends of each individual evaluation in the three IDH subgroups using three hemodynamic graphs of MAP versus CI. Representative lines of both CPI and TPRI are also shown. Hemodynamic ranges for a normal population are illustrated by a dashed octagon. Each arrow provides the MAP and CI changes from pretreatment of all intra- and postdialytic evaluations, which are substantial reductions of CPI and CI for subgroup 1, a substantial reduction of TPRI and an increase of CI in subgroup 2 and substantial reductions of both CPI and TPRI with no change of CI in subgroup 3. DISCUSSION The clinically vital relationship between intradialytic hypotension and consequent morbidity and mortality is unquestioned [16, 17]. The primary objective of this article was to report changes of cardiac output, peripheral resistance and cardiac power during incidents of IDH. The stringent definition of IDH used here includes a nadir SBP <100 mmHg or a nadir MAP <70 mmHg and was used to compensate for the lack of information about intradialytic symptoms, which is included in some definitions of IDH as additional subjective criteria. This study utilized regional impedance cardiographic technology, which has been validated for measurements during chronic hemodialysis [10]. It measures cardiac output, and with the additional information of blood pressure, calculates cardiac power and total peripheral resistance. All parameters were indexed to body surface area. Cardiac power reduction may occur as a result of a preload reduction due to high ultrafiltration rates with inadequate vascular refilling or a low target weight, together resulting in hypovolemia and a consequent reduction in blood pressure and cardiac output. The presence of diastolic dysfunction may increase cardiac sensitivity to preload reduction [18]. Low cardiac power has been shown to be an independent predictor of inhospital mortality in a broad spectrum of patients with primary cardiac disease [12]. As total peripheral resistance is the ratio between blood pressure and cardiac output, a reduction of blood pressure and an increase in cardiac output are the result of a reduction in total peripheral resistance. Such a reduction could be the result of volume reduction without compensatory vasoconstriction, which creates a situation of relative vasodilatation. Autonomic dysfunction, often associated with diabetes, could be the underlying reason for failure of vasoconstriction. Low peripheral resistance has been shown to be an independent predictor of in-hospital mortality in both sepsis and nonseptic patients [19] As hypovolemia (preload reduction) and autonomous dysfunction are independent parameters, both may occur in the same patient during the same dialysis, resulting in unchanged cardiac output. This means that both cardiac power and peripheral resistance may decrease at the same time. The new ability to categorize the underlying hemodynamic abnormalities leading to IDH in each patient and each dialysis session provides significant potential to improve the management of hypotension. A reduction of the ultrafiltration rate or an increase in the prescribed target weight in response to a progressive decrease in cardiac power or by the use of short-acting alpha adrenergic agonists, such as midodrine, in the prophylaxis or treatment of peripheral resistance decrease [20] are rational interventions that could reduce intradialytic episodes and prevent organ ischemia and consequent damage. The unexpected relatively high frequency of vasodilatation as a direct cause of hypotensive episodes either alone or together with decreased cardiac power (collectively 64.7% of all IDH episodes) suggests that more emphasis might be placed on autonomic dysfunctional disorders, particularly in patients with diabetes. The similarity in pretreatment hemodynamics in the IDH and non-IDH groups increases the value of intradialytic characterization of IDH mechanisms for appropriate interventions. Limitations The number of patients studied was small. However, the emphasis of this article is the analysis of all hemodynamic responses occurring during each dialysis, not a comparison of individual patients. Measurements were not made continuously and therefore IDH episodes occurring earlier in treatments would have been missed; however, the hemodynamics of every recorded IDH episode were analyzed. Comorbid and cardiovascular status data and information on inter- and intradialytic drug use were not collected since the main purpose of this study was to characterize IDH episodes in the course of routine dialysis treatments. An increase in the accuracy of blood pressure measurements would improve interpretation of small changes in hemodynamics. In conclusion, intradialytic noninvasive cardiac output measurement has enabled the assessment of changes in cardiac power and peripheral resistance. These have been demonstrated to be distinctly different causes of hypotensive episodes in routine dialysis patients. This information can potentially improve the care of dialysis patients prone to IDH, by interventions based on specific cause. ACKNOWLEDGEMENTS The authors would like to thank Loretta Andersen, RN, Clinical Manager, for facilitating the quality improvement program on which this work was based. FUNDING Partial financial support for the performance of the research was provided by NI Medical. This support consisted solely of the lending of devices and sensors for performance of the research and the financing of a study coordinator. AUTHORS’ CONTRIBUTIONS N.W.L. confirms that he has full access to the data in the study and final responsibility for the decision to submit for publication. Contributing authors were involved in designing the work, acquiring data and interpreting the results. N.W.L. conceived and partially designed the work, acquired data and played an important role in interpreting the results and drafting and revised the manuscript. M.H.F.G. d.A., L.E.B., H.A.T.F. acquired data and approved the final version. R.S. assisted in the design of the work, acquired data and assisted in data coordination. S.G. and T.H. assisted in the design of the work and acquired data. S.L. played a major role in the analysis of the data. C.W. assisted in the design and analysis of the work. CONFLICT OF INTEREST STATEMENT None declared. REFERENCES 1 McIntyre CW , Burton JO , Selby NM et al. Hemodialysis-induced cardiac dysfunction is associated with an acute reduction in global and segmental myocardial blood flow . Clin J Am Nephrol 2008 ; 3 : 19 – 26 Google Scholar CrossRef Search ADS 2 Ritz E , Rambausek M , Mall G et al. Cardiac changes in uremia and their possible relation to cardiovascular instability on dialysis . Contrib Nephrol 1990 ; 78 : 221 – 229 Google Scholar CrossRef Search ADS PubMed 3 Converse RL Jr , Jacobsen TN , Jost CM et al. Paradoxical withdrawal of reflex vasoconstriction as a cause of hemodialysis-induced hypotension . J Clin Invest 1992 ; 90 : 1657 – 1665 Google Scholar CrossRef Search ADS PubMed 4 Shoji T , Tsubakihara Y , Fujii M et al. Hemodialysis-associated hypotension as an independent risk factor for two-year mortality in hemodialysis patients . Kidney Int 2004 ; 66 : 1212 – 1220 Google Scholar CrossRef Search ADS PubMed 5 Zucchelli P , Santoro A. Dialysis-induced hypotension: a fresh look at pathophysiology . Blood Purif 1993 ; 11 : 85 – 98 Google Scholar CrossRef Search ADS PubMed 6 Paredes OL , Shite J , Shinke T et al. Impedance cardiography for cardiac output estimation—reliability of wrist-to-ankle electrode configuration . Circ J 2006 ; 70 : 1164 – 1168 Google Scholar CrossRef Search ADS PubMed 7 Cotter G , Moshkovitz Y , Kaluski E et al. Accurate, non-invasive, continuous monitoring of cardiac output by whole body electrical bio-impedance . Chest 2004 ; 125 : 1431 – 1440 Google Scholar CrossRef Search ADS PubMed 8 Leitman L , Sucher E , Kaluski E et al. Non-invasive measurement of cardiac output by whole-body bio-impedance during dobutamine stress echocardiography: clinical implications in patients with left ventricular dysfunction and ischaemia . Eur J Heart Fail 2006 ; 8 : 136 – 140 Google Scholar CrossRef Search ADS PubMed 9 Cotter G , Schachner A , Sasson L et al. Impedance cardiography revisited . Physiol Meas 2006 ; 27 : 817 – 827 Google Scholar CrossRef Search ADS PubMed 10 Germain M , Nathanson BH , Chait Y et al. Comparison of stroke volume measurements during hemodialysis using bioimpedance cardiography and echocardiography. Hemodial Int 2017 ; doi: 10.1111/hdi.12589 11 Du Bois D , Du Bois EF. A formula to estimate the approximate surface area if height and weight be known . Arch Intern Med (Chic) 1916 ; XVII : 863 – 871 Google Scholar CrossRef Search ADS 12 Mendoza D , Cooper HA , Panza JA. Cardiac power output predicts mortality across a broad spectrum of patients with acute cardiac disease . Am Heart J 2007 ; 153 : 366 – 370 Google Scholar CrossRef Search ADS PubMed 13 Cotter G , Moshkovitz Y , Kaluski E et al. The role of cardiac power and systemic vascular resistance in the pathophysiology and diagnosis of patients with acute congestive heart failure . Eur J Heart Fail 2003 ; 5 : 443 – 451 Google Scholar CrossRef Search ADS PubMed 14 Verbeke G , Molenberghs G. Linear mixed models for longitudinal data . New York : Springer , 2000 15 Tukey JW. Computing individual means in the analysis of variance . Biometrics 1949 ; 5 : 99 – 114 Google Scholar CrossRef Search ADS PubMed 16 Bradshaw W. The importance of mean arterial pressure as a patient assessment tool: in haemodialysis and acute care . Aust Nurs J 2012 ; 20 : 26 – 29 Google Scholar PubMed 17 Flythe JE , Xue H , Lynch KE et al. Association of mortality risk with various definitions of intradialytic hypotension . J Am Soc Nephrol 2015 ; 26 : 724 – 734 Google Scholar CrossRef Search ADS PubMed 18 Acharya P , Ranabhat K , Trikhatri Y et al. Effect of preload reduction by hemodialysis on Doppler indices of diastolic function in patients with end-stage renal disease . Kathmandu Univ Med J 2008 ; 6 : 98 – 101 19 Melo J , Peters J . Low systemic vascular resistance: differential diagnosis and outcome . Crit Care 1999 ; 3 : 71 – 77 Google Scholar CrossRef Search ADS PubMed 20 Prakash S , Garg AX , Heidenheim AP et al. Midodrine appears to be safe and effective for dialysis-induced hypotension: a systematic review . Nephrol Dial Transplant 2004 ; 19 : 2553 – 2558 Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Nephrology Dialysis TransplantationOxford University Press

Published: Sep 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off