Introduction: Clinical studies on differences among changes in cerebral and hepatic oxygenation during hemodialysis (HD) in patients with and without intradialytic hypotension (IDH) are limited. We investigated changes in intradialytic cerebral and hepatic oxygenation before systolic blood pressure (SBP) reached the nadir during HD and compared these differences between patients with and without symptomatic IDH. Methods: We analyzed data from 109 patients with (n = 23) and without (n = 86) symptomatic IDH who were treated with HD. Cerebral and hepatic regional oxygen saturation (rSO2), as a marker of tissue oxygenation and circulation, was monitored during HD using an INVOS 5100c oxygen saturation monitor. Changes in cerebral or hepatic rSO2 when SBP reached the nadir during HD were compared between the groups of patients. Results: The cerebral rSO2 before HD in patients with and without symptomatic IDH was 49.7 ± 11.2% and 51.3 ± 9.1% (p = 0.491). %Changes in cerebral rSO2 did not significantly differ between the two groups from 60 min before the SBP nadir during HD. Hepatic rSO2 before HD in patients with and without symptomatic IDH was 58.5 ± 15.4% and 57.8 ± 15.9% (p = 0.869). The %changes in hepatic rSO2 were significantly lower in patients with symptomatic IDH than in those without throughout the observational period (p < 0.001). We calculated the area under the receiver operating characteristic curve (AUC) and estimated cutoff values for changes in hepatic rSO2 as a symptomatic IDH predictor. The predictive ability at 5 and 40 min before symptomatic IDH onset was excellent, with AUCs and cutoff values of 0.847 and 0.841, and −10.9% and −5.0%, respectively. Conclusions: Hepatic oxygenation significantly decreased more in patients with symptomatic IDH before its onset, than in those without symptomatic IDH, whereas changes in cerebral oxygenation did not differ. Evaluating changes in hepatic oxygenation during HD might help to predict symptomatic IDH.

Ultrafiltration is essential to achieve a target body weight during hemodialysis (HD). However, excessive fluid removal is associated with reduced blood volume (BV) [1]. This frequently leads to the onset of intradialytic hypotension (IDH), which is an established treatment-associated complication of HD. Several harmful events associated with IDH include end-organ ischemia and increased morbidity and mortality rates [2‒5]. A BV shift from splanchnic venous blood pooling to the central circulation plays an important role during HD to relieve the hemodynamic instability and deterioration of systemic tissue oxygenation associated with a BV reduction during HD [6‒8]. Intradialytic splanchnic erythrocyte radioactivity that reflects splanchnic BV decreases in response to ultrafiltration, and this precedes a decrease in mean arterial pressure (MAP) [9]. Intradialytic hepato-splanchnic blood flow evaluated using indocyanine green decreases due to active splanchnic vasoconstriction associated with ultrafiltration [6]. These reports indicated the changes in splanchnic circulation, which could lead to the changes in splanchnic oxygenation, in response to ultrafiltration during HD. However, these methods reported previously [6, 9] were complexed and performed only intermittently. Therefore, in the clinical setting of HD therapy, an intradialytic splanchnic circulation could not be easily evaluated.

Regional oxygen saturation (rSO2) is a marker of tissue oxygenation that has been measured using near-infrared spectroscopy [10, 11]. Cerebral rSO2 during HD is maintained under hemodynamic stability without IDH [12]. However, cerebral rSO2 suddenly decreases at the onset of symptomatic IDH [13, 14]. Hepatic rSO2, as a marker of hepato-splanchnic circulation and oxygenation, has been positively associated with MAP before HD [15]. It is also maintained like cerebral rSO2 during HD without IDH [12]. However, differences in changes in cerebral and hepatic oxygenation during HD between patients with and without IDH, in particular symptomatic IDH, have not been investigated in detail. Therefore, we aimed to clarify differences in altered intradialytic cerebral and hepatic oxygenation during HD between patients with and without symptomatic IDH.

Patients

This study screened 171 patients who were treated by HD between August 1, 2013, and July 31, 2023, and 116 met the inclusion criteria of age more than 20 years, end-stage renal disease managed with HD, started HD at least 1 month before the study, tissue thickness ≤20 mm from the skin to the surface of the liver in the right intercostal area on ultrasonography, and intradialytic cerebral and hepatic rSO2 data monitored using an INVOS 5100c oxygen saturation monitor (Medtronic Japan, Tokyo, Japan). The exclusion criteria were as follows: congestive heart failure requiring oxygen inhalation, neurological disorders such as severe cerebrovascular disease, and cognitive impairment; intradialytic feeding; and systolic blood pressure (SBP) nadir within 60 min after HD initiation. Seven patients were excluded because of a lack of data. We finally analyzed the data from 109 patients.

Ethics

All participants provided written informed consent to analyze and publish their innominate data. The study design was approved by the Institutional Review Board of the Saitama Medical Center at Jichi Medical University (Saitama, Japan) (Approval ID: RIN15-104 and RINS19-HEN007) and proceeded according to the ethical principles enshrined in the Declaration of Helsinki (2013 amendment).

Baseline Characteristics of Patients and Definition of Symptomatic IDH

Patient baseline characteristics and clinical data were retrieved from medical records. Blood pressure (BP) was regularly measured in supine patients at the start of HD and every 60 min thereafter and when intradialytic symptoms such as nausea, cold sweats, and muscle cramps developed. Referring to previous reports, we defined IDH as a decrease in SBP of ≥20 mm Hg after starting HD [16, 17]. Clinical symptoms associated with IDH were muscle cramps, abdominal pain, loss of consciousness, and convulsions [16]. According to our definition, 23 patients treated by HD who developed muscle cramps had symptomatic IDH and 86 patients (asymptomatic IDH, 33 patients; without IDH, 53 patients) did not have symptomatic IDH. Before starting HD, blood samples were collected at room temperature from access points (arteriovenous fistula, n = 108; arteriovenous graft, n = 1) for laboratory analyses (Table 1) and oxygen saturation (SpO2) measurements. Values for oxygen status, including pH, oxygen pressure (pO2 in mm Hg), and SpO2, are similar in blood samples obtained from the radial artery or an arterial line at an arteriovenous fistula [18]. Therefore, we collected all blood samples, including blood gas analysis, from the arterial sites of arteriovenous fistulae in patients before HD. The HD dialysate was composed of 140 mEq/L Na+, 2.0 mEq/L K+, 110 mEq/L Cl, 3.0 mEq/L Ca2+, 1.0 mEq/L Mg2+, 30 mEq/L HCO3, and 100 mg/dL glucose, and the dialysate temperature was set at 36.0°C. A dialyzer composed of polysulfone membrane (16 patients) and cellulose triacetate membrane (7 patients) was used in patients with symptomatic IDH, and polysulfone membrane (73 patients), cellulose triacetate membrane (10 patients), polymethyl methacrylate membrane (1 patient), polyester polymer alloy membrane (1 patient), and ethylene vinyl alcohol copolymer membrane (1 patient) were used in patients without symptomatic IDH.

Table 1.

Characteristics of all patients and patients with and without symptomatic IDH

CharacteristicsAll patientsWith symptomatic IDHWithout symptomatic IDH
N 109 23 86 
Men/women, % 83/26 (79/21) 17/6 (74/26) 66/20 (77/23) 
Age, years 71.0 (63.0–77.0) 70.0 (64.0–76.0) 71.5 (62.0–77.0) 
Body mass index, kg/m2 22.6 (20.1–25.1) 23.6 (21.5–24.4) 22.6 (19.9–25.4) 
Arterial O2 saturation, % 96.3 (94.7–97.4) 96.9 (95.7–97.6) 96.0 (93.9–97.3) 
Time at the SBP nadir after HD initiation, min 148 (104–190) 181 (121–196) 136 (100–183) 
Causes of chronic renal failure 
 Diabetes mellitus 58 (53) 15 (65) 43 (50) 
 Chronic glomerulonephritis 22 (20) 3 (13) 19 (22) 
 Nephrosclerosis 18 (17) 3 (13) 15 (17) 
 Others 11 (10) 2 (9) 9 (11) 
Comorbidities 
 Cardiovascular disease 39 (36) 11 (48) 28 (33) 
 Cerebrovascular disease 20 (18) 3(13) 17 (20) 
 Antihypertensive medication 96 (88) 21 (91) 75 (87) 
 Use of vasopressors before HD 12 (11) 6 (26) 6 (7) 
HD-associated parameters 
 HD vintage, years 1.5 (0.1–7.4) 0.8 (0.2–6.7) 1.6 (0.1–7.6) 
 HD times, 2 times/3 times/week 14/95 0/23 14/72 
 HD duration, h 4.0 (3.0–4.0) 4.0 (3.5–4.0) 4.0 (3.0–4.0) 
 Ultrafiltration rate, mL/kg/h 9.2 (6.1–11.1) 10.1 (7.6–11.1) 8.5 (5.6–11.1) 
 Cumulative ultrafiltration volume at the SBP nadir 1,191 (668–1,733) 1,540 (913–1,958) 1,030 (616–1,705) 
Tissue oxygenation parameters 
 Cerebral rSO2 before HD, % 51.0±9.6 49.7±11.2 51.3±9.1 
 Cerebral rSO2 at the SBP nadir during HD, % 49.8±9.2 47.3±10.0 50.5±8.9 
 %Changes in cerebral rSO2, % −2.0±6.7 −4.2±6.5 −1.4±6.7 
 Hepatic rSO2 before HD, % 58.0±15.8 58.5±15.4 57.8±15.9 
 Hepatic rSO2 at the SBP nadir during HD, % 56.0±15.2 50.7±14.8 57.4±15.1 
 %Changes in hepatic rSO2, % −2.6±11.1 −13.8±8.7 0.3±9.7 
Laboratory findings before HD 
 Hb, g/dL 10.0±1.6 9.9±1.7 10.1±1.6 
 Serum creatinine, mg/dL 8.9 (7.2–10.5) 8.3 (6.1–10.0) 9.1 (7.2–10.7) 
 Serum sodium, mEq/L 137±4 136±4 137±3 
 Serum albumin, g/dL 3.2 (2.8–3.6) 3.2 (2.7–3.5) 3.2 (2.8–3.6) 
CharacteristicsAll patientsWith symptomatic IDHWithout symptomatic IDH
N 109 23 86 
Men/women, % 83/26 (79/21) 17/6 (74/26) 66/20 (77/23) 
Age, years 71.0 (63.0–77.0) 70.0 (64.0–76.0) 71.5 (62.0–77.0) 
Body mass index, kg/m2 22.6 (20.1–25.1) 23.6 (21.5–24.4) 22.6 (19.9–25.4) 
Arterial O2 saturation, % 96.3 (94.7–97.4) 96.9 (95.7–97.6) 96.0 (93.9–97.3) 
Time at the SBP nadir after HD initiation, min 148 (104–190) 181 (121–196) 136 (100–183) 
Causes of chronic renal failure 
 Diabetes mellitus 58 (53) 15 (65) 43 (50) 
 Chronic glomerulonephritis 22 (20) 3 (13) 19 (22) 
 Nephrosclerosis 18 (17) 3 (13) 15 (17) 
 Others 11 (10) 2 (9) 9 (11) 
Comorbidities 
 Cardiovascular disease 39 (36) 11 (48) 28 (33) 
 Cerebrovascular disease 20 (18) 3(13) 17 (20) 
 Antihypertensive medication 96 (88) 21 (91) 75 (87) 
 Use of vasopressors before HD 12 (11) 6 (26) 6 (7) 
HD-associated parameters 
 HD vintage, years 1.5 (0.1–7.4) 0.8 (0.2–6.7) 1.6 (0.1–7.6) 
 HD times, 2 times/3 times/week 14/95 0/23 14/72 
 HD duration, h 4.0 (3.0–4.0) 4.0 (3.5–4.0) 4.0 (3.0–4.0) 
 Ultrafiltration rate, mL/kg/h 9.2 (6.1–11.1) 10.1 (7.6–11.1) 8.5 (5.6–11.1) 
 Cumulative ultrafiltration volume at the SBP nadir 1,191 (668–1,733) 1,540 (913–1,958) 1,030 (616–1,705) 
Tissue oxygenation parameters 
 Cerebral rSO2 before HD, % 51.0±9.6 49.7±11.2 51.3±9.1 
 Cerebral rSO2 at the SBP nadir during HD, % 49.8±9.2 47.3±10.0 50.5±8.9 
 %Changes in cerebral rSO2, % −2.0±6.7 −4.2±6.5 −1.4±6.7 
 Hepatic rSO2 before HD, % 58.0±15.8 58.5±15.4 57.8±15.9 
 Hepatic rSO2 at the SBP nadir during HD, % 56.0±15.2 50.7±14.8 57.4±15.1 
 %Changes in hepatic rSO2, % −2.6±11.1 −13.8±8.7 0.3±9.7 
Laboratory findings before HD 
 Hb, g/dL 10.0±1.6 9.9±1.7 10.1±1.6 
 Serum creatinine, mg/dL 8.9 (7.2–10.5) 8.3 (6.1–10.0) 9.1 (7.2–10.7) 
 Serum sodium, mEq/L 137±4 136±4 137±3 
 Serum albumin, g/dL 3.2 (2.8–3.6) 3.2 (2.7–3.5) 3.2 (2.8–3.6) 

Categorical and continuous data are presented as n (%) and means ± standard deviation or medians and interquartile ranges.

HD, hemodialysis; IDH, intradialytic hypotension; rSO2, regional oxygen saturation.

Monitoring Tissue Oxygenation

We monitored cerebral and hepatic rSO2 values that indicated tissue oxygenation, using an INVOS 5100c saturation monitor, which utilizes near-infrared spectroscopy, as previously described [15]. This instrument uses a light-emitting diode that transmits near-infrared light at wavelengths of 735 and 810 nm to measure oxygenated and deoxygenated hemoglobin (Hb) and two silicon photodiode light detectors. The ratio (%) of oxygenated to total Hb (oxygenated + deoxygenated Hb) signal strength is calculated and presented as a single numerical rSO2 value [10, 11]. The light paths from the emitter are shared between 30- and 40-mm detectors that assess superficial and deep tissues, respectively. We analyzed differential signals collected from deep tissues by the two detectors at distances of 20 and 30 mm from the body surface to determine rSO2 values [19, 20]. All data generated by this instrument were immediately and automatically stored in sequence. The interobserver variance determined that the reproducibility of rSO2 values measured by this instrument was acceptable [21‒23]. Therefore, rSO2 is considered reliable for estimating actual tissue oxygenation. The patients rested for at least 10 min in the supine position before HD to minimize the effects of postural changes on rSO2 values. Cerebral and hepatic rSO2 levels during HD were quantified by sensors attached to the forehead of patients on the side of the dominant brain hemisphere and to the right intercostal area above the liver that was already identified using ultrasonography. Cerebral and hepatic rSO2 levels were monitored thereafter starting 5 min after HD initiation until the HD session ended. The patients remained supine with minimal motion during HD and did not eat food to avoid the influences on changes in BV [24] and hepatic rSO2 levels [25]. To track changes in intradialytic cerebral and hepatic oxygenation, we evaluated the time-course of averaged rSO2 at 5-min intervals during HD, as a single rSO2 value, and rSO2 value at the SBP nadir was determined by a 5-min averaged rSO2 value including the time of SBP nadir. We also calculated and evaluated percentage (%) changes in cerebral and hepatic rSO2 from 60 min before that SBP reached the nadir during HD to the onset of SBP nadir as follows:

%Change in cerebral and hepatic rSO2 at t (min) before SBP nadir or symptomatic IDH onset during HD = (rSO2 at t [min] before SBP nadir or symptomatic IDH onset during HD divided by initial parameter before HD – 1) × 100.

Statistical Analysis

Data are expressed as means ± standard deviation for normally distributed values using Shapiro-Wilk tests or as medians with interquartile ranges with skewed distribution. Differences in SBP before HD, at the SBP nadir during HD, and SBP after HD were evaluated using a one-way analysis of variance and Scheffe test. We compared %changes in cerebral or hepatic rSO2 in patients with and without symptomatic IDH every 5 min from 60 min before reaching the SBP nadir during HD using unpaired Student t tests. We calculated the area under the receiver operating characteristic curve (AUC) with 95% confidence intervals (CIs) and estimated cutoff values for changes in cerebral and hepatic rSO2 every 5 min up to 60 min before the SBP nadir as a predictor of symptomatic IDH. All data were statistically analyzed using SPSS 26 (IBM Corp., Armonk, NY, USA). Values with p < 0.05 were considered statistically significant.

We enrolled 109 patients with (n = 23) and without (n = 86) symptomatic IDH who had been treated by HD (Table 1). The SBP in patients with symptomatic IDH before, at the nadir during HD, and after HD was 145 ± 21, 100 ± 21, and 136 ± 20 mm Hg. Furthermore, the SBP in patients without symptomatic IDH before, at the nadir during HD, and after HD was 147 ± 23, 128 ± 20, and 147 ± 25 mm Hg. The SBP significantly decreased at the time of the SBP nadir, compared with that before and after HD in patients with both groups (p < 0.001; Fig. 1). The SBP at the nadir during and after HD, respectively, indicated significantly decreased SBP during (p < 0.001) and after (p = 0.037) HD in patients with than without symptomatic IDH (Fig. 1).

Fig. 1.

Comparisons of SBP before HD, the SBP nadir during HD, and SBP after HD for patients with or without symptomatic IDH undergoing HD. *p < 0.001 versus before and after HD. p < 0.001 versus patients without symptomatic IDH undergoing HD. p < 0.05 versus patients without symptomatic IDH undergoing HD. IDH, intradialytic hypotension; HD, hemodialysis; SBP, systolic blood pressure.

Fig. 1.

Comparisons of SBP before HD, the SBP nadir during HD, and SBP after HD for patients with or without symptomatic IDH undergoing HD. *p < 0.001 versus before and after HD. p < 0.001 versus patients without symptomatic IDH undergoing HD. p < 0.05 versus patients without symptomatic IDH undergoing HD. IDH, intradialytic hypotension; HD, hemodialysis; SBP, systolic blood pressure.

Close modal

Figure 2 shows the time-course of averaged cerebral and hepatic rSO2 values during HD in patients with and without symptomatic IDH. Cerebral rSO2 values in both groups were similarly maintained during HD (Fig. 2a). Hepatic rSO2 in patients without symptomatic IDH was also kept constant during HD, whereas that in patients with symptomatic IDH gradually decreased until the time of the SBP nadir (Fig. 2b). Cerebral rSO2 before HD in patients with and without symptomatic IDH was 49.7 ± 11.2% and 51.3 ± 9.1%, respectively (p = 0.491). The cerebral rSO2 values at the SBP nadir during HD in patients with and without symptomatic IDH were 47.3 ± 10.0% and 50.5 ± 8.9%, respectively, and %changes in cerebral rSO2 at the SBP nadir did not significantly differ between groups (−4.2 ± 6.5%, vs. −1.4 ± 6.7%, p = 0.072). A comparison of %changes in cerebral rSO2 every 5 min starting 60 min before the SBP nadir in patients with and without symptomatic IDH revealed no significant differences throughout the HD session (Fig. 3a). Hepatic rSO2 before HD in patients with and without symptomatic IDH was 58.5 ± 15.4% and 57.8 ± 15.9%, respectively (p = 0.869). Hepatic rSO2 in patients with and without symptomatic IDH at the SBP nadir was 50.7 ± 14.8% and 57.4 ± 15.1%, respectively. Therefore, %changes in hepatic rSO2 decreased more in patients with symptomatic IDH than in those without (−13.8 ± 8.7% vs. 0.3 ± 9.7%; p < 0.001; Fig. 3b). Furthermore, %changes in hepatic rSO2 significantly differed between patients with and without symptomatic IDH every 5 min starting 60 min before the SBP nadir (p < 0.001; Fig. 3b).

Fig. 2.

Time-course of changes in cerebral and hepatic rSO2 during HD in patients with and without symptomatic IDH. a Changes in cerebral rSO2 during HD. b Changes in hepatic rSO2 during HD. HD, hemodialysis; IDH, intradialytic hypotension; rSO2, regional oxygen saturation.

Fig. 2.

Time-course of changes in cerebral and hepatic rSO2 during HD in patients with and without symptomatic IDH. a Changes in cerebral rSO2 during HD. b Changes in hepatic rSO2 during HD. HD, hemodialysis; IDH, intradialytic hypotension; rSO2, regional oxygen saturation.

Close modal
Fig. 3.

Comparison of %changes in cerebral or hepatic rSO2 in patients with and without symptomatic IDH. a %Changes in cerebral hepatic rSO2. b %Changes in hepatic rSO2. *p < 0.001 versus patients without symptomatic IDH undergoing HD. HD, hemodialysis; IDH, intradialytic hypotension; rSO2, regional oxygen saturation.

Fig. 3.

Comparison of %changes in cerebral or hepatic rSO2 in patients with and without symptomatic IDH. a %Changes in cerebral hepatic rSO2. b %Changes in hepatic rSO2. *p < 0.001 versus patients without symptomatic IDH undergoing HD. HD, hemodialysis; IDH, intradialytic hypotension; rSO2, regional oxygen saturation.

Close modal

The diagnostic ability to identify patients with symptomatic IDH was evaluated using the AUC of %changes in cerebral and hepatic rSO2 every 5 min starting 60 min before the SBP nadir (Table 2). A significant ability to predict %changes in cerebral rSO2 during 60 min before the SBP nadir was not found. However, %changes in hepatic rSO2 every 5 min starting 60 min before the SBP nadir had a good ability to predict symptomatic IDH onset (Table 2), especially at 5 and 40 min (AUC, 0.847; 95% CI, 0.760–0.934; and cutoff value, −10.9%, and AUC, 0.841; 95% CI, 0.751–0.932; and cutoff value, −5.0%, respectively; Table 3; Fig. 4).

Table 2.

Occurrence of symptomatic IDH detected using AUC of %changes in cerebral rSO2 or hepatic rSO2

Presymptomatic IDH, min%Change in cerebral rSO2%Change in hepatic rSO2
AUCp valueAUCp value
0.608 0.112 0.847 <0.001 
10 0.554 0.427 0.804 <0.001 
15 0.551 0.453 0.820 <0.001 
20 0.569 0.309 0.818 <0.001 
25 0.595 0.165 0.812 <0.001 
30 0.584 0.218 0.815 <0.001 
35 0.588 0.196 0.826 <0.001 
40 0.612 0.101 0.841 <0.001 
45 0.570 0.305 0.824 <0.001 
50 0.557 0.406 0.810 <0.001 
55 0.546 0.504 0.787 <0.001 
60 0.562 0.365 0.783 <0.001 
Presymptomatic IDH, min%Change in cerebral rSO2%Change in hepatic rSO2
AUCp valueAUCp value
0.608 0.112 0.847 <0.001 
10 0.554 0.427 0.804 <0.001 
15 0.551 0.453 0.820 <0.001 
20 0.569 0.309 0.818 <0.001 
25 0.595 0.165 0.812 <0.001 
30 0.584 0.218 0.815 <0.001 
35 0.588 0.196 0.826 <0.001 
40 0.612 0.101 0.841 <0.001 
45 0.570 0.305 0.824 <0.001 
50 0.557 0.406 0.810 <0.001 
55 0.546 0.504 0.787 <0.001 
60 0.562 0.365 0.783 <0.001 

AUC, area under the curve; IDH, intradialytic hypotension; rSO2, regional oxygen saturation.

Table 3.

Area under the receiver operator characteristic curve and cutoff values of %change in hepatic rSO2 at 5 and 40 min before symptomatic IDH

Presymptomatic IDH, minAUC95% CIp valueSensitivitySpecificityCutoff value, %
0.847 0.760–0.934 <0.001 0.92 0.65 −10.9 
40 0.841 0.751–0.932 <0.001 0.83 0.78 −5.0 
Presymptomatic IDH, minAUC95% CIp valueSensitivitySpecificityCutoff value, %
0.847 0.760–0.934 <0.001 0.92 0.65 −10.9 
40 0.841 0.751–0.932 <0.001 0.83 0.78 −5.0 

AUC, area under the receiver operator characteristic curve; CI, confidence interval; IDH, intradialytic hypotension; rSO2, regional oxygen saturation.

Fig. 4.

Receiver operating characteristic curve to predict symptomatic IDH based on %changes in hepatic rSO2 at 5 and 40 min before symptomatic IDH onset. IDH, intradialytic hypotension; rSO2, regional oxygen saturation.

Fig. 4.

Receiver operating characteristic curve to predict symptomatic IDH based on %changes in hepatic rSO2 at 5 and 40 min before symptomatic IDH onset. IDH, intradialytic hypotension; rSO2, regional oxygen saturation.

Close modal

This study focused on changes in cerebral and hepatic rSO2 during HD between patients with and without symptomatic IDH. The %changes in cerebral rSO2 did not significantly differ at IDH onset in patients with symptomatic IDH compared with those at the SBP nadir during HD in patients who did not have symptomatic IDH. Furthermore, every 5 min starting 60 min before the SBP nadir, these values did not significantly differ between patients with and without symptomatic IDH. On the other hand, %changes in hepatic rSO2 decreased significantly more in patients with than without symptomatic IDH. The %changes in hepatic rSO2 that significantly differed between the groups every 5 min starting 60 min before the SBP reached the nadir, particularly at 5 and 40 min, had a better ability to predict symptomatic IDH onset than those in cerebral rSO2.

In the clinical setting of HD therapy, IDH-related end-organ ischemia induces cerebral, splanchnic, and myocardial hypoxia that leads to cognitive, gastrointestinal, and myocardial dysfunction [26, 27]. Early prediction methods for IDH onset have been proposed to prevent these events, such as B-line quantitation before HD using lung ultrasound [28], heart rate variability [29, 30], ScvO2 [2, 31], and arterial oxygen saturation [32]. Particularly, as IDH occurrence is associated with changes in systemic oxygenation, ScvO2 significantly decreased in patients with IDH during HD, whereas intradialytic changes in ScvO2 were not significant in patients without IDH [31]. ScvO2 is significantly reduced before IDH onset during HD in patients with than without IDH, and this might be because ScvO2 can reflect a reduction in cardiac output and subsequent inadequate systemic tissue oxygenation during HD [2]. However, this method requires a central venous catheter and intermittent measurements such as repeated blood gas analyses of blood samples. Therefore, ScvO2 is not applicable to all patients undergoing HD.

The rSO2 can be noninvasively determined and continuously evaluated [10, 11], and it reflects organ blood flow [33, 34]. Cerebral blood flow is maintained by cerebral autoregulation between MAP 60 and 150 mm Hg to protect cerebral tissue from fluctuations in systemic BP. The lower limit of the MAP contribution to cerebral autoregulation shifts to 74.1 ± 17.6 mm Hg in patients undergoing HD [35]. We observed an MAP value at symptomatic IDH onset of 72 ± 17 mm Hg; therefore, a decrease in %changes in cerebral rSO2 at the onset of symptomatic IDH, which was not significant, might be attributed to the influence of reduced cerebral blood flow associated with symptomatic IDH. Furthermore, the absence of significant differences in these values before the SBP nadir between patients with and without symptomatic IDH was explained by the protective effect of the cerebral microcirculation associated with cerebral autoregulation. In contrast, we found significantly reduced hepatic rSO2 in patients with symptomatic IDH throughout the HD session. This was similar to intradialytic decreases in splanchnic erythrocyte radioactivity with fluid removal before MAP decreases during the late phase of HD [9]. The reduced hepatic rSO2 confirmed in this study might be attributed to tissue ischemia-associated adenosine release (the adenosine hypothesis) [36]. Adenosine is released into the extracellular fluid from ischemic organs and reduces systemic vessel tone; therefore, increased serum adenosine concentrations could lead to sudden hypotension [36, 37]. Hepato-splanchnic blood flow substantially decreases during ultrafiltration for HD as a result of active splanchnic vasoconstriction. Therefore, adenosine release associated with ischemic hepato-splanchnic tissues might result in sudden IDH [6, 7]. Continuous reductions in hepatic oxygenation before the onset of symptomatic IDH might directly reflect hepato-splanchnic ischemia during HD with ultrafiltration. Therefore, hepatic oxygenation might deteriorate before symptomatic IDH onset. We investigated whether identifying patients with symptomatic IDH using AUC analyses of %changes in hepatic rSO2 before symptomatic IDH onset could determine the probability of symptomatic IDH. The AUCs of %changes in hepatic rSO2 at 5 and 40 min were excellent predictors of symptomatic IDH. Previous studies of IDH prediction using noninvasive and continuous monitoring of tissue oxygenation have not used near-infrared spectroscopy. Based on our findings, monitoring intradialytic hepatic oxygenation in response to ultrafiltration might be useful for predicting symptomatic IDH onset because it is preceded by changes in hepatic oxygenation.

This study was limited by the relatively small sample of patients. Adequately measuring the dry weight of patients using appropriate determination methods is important [1]. We managed body fluid in the patients undergoing HD by physically assessing lower extremity edema, measuring intradialytic BP changes, and monitoring relative BV changes in response to ultrafiltration during HD using a BV monitor [1, 38]. Despite these procedures, we could not prevent symptomatic IDH onset in some patients undergoing HD. Patients without symptomatic IDH included patients with asymptomatic IDH and those without IDH. Hepatic rSO2 before HD did not differ significantly between patients with asymptomatic IDH and those without IDH (58.1 ± 14.8% vs. 57.6 ± 16.7%, p = 0.889), nor did %changes in hepatic rSO2 at the SBP nadir (−1.3 ± 9.2% vs. 1.3 ± 10.0%, p = 0.226). Therefore, patients with asymptomatic IDH and those without IDH were combined into one group and analyzed as patients without symptomatic IDH. We excluded patients whose SBP nadir occurred within 60 min of starting HD to confirm the relationship between changes in hepatic oxygenation and IDH onset associated with ultrafiltration. However, some of these patients might have developed IDH associated with ultrafiltration within 60 min starting HD. We obtained hepatic rSO2 values of deep tissue at points 20–30 mm from the body surface [18, 19]. Therefore, we excluded patients with obesity undergoing HD from the study because the wall thickness from the body surface to the liver surface was ≥20 mm. We could not measure adenosine concentrations at symptomatic IDH onset and therefore cannot directly comment on associations among adenosine, vessels, and symptomatic IDH. We concluded that hepatic oxygenation significantly decreased more in patients with symptomatic IDH than in those without symptomatic IDH before the onset of symptomatic IDH, although cerebral oxygenation did not significantly differ. Evaluating changes in hepatic oxygenation during HD might be useful for predicting symptomatic IDH.

We thank the study participants and staff at the dialysis center at Saitama Medical Center, Jichi Medical University. We would like to thank Editage (www.editage.com) for the English language editing.

The study design was approved by the Institutional Review Board of the Saitama Medical Center at Jichi Medical University (Saitama, Japan; approval numbers: RIN15-104 and RINS19-HEN007) in accordance with the provisions of the Declaration of Helsinki (as revised in Tokyo in 2004). All participants provided written informed consent to analyze and publish their innominate data.

The authors have no conflicts of interest to declare.

This work was partially supported by a grant from the Japanese Association of Dialysis Physicians [JADP Grant IDs 2017-9 and 2018-10] and JSPS KAKENHI [Grant IDs JP20K11534 to S.O. and JP21K16192 to K.I.]. The funder had no role in the design, data collection, data analysis, and reporting of this study.

S.O. and K.I.: conceptualization, methodology, software, validation, formal analysis, investigation, resources, data curation, writing original draft preparation, review and editing, and visualization. Y.S.: formal analysis. Y.U., J.M., S.K., Y.M., T.K., and K.H.: data curation. Y.M.: supervision. S.O.: project administration.

Additional Information

Susumu Ookawara and Kiyonori Ito contributed equally to this work.

All data analyzed during this study are within the manuscript and supplementary material file (for all online suppl. material, see https://doi.org/10.1159/000539277).

1.
Palmer
BF
,
Henrich
WL
.
Recent advances in the prevention and management of intradialytic hypotension
.
J Am Soc Nephrol
.
2008
;
19
(
1
):
8
11
.
2.
Keane
DF
,
Raimann
JG
,
Zhang
H
,
Willetts
J
,
Thijssen
S
,
Kotanko
P
.
The time of onset of intradialytic hypotension during a hemodialysis session associates with clinical parameters and mortality
.
Kidney Int
.
2021
;
99
(
6
):
1408
17
.
3.
Flythe
JE
,
Xue
H
,
Lynch
KE
,
Curhan
GC
,
Brunelli
SM
.
Association of mortality risk with various definitions of intradialytic hypotension
.
J Am Soc Nephrol
.
2015
;
26
(
3
):
724
34
.
4.
Burton
JO
,
Jefferies
HJ
,
Selby
NM
,
McIntyre
CW
.
Hemodialysis-induced repetitive myocardial injury results in global and segmental reduction in systolic cardiac function
.
Clin J Am Soc Nephrol
.
2009
;
4
(
12
):
1925
31
.
5.
Shoji
T
,
Tsubakihara
Y
,
Fujii
M
,
Imai
E
.
Hemodialysis-associated hypotension as an independent risk factor for two-year mortality in hemodialysis patients
.
Kidney Int
.
2004
;
66
(
3
):
1212
20
.
6.
Ribitsch
W
,
Schneditz
D
,
Franssen
CFM
,
Schilcher
G
,
Stadlbauer
V
,
Horina
JH
, et al
.
Increased hepato-splanchnic vasoconstriction in diabetics during regular hemodialysis
.
PLoS One
.
2015
;
10
(
12
):
e0145411
.
7.
Daugirdas
JT
.
Intradialytic hypotension and splanchnic shifting: integrating an overlooked mechanism with the detection of ischemia-related signals during hemodialysis
.
Semin Dial
.
2019
;
32
(
3
):
243
7
.
8.
Sars
B
,
van der Sande
FM
,
Kooman
JP
.
Intradialytic hypotension: mechanisms and outcome
.
Blood Purif
.
2020
;
49
(
1–2
):
158
67
.
9.
Yu
AW
,
Nawab
ZM
,
Barnes
WE
,
Lai
KN
,
Ing
TS
,
Daugirdas
JT
.
Splanchnic erythrocyte content decreases during hemodialysis: a new compensatory mechanism for hypovolemia
.
Kidney Int
.
1997
;
51
(
6
):
1986
90
.
10.
Ferrari
M
,
Mottola
L
,
Quaresima
V
.
Principles, techniques, and limitations of near infrared spectroscopy
.
Can J Appl Physiol
.
2004
;
29
(
4
):
463
87
.
11.
Tobias
JD
.
Cerebral oxygenation monitoring: near-infrared spectroscopy
.
Expert Rev Med Devices
.
2006
;
3
(
2
):
235
43
.
12.
Ookawara
S
,
Ito
K
,
Ueda
Y
,
Miyazawa
H
,
Hayasaka
H
,
Kofuji
M
, et al
.
Differences in tissue oxygenation and changes in total hemoglobin signal strength in the brain, liver, and lower-limb muscle during hemodialysis
.
J Artif Organs
.
2018
;
21
(
1
):
86
93
.
13.
Miyazawa
H
,
Ookawara
S
,
Tabei
K
.
Aggravation of cerebral oxygenation due to intradialytic hypotension induced by blood volume reduction during hemodialysis: a case report
.
Ther Apher Dial
.
2015
;
19
(
5
):
525
7
.
14.
Kitano
T
,
Ito
K
,
Ookawara
S
,
Shindo
M
,
Uchida
T
,
Kofuji
M
, et al
.
Changes in tissue oxygenation in response to sudden intradialytic hypotension
.
J Artif Organs
.
2020
;
23
(
2
):
187
90
.
15.
Ueda
Y
,
Ookawara
S
,
Ito
K
,
Sasabuchi
Y
,
Hayasaka
H
,
Kofuji
M
, et al
.
Association between hepatic oxygenation on near-infrared spectroscopy and clinical factors in patients undergoing hemodialysis
.
PLoS One
.
2021
;
16
(
10
):
e0259064
.
16.
K/DOQI Workgroup
.
K/DOQI clinical practice guidelines for cardiovascular disease in dialysis patients
.
Am J Kidney Dis
.
2005
;
45
(
4 Suppl 3
):
S1
53
.
17.
Ismail
AH
,
Gross
T
,
Schlieper
G
,
Walter
M
,
Eitner
F
,
Floege
J
, et al
.
Monitoring transcellular fluid shifts during episodes of intradialytic hypotension using bioimpedance spectroscopy
.
Clin Kidney J
.
2021
;
14
(
1
):
149
55
.
18.
Nielsen
AL
,
Thunedborg
P
,
Brinkenfeldt
H
,
Hegbrant
J
,
Jensen
HA
,
Wandrup
JH
.
Assessment of pH and oxygen status during hemodialysis using the arterial blood line in patients with an arteriovenous fistula
.
Blood Purif
.
1999
;
17
(
4
):
206
12
.
19.
Maslehaty
,
H
,
Krause-Titz
U
,
Petridis
AK
,
Barth
H
,
Mehdorn
HM
.
K/DOQI clinical practice guidelines for cardiovascular disease in dialysis patients
.
Am J Kidney Dis
.
2005
;
45
(
4 Suppl 3
):
S1
153
.
20.
Hongo
K
,
Kobayashi
S
,
Okudera
H
,
Hokama
M
,
Nakagawa
F
.
Noninvasive cerebral optical spectroscopy: depth-resolved measurements of cerebral haemodynamics using indocyanine green
.
Neurol Res
.
1995
;
17
(
2
):
89
93
.
21.
Lemmers
PMA
,
Toet
MC
,
van Bel
F
.
Impact of patent ductus arteriosus and subsequent therapy with indomethacin on cerebral oxygenation in preterm infants
.
Pediatrics
.
2008
;
121
(
1
):
142
7
.
22.
Hyttel-Sorensen
S
,
Sorensen
LC
,
Riera
J
,
Greisen
G
.
Tissue oximetry: a comparison of mean values of regional tissue saturation, reproducibility and dynamic range of four NIRS-instruments on the human forearm
.
Biomed Opt Express
.
2011
;
2
(
11
):
3047
57
.
23.
Schmitz
J
,
Pichler
G
,
Schwaberger
B
,
Urlesberger
B
,
Baik
N
,
Binder
C
.
Feasibility of long-term cerebral and peripheral regional tissue oxygen saturation measurements
.
Physiol Meas
.
2014
;
35
(
7
):
1349
55
.
24.
Shibagaki
Y
,
Takaichi
K
.
Significant reduction of the large-vessel blood volume by food intake during hemodialysis
.
Clin Nephrol
.
1998
;
49
(
1
):
49
54
.
25.
Imai
S
,
Ito
K
,
Ookawara
S
,
Kiryu
S
,
Iguchi
M
,
Sanayama
H
, et al
.
Does food ingestion during hemodialysis lead to changes in hepatic oxygenation
.
Nefrologia
.
2023
;
43
(
4
):
506
8
.
26.
Stefánsson
BV
,
Brunelli
SM
,
Cabrera
C
,
Rosenbaum
D
,
Anum
E
,
Ramakrishnan
K
, et al
.
Intradialytic hypotension and risk of cardiovascular disease
.
Clin J Am Soc Nephrol
.
2014
;
9
(
12
):
2124
32
.
27.
Yildiz
AB
,
Vehbi
S
,
Covic
A
,
Burlacu
A
,
Covic
A
,
Kanbay
M
.
An update review on hemodynamic instability in renal replacement therapy patients
.
Int Urol Nephrol
.
2023
;
55
(
4
):
929
42
.
28.
Allinovi
M
,
Palazzini
G
,
Lugli
G
,
Gianassi
I
,
Dallari
L
,
Laudicina
S
, et al
.
Pre-dialysis B-line quantification at lung ultrasound is a useful method for evaluating the dry weight and predicting the risk of intradialytic hypotension
.
Diagnostics
.
2022
;
12
(
12
):
2990
.
29.
Mostafa
H
,
Shaban
M
,
Hasanin
A
,
Mohamed
H
,
Fathy
S
,
Abdelreheem
HM
, et al
.
Evaluation of peripheral perfusion index and heart rate variability as early predictors for intradialytic hypotension in critically ill patients
.
BMC Anesthesiol
.
2019
;
19
(
1
):
242
.
30.
Park
S
,
Kim
WJ
,
Cho
NJ
,
Choi
CY
,
Heo
NH
,
Gil
HW
, et al
.
Predicting intradialytic hypotension using heart rate variability
.
Sci Rep
.
2019
;
9
(
1
):
2574
.
31.
Cordtz
J
,
Olde
B
,
Solem
K
,
Ladefoged
SD
.
Central venous oxygen saturation and thoracic admittance during dialysis: new approaches to hemodynamic monitoring
.
Hemodial Int
.
2008
;
12
(
3
):
369
77
.
32.
Mancini
E
,
Perazzini
C
,
Gesualdo
L
,
Aucella
F
,
Limido
A
,
Scolari
F
, et al
.
Intra-dialytic blood oxygen saturation (SO2): association with dialysis hypotension (the SOGLIA Study)
.
J Nephrol
.
2017
;
30
(
6
):
811
9
.
33.
Polinder-Bos
HA
,
Elting
JWJ
,
Aries
MJ
,
Garcia
DV
,
Willemsen
AT
,
van Laar
PJ
, et al
.
Changes in cerebral oxygenation and cerebral blood flow during hemodialysis: a simultaneous near-infrared spectroscopy and positron emission tomography study
.
J Cereb Blood Flow Metab
.
2020
;
40
(
2
):
328
40
.
34.
Alderliesten
T
,
De Vis
JB
,
Lemmers
PMA
,
van Bel
F
,
Benders
MJNL
,
Hendrikse
J
, et al
.
Simultaneous quantitative assessment of cerebral physiology using respiratory-calibrated MRI and near-infrared spectroscopy in healthy adults
.
Neuroimage
.
2014
;
85 Pt 1
(
1
):
255
63
.
35.
MacEwen
CM
,
Sutherland
S
,
Daly
J
,
Pugh
C
,
Tarassenko
L
.
Relationship between hypotension and cerebral ischemia during hemodialysis
.
J Am Soc Nephrol
.
2017
;
28
(
8
):
2511
20
.
36.
Shinzato
T
,
Miwa
M
,
Nakai
S
,
Morita
H
,
Odani
H
,
Inoue
I
, et al
.
Role of adenosine in dialysis-induced hypotension
.
J Am Soc Nephrol
.
1994
;
4
(
12
):
1987
94
.
37.
Daugirdas
JT
.
Pathophysiology of dialysis hypotension: an update
.
Am J Kidney Dis
.
2001
;
38
(
4 Suppl 4
):
S11
7
.
38.
Yoshida
I
,
Ando
K
,
Ando
Y
,
Ookawara
S
,
Suzuki
M
,
Furuya
H
, et al
.
A new device to monitor blood volume in hemodialysis patients
.
Ther Apher Dial
.
2010
;
14
(
6
):
560
5
.