Clinical Research

Relationship between Hypotension and Cerebral Ischemia during Hemodialysis

Clare MacEwen, Sheera Sutherland, Jonathan Daly, Christopher Pugh and Lionel Tarassenko
JASN August 2017, 28 (8) 2511-2520; DOI: https://doi.org/10.1681/ASN.2016060704

Abstract

The relationship between BP and downstream ischemia during hemodialysis has not been characterized. We studied the dynamic relationship between BP, real-time symptoms, and cerebral oxygenation during hemodialysis, using continuous BP and cerebral oxygenation measurements prospectively gathered from 635 real-world hemodialysis sessions in 58 prevalent patients. We examined the relationship between BP and cerebral ischemia (relative drop in cerebral saturation >15%) and explored the lower limit of cerebral autoregulation at patient and population levels. Furthermore, we estimated intradialytic exposure to cerebral ischemia and hypotension for each patient, and entered these values into multivariate models predicting change in cognitive function. In all, 23.5% of hemodialysis sessions featured cerebral ischemia; 31.9% of these events were symptomatic. Episodes of hypotension were common, with mean arterial pressure falling by a median of 22 mmHg (interquartile range, 14.3–31.9 mmHg) and dropping below 60 mmHg in 24% of sessions. Every 10 mmHg drop from baseline in mean arterial pressure associated with a 3% increase in ischemic events (P<0.001), and the incidence of ischemic events rose rapidly below an absolute mean arterial pressure of 60 mmHg. Overall, however, BP poorly predicted downstream ischemia. The lower limit of cerebral autoregulation varied substantially (mean 74.1 mmHg, SD 17.6 mmHg). Intradialytic cerebral ischemia, but not hypotension, correlated with decreased executive cognitive function at 12 months (P=0.03). This pilot study demonstrates that intradialytic cerebral ischemia occurs frequently, is not easily predicted from BP, and may be clinically significant.

Intradialytic hypotension (IDH) is a common problem, associated with disabling symptoms, interrupted treatment, and nursing workload.1 However, the link between IDH and hard outcomes remains elusive, and it is unclear whether there is a “safe” BP. Two large epidemiologic studies did not find an independent relationship between IDH and mortality, and reduced IDH did not translate into reduced morbidity or mortality in a multicenter randomized controlled trial of hemodiafiltration versus hemodialysis.2–4 Only one IDH definition has been independently associated with increased mortality, namely intradialytic systolic BP (SBP) <90 mmHg measured by automated upper-arm cuff, in a retrospective review of datasets with limited physiologic and symptom data.5

Several factors may have obscured any relationship between IDH and adverse outcomes. Firstly, hemodynamic status varies between sessions, particularly after the “long break,” but most studies have derived conclusions using hemodynamic data from only one or two treatments.6 Secondly, most IDH studies have measured BP intermittently using automated upper-arm cuffs, technology known as oscillometry. Although popular in clinical practice, oscillometry is prone to considerable error, particularly in SBP, which is sensitive to cuff characteristics and arterial stiffness.7–10 Thirdly, although SBP is the most popular index in nephrology (reflected in current IDH definitions3,4,11–13), mean arterial pressure (MAP) better reflects organ perfusion, given two-thirds of the cardiac cycle is spent in diastole, and is the preferred index in critical care and tissue oxygenation research.14–17 Fourthly, returning to physiologic principles, BP is useful because it is a surrogate for tissue perfusion, itself a surrogate for cellular oxygen supply.18 However, the relationship between systemic pressures and tissue oxygen balance is complex. Organ-specific autoregulation may preserve regional blood flow (Qb) in the face of changing systemic pressures, and conversely reductions in regional Qb may occur with steady systemic pressures due to changes in vascular tone.

Moreover, reductions in regional Qb do not necessarily lead to cellular ischemia: this depends on blood oxygen content, functional capillary density, oxygen dissociation characteristics, cellular oxygen demand, and mitochondrial health.19,20 Finally, the choice of outcome measure in IDH studies is problematic. The contribution of IDH to mortality may be masked by the heterogeneous comorbidities in this population. Organ-specific adverse events are difficult to quantitate, and in the case of the myocardium the role of the heart in generating BP confounds its utility as an indicator organ for hypotension-induced ischemia.

To address some of these issues we have prospectively and continuously monitored BP, real-time symptoms, and cerebral oxygenation in a large number of hemodialysis sessions. We opted to use the brain as the indicator organ because of the availability of both validated technology suited to large scale, continuous, noninvasive cerebral oxygenation monitoring (near-infrared spectroscopy [NIRS]), and validated cognitive function tests that provide a convenient, clinically important measure of end-organ damage.

NIRS uses the absorption of different wavelengths of near-infrared light by oxygenated and deoxygenated hemoglobin to continuously, noninvasively, and automatically estimate changes in hemoglobin oxygen saturation in tissue 2–3 cm below the skin.21 Infrared light penetrates the skull, allowing NIRS to interrogate the frontal cortex. NIRS-measured changes in frontal lobe oxygenation correlate well with more invasive measures of cerebral perfusion, including functional magnetic resonance imaging (MRI).22–24 Intraoperative desaturation detected by cerebral NIRS is associated with hard outcomes including postoperative cognitive dysfunction and measures of noncerebral organ failure (intensive care days, cardiac events, ventilator days, and renal failure), suggesting that cerebral ischemia indicates systemic insult.25–32 There is increasing experience with NIRS as a monitor of cerebral oxygenation in nonsurgical situations, e.g., traumatic brain injury, exercise, and altitude,33,34 but limited data in patients receiving dialysis to date.35,36

The primary objective of this study was to assess the relationship between BP, symptoms, and cerebral oxygenation during hemodialysis. The secondary objective was to make a preliminary estimate of whether decline in cognitive function is associated with intradialytic physiology.

Results

Seventy representative patients from our prevalent, predominantly white hemodialysis population were screened for inclusion; ten declined, citing the inconvenience of monitoring and concerns that treatment time would extend. Sixty patients underwent baseline assessment; two were unable to progress due to acute illness. Data were gathered from 635 dialysis sessions in the remaining 58 patients.

Each patient underwent monitoring for a median of 12 consecutive sessions (median 51.6 hours) over 4 weeks (interquartile range [IQR], 12–14 over 4–5 weeks; see Table 1). Two patients switched midtrial from volume-clamp BP (continuous BP derived from finger cuff, see Concise Methods) to 15 minutes’ oscillometry (automated arm-cuff pressures) because of severe vascular disease. Two others did not tolerate the NIRS sensor, leaving 583 sessions in 54 patients with the complete set of continuous BP, cerebral NIRS, and symptom data for the primary analysis. Table 1 shows demographics and dialysis parameters.

View this table:
Table 1.

Demographics

The agreement between oscillometry and volume-clamp BP measurements was superior for MAP (mean disagreement −3.5 mmHg; SD 14.0 mmHg) as compared with SBP (mean −6.9 mmHg; SD 19.6 mmHg)—see Supplemental Figures 1 and 2 for Bland–Altman plots. This agreement is similar to that seen between oscillometry and intra-arterial BP measurements.7–10 Adjusting for repeated measures, changes in MAP outperformed changes in SBP for predicting change in cerebral saturation (P<0.001 for comparison of model fit).

Therefore, we used MAP as our primary BP index.

Analysis 1: Association between Hypotension and Adverse Events

The temporal associations of relative MAP and absolute MAP with symptoms, interventions, and cerebral ischemia are shown in Tables 2 and 3, respectively. Of the sessions, 24.9% featured symptoms, including 10.9% cramp, 8.8% nausea, and 7.6% dizziness. Of the hemodialysis sessions, 23.5% featured cerebral ischemia, defined as a drop of 15% of baseline cerebral saturation, of which 31.9% were symptomatic. Sustained (2 minutes minimum) hypotensive episodes were common, with MAP falling by a median of 22 mmHg (IQR, 14.3–31.9 mmHg), and dropping below 60 mmHg in 24% of sessions.

View this table:
Table 2.

Events temporally associated with a fall in MAP from baseline

View this table:
Table 3.

Events temporally associated with a fall in MAP below absolute levels

We examined the relationship between MAP thresholds and incident cerebral ischemia; adjustments were not made for repeated measures because this was an exploratory analysis assessing the validity of population-based IDH definitions. There was a highly significant linear relationship between relative MAP thresholds and contemporaneous ischemia (3% increase in incident cerebral ischemia per 10 mmHg drop from baseline; P<0.001). For absolute MAP, a negative exponential described the relationship, with a sharp increase in incident ischemia below 60 mmHg (doubling per 20 mmHg drop; P<0.001).

There was also a significant linear relationship between relative MAP thresholds and contemporaneous symptoms (4% increase per 10 mmHg drop from baseline; P<0.001). Absolute MAP thresholds again showed an exponential relationship, with a sharp rise in incident symptoms below 60 mmHg (doubling per 24 mmHg drop; P=0.02). Dizziness, gastrointestinal symptoms, and feeling hot/sweaty accounted for this relationship. Conversely, there was no significant correlation between MAP thresholds and cramp, which was instead associated with ultrafiltration volume and serum/dialysate sodium difference (not shown).

Although MAP and cerebral ischemia were significantly related, it was apparent that BP thresholds overall were poorly predictive for cerebral ischemia, and there was no clear “safe” pressure. The sensitivity and specificity of MAP thresholds for predicting cerebral ischemia are depicted visually in Figures 1 and 2. There is no optimal cutoff for sensitivity and specificity.

Figure 1.
Figure 1.

Sensitivity/true positive rate (solid line) and specificity/true negative rate (broken line) of change in MAP from baseline for predicting the onset of cerebral ischemia in patients receiving hemodialysis.

Figure 2.
Figure 2.

Sensitivity/true positive rate (solid line) and specificity/true negative rate (broken line) of absolute MAP thresholds for predicting the onset of cerebral ischemia in patients receiving hemodialysis.

Analysis 2: Lower Limits of Autoregulation

The lower limit of cerebral autoregulation is often considered a surrogate for “safe” BP, and was explored as a complementary analysis. Of the 56 patients who completed NIRS monitoring, 37 had intact autoregulation, 18 had absent autoregulation, and one had insufficient data. Twenty-seven of the 37 patients with autoregulation had an identifiable lower limit; the remainder experienced insufficient range of MAP to allow accurate modeling of the lower limit, but clearly had intact autoregulation. Examples of patients with and without autoregulation, including one with clear autoregulation but no identifiable lower limit, are shown in the online supplement (Supplemental Figures 4–6): the 95% confidence intervals (95% CIs) in these figures illustrate that the relationship between BP and cerebral oxygen saturation had reasonable intrapatient consistency.

Where measurable, the lower limit was highly variable (mean 74.1 mmHg; SD 17.6 mmHg). The relationship between MAP and cerebral oxygen saturation for the whole population is shown in Figure 3: this is essentially an “average” autoregulation curve, and demonstrates that below a MAP of 60 mmHg cerebral saturation becomes linearly dependent on BP for the majority of individuals (the small number of data points below 40 mmHg makes this part of the curve less reliable, as indicated by the large 95% CI).

Figure 3.
Figure 3.

Relationship between MAP and cerebral oxygen saturations at the population level, approximating to an “average” autoregulation curve. The data for all patients was pooled, and BP data were sorted into 5-mmHg-wide bins sliding every 1 mmHg, e.g., 60–65, 61–66, 62–67 mmHg. Each BP data point was associated with a cerebral oxygen saturation. The mean and 95% limits of cerebral oxygen saturation data for each 5-mmHg bin were calculated, and plotted against the midpoint for that bin, e.g., the 60–65 mmHg bin is represented by the point 62.5 mmHg on the x axis.

Analysis 3: Change in Cognitive Function

Of the 58 patients who completed physiologic monitoring, follow-up cognitive function testing was possible on 47 individuals (nine deaths and two geographic moves); 38 remained on hemodialysis and nine had received transplants.

Among the individuals remaining on hemodialysis, Trails Test B (TTB) score deteriorated by median +7.5 seconds (IQR, −6 to +39 seconds), or +8% of baseline score (IQR, −9% to +33%). Among transplanted patients, TTB score improved by median −6 seconds (IQR, −25 to +2 seconds), or 15% of baseline score (IQR, −27% to + 2%). The separation between these groups achieved significance (P=0.02 on Wilcoxon rank sum test): unsurprisingly those transplanted were younger with fewer comorbidities. The 100-point modified mini–mental state test (3MS) score deteriorated by mean −0.84 points (SD 5.4) in the group remaining on hemodialysis and −0.89 (4.1) in those transplanted.

Multivariate linear modeling on predictors of change in cognitive function was performed, excluding those transplanted. After bootstrapping to check robustness, the only significant predictor of deterioration in TTB score was typical exposure to intradialytic cerebral ischemia (P=0.03). The regression equation is in the supplemental material.

Conversely, the only significant predictor of change in 3MS performance was change in Patient Health Questionnaire–9 (PHQ-9) score, with an increase of 0.49 points per unit improvement in depression score (P=0.004).

Discussion

Classic teaching is that cerebral Qb is held steady between MAP 60 and 150 mmHg, so-called cerebral autoregulation, which protects the brain from fluctuating systemic pressures.37 In practice, some patients do not demonstrate autoregulation, and when present, thresholds vary depending on factors such as age and baseline BP.37–39 Multiorgan morbidity and mortality increase with time spent below the lower limit of autoregulation, suggesting it is a reasonable clinical target.31,32

In our study population, the mean lower limit of measurable autoregulation was 74 mmHg, but varied widely, with 95% CI, 38.9 to 109.3 mmHg, a range compatible with earlier studies.40–46

Although we found a significant, temporal relationship between both change in MAP and absolute MAP and incident cerebral ischemia, BP thresholds poorly predicted cerebral ischemia at the population level, with no clinically useful compromise between sensitivity and specificity. For illustration, in Figure 2, a MAP threshold of 60 mmHg has a specificity of >90%, meaning that cerebral ischemia occurs >90% of the time at this BP: treating this BP would rarely be inappropriate. However, the same threshold has a sensitivity <30%, meaning that >70% of ischemic events have occurred at higher pressures, i.e., one cannot be confident that MAP>60 mmHg is adequate for cerebral oxygenation in all patients. This is presumably because of the wide range of lower limits of cerebral autoregulation, further complicated by factors such as varying ability to increase oxygen extraction, meaning that ischemia does not necessarily result as soon as the lower limit is breached.41

The population-level cerebral autoregulation curve (Figure 3) is a somewhat artificial construct as it reflects pooled heterogeneous patient data: its lower limit merely indicates the pressure at which cerebral ischemia becomes universal across the population. However, using the relationship between MAP and SBP described in a large epidemiologic study of pulse pressure in patients receiving hemodialysis (SBP=10.6+1.3×MAP),47 the population lower limit of autoregulation (MAP 60 mmHg) equates to an SBP of 88.6 mmHg, close to the 90 mmHg threshold previously linked with mortality.5 It is perhaps not surprising that the relationship between adverse outcomes and BP at population level only becomes apparent at extreme hypotension, where virtually all patients exhibit downstream ischemia.

It is hypothesized that hemodynamic instability during hemodialysis leads to repetitive end-organ ischemia, in the case of the brain resulting in cognitive decline. Most investigators have found that cognitive dysfunction in patients receiving dialysis is largely executive, e.g., planning and processing speed, areas assessed by the TTB.48,49 There is an association between hemodialysis vintage and both executive cognitive function and white matter abnormalities on MRI.50–53 Proving a causative relationship with IDH is difficult. These imaging techniques do not allow real-time/continuous monitoring, and corresponding hemodynamic data are typically sparse. Moreover, cognitive impairment and white matter abnormalities are independently associated with renal impairment in a dose-related fashion, likely due to increasing burden of small vessel disease.49,54–56

Previous studies have not found a relationship between IDH and either cognitive outcomes or surrogates such as radiologic changes. A recent cross-sectional study found no independent association between intradialytic oscillometric BP and current cognitive function.57 Another cross-sectional study using data from the Frequent Hemodialysis Network trials reported similar findings: although executive dysfunction was common, there was no independent link with intradialytic hemodynamics.58 A small, randomized controlled trial found that cooling on dialysis reduced the development of white matter abnormalities on MRI, but the clinical significance of these lesions was unclear, and there was no difference in hypotension between study groups.53 These studies, although valuable, suffer from many of the issues outlined in the introduction, including no measures of real-time cerebral perfusion/oxygenation and sparse hemodynamic data.

An interesting signal from our pilot study was that typical intradialytic cerebral ischemia, arbitrarily defined as a drop of 15% of baseline value, corresponded to change in executive cognitive function as measured by the TTB (although not global cognitive function measured by the 3MS). Typical intradialytic hypotension, regardless of the threshold used, had no utility in predicting decline in cognition. This is consistent with results of previous studies, and explained by our observation that BP poorly predicts cerebral ischemia.

Our results partially explain why it has proven difficult to link BP (except for extreme hypotension) to hard outcomes, and highlight the shortcomings of current IDH definitions. BP is useful because it affects downstream tissue perfusion, but reduced systemic pressures are only one determinant of tissue ischemia. Therefore, measuring tissue oxygenation directly has advantages. In the environment of critical care and anesthesia, there is increasing use of tissue oxygenation rather than systemic circulatory parameters to guide treatment.20,59

There are a number of caveats to this study. Organs probably have different thresholds for ischemia: adequate cerebral oxygenation may not guarantee adequate cardiac or splanchnic oxygenation. However, there is evidence that cerebral ischemia is a marker of multiorgan insult.29–32 Secondly, we made the assumption that dialysis sessions during the study month were representative of those during follow-up, although we monitored many more sessions per patient than is typical in the literature. Thirdly, the definition of ischemia is arbitrary: there is no consensus on what change in cerebral saturations is harmful. Nevertheless, a convincing body of studies has shown that cerebral saturation changes similar to those reported here correlate with morbidity in other populations.25–32 Finally, cerebral NIRS predominantly measures oxygenation of frontal gray matter, so may not detect subcortical ischemia. It is unlikely that there would be ischemia of the cortex without ischemia of these more vulnerable structures, but we cannot exclude the occurrence of lesser ischemic insults, sufficient only to affect watershed areas.

The small numbers mean that this study can only be considered a pilot. Findings must be confirmed in a larger cohort of patients: statistical techniques to reduce confounding are no substitute for large numbers. Ultimately, a controlled trial pairing continuous cerebral oxygenation monitoring with interventions such as cooling, oxygen supplementation, or biofeedback dialysis might help to establish if such interventions can improve intradialytic cerebral oxygenation; following up with robust functional tests would allow detection of any clinically significant cerebral protection. Data presented here may be helpful in powering such a study. Cerebral NIRS may be the only suitable technology for this task because MRI techniques are not suited to continuous, real-time monitoring in a large number of patients. This study demonstrates that NIRS is feasible and can provide useful data in patients receiving dialysis, currently in the research setting, but possibly in selected clinical circumstances in future.

In summary, this study demonstrates that intradialytic cerebral ischemia occurs frequently, is not easily predicted from BP, and may associate with clinically significant outcomes.

Concise Methods

Ethical approval was obtained from the local research ethics committee; all subjects gave written, informed consent. All procedures were in accordance with the Declaration of Helsinki. Patients were eligible provided they had been established on hemodialysis for at least a month, were >18 years of age, and were able to give consent. The protocol was for each patient to undergo the following physiologic monitoring for 9–12 consecutive sessions: (1) continuous BP (volume-clamp method, see below, Finometer PRO or Finometer MIDI; Finometer Medical Supplies), (2) continuous cerebral tissue oxygenation (NIRS, INVOS system; Covidien), (3) 30 minutes’ oscillometry, i.e., automated upper-arm cuff BP (Colin BX-100; Colin Corporation), (4) continuous relative blood volume and venous saturations (Crit-Line; Fresenius), (5) continuous dialysis machine parameter monitoring including access pressures, ultrafiltration rate/volume, Qb, and dialysate flow (Qd) (Nexadia software; BBraun), and (6) real-time recording of type, onset, and offset of symptoms, with a patient-friendly computer interface and response pad (Cedrus RB-844 handset, Superlab v5 software).

A single computer was used as a time reference to ensure all data streams were tightly synchronized. Dialysis nurses made medical records as per standard practice on a purpose-designed flow sheet: data gathered included nursing interventions (e.g., saline infusion, bed tipping, slowing down ultrafiltration rate) and any concerns about the patient. Clinical staff and patients were blinded to all data except relative blood volume and oscillometry, so this could not influence symptom-reporting or intervention.

Standard (oscillometry) cuff and volume-clamp cuff were placed on the nonaccess arm. For background information on volume-clamp technology, including agreement with oscillometric pressures, please see the full Concise Methods. The INVOS adhesive sensor was placed on one side of the forehead, after cleaning of the skin. In the case of previous anterior circulation stroke, the nonaffected hemisphere was chosen. All patients underwent clinical examination for carotid stenosis at baseline, and if positively identified, the opposite hemisphere was used. For details of data preprocessing and signal quality indices, see full Concise Methods.

Accepting the limitations of any arbitrary threshold, we chose a relative drop of 15% of the baseline value as the cutoff for potential cerebral ischemia on the basis of previous literature (see full Concise Methods for justification).

At baseline and at 12 months from study entry, all patients underwent cognitive function testing consisting of the 3MS and the TTB, and screening for depression using the PHQ-9. Cognitive function tests, including the 3MS and the TTB, have been used extensively in patients receiving dialysis.48,58,60,61 Results are not significantly different before and after dialysis, and repeatability is reasonable.62–64 The 3MS is a well validated test of global cognitive function including concentration, language, praxis, and memory. The TTB is a test of executive function, cognitive flexibility, and processing speed. The PHQ-9 is a validated screening tool for depression, which may affect performance on the 3MS and TTB.

All patients were assessed by the same individual, and in accordance with published protocols. At baseline, all patients were assessed on the dialysis unit, within the first 2 hours of treatment; at 12 months, a small number of individuals had had transplants, and they were assessed either on the dialysis unit or in the outpatient department, in a similar environment. We chose to assess during dialysis on the basis of patient feedback that any extension of time already spent in hospital would be unacceptable; this is a protocol that has been applied by other groups.49 For a small number of individuals whose visual impairment precluded completion of the standard TTB, the oral TTB was administered, and a validated conversion factor applied.65,66

Dialysis Protocol

This was a real-world study and we did not alter usual treatment procedures. Eight individuals underwent hemodiafiltration and the remainder hemodialysis, all with Dialog+ Hemodialysis Systems (BBraun). For hemodialysis, we used high-flux BBraun Diacap polysulfone dialyzers, either the PS18 High (membrane area 1.8 m2, coefficient of ultrafiltration (KUF) 55 ml/h per mmHg, dialyser mass transfer area coefficient for urea (KoA) 911 ml/min) or PS20 High (membrane area 2.0 m2, KUF 58 ml/h per mmHg, K0A 1005 ml/min) depending on clearance requirements. For hemodiafiltration, we used the Fresenius polysulfone FX100 dialyzer (membrane area 2.2 m2, KUF 73 ml/h per mmHg, K0A 1351 ml/min).

Qb, Qd, acid concentrate, and the filter itself were individualized according to patient characteristics and monthly bloods, including Kt/V. Qd was 500–800 ml/min in all cases. Qb for tunneled lines was usually 250–300 ml/min, and for arteriovenous fistulae 250–400 ml/min depending on fistula characteristics and clearance requirements. The acid concentrates in use (BBraun) could provide a dialysate sodium 135–140 mmol/L, potassium 2–4 mmol/L, ionized calcium 1.25–1.75 mmol/L, magnesium 0.5–1.0 mmol/L, acetate 2–3 mmol/L, chloride 100–116 mmol/L, and glucose 1 g/L. Bicarbonate was provided with the BBraun Sol-cart B cartridge, which produced a dialysate bicarbonate of 32–36 mmol/L. Ultrafiltration volume was determined on the basis of the difference between predialysis weight and target dry weight, which was determined at the beginning of the month on clinical grounds.

The temperature of the dialysate was generally matched to the temperature of the patient except for those with a history of repeated IDH, in which case it would be set to approximately 0.5°C lower than body temperature. If IDH persisted despite this, patients were switched to hemodiafiltration.

Therefore, patients had been optimized as per unit protocol before inclusion in the study.

Selection of BP Index for Analyses

In order to determine whether changes in downstream cerebral oxygenation were better explained by changes in MAP or changes in SBP, we compared univariate multilevel models predicting nadir in cerebral oxygenation during hemodialysis where the explanatory variable was either maximum drop in MAP or SBP during the session. We compared goodness of fit using the log likelihood ratio test statistic assessed against a chi-squared distribution.

Analysis 1: Overview of Association between MAP Thresholds and Adverse Events

The aim of this analysis was to get an overview of the frequency of BP drops, and temporally associated adverse events. We identified sustained drops in MAP (minimum duration 2 minutes) below predefined thresholds. We screened for interventions or new onset cerebral desaturation in a window from −10 to +20 minutes around the crossing of MAP threshold. We identified patient symptoms in a window spanning ±20 minutes. The rationale behind these window lengths is outlined in the full Concise Methods. New-onset cerebral ischemia was defined as a sustained (2 minutes minimum) drop in cerebral saturations below the 15% threshold. For comparison, data segments of the same length (30 minutes) where the MAP did not fall below the given threshold were used as controls.

For formal sensitivity and specificity calculation, the median MAP from a window of ±5 minutes around the time cerebral saturations crossed the 15% threshold was taken as the BP at which ischemia occurred. The MAPs at which ischemia occurred were binned into 5 mmHg categories. The cumulative occurrence of cerebral ischemia at or below each 5 mmHg threshold, and above each threshold, was calculated.

Analysis 2: Lower Limits of Cerebral Autoregulation

The lower limit of cerebral autoregulation was explored as a complementary method of assessing the relationship between BP and cerebral ischemia. The cerebral autoregulation curve is best modeled by a third-degree polynomial with positive first coefficient.67–69 In comparison, in the absence of autoregulation, regional Qb and cerebral saturations vary linearly with BP. The MAP data were grouped into 5-mmHg bins (sliding every 1 mmHg) and the mean and 95% CIs of the corresponding cerebral saturation values were found for each bin. For each patient, we assessed whether the relationship between MAP and cerebral saturations was best modeled with an appropriate polynomial or an appropriate (i.e., positive gradient) straight line: goodness of fit was compared using the F test statistic (on the basis of residual sum of squares for each model) against an F distribution. If the polynomial provided a statistically significant better fit, upper and lower autoregulation limits were estimated by drawing a horizontal straight line from the peak of the positive inflection and the nadir of the negative inflection respectively, and identifying where they intersected the ascending and descending limbs of the polynomial (see Supplemental Figure 3 in the full Concise Methods). This is the so-called curve-fitting method and is well documented.37,41

Analysis 3: Relationship between Intradialytic Physiology and Change in Cognitive Function

For each session, the area under the curve (AUC) below the cutoff of a relative 15% drop in cerebral saturations was calculated, known hereafter as the ischemia AUC (minutes×%). The average session ischemia AUC over the month of monitoring was calculated for each patient as an estimation of typical exposure to cerebral ischemia. In addition, the AUCs in mmHg×minutes below different MAP thresholds, relative and absolute, were calculated for each session (absolute thresholds were MAP <70 and <60 mmHg, and relative thresholds MAP drop >10, >20, and >30 mmHg).

Linear models were then constructed to predict the change in TTB and 3MS scores between baseline and 12-month follow-up. The changes in TTB and 3MS were expressed as a ratio of baseline and follow-up scores. Patients who had undergone transplantation during this period were excluded. Covariates tested included hypotension exposure, ischemia exposure, age, level of education, cerebrovascular disease, diabetes vintage, hemodialysis vintage, and PHQ. Variables were transformed to ensure a linear relationship between predictors and outcome, and checks were performed on the residuals to confirm that models were valid. As there were more deaths and transplants than expected, and the number of patients was relatively small, the robustness of all results was tested with boot-strapping (resampling with replacement) and jack-knifing (resampling without replacement). This method also allows for some control of type 1 errors resulting from the multiple comparisons.

Disclosures

None.

Acknowledgments

We thank patients and staff and acknowledge the input of many other members of the team who assisted with data collection, study design, code writing, and data interpretation, including Mauricio Villaroel, Joao Jorge, Alessandro Guazzi, David Meredith, Joanna Carter, Rebecca Ryan, David Clifton, Andy Mosson, Andrew Davenport, Peter Watkinson, and Duncan Young.

C.M. is funded by an National Institute of Health Research doctoral research fellowship. J.D. acknowledges the support of the Research Council United Kingdom Digital Economy Programme grant number EP/G036861/1 (Oxford Centre for Doctoral Training in Healthcare Innovation). The research group additionally receives funding from the Oxford National Institute of Health Research Oxford Biomedical Research Council.

Footnotes

References

  1. ↵
    1. Daugirdas JT
    : Pathophysiology of dialysis hypotension: An update. Am J Kidney Dis 38[Suppl 4]: S11–S17, 2001pmid:11602456
    OpenUrlPubMed
  2. ↵
    1. Tessitore N,
    2. Santoro A,
    3. Panzetta GO,
    4. Wizemann V,
    5. Perez-Garcia R,
    6. Martinez Ara J,
    7. Perrone B,
    8. Mantovani W,
    9. Poli A
    : Acetate-free biofiltration reduces intradialytic hypotension: A European multicenter randomized controlled trial. Blood Purif 34: 354–363, 2012pmid:23406818
    OpenUrlCrossRefPubMed
  3. ↵
    1. Tislér A,
    2. Akócsi K,
    3. Borbás B,
    4. Fazakas L,
    5. Ferenczi S,
    6. Görögh S,
    7. Kulcsár I,
    8. Nagy L,
    9. Sámik J,
    10. Szegedi J,
    11. Tóth E,
    12. Wágner G,
    13. Kiss I
    : The effect of frequent or occasional dialysis-associated hypotension on survival of patients on maintenance haemodialysis. Nephrol Dial Transplant 18: 2601–2605, 2003pmid:14605284
    OpenUrlCrossRefPubMed
  4. ↵
    1. Shoji T,
    2. Tsubakihara Y,
    3. Fujii M,
    4. Imai E
    : Hemodialysis-associated hypotension as an independent risk factor for two-year mortality in hemodialysis patients. Kidney Int 66: 1212–1220, 2004pmid:15327420
    OpenUrlCrossRefPubMed
  5. ↵
    1. Flythe JE,
    2. Xue H,
    3. Lynch KE,
    4. Curhan GC,
    5. Brunelli SM
    : Association of mortality risk with various definitions of intradialytic hypotension. J Am Soc Nephrol 26: 724–734, 2015pmid:25270068
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Kuipers J,
    2. Usvyat LA,
    3. Oosterhuis JK,
    4. Dasselaar JJ,
    5. de Jong PE,
    6. Westerhuis R,
    7. Sands JJ,
    8. Wang Y,
    9. Kotanko P,
    10. Franssen CFM
    : Variability of predialytic, intradialytic, and postdialytic blood pressures in the course of a week: A study of Dutch and US maintenance hemodialysis patients. Am J Kidney Dis 62: 779–788, 2013pmid:23759298
    OpenUrlCrossRefPubMed
  7. ↵
    1. McMahon N,
    2. Hogg LA,
    3. Corfield AR,
    4. Exton AD
    : Comparison of non-invasive and invasive blood pressure in aeromedical care. Anaesthesia 67: 1343–1347, 2012pmid:23033983
    OpenUrlPubMed
    1. Lakhal K,
    2. Macq C,
    3. Ehrmann S,
    4. Boulain T,
    5. Capdevila X
    : Noninvasive monitoring of blood pressure in the critically ill: Reliability according to the cuff site (arm, thigh, or ankle). Crit Care Med 40: 1207–1213, 2012pmid:22425818
    OpenUrlCrossRefPubMed
    1. Ribezzo S,
    2. Spina E,
    3. Di Bartolomeo S,
    4. Sanson G
    : Noninvasive techniques for blood pressure measurement are not a reliable alternative to direct measurement: A randomized crossover trial in ICU. Sci World J 2014: 1–8, 2014pmid:24616624
    OpenUrlPubMed
  10. ↵
    1. Bur A,
    2. Hirschl MM,
    3. Herkner H,
    4. Oschatz E,
    5. Kofler J,
    6. Woisetschläger C,
    7. Laggner AN
    : Accuracy of oscillometric blood pressure measurement according to the relation between cuff size and upper-arm circumference in critically ill patients. Crit Care Med 28: 371–376, 2000pmid:10708169
    OpenUrlCrossRefPubMed
  11. ↵
    1. Burton JO,
    2. Jefferies HJ,
    3. Selby NM,
    4. McIntyre CW
    : Hemodialysis-induced cardiac injury: Determinants and associated outcomes. Clin J Am Soc Nephrol 4: 914–920, 2009pmid:19357245
    OpenUrlAbstract/FREE Full Text
    1. Jefferies HJ,
    2. Burton JO,
    3. McIntyre CW
    : Individualised dialysate temperature improves intradialytic haemodynamics and abrogates haemodialysis-induced myocardial stunning, without compromising tolerability. Blood Purif 32: 63–68, 2011pmid:21346338
    OpenUrlCrossRefPubMed
  13. ↵
    1. Selby NM,
    2. Burton JO,
    3. Chesterton LJ,
    4. McIntyre CW
    : Dialysis-induced regional left ventricular dysfunction is ameliorated by cooling the dialysate. Clin J Am Soc Nephrol 1: 1216–1225, 2006pmid:17699351
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Leone M,
    2. Asfar P,
    3. Radermacher P,
    4. Vincent JL,
    5. Martin C
    : Optimizing mean arterial pressure in septic shock: A critical reappraisal of the literature. Crit Care 19: 101, 2015pmid:25888071
    OpenUrlCrossRefPubMed
    1. Dellinger RP,
    2. Levy MM,
    3. Rhodes A,
    4. Annane D,
    5. Gerlach H,
    6. Opal SM,
    7. Sevransky JE,
    8. Sprung CL,
    9. Douglas IS,
    10. Jaeschke R,
    11. Osborn TM,
    12. Nunnally ME,
    13. Townsend SR,
    14. Reinhart K,
    15. Kleinpell RM,
    16. Angus DC,
    17. Deutschman CS,
    18. Machado FR,
    19. Rubenfeld GD,
    20. Webb SA,
    21. Beale RJ,
    22. Vincent JL,
    23. Moreno R; Surviving Sepsis Campaign Guidelines Committee including the Pediatric Subgroup
    : Surviving sepsis campaign: International guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med 41: 580–637, 2013pmid:23353941
    OpenUrlCrossRefPubMed
    1. Pinsky MR
    : Targets for resuscitation from shock. Minerva Anestesiol 69: 237–244, 2003pmid:12766714
    OpenUrlPubMed
  17. ↵
    1. Perez P,
    2. Kimmoun A,
    3. Blime V,
    4. Levy B
    : Increasing mean arterial pressure in cardiogenic shock secondary to myocardial infarction: Effects on hemodynamics and tissue oxygenation. Shock 41: 269–274, 2014pmid:24509521
    OpenUrlCrossRefPubMed
  18. ↵
    1. De Backer D,
    2. Donadello K,
    3. Cortes DO
    : Monitoring the microcirculation. J Clin Monit Comput 26: 361–366, 2012pmid:22833180
    OpenUrlCrossRefPubMed
  19. ↵
    1. Jakob SM,
    2. Ruokonen E,
    3. Vuolteenaho O,
    4. Lampainen E,
    5. Takala J
    : Splanchnic perfusion during hemodialysis: Evidence for marginal tissue perfusion. Crit Care Med 29: 1393–1398, 2001pmid:11445693
    OpenUrlCrossRefPubMed
  20. ↵
    1. Schober P,
    2. Schwarte LA
    : From system to organ to cell: Oxygenation and perfusion measurement in anesthesia and critical care. J Clin Monit Comput 26: 255–265, 2012pmid:22437884
    OpenUrlCrossRefPubMed
  21. ↵
    1. Bakker A,
    2. Smith B,
    3. Ainslie P,
    4. Smith K
    : Near-Infrared Spectroscopy, 2014. http://wwwintechopencom/books/applied-aspects-of-ultrasonography-inhumans/near-infrared-spectroscopy. Accessed June 12, 2015
  22. ↵
    1. Brady K,
    2. Joshi B,
    3. Zweifel C,
    4. Smielewski P,
    5. Czosnyka M,
    6. Easley RB,
    7. Hogue CW Jr.
    : Real-time continuous monitoring of cerebral blood flow autoregulation using near-infrared spectroscopy in patients undergoing cardiopulmonary bypass. Stroke 41: 1951–1956, 2010pmid:20651274
    OpenUrlAbstract/FREE Full Text
    1. Alderliesten T,
    2. De Vis JB,
    3. Lemmers PMA,
    4. van Bel F,
    5. Benders MJNL,
    6. Hendrikse J,
    7. Petersen ET
    : Simultaneous quantitative assessment of cerebral physiology using respiratory-calibrated MRI and near-infrared spectroscopy in healthy adults. Neuroimage 85: 255–263, 2014pmid:23859925
    OpenUrlCrossRefPubMed
  24. ↵
    1. Ohmae E,
    2. Ouchi Y,
    3. Oda M,
    4. Suzuki T,
    5. Nobesawa S,
    6. Kanno T,
    7. Yoshikawa E,
    8. Futatsubashi M,
    9. Ueda Y,
    10. Okada H,
    11. Yamashita Y
    : Cerebral hemodynamics evaluation by near-infrared time-resolved spectroscopy: Correlation with simultaneous positron emission tomography measurements. Neuroimage 29: 697–705, 2006pmid:16165372
    OpenUrlCrossRefPubMed
  25. ↵
    Slater JP, Guarino T, Stack J, Vinod K, Bustami RT, Brown 3rd JM, Rodriguez AL, Magovern CJ, Zaubler T, Freundlich K, Parr GV: Cerebral oxygen desaturation predicts cognitive decline and longer hospital stay after cardiac surgery. Ann Thorac Surg 87: 36–44; discussion 44–35, 2009
    1. Tang L,
    2. Kazan R,
    3. Taddei R,
    4. Zaouter C,
    5. Cyr S,
    6. Hemmerling TM
    : Reduced cerebral oxygen saturation during thoracic surgery predicts early postoperative cognitive dysfunction. Br J Anaesth 108: 623–629, 2012pmid:22311364
    OpenUrlCrossRefPubMed
    1. de Tournay-Jetté E,
    2. Dupuis G,
    3. Bherer L,
    4. Deschamps A,
    5. Cartier R,
    6. Denault A
    : The relationship between cerebral oxygen saturation changes and postoperative cognitive dysfunction in elderly patients after coronary artery bypass graft surgery. J Cardiothorac Vasc Anesth 25: 95–104, 2011pmid:20650659
    OpenUrlCrossRefPubMed
    1. Suehiro K,
    2. Okutai R
    : Duration of cerebral desaturation time during single-lung ventilation correlates with mini mental state examination score. J Anesth 25: 345–349, 2011pmid:21484501
    OpenUrlCrossRefPubMed
  29. ↵
    1. Murkin JM,
    2. Adams SJ,
    3. Novick RJ,
    4. Quantz M,
    5. Bainbridge D,
    6. Iglesias I,
    7. Cleland A,
    8. Schaefer B,
    9. Irwin B,
    10. Fox S
    : Monitoring brain oxygen saturation during coronary bypass surgery: A randomized, prospective study. Anesth Analg 104: 51–58, 2007pmid:17179242
    OpenUrlCrossRefPubMed
    1. Kazan R,
    2. Bracco D,
    3. Hemmerling TM
    : Reduced cerebral oxygen saturation measured by absolute cerebral oximetry during thoracic surgery correlates with postoperative complications. Br J Anaesth 103: 811–816, 2009pmid:19918024
    OpenUrlCrossRefPubMed
  31. ↵
    1. Ono M,
    2. Arnaoutakis GJ,
    3. Fine DM,
    4. Brady K,
    5. Easley RB,
    6. Zheng Y,
    7. Brown C,
    8. Katz NM,
    9. Grams ME,
    10. Hogue CW
    : Blood pressure excursions below the cerebral autoregulation threshold during cardiac surgery are associated with acute kidney injury. Crit Care Med 41: 464–471, 2013pmid:23263580
    OpenUrlCrossRefPubMed
  32. ↵
    1. Ono M,
    2. Brady K,
    3. Easley RB,
    4. Brown C,
    5. Kraut M,
    6. Gottesman RF,
    7. Hogue CW Jr.
    : Duration and magnitude of blood pressure below cerebral autoregulation threshold during cardiopulmonary bypass is associated with major morbidity and operative mortality. J Thorac Cardiovasc Surg 147: 483–489, 2014pmid:24075467
    OpenUrlCrossRefPubMed
  33. ↵
    1. Smith M
    : Shedding light on the adult brain: A review of the clinical applications of nearinfrared spectroscopy. Philos Trans A Math Phys Eng Sci 369: 4452–4469, 2011pmid:22006901
    OpenUrlCrossRefPubMed
  34. ↵
    1. Hadolt I,
    2. Litscher G
    : Noninvasive assessment of cerebral oxygenation during high altitude trekking in the Nepal Himalayas (2850-5600 m). Neurol Res 25: 183–188, 2003pmid:12635520
    OpenUrlCrossRefPubMed
  35. ↵
    1. Hoshino T,
    2. Ookawara S,
    3. Goto S,
    4. Miyazawa H,
    5. Ito K,
    6. Ueda Y,
    7. Kaku Y,
    8. Hirai K,
    9. Nabata A,
    10. Mori H,
    11. Yoshida I,
    12. Tabei K
    : Evaluation of cerebral oxygenation in patients undergoing long-term hemodialysis. Nephron Clin Pract 126: 57–61, 2014pmid:24526002
    OpenUrlPubMed
  36. ↵
    1. Kashiwazaki D,
    2. Kuroda S,
    3. Terasaka S,
    4. Iwasaki Y
    : Detection of hemodynamic transient ischemic attack during hemodialysis with near-infrared monitoring in a patient with internal carotid artery occlusion. Surg Neurol 68: 292–294, discussion 294–295, 2007pmid:17719970
    OpenUrlCrossRefPubMed
  37. ↵
    1. Paulson OB,
    2. Strandgaard S,
    3. Edvinsson L
    : Cerebral autoregulation. Cerebrovasc Brain Metab Rev 2: 161–192, 1990pmid:2201348
    OpenUrlCrossRefPubMed
    1. Ono M,
    2. Joshi B,
    3. Brady K,
    4. Easley RB,
    5. Zheng Y,
    6. Brown C,
    7. Baumgartner W,
    8. Hogue CW
    : Risks for impaired cerebral autoregulation during cardiopulmonary bypass and postoperative stroke. Br J Anaesth 109: 391–398, 2012pmid:22661748
    OpenUrlCrossRefPubMed
  39. ↵
    1. Drummond JC
    : The lower limit of autoregulation: Time to revise our thinking? Anesthesiology 86: 1431–1433, 1997pmid:9197320
    OpenUrlCrossRefPubMed
  40. ↵
    1. Moyer JH,
    2. Morris G
    : Cerebral hemodynamics during controlled hypotension induced by the continuous infusion of ganglionic blocking agents (hexamethonium, pendiomide and arfonad). J Clin Invest 33: 1081–1088, 1954pmid:13183992
    OpenUrlCrossRefPubMed
  41. ↵
    1. Finnerty FA Jr.,
    2. Witkin L,
    3. Fazekas JF
    : Cerebral hemodynamics during cerebral ischemia induced by acute hypotension. J Clin Invest 33: 1227–1232, 1954pmid:13192186
    OpenUrlCrossRefPubMed
    1. Strandgaard S,
    2. Olesen J,
    3. Skinhoj E,
    4. Lassen NA
    : Autoregulation of brain circulation in severe arterial hypertension. BMJ 1: 507–510, 1973pmid:4692673
    OpenUrlAbstract/FREE Full Text
    1. Olsen KS,
    2. Svendsen LB,
    3. Larsen FS,
    4. Paulson OB
    : Effect of labetalol on cerebral blood flow, oxygen metabolism and autoregulation in healthy humans. Br J Anaesth 75: 51–54, 1995pmid:7669469
    OpenUrlCrossRefPubMed
    1. Olsen KS,
    2. Svendsen LB,
    3. Larsen FS
    : Validation of transcranial near-infrared spectroscopy for evaluation of cerebral blood flow autoregulation. J Neurosurg Anesthesiol 8: 280–285, 1996pmid:8884624
    OpenUrlPubMed
    1. Larsen FS,
    2. Olsen KS,
    3. Hansen BA,
    4. Paulson OB,
    5. Knudsen GM
    : Transcranial Doppler is valid for determination of the lower limit of cerebral blood flow autoregulation. Stroke 25: 1985–1988, 1994pmid:7916502
    OpenUrlAbstract/FREE Full Text
  46. ↵
    1. Joshi B,
    2. Ono M,
    3. Brown C,
    4. Brady K,
    5. Easley RB,
    6. Yenokyan G,
    7. Gottesman RF,
    8. Hogue CW
    : Predicting the limits of cerebral autoregulation during cardiopulmonary bypass. Anesth Analg 114: 503–510, 2012pmid:22104067
    OpenUrlCrossRefPubMed
  47. ↵
    1. Tozawa M,
    2. Iseki K,
    3. Iseki C,
    4. Oshiro S,
    5. Yamazato M,
    6. Higashiuesato Y,
    7. Tomiyama N,
    8. Tana T,
    9. Ikemiya Y,
    10. Takishita S
    : Evidence for elevated pulse pressure in patients on chronic hemodialysis: A case-control study. Kidney Int 62: 2195–2201, 2002pmid:12427145
    OpenUrlCrossRefPubMed
  48. ↵
    1. Pereira AA,
    2. Weiner DE,
    3. Scott T,
    4. Chandra P,
    5. Bluestein R,
    6. Griffith J,
    7. Sarnak MJ
    : Subcortical cognitive impairment in dialysis patients. Hemodial Int 11: 309–314, 2007pmid:17576295
    OpenUrlCrossRefPubMed
  49. ↵
    1. Weiner DE,
    2. Scott TM,
    3. Giang LM,
    4. Agganis BT,
    5. Sorensen EP,
    6. Tighiouart H,
    7. Sarnak MJ
    : Cardiovascular disease and cognitive function in maintenance hemodialysis patients. Am J Kidney Dis 58: 773–781, 2011pmid:21778003
    OpenUrlCrossRefPubMed
  50. ↵
    1. Hsieh T-J,
    2. Chang J-M,
    3. Chuang H-Y,
    4. Ko C-H,
    5. Hsieh M-L,
    6. Liu G-C,
    7. Hsu J-S
    : End-stage renal disease: In vivo diffusion-tensor imaging of silent white matter damage. Radiology 252: 518–525, 2009pmid:19528357
    OpenUrlCrossRefPubMed
    1. Zhang R,
    2. Liu K,
    3. Yang L,
    4. Zhou T,
    5. Qian S,
    6. Li B,
    7. Peng Z,
    8. Li M,
    9. Sang S,
    10. Jiang Q,
    11. Sun G
    : Reduced white matter integrity and cognitive deficits in maintenance hemodialysis ESRD patients: A diffusion-tensor study. Eur Radiol 25: 661–668, 2015pmid:25326436
    OpenUrlCrossRefPubMed
    1. Chou M-C,
    2. Hsieh T-J,
    3. Lin Y-L,
    4. Hsieh Y-T,
    5. Li W-Z,
    6. Chang J-M,
    7. Ko C-H,
    8. Kao E-F,
    9. Jaw T-S,
    10. Liu G-C
    : Widespread white matter alterations in patients with end-stage renal disease: A voxelwise diffusion tensor imaging study. AJNR Am J Neuroradiol 34: 1945–1951, 2013pmid:23788598
    OpenUrlAbstract/FREE Full Text
  53. ↵
    1. Eldehni MT,
    2. Odudu A,
    3. McIntyre CW
    : Randomized clinical trial of dialysate cooling and effects on brain white matter. J Am Soc Nephrol 26: 957–965, 2015pmid:25234925
    OpenUrlAbstract/FREE Full Text
  54. ↵
    1. Slinin Y,
    2. Paudel ML,
    3. Ishani A,
    4. Taylor BC,
    5. Yaffe K,
    6. Murray AM,
    7. Fink HA,
    8. Orwoll ES,
    9. Cummings SR,
    10. Barrett-Connor E,
    11. Jassal S,
    12. Ensrud KE; Osteoporotic Fractures in Men Study Group
    : Kidney function and cognitive performance and decline in older men. J Am Geriatr Soc 56: 2082–2088, 2008pmid:18795984
    OpenUrlCrossRefPubMed
    1. Kurella M,
    2. Chertow GM,
    3. Luan J,
    4. Yaffe K
    : Cognitive impairment in chronic kidney disease. J Am Geriatr Soc 52: 1863–1869, 2004pmid:15507063
    OpenUrlCrossRefPubMed
  56. ↵
    1. Weiner DE,
    2. Seliger SL
    : Cognitive and physical function in chronic kidney disease. Curr Opin Nephrol Hypertens 23: 291–297, 2014pmid:24638060
    OpenUrlCrossRefPubMed
  57. ↵
    1. Wolfgram DF,
    2. Sunio L,
    3. Vogt E,
    4. Smith HM,
    5. Visotcky A,
    6. Laud P,
    7. Whittle J
    : Haemodynamics during dialysis and cognitive performance. Nephrology (Carlton) 19: 771–776, 2014pmid:25103846
    OpenUrlPubMed
  58. ↵
    1. Kurella Tamura M,
    2. Larive B,
    3. Unruh ML,
    4. Stokes JB,
    5. Nissenson A,
    6. Mehta RL,
    7. Chertow GM; Frequent Hemodialysis Network Trial Group
    : Prevalence and correlates of cognitive impairment in hemodialysis patients: The frequent hemodialysis network trials. Clin J Am Soc Nephrol 5: 1429–1438, 2010pmid:20576825
    OpenUrlAbstract/FREE Full Text
  59. ↵
    1. Lehmann C,
    2. Cerny V,
    3. Abdo I,
    4. Kern H,
    5. Sander M; Microcirculation Diagnostics and Applied Studies (MiDAS) Investigators
    : Microcirculation diagnostics and applied studies in circulatory shock - research from the bench to the bedside. Clin Hemorheol Microcirc 52: 131–139, 2012pmid:22975933
    OpenUrlPubMed
  60. ↵
    1. Sarnak MJ,
    2. Tighiouart H,
    3. Scott TM,
    4. Lou KV,
    5. Sorensen EP,
    6. Giang LM,
    7. Drew DA,
    8. Shaffi K,
    9. Strom JA,
    10. Singh AK,
    11. Weiner DE
    : Frequency of and risk factors for poor cognitive performance in hemodialysis patients. Neurology 80: 471–480, 2013pmid:23303848
    OpenUrlCrossRefPubMed
  61. ↵
    1. Rodriguez L,
    2. Tighiouart H,
    3. Scott T,
    4. Lou K,
    5. Giang L,
    6. Sorensen E,
    7. Weiner DE,
    8. Sarnak MJ
    : Association of sleep disturbances with cognitive impairment and depression in maintenance hemodialysis patients. J Nephrol 26: 101–110, 2013pmid:22641565
    OpenUrlCrossRefPubMed
  62. ↵
    1. Dahbour SS,
    2. Wahbeh AM,
    3. Hamdan MZ
    : Mini mental status examination (MMSE) in stable chronic renal failure patients on hemodialysis: The effects of hemodialysis on the MMSE score. A prospective study. Hemodial Int 13: 80–85, 2009pmid:19210282
    OpenUrlCrossRefPubMed
    1. Tombaugh TN
    : Test-retest reliable coefficients and 5-year change scores for the MMSE and 3MS. Arch Clin Neuropsychol 20: 485–503, 2005pmid:15896562
    OpenUrlCrossRefPubMed
  64. ↵
    1. Amodio P,
    2. Wenin H,
    3. Del Piccolo F,
    4. Mapelli D,
    5. Montagnese S,
    6. Pellegrini A,
    7. Musto C,
    8. Gatta A,
    9. Umiltà C
    : Variability of trail making test, symbol digit test and line trait test in normal people. A normative study taking into account age-dependent decline and sociobiological variables. Aging Clin Exp Res 14: 117–131, 2002pmid:12092785
    OpenUrlPubMed
  65. ↵
    1. Ricker JH,
    2. Axelrod BN
    : Analysis of an oral paradigm for the trail making test. Assessment 1: 47–52, 1994pmid:9463499
    OpenUrlCrossRefPubMed
  66. ↵
    1. Abraham E,
    2. Axelrod B,
    3. Ricker J
    : Application of the oral trail making test to a mixed clinical sample. Arch Clin Neuropsychol 11(8): 697–701, 1996
    OpenUrlCrossRef
  67. ↵
    1. Hernández MJ,
    2. Brennan RW,
    3. Bowman GS
    : Cerebral blood flow autoregulation in the rat. Stroke 9: 150–154, 1978pmid:644608
    OpenUrlAbstract/FREE Full Text
    1. Sato M,
    2. Pawlik G,
    3. Heiss WD
    : Comparative studies of regional CNS blood flow autoregulation and responses to CO2 in the cat. Effects of altering arterial blood pressure and PaCO2 on rCBF of cerebrum, cerebellum, and spinal cord. Stroke 15: 91–97, 1984pmid:6420948
    OpenUrlAbstract/FREE Full Text
  69. ↵
    1. Dirnagl U,
    2. Pulsinelli W
    : Autoregulation of cerebral blood flow in experimental focal brain ischemia. J Cereb Blood Flow Metab 10: 327–336, 1990pmid:2329121
    OpenUrlCrossRefPubMed
Back to top

In this issue

View Selected Citations (0)
Print
Download PDF
Email Article
Citation Tools
Request Permissions
Relationship between Hypotension and Cerebral Ischemia during Hemodialysis
Clare MacEwen, Sheera Sutherland, Jonathan Daly, Christopher Pugh, Lionel Tarassenko
JASN Aug 2017, 28 (8) 2511-2520; DOI: 10.1681/ASN.2016060704
del.icio.us logo Digg logo Reddit logo Twitter logo Facebook logo Google logo Mendeley logo

Jump to section

Keywords