- Email: [email protected]
Microcirculatory alterations during continuous renal replacement therapy in ICU: A novel view on the ‘dialysis trauma’ concept
Microcirculatory alterations during continuous renal replacement therapy in ICU: A novel view on the ‘dialysis trauma’ concept Chrysoula Pipili, Ioannis Vasileiadis, Eirini Grapsa, Elli-Sophia Tripodaki, Sophia Ioannidou, Adroula Papastylianou, Stelios Kokkoris, Christina Routsi, Marianna Politou, Serafeim Nanas PII: DOI: Reference:
S0026-2862(15)30026-1 doi: 10.1016/j.mvr.2015.09.004 YMVRE 3573
To appear in:
Microvascular Research
Received date: Revised date: Accepted date:
12 February 2015 27 September 2015 28 September 2015
Please cite this article as: Pipili, Chrysoula, Vasileiadis, Ioannis, Grapsa, Eirini, Tripodaki, Elli-Sophia, Ioannidou, Sophia, Papastylianou, Adroula, Kokkoris, Stelios, Routsi, Christina, Politou, Marianna, Nanas, Serafeim, Microcirculatory alterations during continuous renal replacement therapy in ICU: A novel view on the ‘dialysis trauma’ concept, Microvascular Research (2015), doi: 10.1016/j.mvr.2015.09.004
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Title: Microcirculatory alterations during continuous renal replacement therapy
PT
in ICU: a novel view on the ‘dialysis trauma’ concept Chrysoula Pipili1, Ioannis Vasileiadis1, Eirini Grapsa1, Elli-Sophia Tripodaki1, Sophia
RI
Ioannidou2, Adroula Papastylianou1, Stelios Kokkoris1, Christina Routsi1, Marianna
First Critical Care Department, ‘Evangelismos’ General Hospital, National and
NU
1
SC
Politou3, Serafeim Nanas1
Kapodistrian University of Athens
3
Laboratory of Biochemistry, ‘Evangelismos’ Hospital, Athens, Greece
MA
2
Blood Transfusion Department, Aretaieion Hospital, Athens University Medical
AC CE P
Prof. Serafeim Nanas
TE
Corresponding Author:
D
School, Athens, Greece
First Critical Care Department, National and Kapodistrian University of Athens Evangelismos Hospital
Ypsilantou 45-47, 106 75, Athens, Greece Email: [email protected] Mobile: +306973036448
Fax: +302132043385
ACCEPTED MANUSCRIPT ABSTRACT
renal replacement therapy (CRRT) in critically ill patients.
PT
Objective: The purpose of this study was to evaluate microcirculation over 24hours
RI
Methods: We conducted a single-center, prospective, observational study, measuring
SC
microcirculation parameters, monitored by near infrared spectroscopy (NIRS) before hemodiafiltration onset (H0), and at six (H6) and 24 hours (H24) during CRRT in
NU
critically ill patients. Serum Cystatin C (sCysC) and soluble (s)E-selectin levels were
MA
measured at the same time points. Twenty-eight patients [19 men (68%)] were included in the study.
D
Results: Tissue oxygen saturation (StO2, %) [76.5±12.5 (H0) vs 75±11 (H6) vs 70±16
TE
(H24), p=0.04], reperfusion rate, indicating endothelial function (EF, %/sec)
AC CE P
[2.25±1.44 (H0) vs 2.1±1.8 (H6) vs 1.6 ±1.4 (H24), p=0.02] and sCysC (mg/L) [2.7±0.8 (H0) vs 2.2±0.6 (H6) vs 1.8±0.8 (H24), p<0.0001] significantly decreased within the 24hours CRRT. Change of EF positively correlated with changes of sCysC within 24hours CRRT (r= 0.464, p=0.013) while in patients with diabetes the change of StO2 correlated with dose (r=-0.8, p=0.01). No correlation existed between hemoglobin and temperature changes with the deteriorated microcirculation indices. sE-Selectin levels in serum were elavated; no difference was established over the 24h CRRT period. A strong correlation existed between the sE-Selectin concentration change at H6 and H24 and the mean arterial pressure change in the same period (r=0.77, p<0.001).
Conclusions: During the first 24 hours of CRRT implementation in critically ill patients, deterioration of microcirculation parameters was noted. Microcirculatory
ACCEPTED MANUSCRIPT alterations correlated with sCysC changes and with dose in patients with diabetes. Keywords:
Dialytrauma,
Near
infrared
spectroscopy,
AC CE P
TE
D
MA
NU
SC
RI
PT
Continuous renal replacement therapy, Cystatin C
Tissue
oxygenation,
ACCEPTED MANUSCRIPT
Abbreviations
PT
CRRT: continuous renal replacement therapy
SC
CVVHDF: continuous veno-venous dia-hemofiltration
RI
NIRS: near infrared spectroscopy
NU
ICU: intensive care unit
RIFLE: risk-injury-failure-loss-end stage renal disease
MA
StO2: tissue oxygen saturation
D
Hb: hemoglobin
TE
OCR: oxygen consumption rate
AC CE P
EF: endothelial function
sCysC: serum cystatin C sCr: serum creatinine Lac: lactate
APACHE: acute physiology and chronic health evaluation SOFA: sequential organ failure assessment sE-Selectin: soluble E-Selectin
ACCEPTED MANUSCRIPT INTRODUCTION Continuous renal replacement therapy (CRRT) has proved to be a competent
PT
therapeutic tool for patients in Intensive Care Unit (ICU), increasing significantly
RI
their quality of care. However, optimal monitoring for CRRT has not yet been
SC
determined and the mortality has not decreased notably (Clec'h et al., 2012). Although CRRT implementation is crucial to reverse situations posing an immediate danger to
NU
survival (as severe hyperkalemia and volume overload with impending pulmonary edema), adverse effects –a related morbidity– can also occur (Maynar Moliner et al.,
MA
2012 and Selby & McIntyre, 2012). During CRRT, the function of the failed nephron has been replaced by inflexible, largely, prescription orders, regarding blood flow
D
rate, type of replacement or dialysis fluids and the dose (effluent flow rate).
TE
Adjustments of the prescribed treatment are made, however, not in the constant basis
AC CE P
that the changing metabolic requirements would demand, so as to adapt the microenvironment and achieve a fine tuning. Moreover, the physician may not necessarily know or control all the affected variables. Thus, CRRT may prove inadequate or act as an exogenous factor that overcomes the natural resistance of the tissues and organs and causes ‘trauma’ (Maynar Moliner et al., 2012 and Selby & McIntyre, 2012). CRRT has a direct impact on microcirculation where the removal of waste products and supply of essential nutrients take place. Studies for predicting and managing complications of RRT in the critically ill recommend the assessment of microcirculatory function to estimate the need for evolved therapies (Creteur et al., 2007, Ruiz et al., 2010 and Veenstra et al., 2014).
ACCEPTED MANUSCRIPT Technological progress enables assessment of skeletal microcirculation, which exhibits higher prognostic capability than broad indices of hemodynamic, ischemic
PT
and septic alterations in critically ill patients (Creteur et al., 2007 and Nanas et al., 2009). Near infrared spectroscopy (NIRS) methodology can reliably estimate
RI
microcirculation in this patient group. Furthermore, biomarkers of endothelial
SC
activation and/or damage, such as soluble (s)E-Selectin levels in serum, (Xing et al., 2012 and Zonneveld et al., 2014), could provide additional information. Although
NU
microcirculatory abnormalities have been involved in the pathophysiology of acute
MA
kidney injury (AKI), few data exist regarding their monitoring during CRRT in critically ill patients. Serum Cystatin C (sCysC) has been used as a biomarker for AKI development, tested for early diagnosis and monitoring of disease progression
TE
D
(Kokkoris et al., 2013 and Nejat et al., 2010). We hypothesized that assessment of microcirculation together with an endothelial and
AC CE P
a renal marker over 24hours continuous hemodiafiltration would result to a more complete evaluation of the CRRT process. We assessed, accordingly, the impact of CRRT on skeletal microcirculation utilizing NIRS technology in parallel with endothelial and renal marker sampling during CRRT, at predetermined time points. Assessment of microcirculation by the bedside, alone or in combination with renal markers, may prove useful in monitoring disease progression and guiding targeted intervention during CRRT.
ACCEPTED MANUSCRIPT
PT
MATERIAL AND METHODS
RI
Study design
This is a pilot, prospective, observational study in which renal, perfusion and
SC
microcirculation parameters were monitored during CRRT. The study was performed
NU
at the two ICUs of a tertiary hospital, comprising 30 beds, from October 2009 to July 2010. The procedures followed were in accord with the Declaration of Helsinki and
MA
its revisions, and written informed consent was obtained from patients’ relatives since all patients were sedated, under mechanical ventilation. All consecutive patients who
D
met both, the inclusion criteria for CRRT initiation, according to the attending
TE
intensivists, and the criteria for NIRS implementation, and completed, also, 24 hours
AC CE P
CRRT were included in the study. More specifically, patients without the exclusion criteria listed below could be enrolled: 1) Exclusion criteria before the CRRT onset: age ≤18 years, BMI ≥ 35kg/m2, chronic renal failure under hemodialysis, pregnancy, brain death, metastatic cancer, ICU re-admission, missing baseline sCr values 2) Exclusion criteria after the CRRT onset: To avoid a major ‘down-time’ effect ( the time spent off filter) and also enrolling patients with major blood loss, both of which could affect the results of our study, we included patients in whom CRRT was interrupted only to change a clotted filter, i.e. less than one hour (CRRT achieved > 23 h). More prolonged CRRT interruption and/or repeated filter change disqualified a patient from participation in the study. 3) Exclusion criteria for NIRS implementation: musculoskeletal and connective tissue diseases, upper limb fractures. The treating physician was aware that the subject was participating in the study. However, there was no special intervention performed that could theoretically impact
ACCEPTED MANUSCRIPT the study results. At enrollment, demographic data, acute physiology and chronic health evaluation (APACHE II) (Knaus et al., 1985) and the sequential organ failure
PT
assessment (SOFA) scores (Vincent et al., 1996) were recorded. AKI was defined using both the serum creatinine (sCr) and urine output criteria of RIFLE classification
RI
(Bellomo et al., 2004). Diagnosis of sepsis was made by the ICU team according to
SC
American College of Chest Physicians/Society of Critical Care Medicine Consensus
NU
Conference (Bone et al., 2009). The CRRT dose used was the prescribed dose.
MA
Endothelial and Renal marker sampling
Data collection included sE-Selectin, sCysC and sCr before CRRT initiation (H0), at six hours (H6) and at 24 hours (H24) during CRRT. All blood samples were taken
TE
D
through a central venous catheter (not the same used for dialysis). sCr levels were quantified by a modified Jaffe method with protein precipitation using an alkaline
AC CE P
picrate reaction while sCysC by using a MNII nephelometer (Dade Behring GmbH, Marburg, Germany). sE-selectin was accurately measured by enzyme-linked immunosorbent assay (ELISA) purchased from R&D Systems (R&D Systems Inc., Minneapolis, MN, USA).
Description of Hemodiafiltration technique All patients received continuous veno-venous hemodiafiltration (CVVHDF) performed by pump-driven machines. The CRRT machines used in the study were of two types: (1) Prisma Gambro, Lakewood, Co. and (2) Kimal, Nephrotech Co., Ltd., both with a 0.9 m2 polysulfone filter. Blood flow rate was maintained between 100 and 130 ml/min and predilution or postdilution replacement fluid was administered.
ACCEPTED MANUSCRIPT Sterile Bicarbonate solution (ACCUSOL Dialysis Solution 2.5L, 35 Bicarbonate, K4, Baxter International Inc.) was used as the dialysate and the replacement fluid for
PT
CVVHDF. Replacement flow rate was set at 1-2 L/h and dialysate flow rate at 1-1.3 L/h. The ultrafiltrate was adjusted to achieve fluid balance for most patients. Heparin
RI
was administered unless a severe bleeding risk existed. Then, predilution mode and
SC
periodic flushing with saline were administered to prevent filter clotting. In case of heparin induced thrombocytopenia, fondaparinux was utilized with close monitoring
NU
of the levels of anti-Xa antibodies to assess anticoagulant activity. Clotting of the
the first 24 hours of treatment.
MA
extracorporeal system was the most common reason of CVVHDF interruption during
TE
D
Microcirculatory assessment and analysis
NIRS constitutes an easily applicable, non-invasive monitoring system that can assess
AC CE P
skeletal muscle microcirculation in health and disease (Myers et al., 2006, Nanas et al., 2009 and Siafaka et al., 2007). The principles of NIRS utilization, along with the vascular occlusion technique, have been described elsewhere (Gerovasili et al., 2009, Manetos et al., 2011 and Tripodaki et al., 2012). In the present study, thenar muscle StO2 was measured non-invasively, by a second generation NIRS device (Inspectra Model 325, Hutchinson Technology, USA). Initially, a sphygmomanometer cuff was placed around the patient's arm and a transdermal light probe was stuck on the patient’s homolateral thenar. Baseline arterial pressure and baseline StO2 were recorded after a three minute stabilization period. Subsequently, ischemia was introduced as the pneumatic cuff was rapidly inflated to 50 mmHg above the patient’s systolic blood pressure. The vascular blockade (mostly of the brachial artery) lasted three minutes, during which, the fall of
ACCEPTED MANUSCRIPT thenar StO2 was recorded (ischemia phase). Then, the cuff was rapidly deflated, and the rise of the thenar StO2 (reperfusion phase) was recorded as well. The monitoring
PT
continued until the return of StO2 at its baseline value (hyperemia phase). The vascular occlusion derived curves were stored using InSpectra software.
RI
StO2 curves were analyzed offline, blindly and in random order (InSpectra Analysis
SC
Program, version 2.0; Hutchinson Technology; Hutchinson, MN; running in MatLab 7.0; The MathWorks; Novi, MI). The first degree slope of the hemoglobin (Hb)
NU
desaturation curve, during stagnant limb ischemia, reflects the tissue oxygen
MA
consumption rate (OCR, %/min) and the slope of the increase of StO2, after the release of the vascular occlusion, is indicative of the endothelial function (EF,%/sec). NIRS measurements were performed at H0, H6 and H24. Differences in StO2
TE
D
(∆StO2), OCR (∆OCR) and endothelial function (∆EF) values were calculated before
AC CE P
CVVHDF initiation and after the 24hours monitoring period.
STATISTICAL ANALYSIS Analysis of variance (repeated measures) and subsequent post hoc test were used to establish differences in biomarker values and microcirculatory, hemodynamic and perfusion parameters. The associations between variables were assessed using the Pearson or Spearman correlation coefficients. Parametric and non-parametric tests were used to check for significant differences between demographic and baseline characteristics of study groups. The analyses were performed using statistical software SPSS version 17 (SPSS, Chicago IL). Two sided p-values of less than 0.05 were required to establish statistical significance. Results were expressed as frequencies, mean±SD and medians with interquartile range.
ACCEPTED MANUSCRIPT
PT
RESULTS
RI
Study population
57 critically ill patients were eligible for the study but only 28 received CVVHDF for
SC
24 hours without interruption and met the NIRS inclusion criteria. All patients
NU
included in the study were mechanically ventilated. Patient characteristics are shown in Table 1. 16/28 (57%) of patients were treated with CVVHDF on postdilution
MA
mode. CRRT was initiated at 8.5±2 days of ICU hospitalization, when the patients fulfilled at least one of the following criteria (Gibney et al., 2008): oliguria (urine
D
output<200ml/12h) despite fluid resuscitation and intravenous diuretic treatment: 4
TE
patients (14%), anuria (urine output<50ml/12h): 6 patients (21.5%), volume overload:
AC CE P
1 patient (3.5%), azotemia (blood urea nitrogen 80>mg/dL): 7 patients (25%), hypernatremia: 3 patients (11%), sepsis: 3 patients (11%), hyperkalemia (potassium ≥7.5mmol/L): 1 patient (3.5%), metabolic acidosis (pH less than 7.10) with oliguria: 2 patients (7%) and hypernatremia with oliguria: 1 patient (3.5%). The mean prescribed CVVHDF dose was 23±7 ml/kg/h. During the monitoring period, all patients remained euvolemic, mean arterial pressure did not change significantly and blood transfusions were not performed. Major circuit ‘down-time’ effect was avoided by immediate replacement of the clotted filters. All patients required heparin for anticoagulation except one patient that required fondaparinux and three patients needed saline flashes. No differences were noted concerning the filter lifetimes in each anti-coagulation group.
ACCEPTED MANUSCRIPT Microcirculatory, endothelial and renal marker measurements during 24-hour RRT
PT
StO2 and EF significantly decreased within 24hours CRRT (Table 2). Also, sCysC,
RI
sCr, Hb, lactate and peripheral temperature showed significant differentiation during
SC
the monitoring period. sCysC decreased at H6 (H0-H6, p<0.0001) and H24 (H0-H24, p<0.0001) as sCr, Hb and peripheral temperature did, while lactate levels increased at
EF
positively
correlated
with
NU
H6 (H0-H6, p=0.019) and H24 (H0-H24, p=0.017) (Table 2). In parallel, change of changes
of
sCysC
within
24hours
CRRT
MA
(r= 0.464, p=0.013) and ∆OCR negatively correlated with lactate changes within the same 24hours period (r= -0.44, p=0.036). In the subgroup of patients with diabetes
D
CRRT dose correlated with ∆StO2 (r=-0.8, p=0.01). Statistical analysis showed no
TE
influence in the microcirculation parameters when pre- or post -dilution method were
AC CE P
used. Hb drop was most probably due to clotting; no major hemorrhagic episodes were noticed.
The patients with low Hb (≤7 g/dL) or Hb fall more than 1.5 g/dL over the monitoring 24hours and the patients with low temperature (≤ 35.5 oC after six hours of CRRT) did not present worse microcirculatory parameters when compared with patients with Hb >7 g/dL or temperature >35.5 oC respectively. In addition, in all subgroups (low/high Hb and low/high temperature) StO2 and EF significantly deteriorated during the monitoring period (Tables 3,4). No correlation existed between temperature and Hb changes with the deteriorated microcirculation indices. sE-Selectin levels were obtained in the serum of 23 subjects in total. 19 patients had sE-Selectin measurements at H6 and H24 (55.85±25.79 vs 56.82±25.84 ng/ml respectively, p=0.827) and three had sE-Selectin values at H0, H6 and H24
ACCEPTED MANUSCRIPT (62.27±34.64 vs 53.12±19.8 vs 63.72±10.09 ng/ml respectively, p=0.86). As shown, no difference in sE-Selectin levels was evidenced during the CRRT process. Also, no
PT
correlation was found between the sE-Selectin levels and the microcirculation indices. Interestingly, a strong correlation was noted between the change of sE-Selectin levels
SC
RI
at H6 and H24 with the change of MAP at the same time period (r=0.77, p<0.001).
NU
Discussion
In this pilot study, we assessed skeletal muscle microcirculation and a renal marker to
MA
track the course of continuous hemodiafiltration in critically ill patients. We found that during the first 24 hours of CRRT implementation, microcirculatory parameters
D
deteriorated, sCysC decreased and changes of EF correlated positively with changes
AC CE P
correlated with dose.
TE
of sCysC. Moreover, in patients with diabetes the differences in microcirculation
There is little information concerning the potential impact of CVVHDF on microcirculation. StO2 and EF deteriorated in our heterogeneous ICU population who received hemodiafiltration for 24 hours. StO2 is indicative of the dynamic balance between the regional oxygen delivery and oxygen utilization, while EF reflects the ability of microvasculature to react appropriately to hypoxia. CVVHDF may be responsible for the observed changes as it has been shown that the dialysis process can induce pre-capillary sphincter constriction due to changes in electrolyte concentrations and the cooler dialysate with decreased core temperature (Kong & Farrington, 2002). In our study, although temperature decreased, it is questionable whether the recorded difference (0.6 oC) would be of any clinical relevance. None of the patients was severely hypothermic so as the peripheral microcirculation and tissue
ACCEPTED MANUSCRIPT metabolism being significantly affected. The correlation of ∆OCR with lactate changes (perfusion marker) suggests a deteriorated microcirculation that does not meet the
PT
tissue oxygen requirements. In point of fact, StO2 and OCR decrease should be greater than that recorded, since the Hb reduction was substantial. Although in the
RI
present pilot study it is difficult to draw sound conclusions regarding the exact
SC
pathophysiologic consequences of CRRT implementation on microcirculation, significant deterioration of patient microcirculation was noted, without obvious
NU
reasons having appeared (as a dramatic decrease of hemodynamic variables, significant
MA
increase of vasopressors, clinically significant hypothermia) that could account for this deterioration. Modest drop of Hb (probably due to clotting and filter change as no major hemorrhagic complications were noted) could be a relevant factor, being
TE
D
however among the CRRT complications. To test the effect of the Hb fall to the deterioration of microcirculation parameters, we
AC CE P
divided patients into two groups according to Hb levels at the end of the 24hours period (more or less than 7 g/dL, a cut-off point for transfusion requirement). No statistically significant difference was documented between the two groups for any of the microcirculation indices, in different time points. Furthermore, EF, as well as StO2, significantly deteriorated in both groups. The same results were observed when we categorized patients according to a significant Hb fall (1.5 g/dl or more over the 24hours monitoring period, data not shown). As for the temperature changes, similarly, there was no difference in the microcirculation indices in the various time points, when comparing patients with a significant or not temperature drop (more or less than 35.5 oC). In both groups, also, EF and StO2 significantly decreased. Finally, no correlation was noted between the Hb levels or the temperature drop and the deteriorated microcirculatory indices. The above results clearly show that the
ACCEPTED MANUSCRIPT deterioration of microcirculation during the 24hours CRRT could not be attributed to the above mentioned confounding factors.
PT
Most importantly, there was a significant correlation between the decreasing sCysC
RI
levels and worsening of EF. The fact that the sCysC decrease, which marks the clearance achieved with CRRT, correlates with the deterioration of the EF, might
SC
indicate the negative impact of CRRT on the microcirculation of the critically ill.
NU
Also, microcirculatory alterations were worse in higher dosed CRRT in diabetics. On the whole, it seems that with a higher CRRT effect, the microcirculation worsened
MA
more during the intervention.
Furthermore, serum sE-Selectin levels were obtained over the 24-hour CRRT process,
D
as a biomarker of endothelial activation and/or damage. In general, sE-Selectin is
TE
present in low levels in healthy individuals but it increases in inflammatory states,
AC CE P
including sepsis (Xing et al., 2012 and Zonneveld et al., 2014). However, its precise pathophysiological significance has not been fully elucidated, relating its presence as a collateral finding in inflammatory tissue damage, its causal relationship to the inflammatory process or, contrariwise, its potential anti-inflammatory activity (Smith, 1997, Zonneveld et al., 2014). In the present study, serum sE-Selectin levels were high, with the sE-Selectin concentration change to be strongly correlated with the MAP change in the same period. Likewise, another recent study (Vassiliou et al., 2014) demonstrated high serum sE-Selectin levels in critically ill patients. In addition, a similar dependence of sE-Selectin on the systemic arterial pressure has been proved in the study of Buemi et al. (1997), where, a slight and transient increase of arterial pressure (induced by the cold pressure test) was in itself enough to increase sE-Selectin concentration, which,
ACCEPTED MANUSCRIPT in turn, decreased to basal levels after the vasopressor stimulus was lifted. The authors suggested that the increased arterial pressure induces increased vascular distending
PT
and transmural pressures that may cause endothelial cell activation with a consequent
RI
change in sE-Selectin levels.
Nevertheless, no difference of the sE-Selectin levels was established during the 24h
SC
CRRT period in our study. The fact that, in our study, sE-Selectin levels were
NU
sensitive enough to follow the changes of MAP, where, at the same time, did not show to correlate with the deteriorating microcirculation, indicates that the
MA
microcirculation effect (observed in the defined time period) is potentially mediated by a process not specifically influenced or marked by the circulating levels of
D
sE-Selectin. On the other hand, it cannot be excluded that the microcirculation may
TE
have been already influenced by a systemic inflammatory response (indicated by the sE-Selectin levels that were elevated before RRT onset), while further deterioration
AC CE P
probably involves other mechanisms. Here, the results of our study implicate the CRRT implementation.
In the same direction, study results did not indicate correlation of microcirculation with changes of global hemodynamic variables (such as MAP), which is not unexpected given that global hemodynamic parameters are influenced by many factors and are not specific and sensitive variables to follow tissue oxygenation; this is probably the reason that the goals of hemodynamic optimization in the ICU patients are a constant subject of debate (Vincent & De Backer, 2004 and Rampal et al., 2010). Microcirculatory disorders over the 24hours CRRT may be seen as part of the ‘dialytrauma’. This concept recognizes and bands together all the potential
ACCEPTED MANUSCRIPT CRRT-related adverse effects, which may have negative impact on recovery of critically ill patients (Maynar Moliner et al., 2012). Although the mortality rates of
PT
ICU patients receiving CRRT are high (50-80%) (Fonseca Ruiz et al., 2011) no association with ‘dialytrauma’ has been confirmed yet. Microcirculation evaluation is
RI
a dynamic approach which might provide a window to critical central vascular bed
SC
[assessment of subcutaneous dermal capillaries are representative of microvascular supply to heart (Pignocchino et al., 1994) and kidney (Economides et al., 2004)] and
NU
might contribute to CRRT individualization with prevention of its hemodynamic and
MA
metabolic complications (Selby et al., 2012).
So far, two studies evaluating potential changes of microcirculation during CRRT in
D
ICU did not observe significant alterations (Ruiz et al., 2010 and Veenstra et al.,
TE
2014). Differences in patient population, prescribed treatment, and timing of follow
AC CE P
up may explain the differences compared with the results of our study. In conclusion, our findings may simply show that microcirculation is stressed in promoting solute clearance during hemodiafiltration. This view is in accord with the report of Kong & Farrington (2002) who suggested that two levels of dialysis are going on, one at the capillary membrane and another at the dialyzer membrane and if one intend to increase dialysis efficiency should not only concentrate in the dialysis process within the dialyzer, but harness microcirculation as well. Limitations A limitation of our study is the small number of patients included, which hampers the generalization of the study results; however it offers information to guide new research in the field. Other limitations are the lack of baseline microcirculation measurement at ICU admission and the lack of core temperature recordings;
ACCEPTED MANUSCRIPT assessments that would potentially better elucidate the microcirculation alterations
AC CE P
TE
D
MA
NU
SC
RI
PT
during CRRT.
ACCEPTED MANUSCRIPT Conclusions The present study assessed the impact of CRRT on microcirculation and added
PT
information regarding the CRRT- related adverse effects in critically ill patients, or
RI
‘dialytrauma’. In a multidisciplinary ICU, deterioration of microcirculation parameters during the first 24hours of CVVHDF and a correlation between changes in
SC
endothelial function with the marked reduction of sCysC were noted. Moreover,
NU
microcirculation alterations correlated with dose in patients with diabetes. Further research will clarify the potential of microcirculation imaging technologies, as CRRT
MA
prognostic and/or monitoring tools.
D
Acknowledgments
TE
The present study was partly founded by a grant from the Special Account
AC CE P
for Research Grants of the National and Kapodistrian University of Athens, Greece.
ACCEPTED MANUSCRIPT References 1. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P. Acute renal failure -definition, outcome measures, animal models, fluid therapy and information
PT
technology needs: the Second International Consensus Conference of the Acute
RI
Dialysis Quality Initiative (ADQI) Group. Crit Care 8: R204-212, 2004.
SC
2. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, Schein RM, Sibbald WJ. Definitions for sepsis and organ failure and guidelines for the
NU
use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care
MA
Medicine. 1992. Chest 136(5 Suppl): e28, 2009.
D
3. Buemi M, Allegra A, Aloisi C, Corica F, Alonci A, Ruello A, Montalto G,
TE
Frisina N. Cold pressor test raises serum concentrations of ICAM-1, VCAM-1, and E-selectin in normotensive and hypertensive patients. Hypertension
AC CE P
30:845-847, 1997.
4. Clec'h C, Darmon M, Lautrette A, Chemouni F, Azoulay E, Schwebel C, Dumenil AS, Garrouste-Orgeas M, Goldgran-Toledano D, Cohen Y, Timsit JF. Efficacy of renal replacement therapy in critically ill patients: a propensity analysis. Crit Care 16: R236, 2012. 5. Creteur J, Carollo T, Soldati G, Buchele G, De Backer D, Vincent JL. The prognostic value of muscle StO2 in septic patients. Intensive Care Med 33: 1549-1556, 2007. 6. Economides PA, Caselli A, Zuo CS, Sparks C, Khaodhiar L, Katsilambros N,Horton ES, Veves A. Kidney oxygenation during water diuresis and
ACCEPTED MANUSCRIPT endothelial function in patients with type 2 diabetes and subjects at risk to develop diabetes. Metabolism 53: 222-227, 2004. 7. Fonseca Ruiz NJ, Castro DP, Guerra AM, Saldarriaga FM, Hernandez JD.
PT
Renal injury study in critical ill patients in accordance with the new definition
RI
given by the Acute Kidney Injury Network. J Crit Care 26: 206-212, 2011.
SC
8. Gerovasili V, Tripodaki E, Karatzanos E, Pitsolis T, Markaki V, Zervakis D, Routsi C, Roussos C, Nanas S. Short-term systemic effect of electrical muscle
NU
stimulation in critically ill patients. Chest 136: 1249-1256, 2009.
MA
9. Gibney N, Hoste E, Burdmann EA, Bunchman T, Kher V, Viswanathan R, Mehta RL, Ronco C. Timing of initiation and discontinuation of renal
D
replacement therapy in AKI: unanswered key questions. Clin J Am Soc Nephrol
TE
3: 876-880, 2008.
AC CE P
10. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med 13: 818-829, 1985. 11. Kokkoris S, Pipili C, Grapsa E, Kyprianou T, Nanas S. Novel Biomarkers of Acute Kidney Injury in the General Adult ICU: A Review. Ren Fail 35: 579-591, 2013.
12. Kong CH, Farrington K. Harnessing the microcirculation to increase dialysis efficiency. Nephron 92: 951, 2002. 13. Manetos C, Dimopoulos S, Tzanis G, Vakrou S, Tasoulis A, Kapelios C, Agapitou
V,
Ntalianis
A,
Terrovitis
J,
Nanas
S.
Skeletal
muscle
microcirculatory abnormalities are associated with exercise intolerance,
ACCEPTED MANUSCRIPT ventilatory inefficiency, and impaired autonomic control in heart failure. J Heart Lung Transplant 30: 1403-1408, 2011. 14. Maynar Moliner J, Honore PM, Sanchez-Izquierdo Riera JA, Herrera M,
Spapen
HD.
Handling
continuous
renal
replacement
PT
Gutierrez
RI
therapy-related adverse effects in intensive care unit patients: the dialytrauma
SC
concept. Blood Purif 34: 177-185, 2012.
15. Myers DE, Cooper CE, Beilman GJ, Mowlem JD, Anderson LD, Seifert RP,
NU
Ortner JP. A wide gap second derivative NIR spectroscopic method for
MA
measuring tissue hemoglobin oxygen saturation. Adv Exp Med Biol 578: 217-222, 2006.
D
16. Nanas S, Gerovasili V, Renieris P, Angelopoulos E, Poriazi M, Kritikos K,
TE
Siafaka A, Baraboutis I, Zervakis D, Markaki V, Routsi C, Roussos C.
AC CE P
Non-invasive assessment of the microcirculation in critically ill patients. Anaesth Intensive Care 37: 733-739, 2009. 17. Nejat M, Pickering JW, Walker RJ, Endre ZH. Rapid detection of acute kidney injury by plasma cystatin C in the intensive care unit. Nephrol Dial Transplant 25: 3283-3289, 2010. 18. Pignocchino P, Conte MR, Scarnato S, Grande A. [Study of cutaneous microcirculation using the laser-Doppler method in syndrome X]. Cardiologia 39: 193-197, 1994. 19. Rampal T, Jhanji S, Pearse RM. Using oxygen delivery targets to optimize resuscitation in critically ill patients. Curr Opin Crit Care 16:244-249, 2010.
ACCEPTED MANUSCRIPT 20. Ruiz C, Hernandez G, Godoy C, Downey P, Andresen M, Bruhn A. Sublingual microcirculatory changes during high-volume hemofiltration in hyperdynamic septic shock patients. Crit Care 14: R170, 2010.
PT
21. Selby NM, McIntyre CW. Predicting and managing complications of renal
RI
replacement therapy in the critically ill. Blood Purif 34: 171-176, 2012.
SC
22. Siafaka A, Angelopoulos E, Kritikos K, Poriazi M, Basios N, Gerovasili V, Andreou A, Roussos C, Nanas S. Acute effects of smoking on skeletal muscle spectroscopy. Chest 131:
NU
microcirculation monitored by near-infrared
MA
1479-1485, 2007.
23. Smith CW. Potential significance of circulating E-selectin. Circulation
D
95:1986-1988, 1997.
TE
24. Tripodaki ES, Tasoulis A, Vasileiadis I, Vastardis L, Skampas N,
AC CE P
Sakellaridis T, Argiriou M, Charitos C, Nanas S. Microcirculatory alterations after cardiopulmonary bypass as assessed with near infrared spectroscopy: a pilot study. Can J Anaesth 59: 620-621, 2012. 25. Vassiliou AG, Mastora Z, Orfanos SE, Jahaj E, Maniatis NA, Koutsoukou A, Armaganidis A, Kotanidou A. Elevated biomarkers of endothelial dysfunction/activation at ICU admission are associated with sepsis development. Cytokine 69:240-247, 2014. 26. Veenstra G, Scheenstra B, Koopmans M, Kingma WP, Buter H, Boerma EC. Microcirculatory perfusion derangements during continuous hemofiltration with fixed dose of ultrafiltration in stabilized intensive care unit patients. J Crit Care 29: 478-481, 2014.
ACCEPTED MANUSCRIPT 27. Vincent JL, De Backer D. Oxygen transport-the oxygen delivery controversy. Intensive Care Med 30:1990-1996, 2004. 28. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H, The SOFA (Sepsis-related Organ Failure
PT
Reinhart CK, Suter PM, Thijs LG.
RI
Assessment) score to describe organ dysfunction/failure. On behalf of the
SC
Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 22: 707-710, 1996.
NU
29. Xing K, Murthy S, Liles WC, Singh JM. Clinical utility of biomarkers of
MA
endothelial activation in sepsis--a systematic review. Crit Care 2012;16(1):R7. 30. Zonneveld R, Martinelli R, Shapiro NI, Kuijpers TW, Plötz FB, Carman CV.
D
Soluble adhesion molecules as markers for sepsis and the potential
AC CE P
18:204, 2014.
TE
pathophysiological discrepancy in neonates, children and adults. Crit Care
ACCEPTED MANUSCRIPT Highlights CRRT adversely affected microcirculation in the critically ill.
•
Microcirculation worsened more with a higher CRRT effect.
•
Microcirculation monitoring might contribute to CRRT individualization.
AC CE P
TE
D
MA
NU
SC
RI
PT
•
ACCEPTED MANUSCRIPT Tables
19/9
Age (years)
67.5±17
APACHE II
19 ±7
SOFA
9.8 ± 2.2
SC
Reason for ICU admission 15 (53.6%)
Cardiovascular
6 (21.4%)
Respiratory
7 (25%)
NU
Post-operative
19 (68%)
MA
Sepsis RIFLE No
5 (17.8%)
AC CE P
Diabetes
TE
Failure
D
Risk Injury
RI
Gender (Male/Female)
PT
Table 1. Patient characteristics (N=28)
9 (32.4%) 13 (46.8%) 1 (3.6%) 8 (28.5%)
ACCEPTED MANUSCRIPT
H6
H24
p-value*
MAP (mmHg)
82.3± 18.5
91.6 ± 15
84.7 ± 21.3
0.150
Periph. Temp. (°C)
36.6± 1.3
36.2 ± 1.3
36 ± 1.1
0.020
Lac (mmol/L)
3.2 ± 5.4
4.4 ± 6.8
4.8 ±7.3
Hb (g/dL)
9.5 ± 2
9±2
Norepinephrine (µg/kg/h)
0.65 ± 0.66
0.68 ± 0.72
sCr (mg/dL)
2.5 ± 1.5
2±1
sCysC (mg/L)
2.7 ± 0.8
StO2 (%)
76.5±12.5
OCR (%/min)
-12 ± 7
EF (%/sec)
2.25 ± 1.44
RI
SC
0.015 0.004
0.81 ± 0.86
0.280
1.7 ± 0.9
<0.0001
2.2 ± 0.6
1.8 ± 0.8
<0.0001
75 ± 11
70 ±16
0.040
-11 ± 7
-9 ± 4
0.270
2.1 ± 1.8
1.6 ±1.4
0.020
NU
8±2
D
TE
PT
H0
MA
Table 2. RRT characteristics before RRT initiation (H0), at six (H6) and 24 hours (H24) during RRT (N=28).
AC CE P
Data are expressed as mean ± SD
Definition of abbreviations: RRT, Renal replacement therapy, MAP, Mean Arterial Pressure; Periph Temp, peripheral temperature; Lac, Lactate; sCr, serum creatinine; sCysC, serum cystatin C; StO2, tissue oxygen saturation; OCR Oxygen consumption rate; EF, endothelial function * p -value for statistical significance of repeated measures ANOVA set at <0.05
ACCEPTED MANUSCRIPT
Hb > 7 g/dL (N=20)
76.1±12.1
76.7±12.4
StO2 (%)
at H6
74.8±10.4
StO2 (%)
at H24
69.6±15.7
RI
before RRT
Hb ≤7 g/dL (N=8)
75.5±10.8
SC
StO2 (%)
PT
Table 3. Microcirculatory parameters in patients with hemoglobin levels more or less than 7 g/dL after 24 hours of RRT.
70.3±16.2 -12.8±6
NU
OCR(%/min)before RRT -11.9±5.6 -10.2±6
OCR (%/min) at H24
-10±3
EF (%/sec) before RRT
1.8±1
2.2±1.5
EF (%/sec) at H6
1.6±1.2
2±1.6
EF (%/sec) at H24
1.3±0.9
TE
D
MA
OCR (%/min) at H6
-10.4±6.3 -9.8±2.7
1.5±1.1
AC CE P
Data are expressed as mean ± SD; p-value was not statistically significant in any of the comparisons between the two subgroups. In the subgroup of patients with Hb ≤ 7.0g/dL at 24 h, StO2 and EF significantly decreased during the monitoring period (H0-H6-H24) (p= 0.030 and p=0.035 respectively). In the subgroup of patients with Hb > 7.0 g/dL StO2 and EF also significantly decreased during the monitoring period (p= 0.034 and p=0.038 respectively). Definition of abbreviations: RRT, Renal replacement therapy; Hb, hemoglobin; StO2, tissue oxygen saturation; OCR, Oxygen consumption rate; EF, endothelial function; H0, measurements recorded before RRT initiation; H6, measurements recorded at six hours of RRT; H24, measurements recorded at 24 hours of RRT
PT
ACCEPTED MANUSCRIPT
Table 4. Microcirculatory parameters in patients with peripheral temperature more or less than 35.5 °C after six hours of RRT.
StO2 (%)
at H6
74.7±10.4
StO2 (%)
at H24
69.5±15.1
75.4±11.7 70.3±15.8 -11.8±5.9
-10.3±6.3
-10.7±6.7
OCR (%/min) at H24
-8.8±3.1
-8.6±3.4
EF (%/sec) before RRT
2±0.8
2.2±1.4
1.8±1.3
1.9±1.6
1.3±0.8
1.5±1.2
EF (%/sec) at H6
AC CE P
EF (%/sec) at H24
D
OCR (%/min) at H6
TE
MA
OCR(%/min)before RRT -11.5±5.4
76.2±12.4
SC
75.9±12.1
before RRT
NU
StO2 (%)
Temp > 35.5 oC (N=16)
RI
Temp ≤ 35.5 °C (N=12)
Data are expressed as mean ± SD; p-value was not statistically significant in any of the comparisons between the two subgroups. In the subgroup of patients with Temperature ≤ 35.5 °C after six hours of RRT, StO2 and EF significantly decreased during the monitoring period (H0-H6-H24) (p= 0.040 and p=0.038 respectively). In the subgroup of patients with Temperature > 35.5 °C, StO2 and EF also significantly decreased during the monitoring period (p= 0.037 and p=0.040 respectively). Definition of abbreviations: RRT, Renal replacement therapy; Temp, peripheral temperature; StO2, tissue oxygen saturation; OCR, Oxygen consumption rate; EF, endothelial function; H0, measurements recorded before RRT initiation; H6, measurements recorded at six hours of RRT; H24, measurements recorded at 24 hours of RRT
S0026-2862(15)30026-1 doi: 10.1016/j.mvr.2015.09.004 YMVRE 3573
To appear in:
Microvascular Research
Received date: Revised date: Accepted date:
12 February 2015 27 September 2015 28 September 2015
Please cite this article as: Pipili, Chrysoula, Vasileiadis, Ioannis, Grapsa, Eirini, Tripodaki, Elli-Sophia, Ioannidou, Sophia, Papastylianou, Adroula, Kokkoris, Stelios, Routsi, Christina, Politou, Marianna, Nanas, Serafeim, Microcirculatory alterations during continuous renal replacement therapy in ICU: A novel view on the ‘dialysis trauma’ concept, Microvascular Research (2015), doi: 10.1016/j.mvr.2015.09.004
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Title: Microcirculatory alterations during continuous renal replacement therapy
PT
in ICU: a novel view on the ‘dialysis trauma’ concept Chrysoula Pipili1, Ioannis Vasileiadis1, Eirini Grapsa1, Elli-Sophia Tripodaki1, Sophia
RI
Ioannidou2, Adroula Papastylianou1, Stelios Kokkoris1, Christina Routsi1, Marianna
First Critical Care Department, ‘Evangelismos’ General Hospital, National and
NU
1
SC
Politou3, Serafeim Nanas1
Kapodistrian University of Athens
3
Laboratory of Biochemistry, ‘Evangelismos’ Hospital, Athens, Greece
MA
2
Blood Transfusion Department, Aretaieion Hospital, Athens University Medical
AC CE P
Prof. Serafeim Nanas
TE
Corresponding Author:
D
School, Athens, Greece
First Critical Care Department, National and Kapodistrian University of Athens Evangelismos Hospital
Ypsilantou 45-47, 106 75, Athens, Greece Email: [email protected] Mobile: +306973036448
Fax: +302132043385
ACCEPTED MANUSCRIPT ABSTRACT
renal replacement therapy (CRRT) in critically ill patients.
PT
Objective: The purpose of this study was to evaluate microcirculation over 24hours
RI
Methods: We conducted a single-center, prospective, observational study, measuring
SC
microcirculation parameters, monitored by near infrared spectroscopy (NIRS) before hemodiafiltration onset (H0), and at six (H6) and 24 hours (H24) during CRRT in
NU
critically ill patients. Serum Cystatin C (sCysC) and soluble (s)E-selectin levels were
MA
measured at the same time points. Twenty-eight patients [19 men (68%)] were included in the study.
D
Results: Tissue oxygen saturation (StO2, %) [76.5±12.5 (H0) vs 75±11 (H6) vs 70±16
TE
(H24), p=0.04], reperfusion rate, indicating endothelial function (EF, %/sec)
AC CE P
[2.25±1.44 (H0) vs 2.1±1.8 (H6) vs 1.6 ±1.4 (H24), p=0.02] and sCysC (mg/L) [2.7±0.8 (H0) vs 2.2±0.6 (H6) vs 1.8±0.8 (H24), p<0.0001] significantly decreased within the 24hours CRRT. Change of EF positively correlated with changes of sCysC within 24hours CRRT (r= 0.464, p=0.013) while in patients with diabetes the change of StO2 correlated with dose (r=-0.8, p=0.01). No correlation existed between hemoglobin and temperature changes with the deteriorated microcirculation indices. sE-Selectin levels in serum were elavated; no difference was established over the 24h CRRT period. A strong correlation existed between the sE-Selectin concentration change at H6 and H24 and the mean arterial pressure change in the same period (r=0.77, p<0.001).
Conclusions: During the first 24 hours of CRRT implementation in critically ill patients, deterioration of microcirculation parameters was noted. Microcirculatory
ACCEPTED MANUSCRIPT alterations correlated with sCysC changes and with dose in patients with diabetes. Keywords:
Dialytrauma,
Near
infrared
spectroscopy,
AC CE P
TE
D
MA
NU
SC
RI
PT
Continuous renal replacement therapy, Cystatin C
Tissue
oxygenation,
ACCEPTED MANUSCRIPT
Abbreviations
PT
CRRT: continuous renal replacement therapy
SC
CVVHDF: continuous veno-venous dia-hemofiltration
RI
NIRS: near infrared spectroscopy
NU
ICU: intensive care unit
RIFLE: risk-injury-failure-loss-end stage renal disease
MA
StO2: tissue oxygen saturation
D
Hb: hemoglobin
TE
OCR: oxygen consumption rate
AC CE P
EF: endothelial function
sCysC: serum cystatin C sCr: serum creatinine Lac: lactate
APACHE: acute physiology and chronic health evaluation SOFA: sequential organ failure assessment sE-Selectin: soluble E-Selectin
ACCEPTED MANUSCRIPT INTRODUCTION Continuous renal replacement therapy (CRRT) has proved to be a competent
PT
therapeutic tool for patients in Intensive Care Unit (ICU), increasing significantly
RI
their quality of care. However, optimal monitoring for CRRT has not yet been
SC
determined and the mortality has not decreased notably (Clec'h et al., 2012). Although CRRT implementation is crucial to reverse situations posing an immediate danger to
NU
survival (as severe hyperkalemia and volume overload with impending pulmonary edema), adverse effects –a related morbidity– can also occur (Maynar Moliner et al.,
MA
2012 and Selby & McIntyre, 2012). During CRRT, the function of the failed nephron has been replaced by inflexible, largely, prescription orders, regarding blood flow
D
rate, type of replacement or dialysis fluids and the dose (effluent flow rate).
TE
Adjustments of the prescribed treatment are made, however, not in the constant basis
AC CE P
that the changing metabolic requirements would demand, so as to adapt the microenvironment and achieve a fine tuning. Moreover, the physician may not necessarily know or control all the affected variables. Thus, CRRT may prove inadequate or act as an exogenous factor that overcomes the natural resistance of the tissues and organs and causes ‘trauma’ (Maynar Moliner et al., 2012 and Selby & McIntyre, 2012). CRRT has a direct impact on microcirculation where the removal of waste products and supply of essential nutrients take place. Studies for predicting and managing complications of RRT in the critically ill recommend the assessment of microcirculatory function to estimate the need for evolved therapies (Creteur et al., 2007, Ruiz et al., 2010 and Veenstra et al., 2014).
ACCEPTED MANUSCRIPT Technological progress enables assessment of skeletal microcirculation, which exhibits higher prognostic capability than broad indices of hemodynamic, ischemic
PT
and septic alterations in critically ill patients (Creteur et al., 2007 and Nanas et al., 2009). Near infrared spectroscopy (NIRS) methodology can reliably estimate
RI
microcirculation in this patient group. Furthermore, biomarkers of endothelial
SC
activation and/or damage, such as soluble (s)E-Selectin levels in serum, (Xing et al., 2012 and Zonneveld et al., 2014), could provide additional information. Although
NU
microcirculatory abnormalities have been involved in the pathophysiology of acute
MA
kidney injury (AKI), few data exist regarding their monitoring during CRRT in critically ill patients. Serum Cystatin C (sCysC) has been used as a biomarker for AKI development, tested for early diagnosis and monitoring of disease progression
TE
D
(Kokkoris et al., 2013 and Nejat et al., 2010). We hypothesized that assessment of microcirculation together with an endothelial and
AC CE P
a renal marker over 24hours continuous hemodiafiltration would result to a more complete evaluation of the CRRT process. We assessed, accordingly, the impact of CRRT on skeletal microcirculation utilizing NIRS technology in parallel with endothelial and renal marker sampling during CRRT, at predetermined time points. Assessment of microcirculation by the bedside, alone or in combination with renal markers, may prove useful in monitoring disease progression and guiding targeted intervention during CRRT.
ACCEPTED MANUSCRIPT
PT
MATERIAL AND METHODS
RI
Study design
This is a pilot, prospective, observational study in which renal, perfusion and
SC
microcirculation parameters were monitored during CRRT. The study was performed
NU
at the two ICUs of a tertiary hospital, comprising 30 beds, from October 2009 to July 2010. The procedures followed were in accord with the Declaration of Helsinki and
MA
its revisions, and written informed consent was obtained from patients’ relatives since all patients were sedated, under mechanical ventilation. All consecutive patients who
D
met both, the inclusion criteria for CRRT initiation, according to the attending
TE
intensivists, and the criteria for NIRS implementation, and completed, also, 24 hours
AC CE P
CRRT were included in the study. More specifically, patients without the exclusion criteria listed below could be enrolled: 1) Exclusion criteria before the CRRT onset: age ≤18 years, BMI ≥ 35kg/m2, chronic renal failure under hemodialysis, pregnancy, brain death, metastatic cancer, ICU re-admission, missing baseline sCr values 2) Exclusion criteria after the CRRT onset: To avoid a major ‘down-time’ effect ( the time spent off filter) and also enrolling patients with major blood loss, both of which could affect the results of our study, we included patients in whom CRRT was interrupted only to change a clotted filter, i.e. less than one hour (CRRT achieved > 23 h). More prolonged CRRT interruption and/or repeated filter change disqualified a patient from participation in the study. 3) Exclusion criteria for NIRS implementation: musculoskeletal and connective tissue diseases, upper limb fractures. The treating physician was aware that the subject was participating in the study. However, there was no special intervention performed that could theoretically impact
ACCEPTED MANUSCRIPT the study results. At enrollment, demographic data, acute physiology and chronic health evaluation (APACHE II) (Knaus et al., 1985) and the sequential organ failure
PT
assessment (SOFA) scores (Vincent et al., 1996) were recorded. AKI was defined using both the serum creatinine (sCr) and urine output criteria of RIFLE classification
RI
(Bellomo et al., 2004). Diagnosis of sepsis was made by the ICU team according to
SC
American College of Chest Physicians/Society of Critical Care Medicine Consensus
NU
Conference (Bone et al., 2009). The CRRT dose used was the prescribed dose.
MA
Endothelial and Renal marker sampling
Data collection included sE-Selectin, sCysC and sCr before CRRT initiation (H0), at six hours (H6) and at 24 hours (H24) during CRRT. All blood samples were taken
TE
D
through a central venous catheter (not the same used for dialysis). sCr levels were quantified by a modified Jaffe method with protein precipitation using an alkaline
AC CE P
picrate reaction while sCysC by using a MNII nephelometer (Dade Behring GmbH, Marburg, Germany). sE-selectin was accurately measured by enzyme-linked immunosorbent assay (ELISA) purchased from R&D Systems (R&D Systems Inc., Minneapolis, MN, USA).
Description of Hemodiafiltration technique All patients received continuous veno-venous hemodiafiltration (CVVHDF) performed by pump-driven machines. The CRRT machines used in the study were of two types: (1) Prisma Gambro, Lakewood, Co. and (2) Kimal, Nephrotech Co., Ltd., both with a 0.9 m2 polysulfone filter. Blood flow rate was maintained between 100 and 130 ml/min and predilution or postdilution replacement fluid was administered.
ACCEPTED MANUSCRIPT Sterile Bicarbonate solution (ACCUSOL Dialysis Solution 2.5L, 35 Bicarbonate, K4, Baxter International Inc.) was used as the dialysate and the replacement fluid for
PT
CVVHDF. Replacement flow rate was set at 1-2 L/h and dialysate flow rate at 1-1.3 L/h. The ultrafiltrate was adjusted to achieve fluid balance for most patients. Heparin
RI
was administered unless a severe bleeding risk existed. Then, predilution mode and
SC
periodic flushing with saline were administered to prevent filter clotting. In case of heparin induced thrombocytopenia, fondaparinux was utilized with close monitoring
NU
of the levels of anti-Xa antibodies to assess anticoagulant activity. Clotting of the
the first 24 hours of treatment.
MA
extracorporeal system was the most common reason of CVVHDF interruption during
TE
D
Microcirculatory assessment and analysis
NIRS constitutes an easily applicable, non-invasive monitoring system that can assess
AC CE P
skeletal muscle microcirculation in health and disease (Myers et al., 2006, Nanas et al., 2009 and Siafaka et al., 2007). The principles of NIRS utilization, along with the vascular occlusion technique, have been described elsewhere (Gerovasili et al., 2009, Manetos et al., 2011 and Tripodaki et al., 2012). In the present study, thenar muscle StO2 was measured non-invasively, by a second generation NIRS device (Inspectra Model 325, Hutchinson Technology, USA). Initially, a sphygmomanometer cuff was placed around the patient's arm and a transdermal light probe was stuck on the patient’s homolateral thenar. Baseline arterial pressure and baseline StO2 were recorded after a three minute stabilization period. Subsequently, ischemia was introduced as the pneumatic cuff was rapidly inflated to 50 mmHg above the patient’s systolic blood pressure. The vascular blockade (mostly of the brachial artery) lasted three minutes, during which, the fall of
ACCEPTED MANUSCRIPT thenar StO2 was recorded (ischemia phase). Then, the cuff was rapidly deflated, and the rise of the thenar StO2 (reperfusion phase) was recorded as well. The monitoring
PT
continued until the return of StO2 at its baseline value (hyperemia phase). The vascular occlusion derived curves were stored using InSpectra software.
RI
StO2 curves were analyzed offline, blindly and in random order (InSpectra Analysis
SC
Program, version 2.0; Hutchinson Technology; Hutchinson, MN; running in MatLab 7.0; The MathWorks; Novi, MI). The first degree slope of the hemoglobin (Hb)
NU
desaturation curve, during stagnant limb ischemia, reflects the tissue oxygen
MA
consumption rate (OCR, %/min) and the slope of the increase of StO2, after the release of the vascular occlusion, is indicative of the endothelial function (EF,%/sec). NIRS measurements were performed at H0, H6 and H24. Differences in StO2
TE
D
(∆StO2), OCR (∆OCR) and endothelial function (∆EF) values were calculated before
AC CE P
CVVHDF initiation and after the 24hours monitoring period.
STATISTICAL ANALYSIS Analysis of variance (repeated measures) and subsequent post hoc test were used to establish differences in biomarker values and microcirculatory, hemodynamic and perfusion parameters. The associations between variables were assessed using the Pearson or Spearman correlation coefficients. Parametric and non-parametric tests were used to check for significant differences between demographic and baseline characteristics of study groups. The analyses were performed using statistical software SPSS version 17 (SPSS, Chicago IL). Two sided p-values of less than 0.05 were required to establish statistical significance. Results were expressed as frequencies, mean±SD and medians with interquartile range.
ACCEPTED MANUSCRIPT
PT
RESULTS
RI
Study population
57 critically ill patients were eligible for the study but only 28 received CVVHDF for
SC
24 hours without interruption and met the NIRS inclusion criteria. All patients
NU
included in the study were mechanically ventilated. Patient characteristics are shown in Table 1. 16/28 (57%) of patients were treated with CVVHDF on postdilution
MA
mode. CRRT was initiated at 8.5±2 days of ICU hospitalization, when the patients fulfilled at least one of the following criteria (Gibney et al., 2008): oliguria (urine
D
output<200ml/12h) despite fluid resuscitation and intravenous diuretic treatment: 4
TE
patients (14%), anuria (urine output<50ml/12h): 6 patients (21.5%), volume overload:
AC CE P
1 patient (3.5%), azotemia (blood urea nitrogen 80>mg/dL): 7 patients (25%), hypernatremia: 3 patients (11%), sepsis: 3 patients (11%), hyperkalemia (potassium ≥7.5mmol/L): 1 patient (3.5%), metabolic acidosis (pH less than 7.10) with oliguria: 2 patients (7%) and hypernatremia with oliguria: 1 patient (3.5%). The mean prescribed CVVHDF dose was 23±7 ml/kg/h. During the monitoring period, all patients remained euvolemic, mean arterial pressure did not change significantly and blood transfusions were not performed. Major circuit ‘down-time’ effect was avoided by immediate replacement of the clotted filters. All patients required heparin for anticoagulation except one patient that required fondaparinux and three patients needed saline flashes. No differences were noted concerning the filter lifetimes in each anti-coagulation group.
ACCEPTED MANUSCRIPT Microcirculatory, endothelial and renal marker measurements during 24-hour RRT
PT
StO2 and EF significantly decreased within 24hours CRRT (Table 2). Also, sCysC,
RI
sCr, Hb, lactate and peripheral temperature showed significant differentiation during
SC
the monitoring period. sCysC decreased at H6 (H0-H6, p<0.0001) and H24 (H0-H24, p<0.0001) as sCr, Hb and peripheral temperature did, while lactate levels increased at
EF
positively
correlated
with
NU
H6 (H0-H6, p=0.019) and H24 (H0-H24, p=0.017) (Table 2). In parallel, change of changes
of
sCysC
within
24hours
CRRT
MA
(r= 0.464, p=0.013) and ∆OCR negatively correlated with lactate changes within the same 24hours period (r= -0.44, p=0.036). In the subgroup of patients with diabetes
D
CRRT dose correlated with ∆StO2 (r=-0.8, p=0.01). Statistical analysis showed no
TE
influence in the microcirculation parameters when pre- or post -dilution method were
AC CE P
used. Hb drop was most probably due to clotting; no major hemorrhagic episodes were noticed.
The patients with low Hb (≤7 g/dL) or Hb fall more than 1.5 g/dL over the monitoring 24hours and the patients with low temperature (≤ 35.5 oC after six hours of CRRT) did not present worse microcirculatory parameters when compared with patients with Hb >7 g/dL or temperature >35.5 oC respectively. In addition, in all subgroups (low/high Hb and low/high temperature) StO2 and EF significantly deteriorated during the monitoring period (Tables 3,4). No correlation existed between temperature and Hb changes with the deteriorated microcirculation indices. sE-Selectin levels were obtained in the serum of 23 subjects in total. 19 patients had sE-Selectin measurements at H6 and H24 (55.85±25.79 vs 56.82±25.84 ng/ml respectively, p=0.827) and three had sE-Selectin values at H0, H6 and H24
ACCEPTED MANUSCRIPT (62.27±34.64 vs 53.12±19.8 vs 63.72±10.09 ng/ml respectively, p=0.86). As shown, no difference in sE-Selectin levels was evidenced during the CRRT process. Also, no
PT
correlation was found between the sE-Selectin levels and the microcirculation indices. Interestingly, a strong correlation was noted between the change of sE-Selectin levels
SC
RI
at H6 and H24 with the change of MAP at the same time period (r=0.77, p<0.001).
NU
Discussion
In this pilot study, we assessed skeletal muscle microcirculation and a renal marker to
MA
track the course of continuous hemodiafiltration in critically ill patients. We found that during the first 24 hours of CRRT implementation, microcirculatory parameters
D
deteriorated, sCysC decreased and changes of EF correlated positively with changes
AC CE P
correlated with dose.
TE
of sCysC. Moreover, in patients with diabetes the differences in microcirculation
There is little information concerning the potential impact of CVVHDF on microcirculation. StO2 and EF deteriorated in our heterogeneous ICU population who received hemodiafiltration for 24 hours. StO2 is indicative of the dynamic balance between the regional oxygen delivery and oxygen utilization, while EF reflects the ability of microvasculature to react appropriately to hypoxia. CVVHDF may be responsible for the observed changes as it has been shown that the dialysis process can induce pre-capillary sphincter constriction due to changes in electrolyte concentrations and the cooler dialysate with decreased core temperature (Kong & Farrington, 2002). In our study, although temperature decreased, it is questionable whether the recorded difference (0.6 oC) would be of any clinical relevance. None of the patients was severely hypothermic so as the peripheral microcirculation and tissue
ACCEPTED MANUSCRIPT metabolism being significantly affected. The correlation of ∆OCR with lactate changes (perfusion marker) suggests a deteriorated microcirculation that does not meet the
PT
tissue oxygen requirements. In point of fact, StO2 and OCR decrease should be greater than that recorded, since the Hb reduction was substantial. Although in the
RI
present pilot study it is difficult to draw sound conclusions regarding the exact
SC
pathophysiologic consequences of CRRT implementation on microcirculation, significant deterioration of patient microcirculation was noted, without obvious
NU
reasons having appeared (as a dramatic decrease of hemodynamic variables, significant
MA
increase of vasopressors, clinically significant hypothermia) that could account for this deterioration. Modest drop of Hb (probably due to clotting and filter change as no major hemorrhagic complications were noted) could be a relevant factor, being
TE
D
however among the CRRT complications. To test the effect of the Hb fall to the deterioration of microcirculation parameters, we
AC CE P
divided patients into two groups according to Hb levels at the end of the 24hours period (more or less than 7 g/dL, a cut-off point for transfusion requirement). No statistically significant difference was documented between the two groups for any of the microcirculation indices, in different time points. Furthermore, EF, as well as StO2, significantly deteriorated in both groups. The same results were observed when we categorized patients according to a significant Hb fall (1.5 g/dl or more over the 24hours monitoring period, data not shown). As for the temperature changes, similarly, there was no difference in the microcirculation indices in the various time points, when comparing patients with a significant or not temperature drop (more or less than 35.5 oC). In both groups, also, EF and StO2 significantly decreased. Finally, no correlation was noted between the Hb levels or the temperature drop and the deteriorated microcirculatory indices. The above results clearly show that the
ACCEPTED MANUSCRIPT deterioration of microcirculation during the 24hours CRRT could not be attributed to the above mentioned confounding factors.
PT
Most importantly, there was a significant correlation between the decreasing sCysC
RI
levels and worsening of EF. The fact that the sCysC decrease, which marks the clearance achieved with CRRT, correlates with the deterioration of the EF, might
SC
indicate the negative impact of CRRT on the microcirculation of the critically ill.
NU
Also, microcirculatory alterations were worse in higher dosed CRRT in diabetics. On the whole, it seems that with a higher CRRT effect, the microcirculation worsened
MA
more during the intervention.
Furthermore, serum sE-Selectin levels were obtained over the 24-hour CRRT process,
D
as a biomarker of endothelial activation and/or damage. In general, sE-Selectin is
TE
present in low levels in healthy individuals but it increases in inflammatory states,
AC CE P
including sepsis (Xing et al., 2012 and Zonneveld et al., 2014). However, its precise pathophysiological significance has not been fully elucidated, relating its presence as a collateral finding in inflammatory tissue damage, its causal relationship to the inflammatory process or, contrariwise, its potential anti-inflammatory activity (Smith, 1997, Zonneveld et al., 2014). In the present study, serum sE-Selectin levels were high, with the sE-Selectin concentration change to be strongly correlated with the MAP change in the same period. Likewise, another recent study (Vassiliou et al., 2014) demonstrated high serum sE-Selectin levels in critically ill patients. In addition, a similar dependence of sE-Selectin on the systemic arterial pressure has been proved in the study of Buemi et al. (1997), where, a slight and transient increase of arterial pressure (induced by the cold pressure test) was in itself enough to increase sE-Selectin concentration, which,
ACCEPTED MANUSCRIPT in turn, decreased to basal levels after the vasopressor stimulus was lifted. The authors suggested that the increased arterial pressure induces increased vascular distending
PT
and transmural pressures that may cause endothelial cell activation with a consequent
RI
change in sE-Selectin levels.
Nevertheless, no difference of the sE-Selectin levels was established during the 24h
SC
CRRT period in our study. The fact that, in our study, sE-Selectin levels were
NU
sensitive enough to follow the changes of MAP, where, at the same time, did not show to correlate with the deteriorating microcirculation, indicates that the
MA
microcirculation effect (observed in the defined time period) is potentially mediated by a process not specifically influenced or marked by the circulating levels of
D
sE-Selectin. On the other hand, it cannot be excluded that the microcirculation may
TE
have been already influenced by a systemic inflammatory response (indicated by the sE-Selectin levels that were elevated before RRT onset), while further deterioration
AC CE P
probably involves other mechanisms. Here, the results of our study implicate the CRRT implementation.
In the same direction, study results did not indicate correlation of microcirculation with changes of global hemodynamic variables (such as MAP), which is not unexpected given that global hemodynamic parameters are influenced by many factors and are not specific and sensitive variables to follow tissue oxygenation; this is probably the reason that the goals of hemodynamic optimization in the ICU patients are a constant subject of debate (Vincent & De Backer, 2004 and Rampal et al., 2010). Microcirculatory disorders over the 24hours CRRT may be seen as part of the ‘dialytrauma’. This concept recognizes and bands together all the potential
ACCEPTED MANUSCRIPT CRRT-related adverse effects, which may have negative impact on recovery of critically ill patients (Maynar Moliner et al., 2012). Although the mortality rates of
PT
ICU patients receiving CRRT are high (50-80%) (Fonseca Ruiz et al., 2011) no association with ‘dialytrauma’ has been confirmed yet. Microcirculation evaluation is
RI
a dynamic approach which might provide a window to critical central vascular bed
SC
[assessment of subcutaneous dermal capillaries are representative of microvascular supply to heart (Pignocchino et al., 1994) and kidney (Economides et al., 2004)] and
NU
might contribute to CRRT individualization with prevention of its hemodynamic and
MA
metabolic complications (Selby et al., 2012).
So far, two studies evaluating potential changes of microcirculation during CRRT in
D
ICU did not observe significant alterations (Ruiz et al., 2010 and Veenstra et al.,
TE
2014). Differences in patient population, prescribed treatment, and timing of follow
AC CE P
up may explain the differences compared with the results of our study. In conclusion, our findings may simply show that microcirculation is stressed in promoting solute clearance during hemodiafiltration. This view is in accord with the report of Kong & Farrington (2002) who suggested that two levels of dialysis are going on, one at the capillary membrane and another at the dialyzer membrane and if one intend to increase dialysis efficiency should not only concentrate in the dialysis process within the dialyzer, but harness microcirculation as well. Limitations A limitation of our study is the small number of patients included, which hampers the generalization of the study results; however it offers information to guide new research in the field. Other limitations are the lack of baseline microcirculation measurement at ICU admission and the lack of core temperature recordings;
ACCEPTED MANUSCRIPT assessments that would potentially better elucidate the microcirculation alterations
AC CE P
TE
D
MA
NU
SC
RI
PT
during CRRT.
ACCEPTED MANUSCRIPT Conclusions The present study assessed the impact of CRRT on microcirculation and added
PT
information regarding the CRRT- related adverse effects in critically ill patients, or
RI
‘dialytrauma’. In a multidisciplinary ICU, deterioration of microcirculation parameters during the first 24hours of CVVHDF and a correlation between changes in
SC
endothelial function with the marked reduction of sCysC were noted. Moreover,
NU
microcirculation alterations correlated with dose in patients with diabetes. Further research will clarify the potential of microcirculation imaging technologies, as CRRT
MA
prognostic and/or monitoring tools.
D
Acknowledgments
TE
The present study was partly founded by a grant from the Special Account
AC CE P
for Research Grants of the National and Kapodistrian University of Athens, Greece.
ACCEPTED MANUSCRIPT References 1. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P. Acute renal failure -definition, outcome measures, animal models, fluid therapy and information
PT
technology needs: the Second International Consensus Conference of the Acute
RI
Dialysis Quality Initiative (ADQI) Group. Crit Care 8: R204-212, 2004.
SC
2. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, Schein RM, Sibbald WJ. Definitions for sepsis and organ failure and guidelines for the
NU
use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care
MA
Medicine. 1992. Chest 136(5 Suppl): e28, 2009.
D
3. Buemi M, Allegra A, Aloisi C, Corica F, Alonci A, Ruello A, Montalto G,
TE
Frisina N. Cold pressor test raises serum concentrations of ICAM-1, VCAM-1, and E-selectin in normotensive and hypertensive patients. Hypertension
AC CE P
30:845-847, 1997.
4. Clec'h C, Darmon M, Lautrette A, Chemouni F, Azoulay E, Schwebel C, Dumenil AS, Garrouste-Orgeas M, Goldgran-Toledano D, Cohen Y, Timsit JF. Efficacy of renal replacement therapy in critically ill patients: a propensity analysis. Crit Care 16: R236, 2012. 5. Creteur J, Carollo T, Soldati G, Buchele G, De Backer D, Vincent JL. The prognostic value of muscle StO2 in septic patients. Intensive Care Med 33: 1549-1556, 2007. 6. Economides PA, Caselli A, Zuo CS, Sparks C, Khaodhiar L, Katsilambros N,Horton ES, Veves A. Kidney oxygenation during water diuresis and
ACCEPTED MANUSCRIPT endothelial function in patients with type 2 diabetes and subjects at risk to develop diabetes. Metabolism 53: 222-227, 2004. 7. Fonseca Ruiz NJ, Castro DP, Guerra AM, Saldarriaga FM, Hernandez JD.
PT
Renal injury study in critical ill patients in accordance with the new definition
RI
given by the Acute Kidney Injury Network. J Crit Care 26: 206-212, 2011.
SC
8. Gerovasili V, Tripodaki E, Karatzanos E, Pitsolis T, Markaki V, Zervakis D, Routsi C, Roussos C, Nanas S. Short-term systemic effect of electrical muscle
NU
stimulation in critically ill patients. Chest 136: 1249-1256, 2009.
MA
9. Gibney N, Hoste E, Burdmann EA, Bunchman T, Kher V, Viswanathan R, Mehta RL, Ronco C. Timing of initiation and discontinuation of renal
D
replacement therapy in AKI: unanswered key questions. Clin J Am Soc Nephrol
TE
3: 876-880, 2008.
AC CE P
10. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med 13: 818-829, 1985. 11. Kokkoris S, Pipili C, Grapsa E, Kyprianou T, Nanas S. Novel Biomarkers of Acute Kidney Injury in the General Adult ICU: A Review. Ren Fail 35: 579-591, 2013.
12. Kong CH, Farrington K. Harnessing the microcirculation to increase dialysis efficiency. Nephron 92: 951, 2002. 13. Manetos C, Dimopoulos S, Tzanis G, Vakrou S, Tasoulis A, Kapelios C, Agapitou
V,
Ntalianis
A,
Terrovitis
J,
Nanas
S.
Skeletal
muscle
microcirculatory abnormalities are associated with exercise intolerance,
ACCEPTED MANUSCRIPT ventilatory inefficiency, and impaired autonomic control in heart failure. J Heart Lung Transplant 30: 1403-1408, 2011. 14. Maynar Moliner J, Honore PM, Sanchez-Izquierdo Riera JA, Herrera M,
Spapen
HD.
Handling
continuous
renal
replacement
PT
Gutierrez
RI
therapy-related adverse effects in intensive care unit patients: the dialytrauma
SC
concept. Blood Purif 34: 177-185, 2012.
15. Myers DE, Cooper CE, Beilman GJ, Mowlem JD, Anderson LD, Seifert RP,
NU
Ortner JP. A wide gap second derivative NIR spectroscopic method for
MA
measuring tissue hemoglobin oxygen saturation. Adv Exp Med Biol 578: 217-222, 2006.
D
16. Nanas S, Gerovasili V, Renieris P, Angelopoulos E, Poriazi M, Kritikos K,
TE
Siafaka A, Baraboutis I, Zervakis D, Markaki V, Routsi C, Roussos C.
AC CE P
Non-invasive assessment of the microcirculation in critically ill patients. Anaesth Intensive Care 37: 733-739, 2009. 17. Nejat M, Pickering JW, Walker RJ, Endre ZH. Rapid detection of acute kidney injury by plasma cystatin C in the intensive care unit. Nephrol Dial Transplant 25: 3283-3289, 2010. 18. Pignocchino P, Conte MR, Scarnato S, Grande A. [Study of cutaneous microcirculation using the laser-Doppler method in syndrome X]. Cardiologia 39: 193-197, 1994. 19. Rampal T, Jhanji S, Pearse RM. Using oxygen delivery targets to optimize resuscitation in critically ill patients. Curr Opin Crit Care 16:244-249, 2010.
ACCEPTED MANUSCRIPT 20. Ruiz C, Hernandez G, Godoy C, Downey P, Andresen M, Bruhn A. Sublingual microcirculatory changes during high-volume hemofiltration in hyperdynamic septic shock patients. Crit Care 14: R170, 2010.
PT
21. Selby NM, McIntyre CW. Predicting and managing complications of renal
RI
replacement therapy in the critically ill. Blood Purif 34: 171-176, 2012.
SC
22. Siafaka A, Angelopoulos E, Kritikos K, Poriazi M, Basios N, Gerovasili V, Andreou A, Roussos C, Nanas S. Acute effects of smoking on skeletal muscle spectroscopy. Chest 131:
NU
microcirculation monitored by near-infrared
MA
1479-1485, 2007.
23. Smith CW. Potential significance of circulating E-selectin. Circulation
D
95:1986-1988, 1997.
TE
24. Tripodaki ES, Tasoulis A, Vasileiadis I, Vastardis L, Skampas N,
AC CE P
Sakellaridis T, Argiriou M, Charitos C, Nanas S. Microcirculatory alterations after cardiopulmonary bypass as assessed with near infrared spectroscopy: a pilot study. Can J Anaesth 59: 620-621, 2012. 25. Vassiliou AG, Mastora Z, Orfanos SE, Jahaj E, Maniatis NA, Koutsoukou A, Armaganidis A, Kotanidou A. Elevated biomarkers of endothelial dysfunction/activation at ICU admission are associated with sepsis development. Cytokine 69:240-247, 2014. 26. Veenstra G, Scheenstra B, Koopmans M, Kingma WP, Buter H, Boerma EC. Microcirculatory perfusion derangements during continuous hemofiltration with fixed dose of ultrafiltration in stabilized intensive care unit patients. J Crit Care 29: 478-481, 2014.
ACCEPTED MANUSCRIPT 27. Vincent JL, De Backer D. Oxygen transport-the oxygen delivery controversy. Intensive Care Med 30:1990-1996, 2004. 28. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H, The SOFA (Sepsis-related Organ Failure
PT
Reinhart CK, Suter PM, Thijs LG.
RI
Assessment) score to describe organ dysfunction/failure. On behalf of the
SC
Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 22: 707-710, 1996.
NU
29. Xing K, Murthy S, Liles WC, Singh JM. Clinical utility of biomarkers of
MA
endothelial activation in sepsis--a systematic review. Crit Care 2012;16(1):R7. 30. Zonneveld R, Martinelli R, Shapiro NI, Kuijpers TW, Plötz FB, Carman CV.
D
Soluble adhesion molecules as markers for sepsis and the potential
AC CE P
18:204, 2014.
TE
pathophysiological discrepancy in neonates, children and adults. Crit Care
ACCEPTED MANUSCRIPT Highlights CRRT adversely affected microcirculation in the critically ill.
•
Microcirculation worsened more with a higher CRRT effect.
•
Microcirculation monitoring might contribute to CRRT individualization.
AC CE P
TE
D
MA
NU
SC
RI
PT
•
ACCEPTED MANUSCRIPT Tables
19/9
Age (years)
67.5±17
APACHE II
19 ±7
SOFA
9.8 ± 2.2
SC
Reason for ICU admission 15 (53.6%)
Cardiovascular
6 (21.4%)
Respiratory
7 (25%)
NU
Post-operative
19 (68%)
MA
Sepsis RIFLE No
5 (17.8%)
AC CE P
Diabetes
TE
Failure
D
Risk Injury
RI
Gender (Male/Female)
PT
Table 1. Patient characteristics (N=28)
9 (32.4%) 13 (46.8%) 1 (3.6%) 8 (28.5%)
ACCEPTED MANUSCRIPT
H6
H24
p-value*
MAP (mmHg)
82.3± 18.5
91.6 ± 15
84.7 ± 21.3
0.150
Periph. Temp. (°C)
36.6± 1.3
36.2 ± 1.3
36 ± 1.1
0.020
Lac (mmol/L)
3.2 ± 5.4
4.4 ± 6.8
4.8 ±7.3
Hb (g/dL)
9.5 ± 2
9±2
Norepinephrine (µg/kg/h)
0.65 ± 0.66
0.68 ± 0.72
sCr (mg/dL)
2.5 ± 1.5
2±1
sCysC (mg/L)
2.7 ± 0.8
StO2 (%)
76.5±12.5
OCR (%/min)
-12 ± 7
EF (%/sec)
2.25 ± 1.44
RI
SC
0.015 0.004
0.81 ± 0.86
0.280
1.7 ± 0.9
<0.0001
2.2 ± 0.6
1.8 ± 0.8
<0.0001
75 ± 11
70 ±16
0.040
-11 ± 7
-9 ± 4
0.270
2.1 ± 1.8
1.6 ±1.4
0.020
NU
8±2
D
TE
PT
H0
MA
Table 2. RRT characteristics before RRT initiation (H0), at six (H6) and 24 hours (H24) during RRT (N=28).
AC CE P
Data are expressed as mean ± SD
Definition of abbreviations: RRT, Renal replacement therapy, MAP, Mean Arterial Pressure; Periph Temp, peripheral temperature; Lac, Lactate; sCr, serum creatinine; sCysC, serum cystatin C; StO2, tissue oxygen saturation; OCR Oxygen consumption rate; EF, endothelial function * p -value for statistical significance of repeated measures ANOVA set at <0.05
ACCEPTED MANUSCRIPT
Hb > 7 g/dL (N=20)
76.1±12.1
76.7±12.4
StO2 (%)
at H6
74.8±10.4
StO2 (%)
at H24
69.6±15.7
RI
before RRT
Hb ≤7 g/dL (N=8)
75.5±10.8
SC
StO2 (%)
PT
Table 3. Microcirculatory parameters in patients with hemoglobin levels more or less than 7 g/dL after 24 hours of RRT.
70.3±16.2 -12.8±6
NU
OCR(%/min)before RRT -11.9±5.6 -10.2±6
OCR (%/min) at H24
-10±3
EF (%/sec) before RRT
1.8±1
2.2±1.5
EF (%/sec) at H6
1.6±1.2
2±1.6
EF (%/sec) at H24
1.3±0.9
TE
D
MA
OCR (%/min) at H6
-10.4±6.3 -9.8±2.7
1.5±1.1
AC CE P
Data are expressed as mean ± SD; p-value was not statistically significant in any of the comparisons between the two subgroups. In the subgroup of patients with Hb ≤ 7.0g/dL at 24 h, StO2 and EF significantly decreased during the monitoring period (H0-H6-H24) (p= 0.030 and p=0.035 respectively). In the subgroup of patients with Hb > 7.0 g/dL StO2 and EF also significantly decreased during the monitoring period (p= 0.034 and p=0.038 respectively). Definition of abbreviations: RRT, Renal replacement therapy; Hb, hemoglobin; StO2, tissue oxygen saturation; OCR, Oxygen consumption rate; EF, endothelial function; H0, measurements recorded before RRT initiation; H6, measurements recorded at six hours of RRT; H24, measurements recorded at 24 hours of RRT
PT
ACCEPTED MANUSCRIPT
Table 4. Microcirculatory parameters in patients with peripheral temperature more or less than 35.5 °C after six hours of RRT.
StO2 (%)
at H6
74.7±10.4
StO2 (%)
at H24
69.5±15.1
75.4±11.7 70.3±15.8 -11.8±5.9
-10.3±6.3
-10.7±6.7
OCR (%/min) at H24
-8.8±3.1
-8.6±3.4
EF (%/sec) before RRT
2±0.8
2.2±1.4
1.8±1.3
1.9±1.6
1.3±0.8
1.5±1.2
EF (%/sec) at H6
AC CE P
EF (%/sec) at H24
D
OCR (%/min) at H6
TE
MA
OCR(%/min)before RRT -11.5±5.4
76.2±12.4
SC
75.9±12.1
before RRT
NU
StO2 (%)
Temp > 35.5 oC (N=16)
RI
Temp ≤ 35.5 °C (N=12)
Data are expressed as mean ± SD; p-value was not statistically significant in any of the comparisons between the two subgroups. In the subgroup of patients with Temperature ≤ 35.5 °C after six hours of RRT, StO2 and EF significantly decreased during the monitoring period (H0-H6-H24) (p= 0.040 and p=0.038 respectively). In the subgroup of patients with Temperature > 35.5 °C, StO2 and EF also significantly decreased during the monitoring period (p= 0.037 and p=0.040 respectively). Definition of abbreviations: RRT, Renal replacement therapy; Temp, peripheral temperature; StO2, tissue oxygen saturation; OCR, Oxygen consumption rate; EF, endothelial function; H0, measurements recorded before RRT initiation; H6, measurements recorded at six hours of RRT; H24, measurements recorded at 24 hours of RRT