Edited by: Amre Nouh, Hartford Hospital and University of Connecticut, USA
Reviewed by: Olimpia Mihaela Carbunar, Baptist Hospital, USA; Christoph Stretz, Hartford Hospital and University of Connecticut, USA; Mohammed Arif Hussain, Hartford Hospital, USA
Specialty section: This article was submitted to Neurocritical and Neurohospitalist Care, a section of the journal Frontiers in Neurology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Anemia is accepted among critically ill patients as an alternative to elective blood transfusion. This practice has been extrapolated to head injury patients with only one study comparing the effects of mild anemia on neurological outcome. There are no studies quantifying microcirculation during anemia. Experimental studies suggest that anemia leads to cerebral hypoxia and increased rates of infarction, but the lack of clinical equipoise, when testing the cerebral effects of transfusion among critically injured patients, supports the need of experimental studies. The aim of this study was to quantify cerebral microcirculation and the potential presence of axonal damage in an experimental model exposed to normovolaemic anemia, with the intention of describing possible limitations within management practices in critically ill patients. Under non-recovered anesthesia, six Merino sheep were instrumented using an intracardiac transeptal catheter to inject coded microspheres into the left atrium to ensure systemic and non-chaotic distribution. Cytometric analyses quantified cerebral microcirculation at specific regions of the brain. Amyloid precursor protein staining was used as an indicator of axonal damage. Animals were exposed to normovolaemic anemia by blood extractions from the indwelling arterial catheter with simultaneous fluid replacement through a venous central catheter. Simultaneous data recording from cerebral tissue oxygenation, intracranial pressure, and cardiac output was monitored. A regression model was used to examine the effects of anemia on microcirculation with a mixed model to control for repeated measures. Homogeneous and normal cerebral microcirculation with no evidence of axonal damage was present in all cerebral regions, with no temporal variability, concluding that acute normovolaemic anemia does not result in short-term effects on cerebral microcirculation in the ovine brain.
Anemia and its pathophysiological implications have been investigated in recent years (
In light of this controversy, this novel study aimed to directly quantify regional microvascular blood flow (RMBF) using cytometric counting of color-coded microspheres with the potential neurological injury assessed using amyloid precursor protein (APP) staining of brain tissue of sheep exposed to normovolemic anemia.
All experimental procedures were approved by the Animal Ethics Committee of the Queensland University of Technology and University of Queensland Committee and complied with local guidelines. Sheep were selected as the preferred animal model because of their cerebral anatomical similarities with humans, specifically within the gyrencephalic structure, allowing better examination of gray–white matter. In addition, the hemoglobin dissociation curve in sheep is comparable to that of human with a considerable amount of experimental studies in ovine neurosciences. A triple lumen central line (Cook Medical INC., QLD, Australia) and two 16-Fr introducer sheaths were placed in the right internal jugular (RIJ) vein in a convenience sample of six Merino sheep wethers weighing 65 ± 6.01 kg. Via the central line, the sheep were placed under general anesthesia with an initial bolus of 5 mg/kg ketamine and a maintenance infusion between 0.5 and 1 mg/kg/h. Sedation was achieved with a combined infusion of midazolam (0.5 mg/kg/h), fentanyl (10 mcg/kg/h), and alfaxalone (6 mg/kg/h). This is a common veterinarian anesthetic regimen that has stable cardiovascular effects. Hydration was maintained with an infusion of Hartmann’s solution at a rate up to 2 mL/kg/h titrated to a central venous pressure (CVP) of 6–10 mmHg. Cardiovascular monitoring included cardiac output and vascular resistances via a Swan–Ganz catheter, as in previous models (
To avoid self-transfusion from the ovine spleen (
Acute normovolaemic anemia was achieved by sequential blood extractions from the indwelling arterial catheter performed simultaneously with isovolaemic saline infusions. The aim was to achieve a 30% reduction of baseline hemoglobin to reproduce the fall in hemoglobin concentration that was achieved in the TRICC trial. The targeted hemoglobin was monitored using cardiac index, blood pressure, systemic vascular resistances, and arterial blood gas sampling to ensure stable and normal lactate levels, every 15 min.
Following anesthesia, a transeptal catheter was inserted into the left atrium (LA) under intracardiac echocardiography (ICE) surveillance. Two 11-Fr Terumo sheaths located in the RIJ allowed for the insertion of the intracardiac ultrasound probe (Acuson AcuNav™ probe, California®) and the transeptal catheter (Mullins TS introducer, Medtronic®). Echocardiography images were obtained using an Acuson Sequoia C512 scanner (Siemens, CA, USA). Transeptal puncture and insertion of a pigtail catheter into the LA followed previously described methods (
Randomly assigned color-coded microspheres (E-Z TRAC; Interactive Medical Technology, Los Angeles, CA, USA) were injected hourly via the LA pigtail catheter using the following protocol. Six different colors (
Approximately, five million color-coded spheres of one color were injected at each hour with the color of the microspheres randomly assigned at each injection time, to minimize selection biases and to enable the quantification of flow at each defined time points. Each sheep had a different microsphere color injected at defined time points to assess temporal changes in RMBF by anatomical area. As per the manufacturer’s recommendations, the microspheres were maintained at room temperature, were not exposed to sunlight, heat, or vibration, and were mixed manually to minimize foaming and avoid non-uniform concentrations of microspheres throughout each injection.
After 5 h of continuous monitoring and microsphere injection, sheep were euthanized under non-recovered anesthesia with a bolus injection of 0.5 mL/kg of sodium pentobarbitone. After confirmation of death, the brain was harvested by craniotomy, weighed, and then fixed in 10% formalin for 3 weeks.
Brains were harvested using a deep incision of the skin and underlying tissue planes between the first and second occipital vertebral body with the intention of sectioning the spinal cord at that level. A round reciprocating saw was used to create a wide bitemporal bone incision. Rectangular bone sections of ~5 cm were made anteriorly to the bitemporal incision to the frontal sinuses. These bone sections were removed with retractors with simultaneous dissection of the dura. Once the brain was fully exposed, the olfactory bulbs, optic chiasm, tentorium, and cranial nerves were sectioned to liberate the brain from the cranium.
A novel design for tissue sampling was developed to capture the specific anatomical regions of interest (Figure
Samples of skin, gut, kidney, spleen, and heart were harvested to demonstrate the systemic distribution of the microspheres. The spleen was sampled to demonstrate splenic infarct due to successful arterial spleen ligation and the absence of microspheres (Figure
Cytometric counts of the microspheres were performed using a validated technique (
This cytometric analysis was performed at Interactive Medical Technology (IMT), Los Angeles, CA, USA.
Immunohistochemistry analysis was performed at the neuropathology laboratory, Royal Brisbane and Women’s Hospital, QLD, Australia, using a Leica Novolink Polymer Detection Systems Kit (Leica Microsystems Pty Ltd., North Ryde, 2113 Australia) as per the manufacturer’s instructions. Paraffin was removed from the sections through a series of xylene immersions and re-hydrations. Antigen retrieval was carried out using Leica BOND ER1 solution. Endogenous peroxidase was neutralized. Sections were incubated with a protein block. The primary antiserum made up in Leica BOND Antibody Diluent was applied to the sections.
Immunohistochemistry analysis was applied to the sheep brains for the baseline presence and location of APP antibodies. APP antibody staining was used because it is considered to be a very early marker of neuronal damage (
Regional microvascular blood flow was measured at baseline with no anemia (corresponding to time zero – T0) and at four follow-up times following the induction of anemia (T1–T4 corresponding to first to the fourth hour post intervention). A repeated-measure mixed regression model was used to examine differences in RMBF associated with anemia, and a mixed model with a random intercept was used for each sheep to control for repeated results from the same sheep.
To examine the variability in RMBF, all results were plotted using box-plots. We also calculated the percentage of the total variation in the mixed models due to the between-sheep variability. A high percentage (above 50%) indicated that the variance in RMBF was due to differences between sheep, whereas a lower percentage indicated that the differences between sheep were small. The analysis was performed used R software version 3.0.2.
Despite a fall in hemoglobin to levels between 6 and 7 g/dL (a 30% of the baseline hemoglobin in each animal) with normovolaemia, cerebral RMBF throughout all anatomical areas of interest was maintained in a physiological range after correction to conventional units (mL/100 g/min) during the 4 h of studying time. However, subject number 3 had a significant increase in RMBF at 4 h (T4) with otherwise similar RMBF values at earlier time points (Figure
Box-plots (Figure
The effect of anemia on RMBF during the 4 h of study time is presented in Table
Tissue | Mean | Lower | Upper | Ratio | |
---|---|---|---|---|---|
AI | 0.045 | −0.166 | 0.256 | 0.676 | 7 |
BI | 0.053 | −0.150 | 0.258 | 0.609 | 19 |
AR | 0.075 | −0.203 | 0.354 | 0.6 | 3 |
BR | 0.067 | −0.186 | 0.322 | 0.601 | 29 |
C | 0.050 | −0.148 | 0.247 | 0.619 | 28 |
D | 0.077 | −0.221 | 0.375 | 0.613 | 11 |
Slice | 0.082 | −0.189 | 0.353 | 0.553 | 26 |
Skin | −0.042 | −0.101 | 0.016 | 0.164 | 97 |
Spleen | 0.000 | −0.001 | 0.001 | 0.989 | 42 |
Kidney | −1.371 | −2.505 | −0.229 | 0.0248 | 30 |
Heart | 0.089 | −0.508 | 0.690 | 0.769 | 17 |
The cardiovascular responses to acute normovolaemic anemia are presented in Table
Hours post intervention | T 0 | Goal: reduction of 30% baseline Hbl | T1 | T2 | T3 | T4 |
---|---|---|---|---|---|---|
Baseline time | First hour post intervention | Second hour post intervention | Third hour post intervention | Fourth hour post intervention | ||
CO (L) | 4.8 | 4.1 | 2.7 | 2.5 | 2.5 | |
SV02 (%) | 75 | 71 | 46 | 55 | 50 | |
SVR (dynes s/cm/m2) | ||||||
Hbl (g/dL) | 10.0 | 7.6 | 7.8 | 8.1 | ||
CPP | 100 | 80 | 70 | 65 | 65 | |
PTiO2 | 17.3 | 8.2 | 10 | 5.4 | 6.1 | |
CO | 5 | 4.5 | 3.6 | 3.6 | 3.7 | |
SV02 (%) | 85 | 75 | 71 | 75 | 76 | |
SVR (dynes s/cm/m2) | 1445 | 1682 | 1865 | 1817 | 1886 | |
Hbl | 7.8 | 8.1 | 8.1 | 8.1 | ||
CPP | 90 | 84 | 84 | 75 | 83 | |
PTiO2 | 2.65 | 2.89 | 3.26 | 4.90 | 5.21 | |
CO | 4.3 | 4.0 | 4.7 | 4.2 | 4.3 | |
SV02 (%) | 66 | 70 | 69 | 67 | 67 | |
SVR (dynes s/cm/m2) | 1250 | 1060 | 1240 | 1270 | 1280 | |
Hbl | 7.9 | 8.1 | 7.9 | 7.9 | ||
CPP | 122 | 118 | 118 | 107 | 106 | |
PTiO2 | 25 | 43 | 45 | 42 | 44 | |
CO | 4.1 | 3.9 | 3.2 | 2.7 | 3.2 | |
SV02 (%) | 62 | 62 | 55 | 49 | 56 | |
SVR (dynes s/cm/m2) | 2245 | 2258 | 2460 | 2622 | 2367 | |
Hbl | 7.5 | 6.9 | 7.1 | 7.0 | ||
CPP | 109 | 110 | 93 | 100 | 80 | |
PTiO2 | 26 | 16 | 12 | 13.6 | 14.1 | |
CO | 3.4 | 3.3 | 3.7 | 3.0 | 3.8 | |
SV02 (%) | 77 | 72 | 66 | 66 | 71 | |
SVR (dynes s/cm/m2) | 2226 | 2133 | 1678 | 1601 | 1228 | |
Hbl | 6.7 | 7.5 | 7.5 | 7.0 | ||
CPP | 100 | 82 | 93 | 92 | 92 | |
PTiO2 | 5.7 | 2.15 | 2 | 1.8 | 2.1 | |
CO | 4.0 | 2.6 | 3.3 | 3.9 | 3.8 | |
SV02 (%) | 83 | 70 | 71 | 75 | 75 | |
SVR (dynes s/cm/m2) | 1490 | 3196 | 2064 | 1694 | 1682 | |
Hbl | 7.9 | 7.5 | 7.5 | 7.2 | ||
CPP | 99 | 98 | 90 | 85 | 85 | |
PTiO2 | 37 | 5.8 | 9.5 | 10.2 | 7.4 |
The cerebral PTiO2, CPP, and CO responses to acute anemia are described in Table
Subject number | Region of interest | H&E staining | APP staining |
---|---|---|---|
Sheep 01 | AL | Normal | 0 |
AR | Normal | 1 | |
BL | Normal | 0 | |
BR | Microglial activation | 2 | |
C | Focal hemorrhagic necrosis | 2 | |
D | Normal | 0 | |
Sheep 03 | AL | Normal | 0 |
AR | Normal | 0 | |
BL | Normal | 0 | |
BR | Normal | 0 | |
C | Normal | 0 | |
D | Small perivascular petechial hemorrhage | 1 | |
Sheep 04 | AL | Normal | 0 |
AR | Normal | 0 | |
BL | Normal | 0 | |
BR | Small perivascular petechial hemorrhage at edge, artifact? | 0 | |
C | Small perivascular petechial hemorrhage at edge, artifact? | 0 | |
D | Normal | 0 | |
Sheep 05 | AL | Normal | 0 |
AR | Normal | 0 | |
BL | Normal | 0 | |
BR | Normal | 0.5 | |
C | Normal | 0 | |
D | Normal | 0 | |
Sheep 06 | AL | Localized focus of acute meningitis and SAH in depth of sulcus | 0 |
AR | Normal | 0 | |
BL | Normal | 0 | |
BR | Normal | 0 | |
C | Focal hemorrhage near deep gray nuclei | 2 | |
D | Normal | 0 | |
Sheep 07 | AL | Normal | 0 |
AR | Normal | 0 | |
BL | Normal | 0 | |
BR | Normal | 0 | |
C | Normal | 0 | |
D | Normal | 0 |
This study demonstrates the maintenance of stable cerebral microcirculation in Merino sheep during and after 4 h of sustained normovolaemic anemia. After the achievement of a 30% reduction of baseline hemoglobin, as in the TRICC trial, despite a reduction in cardiac output and an increase in vascular resistances, RMBF remained unchanged in both intra- and inter-subject comparisons. This observation suggests that despite the cardiovascular impact of anemia, cerebral autoregulation in an otherwise intact brain is perhaps the mechanism involved in the preservation of microcirculation throughout different anatomical regions; however, specific measurement of cerebral autoregulation was not the focus of this study.
Normovolaemic anemia alone in the short term does not appear to have objective effect on brain tissue histo-anatomy. However, cerebral microcirculation and cerebral autoregulation have heterogeneous behaviors that can lead to perfusion mismatch. This characteristic may be magnified in situations where normal perfusion is compromised, such as in head injury and hemorrhagic shock (
The experimental strategy used for the direct measurement of RBMF and tissue sampling in this study has proven to be feasible. Quantification of RMBF in specific brain regions allows for intra- and inter-subject comparisons and is a significant improvement on previous approaches using predefined brain thickness sectioning (
Limitations of this study are inherent to the experimental nature of the study. Regional microcirculatory blood flow cytometric count in targeted regions of the brain, before and after an insult is not a feasible method in humans. Cerebral cytometric count of microspheres is mainly an experimental concept although it shares the same physical principle of flow cytometry extensively applied in diagnostic medicine (
This study achieved predefined aims and followed a rigorous methodology, demonstrating that short-term normovolemic anemia does not impair cerebral RMBF at specific anatomical regions, a model not feasible in the clinical arena, as an adjunct data to future head injury studies.
Acute normovolaemic anemia replicating a restrictive transfusion strategy in ovine models without head injury does not impair cerebral microcirculation or induce axonal damage.
Primary roles: JB, Study design, surgical procedures and data collection, data interpretation, and manuscript preparation. KC, Histopathology analysis. KD, Laboratory support and manuscript preparation. SD, surgical procedures and data collection. DP and OR, intracardiac echography and transeptal catheterization. LG, manuscript preparation. AB, statistical analysis. JP, RB and JF, data interpretation and manuscript preparation.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
This work was supported by the Australian Defence Health Foundation 2012 Research grant – ADHREC number: 2012/1175375. JF acknowledges the support provided by Queensland Health Research Fellowship.