Cortical Venous Redness Represents Tissue Circulation Status in Patients With Moyamoya Disease
Background and Purpose—Venous oxygen saturation (SO2) is measured in medical fields to assess tissue circulation insufficiency. This study aimed to elucidate the use of a cortical venous redness measurement to evaluate hemodynamic changes during revascularization surgery for patients with moyamoya disease.
Methods—In this retrospective case-series analysis, we first quantitatively measured and correlated SO2 and R intensity of 24-bit color digital red–green–blue pictures of blood samples from 3 volunteers. Subsequently, based on intraoperative digital pictures of 29 patients with moyamoya disease, we measured the R intensities of a cortical vein near the anastomosis site before and after anastomosis. Cerebral blood flow (CBF) at the site was measured using a single-photon emission computed tomography before and 1 to 3 days after surgery. Venous R intensity and CBF were measured twice by 4 raters, and their correlations were examined using generalized linear mixed effect model and linear regression analysis.
Results—A strong linear correlation was found between blood R intensity and its SO2 (coefficients, 0.522; 95% confidence interval, 0.364–0.680, using generalized linear mixed effect model). Venous R intensity before the anastomosis was not correlated with preoperative CBF (coefficients, 0.000352; 95% confidence interval, −0.000369 to 0.00107, by generalized linear mixed effect); however, the increases in venous R intensity after anastomosis were correlated with postoperative increases in CBF (R2, 0.367; 95% confidence interval, 0.116–0.618 to 0.548; 95% confidence interval, 0.331–0.764, by linear regression analysis).
Conclusions—Cortical venous redness represented impaired CBF and could be a useful parameter for assessing hemodynamic changes during revascularization surgery.
Venous oxygen content depends on the balance of tissue oxygen supply and its consumption; venous oxygen saturation (SO2) intensity has been used as a parameter to evaluate tissue circulation insufficiency of its upstream organs. SO2 of the vena cava and jugular vein reflects the circulatory condition of the whole body and brain, respectively.1,2 Similarly, venous SO2 of the cerebral cortex is speculated to represent the circulatory condition of its perfused region. On the basis of this hypothesis, we sought to determine whether cortical venous redness represents its SO2 and becomes a parameter for assessing hemodynamic change during revascularization surgery for moyamoya disease (MMD).
This retrospective case-series analysis was approved by the institutional review board of our hospital.
Quantification of Blood Color and SO2
We first examined the association between venous color and its SO2. Both arterial and venous blood samples were collected from 3 volunteers and were mixed in various proportions (5:1–1:5). After taking 24-bit color digital pictures of the samples, their red (R), green (G), and blue (B) signal intensities were quantitatively measured using the histogram function of the Photoshop software (Adobe). Thus, the blood color was dispersed and digitalized to R, G, and B intensities with values that range from 0 to 255. The correlation between blood SO2 measured by blood gas analyzer and its R, G, and B intensities was assessed.
Operative Procedure and Management
We enrolled consecutive adult patients with MMD who underwent superficial temporal artery–middle cerebral artery anastomosis at our hospital between June 2010 and August 2016. Surgery was performed under total intravenous anesthesia with propofol and fentanyl. Arterial carbon dioxide pressure was maintained at 35 to 45 mm Hg.
Before surgery, cerebral blood flow (CBF) was quantitatively measured using 123I-N-isopropyl-p-iodoamphetamine or 99mTc-ethyl cysteinate dimer single-photon emission computed tomography. Postoperative single-photon emission computed tomography images were obtained within 3 days of surgery. After fusing single-photon emission computed tomography images with postoperative magnetic resonance or computed tomographic angiograms, raters 1 and 2 (J.S. and Y.A.), who were blinded to the operative findings, set 20 mm2 region of interest at the anastomosis site and ipsilateral cerebellar hemisphere (internal standard) and measured their radioisotope count twice. Regional CBF was calculated as the radioisotope count ratio against the internal standard, and the change in CBF (ΔCBF) was calculated as postoperative CBF/preoperative CBF.3
Quantification of Cortical Venous Color
Intraoperative digital video recording was performed using a microsurgical system (Contavas and OPMI PENTERO 900; Carl Zeiss, Germany). By reviewing intraoperative digital pictures pre- and post-anastomosis, raters 3 and 4 (Y.Y. and S.N.), who were blinded to the CBF values, selected a cortical vein near the anastomosis point that demonstrated the most drastic changes in redness. Venous R intensity was measured twice by the same way as the measurement of blood R intensity (Figure 1). The change in venous R intensity (ΔR) was calculated as postanastomosis R intensity/preanastomosis R intensity.
Associations between numeric data were assessed by a generalized linear mixed effect model and linear regression analysis. Intra- and inter-rater reliabilities were assessed by an intrarater correlation coefficient. Data of raters 1 and 3 and raters 2 and 4 were paired to analyze the association between CBF and R intensity. Analyses were performed using R version 3.3.2 (R Foundation for Statistical Computing, Vienna, Austria).
Correlation Between Blood R, G, and B Intensities and SO2
Blood R intensity was linearly correlated with its SO2 (coefficients, 0.522; 95% confidence interval [CI], 0.364–0.680, by generalized linear mixed effect model; R2, 0.838; 95% CI, 0.712–0.964, to R2, 0.980; 95% CI, 0.963–0.997, by linear regression analysis; Figure IA in the online-only Data Supplement). Neither G nor B intensity was correlated with SO2 (coefficients, 1.30; 95% CI, −3.89 to 6.50 and coefficients, 0.565; 95% CI, −0.808 to 1.90, by generalized linear mixed effect model, respectively; Figure IB and IC in the online-only Data Supplement).
Intra- and Interrater Reliability for Regional CBF and Venous R Intensity
Intra-rater reliability of CBF, ΔCBF, venous R intensity, and ΔR showed good agreement (intrarater correlation coefficient, 0.748; 95% CI, 0.528–0.874, to intrarater correlation coefficient, 0.971; 95% CI, 0.939–0.981; Table I in the online-only Data Supplement). Interrater reliability of CBF and the other 3 parameters showed moderate (intrarater correlation coefficient, 0.475; 95% CI, 0.127–0.718, to intrarater correlation coefficient, 0.658; 0.390–0.825) and good (intrarater correlation coefficient, 0.653; 95% CI, 0.378–0.823, to intrarater correlation coefficient, 0.914; 95% CI, 0.823–0.959) agreements, respectively.
Correlation Between Venous R Intensity and Hemodynamic Parameters
During the study period, we performed 29 revascularization surgeries. Patient details are summarized in Table. Venous R intensity before anastomosis and preoperative CBF ranged from 40.0 to 218 and 0.55 to 1.2, respectively. There was no significant correlation between them (coefficient, 0.000352; 95% CI, −0.000368 to 0.00107, by the generalized linear mixed effect model; Figure 1A). ΔR and ΔCBF ranged from 0.95 to 1.90 and 0.67 to 2.3, respectively, and there was a significant correlation (coefficient, 1.12; 95% CI, 0.895–1.34, by the generalized linear mixed effect model; R2, 0.367; 95% CI, 0.116–0.618, to R2, 0.548; 95% CI, 0.331–0.764, by linear regression analysis; Table II in the online-only Data Supplement; Figure 2B).
We demonstrated that cortical venous redness indicates hemodynamic changes during revascularization surgery for MMD. MMD is a steno-occlusive cerebrovascular disease. Narrowing of the internal carotid artery reduces cerebral perfusion pressure, which is compensated by increase in cerebral blood volume and oxygen extraction fraction (OEF).4 Revascularization surgery conversely decreases OEF and cerebral blood volume via increasing cerebral perfusion pressure. When we observe these mechanisms from the perspective of venous SO2, reduction of OEF increases venous SO2. As assumed, ΔCBF significantly correlated with ΔR (Figure 2B). Furthermore, cortical venous R intensity before anastomosis would correlate with preoperative CBF; however, they did not (Figure 2A). One of the reasons of this could be because raw blood R value did not accurately represent its SO2 because of variability of original blood redness among individuals (Figure IA in the online-only Data Supplement). Uncertainty of the raw value of R intensity and CBF, which may be because of differences in the location of region of interest among raters, could have also negatively affected our statistical calculations (Table I in the online-only Data Supplement). Improving accuracy of venous SO2 measurement using novel techniques, such as pulse oximetry, may resolve these limitations.
To date, various monitoring techniques, such as fluorescence angiography and Doppler flow meter, have been applied to evaluate hemodynamic changes during surgery.5,6 The principle of these modalities is mainly based on blood flow assessment; however, venous redness reflects tissue oxygen metabolism and can provide complemental information. For example, an area with a dark cortical vein represents a severe hemodynamic impairment with an elevated OEF (misery perfusion) and can be a valid indication for revascularization. Conversely, area with reddish vein will represent deceased OEF (luxury perfusion) because of a pathological condition such as cerebral infarction and may not need revascularization. Marked venous reddening after anastomosis would represent hyperperfusion syndrome, which needs strict blood pressure control (Figure II in the online-only Data Supplement).7 Thus, the intraoperative assessment of venous redness can provide available information for the management of surgery.
Venous redness is a potential parameter that can be used to assess cerebral hemodynamics during revascularization surgery for MMD.
Dr Machida conceived and designed the study. Dr Machida, Dr Nakano, Dr Ishige, Dr Ono, and Dr Fujikawa performed the surgery and managed perioperative course. S. Maru, Dr Shimada, and Dr Akaogi measured cerebral blood flow. Dr Nakano and Dr Yoshida measured venous R intensity. Dr Machida, Dr Higuchi, and Dr Ishige analyzed data. Dr Machida and Dr Ono drafted the article.
Sources of Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Portions of this work were published in the Japanese journal "Surgery for Cerebral Stroke" 2015;43:448–453.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.116.015991/-/DC1.
- Received September 4, 2016.
- Revision received March 8, 2017.
- Accepted March 14, 2017.
- © 2017 American Heart Association, Inc.
- Hamano E,
- Kataoka H,
- Morita N,
- Maruyama D,
- Satow T,
- Iihara K,
- et al
- Kawamata T,
- Kawashima A,
- Yamaguchi K,
- Hori T,
- Okada Y
- Machida T,
- Ono J,
- Nomura R,
- Fujikawa A,
- Nagano O,
- Higuchi Y