- Email: [email protected]
The monitoring of microvascular liver blood flow changes during ischemia and reperfusion using laser speckle contrast imaging
The monitoring of microvascular liver blood flow changes during ischemia and reperfusion using laser speckle contrast imaging Chong Hui Li, Hong Dong Wang, Jian Jun Hu, Xin Lan Ge, Ke Pan, Ai Qun Zhang, Jia Hong Dong PII: DOI: Reference:
S0026-2862(14)00070-3 doi: 10.1016/j.mvr.2014.04.010 YMVRE 3423
To appear in:
Microvascular Research
Received date: Revised date: Accepted date:
16 November 2013 14 April 2014 24 April 2014
Please cite this article as: Li, Chong Hui, Wang, Hong Dong, Hu, Jian Jun, Ge, Xin Lan, Pan, Ke, Zhang, Ai Qun, Dong, Jia Hong, The monitoring of microvascular liver blood flow changes during ischemia and reperfusion using laser speckle contrast imaging, Microvascular Research (2014), doi: 10.1016/j.mvr.2014.04.010
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 The monitoring of microvascular liver blood flow changes during ischemia
T
and reperfusion using laser speckle contrast imaging
IP
Chong Hui Li, Hong Dong Wang, Jian Jun Hu, Xin Lan Ge, Ke Pan, Ai Qun Zhang, Jia Hong
SC R
Dong*
MA
PLA Medical College，Beijing, China
NU
Department and Institute of Hepatobiliary Surgery, Chinese PLA General Hospital, Chinese
*Corresponding author: Department of Hepatobiliary Surgery, Chinese PLA General Hospital,
TE
AC
CE P
[email protected]
D
28 Fuxing Road, Beijing 100853 China. Tel.: 86-010-66938030; E-mail address:
1
ACCEPTED MANUSCRIPT Abstract Objective: The recovery of microvascular liver blood flow (LBF) after ischemia is an
T
important determinant of the degree of hepatocellular injury. Laser speckle contrast imaging
IP
(LSCI) was recently suggested to be a suitable instrument for monitoring the LBF. This study
SC R
was designed to evaluate LSCI in monitoring the LBF changes during liver ischemia and reperfusion (IR).
Methods: A rat model with 120-min ischemia and 60-min reperfusion to 90% of the liver
NU
(entire liver except the caudate lobe, which was kept as portal blood bypass) was used.
The
MA
LBF of the sham operation (SO) group and the IR group was measured with LSCI at the following time points: before ischemia (Baseline), 5 min after the start of ischemia (I-5 min), 5 min before the end of ischemia (I-115 min) and 5 and 60 min after the start of reperfusion
D
The reproducibility among different rats or repeated measurements,
TE
(R-5 min and R-60 min).
the liver histopathology, the liver biological zero (BZ) and the influence of liver movement on
CE P
the LSCI measurements were investigated. Results: The entire exposed liver surface after laparotomy was suitable for full-view LSCI Establishing many circular or oval regions of interest (ROIs) on the LSCI flux
AC
imaging.
image was a simple and convenient method for calculating and comparing the LBF of different ROIs and different liver lobes. There was good-to-moderate intra-individual and inter-individual reproducibility for the LSCI measurements of the LBF in the rats of the SO group. In the IR group, the total blood inflow occlusion resulted in a notable drop of the LBF from baseline (P<0.05) that remained for the 120 min of ischemia.
The LBF decreased
further after reperfusion (P<0.05), reflecting the IR-induced liver microcirculation dysfunction.
The histopathological examination revealed severe hepatic sinus congestion
and damaged hepatocytes in the IR group.
The no flow BZ and liver movement contributed
to the LBF values.
2
ACCEPTED MANUSCRIPT Conclusions: LSCI technology is a simple, convenient and accurate method for the real-time monitoring of microvascular LBF changes during ischemia and reperfusion, regardless of the This finding suggests the possible
T
contribution of biological zero and liver movement.
SC R
IP
application of LSCI for monitoring the microvascular LBF changes intraoperatively.
Keywords: microvascular liver blood flow, laser speckle contrast imaging, ischemia and
NU
reperfusion, liver microcirculation
MA
Abbreviations: laser speckle contrast imaging (LSCI); ischemia and reperfusion (IR); liver blood flow (LBF); biological zero (BZ); regions of interest (ROI)；time of interest (TOI); laser
AC
CE P
TE
D
speckle perfusion unit (LSPU); arbitrary unit (AU)
3
ACCEPTED MANUSCRIPT Introduction The Pringle maneuver (temporary occlusion of the hepatoduodenal ligament) is widely This
T
employed during hepatectomy to reduce intraoperative blood loss (Dixon et al., 2005).
IP
maneuver inevitably results in ischemia and subsequent reperfusion injury, which could cause The ischemic time and
SC R
significant postoperative complications (Lesurtel et al., 2009).
restoration of liver perfusion is an important determinant of the degree of hepatocellular injury because microcirculatory collapse corresponds to a profound reduction in tissue It would be helpful to monitor the microvascular liver
NU
oxygenation (Vollmar et al., 1994).
MA
blood flow (LBF) changes during ischemia and reperfusion (IR) intraoperatively with an appropriate and convenient instrument in a surgical environment. Laser speckle contrast imaging (LSCI) is a recently marketed technique that is based High frame rate LSCI provides
TE
D
on speckle contrast analysis (Basak et al., 2012).
non-contact full-field imaging over wide areas with excellent spatial and temporal resolutions
CE P
and theoretically combines the advantages of laser Doppler flowmetry (LDF) and laser Doppler imaging (LDI) (Roustit et al., 2010; Puissant et al., 2013). Applications of LSCI
AC
include pre-clinical studies of neurological disorders and clinical applications, including dermatological (Kernick and Shore, 2000), neurosurgical and endoscopic studies (Boas and Dunn, 2010; Dunn, 2012). LSCI was recently used to assess the LBF during sequential liver inflow occlusions, and the method was able to produce reproducible real-time blood perfusion measurements of hepatic microcirculation that correlated well with sidestream dark field imaging-derived sinusoidal blood flow velocity measurements (Sturesson et al., 2013).
Compared with the
established techniques for LBF measurements, LSCI has the advantage of non-contact measurement over a large surface with high-speed data acquisition (Richter et al., 2010).
In
Sturesson’s (2013) study, the liver blood inflow was occluded for only 3 min, and there was
4
ACCEPTED MANUSCRIPT no obvious IR injury to the liver.
The LBF measurements were obtained during a period of
apnea to minimize movement artifacts.
Apnea affects blood pressure and circulation and is It is well known that the laser signal never
T
not regularly used in small animal experiments.
IP
reaches zero in skin recordings under a situation of arrested flow (tourniquet ischemia)
cells in venules.
SC R
because of spontaneous Brownian motion of macromolecules and the remaining red blood The remaining non-zero signal is called biological zero (BZ) (Kemick et al.,
1999). The liver BZ in LSCI measurement has not been addressed.
NU
The present study was designed to evaluate the application of LSCI in monitoring the
MA
microvascular LBF changes during ischemia and reperfusion in a condition of normal anesthesia and spontaneous breathing in rats.
The experimental factors influencing the
AC
CE P
TE
D
accuracy of the LSCI measurement of LBF were analyzed.
5
ACCEPTED MANUSCRIPT Methods Animals
T
Male Wistar rats weighing 240 to 270 g were obtained from the Experimental Animal The rats were
IP
Center of the Academy of Military Medical Science (Beijing, China).
allowed food and water ad libitum.
SC R
maintained at 24°C under pathogen-free conditions with a 12/12-hour dark/light cycle and All the experiments performed in this study were
NU
approved by the Animal Research Committee of Chinese PLA General Hospital.
MA
Research design
The rats were divided into three experimental groups: the sham operation group (SO, n=10), the ischemia and reperfusion group (IR, n = 10) and the liver biological zero group
TE
D
(BZ, n=5). Ischemia and reperfusion was performed in the SO and IR groups, and the LBF was measured with LSCI at 5 time points (Fig. 1): 5 min after laparotomy, which was taken as
CE P
the baseline of LBF (Baseline); 5 min after the start of ischemia (I-5 min); 5 min before the end of 120 min of ischemia (i.e., the start of reperfusion) (I-115 min); and 5 min and 60 min The study of liver biological zero signals
AC
after the start of reperfusion (R-5 min, R-60 min).
and the effect of liver movement on LSCI measurements were performed in the BZ group. The left liver lobe was ligated at the base (Fig. 1B) to occlude the blood inflow and outflow. After the LSCI measurement, the rat was assigned to the induction of apnea under continuous LSCI monitoring.
The surgical procedures A rat model with IR to 90% of the liver was used (Dong et al., 2002) (Fig. 1A). In this rat model, the caudal lobe, which represented 10% of the liver, was not occluded but was kept as a passage of the portal blood to reduce intestinal congestion during liver ischemia (also see
6
ACCEPTED MANUSCRIPT the supplemental materials). After overnight fasting but with free access to water, the rats were anaesthetized with 1.5% isoflurane inhalation and placed on a thermostatically
T
controlled heating pad that maintained the body temperature at 37°C throughout the operation
IP
and monitoring periods.
SC R
After a midline laparotomy, the common hepatic pedicles of the left and median lobes and the right lobe were gently isolated and temporally clamped with two lengths of “0” surgical suture tied in bowknots. In the SO group, the hepatic pedicles were isolated but not
NU
clamped. A continuous suture was used to close the peritoneal skin, and the rat was returned
MA
to a cage. After 120 min of ischemia, the peritoneal skin was reopened, and reperfusion of the liver was achieved by untying the bowknots, followed by re-closure of the peritoneal skin. In LDF and LSCI research, the BZ of a specific organ or tissue is defined as the signal
TE
D
obtained in the absence of vascular flow. When the hepatic BZ for LSCI was measured in the BZ group, the pedicle of the left lobe of the liver was first ligated with 3-0 silk sutures to
CE P
occlude the blood inflow and outflow of the left lobe (Fig. 1B). The LBF of the left lobe was then measured with LSCI and taken as the BZ of the rat liver. Three of the BZ rats were given
AC
deep ether anesthesia to induce apnea. After approximately 20 seconds of apnea, the ether was removed, and the rats were given fresh air to promote recovery. This procedure was used to evaluate the influence of respiration-induced liver movement on LSCI measurements. Ether was used to induce deep anesthesia because of its rapid anesthesia and recovery.
Laser speckle contrast imaging procedures The LBF was assessed using the moorFLPI-2 Full-Field Laser Perfusion Imager (Moor Instruments, Axminster, UK). The auto focus function was used to automatically confirm the distance between the scan head and the rat liver, and the distance was 17.11.3 cm. scanning area was approximately 57 cm.
The
The sampling frequency was 25 Hz, and the
7
ACCEPTED MANUSCRIPT image acquisition rate was one frame per second in the high-resolution (752580 pixels) mode. The duration of the recording was 15 seconds.
There was no direct sunlight or
T
infrared radiation, and the room temperature was maintained at 26°C. During the LSCI
IP
measurement, the skin on both sides of the xiphoid was pulled upwards with two surgical
SC R
retractors to reduce the influence of respiratory diaphragmatic movements.
Region of interest (ROI) settings
NU
After the laparotomy, the right median lobe (RM), the left median lobe (LM) and the The LSCI imaging was undertaken by direct
MA
lower part of the left lobe (L) were exposed.
laser exposure of the three liver parts at their original positions to maintain the normal liver The surfaces of the three liver lobes were not at the same level, but the
D
blood flow (Fig. 2).
TE
majority of areas of the LSCI image were homogeneous, except for some small blue artifact spots or areas resulting from specular reflection of the liver surface. A total of 6 to 10 circular
CE P
or oval ROIs approximately 25 mm2 each were placed on the LSCI flux image of each rat liver, and as large of an area as possible for the analysis was used (Fig. 2).
The ROIs were
AC
simultaneously displayed on the photo image and the color image. The artifact spots or areas where the flux value was not accurately measured were excluded from the ROIs. This ROI setting method facilitates a comparison of the flux value of different ROIs and different liver lobes.
Data processing and analysis The manufacturer's software (moorFLPI-2 Review V4.0, Moor Instruments, Axminster, UK) was used to analyze the images. The LSCI quantifies the perfusion in the arbitrary laser speckle perfusion unit (LSPU).
The software enabled the mean LSPU to be calculated
over a defined ROI and time of interest (TOI). The flux in the LSPU of each ROI was
8
ACCEPTED MANUSCRIPT averaged over 15 seconds of recording (i.e., the TOI was 15 seconds) (Rousseau et al., 2011). Next, the flux of all the ROIs was averaged to obtain the LBF of each rat liver.
The LBF
T
was expressed as a percentage of the baseline value when needed.
IP
The offline analysis consisted of calculations of the reproducibility and heterogeneity,
SC R
which were expressed as the coefficient of variation (CV), calculated as the standard deviation (SD) divided by the mean, with ≤10%, 10-25% and ≥25% considered good,
NU
moderate and poor reproducibility, respectively (Richter et al., 2010; Rousseau et al., 2011; Tew et al., 2011). The CV for the inter-individual reproducibility was calculated as the SD for
MA
the baseline LBF of each group divided by the mean. The CV for the intra-individual reproducibility was calculated as the SD for the repeatedly measured LBF at the five time
D
points of each rat in the SO group divided by the mean. The CV for the spatial heterogeneity
CE P
liver divided by the mean.
TE
among different ROIs of each rat was calculated as the SD for the LBF of all ROIs in each rat
Histopathological assessment
AC
For the SO and IR groups, the rats were sacrificed after the final LSCI measurements, and the liver tissues were removed and fixed in formalin. The paraffin embedded liver tissues were cut into 5 μm sections and stained with hematoxylin and eosin for the histopathological assessment. Light microscopy examination was performed to evaluate the IR injury.
Statistical analysis The data were expressed as the mean ± SD. The normality of the LBF values for the SO and IR groups were tested with the Kolmogorov-Smirnov test and the Shapiro-Wilk test (P>0.05). The means of the LBF values recorded at each time point within a group were compared with the ANOVA test for repeated measurements. Comparisons of the individual 9
ACCEPTED MANUSCRIPT time points between the two groups were made by Student’s t-test. The differences were considered statistically significant at P<0.05. The statistical analyses were accomplished using
AC
CE P
TE
D
MA
NU
SC R
IP
T
SPSS version 17.0 (SPSS Inc., Chicago, IL, USA).
10
ACCEPTED MANUSCRIPT Results The baseline measurement of the LBF in rats
T
Under these experimental LSCI monitoring and analysis conditions, the LBF values of the
The baseline LBF of the two groups did not show a significant
SC R
LSPU (n=10), respectively.
IP
SO group and IR group at baseline were 713.99106.07 LSPU (n=10) and 728.0788.92
difference (P>0.05).
The CV for the baseline LBF of the SO and IR groups were 14.86%
and 12.21%, respectively, exhibiting moderately acceptable inter-individual variability.
In
NU
the SO group, each rat was repeatedly measured five times at different time points, and the
MA
CV of the repeated measurements for the 10 rats was 4.62%2.44% (1.93%-9.36%),
D
indicating good intra-individual reproducibility.
TE
Ischemia and reperfusion-induced LBF changes measured with LSCI In the IR group, the total blood inflow occlusion to 90% of the liver (except the caudal
CE P
lobe) resulted in a marked drop in the LBF from baseline (P<0.05) (Fig. 3). As illustrated in Fig.3B, the LBF changes of all the 10 rats in IR group were given. In some rats, the LBF
AC
values of I-115 min were lower than the values of I-5 min, and in other rats, these values were higher. But there was no statistically significant change between the mean LBF values of the two time points. The LBF further decreased after the start of reperfusion (P<0.05), reflecting that reperfusion induced microcirculation dysfunction.
At 60 min after the start of
reperfusion, the LBF recovered slightly but remained much lower than baseline, showing severe IR injury.
The LBF in the SO group did not show significant changes at any
monitored time points (P>0.05).
When the raw arbitrary LSCI value of each rat liver was
expressed as a percentage of its baseline value, as shown in Fig. 3D, the changes of the LBF following ischemia and reperfusion displayed a similar trend as the raw arbitrary LSCI value in the SO and IR groups.
11
ACCEPTED MANUSCRIPT
The measured spatial heterogeneity
T
The spatial heterogeneity of the LSCI measurement was analyzed by comparing the
IP
variability of different ROIs in each rat in the IR group. The CV values are shown in Fig. 4.
SC R
The measured spatial heterogeneity increased following the occlusion of liver blood inflow and the start of reperfusion, but only the CV value of the LBF at R-60 min was significantly higher than that of other time points (P<0.05 vs. baseline, I-5 min, I-115 min and R-5 min),
NU
which reflected the heterogeneity of the LBF due to IR injury.
MA
A comparison of the LBF of the different liver lobes in the IR group found that the mean LBF of the LM was lower than the mean LBF of the RM and LM at baseline, but there was
D
no significant difference between the median lobe and the left lobes (L: 663.0163.87 LSPU, After
TE
LM: 739.9091.21 LSPU, RM: 745.82119.27 LSPU, L vs. LM and RM, P>0.05).
the occlusion of the blood inflow, the LBF of the left lobe was significantly lower than the
CE P
LBF of the right median and left median lobes (L: 289.1335.77 LSPU, LM: 370.7449.30 LSPU, RM: 332.0140.36 LSPU, L vs. LM and RM, P<0.05).
The LBF of the left lobe was
AC
lower than the right median and left median lobes at I-115 min (L: 265.3429.35 LSPU, LM: 349.3551.43 LSPU, RM: 318.0545.46 LSPU, L vs. LM and RM, P<0.05). There were no statistical significances between the LBF of the left lobe and the LBF of the right median and left median lobes at 5 min and 60 min after the start of reperfusion.
The effects of biological zero and movement of the liver on LSCI measurements The BZ was 244.0959.80 LSPU for the normal rat livers, which accounts for approximately 35% of the baseline LBF.
The remaining LBF signal during ischemia
accounts for 43%-45% of the baseline LBF. When subtracting the liver BZ from the
12
ACCEPTED MANUSCRIPT remaining LBF signal during ischemia, the residual LBF signal was 8%-10% of the baseline LBF.
As illustrated in Fig. 5A, the left lobe and left median lobe
IP
anesthesia-induced apnea.
T
The influence of movement on LSCI measurements was evaluated by deep
SC R
showed lower LBF values than the right lobe and right median lobe at the beginning of monitoring because the occlusion of the blood inflow and outflow of the left lobe stopped the blood supply to the left median lobe.
Following the inhalation of a high concentration of
NU
ether, the LBF values of the four liver lobes decreased gradually (Fig. 5C).
At
MA
approximately 2:30 (2.5 min after the start of the LSCI measurement), respiratory movement stopped, and the LBF graph line became smooth; the LBF of the right lobes dropped quickly. Following this period of apnea, all of the LBF graph lines reached their lowest levels.
The decrease in
TE
D
cessation of ether inhalation, the respiration and LBF gradually recovered.
After
the LBF in the left lobe, following the induction of apnea, represented the effect of
CE P
diaphragm-induced liver movement because there was no blood flow in the left lobe. liver movement led to an approximate 150 to 200 LSPU increase in the LBF value.
This
During
AC
apnea, the LBF value of the left lobe represented the actual BZ of the rat liver, and it was approximately 150 LSPU.
Apnea resulted in a significant decrease in the systemic and
microcirculation blood flow, and the LBF of the right lobe dropped quickly.
Because of the
difficulty in avoiding diaphragmatic movement and apnea-induced blood flow decreases at the same time, the value of the liver BZ consisted of the actual BZ and the movement-induced LBF increase.
Histopathological changes of the rat liver The histopathological analysis of liver tissues from the SO group revealed a normal liver lobule structure and clear and complete hepatic cords (Fig. 6). 13
There was no sinusoidal
ACCEPTED MANUSCRIPT congestion, expansion or necrosis of the hepatocytes.
In the liver tissues from rats that
experienced ischemia and reperfusion, the lobular structure remained intact, but some hepatic
Hepatic subcapsular sinus congestion and thrombin were
IP
and swelling were observed.
T
cord structures were not clear, and sinusoidal congestion and hepatocytes with karyopyknosis
AC
CE P
TE
D
MA
NU
SC R
observed.
14
ACCEPTED MANUSCRIPT Discussion The liver blood flow is dually supplied by the hepatic artery and the portal vein, and the LDF and intravital fluorescence microscopy have
T
hepatic sinusoid is the capillary network.
IP
frequently been used to investigate the microvascular LBF during ischemia and reperfusion
SC R
(Tawadrous et al., 2001; Nishida et al., 2000). Typically, LDF measures the LBF at a single point, resulting in large inter-site and inter-individual variability (Richter et al., 2010; Wheatley et al., 1993).
Using intravital fluorescence microscopy, it was found that hepatic
NU
microcirculatory perfusion failure was a determinant of liver dysfunction in warm Intravital fluorescence microscopy requires
MA
ischemia-reperfusion (Vollmar et al., 1994).
toxic fluorescent dyes for contrast enhancement and is not suitable for monitoring the intraoperative microvascular LBF in humans or large animals because of their marked liver
D
In this study, we evaluated the use of LSCI in the assessment of IR-induced
TE
thickness.
microvascular LBF changes in rats. The results showed that LSCI is a useful instrument to
decrease
that
CE P
measure the LBF intraoperatively. resulted
from
LSCI could effectively and accurately monitor the LBF
liver
blood
inflow
occlusion
and
reflected
the
AC
ischemia/reperfusion-induced liver microcirculation dysfunction. According to a recent study, increases in the "region of interest" and "time of interest" reduce the variability of blood flow measurements using LSCI (Rousseau et al., 2011).
This
study suggested that an ROI larger than 10 mm2 and a TOI longer than 1 second were required to reduce the variability in LSCI measurements. lobes were simultaneously monitored with LSCI.
In our study, three exposed liver
A circular or oval ROI of approximately
25 mm2 was selected as the appropriate size for the rat liver LSCI analysis.
By copying the
first ROI, 6 to 10 ROIs were placed on the LSCI image of each rat liver to cover almost all of the homogenous imaging area.
The LSCI measurements and ROI setting method in our
study could increase the region of interest while avoiding the artifact areas, and are
15
ACCEPTED MANUSCRIPT convenient for the comparison of the LBF values within different ROIs or different liver lobes. This is an alternative, simple method for monitoring the microvascular LBF, and has the
T
advantage of not requiring a breathing machine and apnea, as was necessary in a study by
IP
Sturesson C et al. (2013). The surfaces of the three exposed liver lobes were not at the same
SC R
level, but the liver LSCI images were primarily homogeneous except for some small blue artifact spots that resulted from the specular reflection. This observation supports the finding that the distance between the laser head and the skin does not influence the skin blood flow
NU
values recorded by LSCI (Mahé et al., 2011a).
MA
Preliminary studies showed that the hepatic microcirculation is homogenous with no differences occurring between the single lobes, especially when assessed using a tissue depth of 2-mm (Ladurner, 2009). We found that there was no significant difference in the LBF
TE
D
between the median lobe and the left lobe at baseline and during reperfusion.
However, the
LBF of the left lobe was significantly lower than the median lobe during ischemia.
These
CE P
results might reflect the different anatomical location of the left and median lobes, which might influence the compensatory response of liver perfusion, such as back flow from the
AC
hepatic vein (Martins and Neuhaus, 2007). Similar to LDF, although the measured LSCI microvascular LBF is not the arbitrary blood flow and is a relative value (Mahé et al., 2012), our results showed that the LBF of the rat liver was relatively stable at a specific anesthesia level, environmental conditions and instrumental measuring parameters. Good-to-moderate intra-individual and inter-individual reproducibilities were observed in the LBF measurements with LSCI for the baseline and repeated measurements in the SO group. These results demonstrated the possibility to compare the arbitrary LBF values from multiple measurements and multiple experimental groups without the need to depend on relative measurements. In the SO group, the LBF did not show a significant difference at each measuring time
16
ACCEPTED MANUSCRIPT point. In the IR group, the LBF decreased significantly after total blood inflow occlusion, and there was a further decrease in the LBF after the beginning of reperfusion, reflecting the In a study by
T
microcirculation dysfunction caused by hepatic ischemia-reperfusion injury.
IP
Sturesson C et al. (2013), the total liver blood inflow was only clamped for 5 min, and the
SC R
LBF could recover to the original level. By prolonging the ischemic time, as was performed in our study, the LBF could not recover to the original level because of severe IR injury. This finding was consistent with the results of previous studies with intravital fluorescence
NU
microscopy, which revealed that reperfusion injury of the liver after warm ischemia was
et al., 1994, 1996).
MA
characterized by impaired sinusoidal perfusion with a reduction of erythrocyte flux (Vollmar Our results from the liver tissue pathological analysis showed sinusoidal
congestion within the liver and under the liver capsule.
Intracellular edema and
TE
D
hepatocellular and sinusoidal endothelial cell injury formed the structural basis for the decline in the microvascular LBF (Pannen, 2002).
CE P
After the total liver blood inflow occlusion, the LBF decreased to approximately 45% of baseline. Theoretically, when the hepatic blood inflow is completely blocked, the
result.
AC
microvascular LBF should drop to zero. We hypothesize three possible explanations for this First, the laser speckle and laser Doppler technologies have a BZ phenomenon.
For
the laser Doppler measurements, the skin blood flow does not reach the value of zero when perfusion is absent because of the Brownian motion of macromolecules arising from the interstitial space (Kernick, et al., 1999).
A portion of this signal might be attributed to the
remaining red blood cells in the venules under the condition of arrested flow (tourniquet ischemia) (Kernick and Shore, 2000).
A recent skin blood flow study showed a higher BZ
with LSCI than with LDI (Millet et al., 2011).
Regarding the liver tissue, the origin of the
BZ with LSCI remains unknown, but possible sources for the liver BZ include the Brownian motion or other passive movements of fluid, macromolecules and red blood cells.
17
Residual
ACCEPTED MANUSCRIPT vasomotion of the hepatic sinusoidal was not observed.
The blood supply to the liver is
much richer than the supply to other organs, which indicates that when the blood inflow and
T
outflow are occluded, there is more blood remaining in the hepatic sinusoid, and the
At 5 min after the start of reperfusion, the LDF further decreased to
SC R
signals during ischemia.
IP
Brownian motion of the remaining red cells might contribute largely to the remaining LSCI
approximately 30% of the baseline, indicating that the BZ of a rat liver that undergoes IR was much lower than the BZ during ischemia because of the IR injury. This might be related to the
NU
microcirculation dysfunction due to hepatic sinusoidal congestion and hemagglutination. The regular movement of the liver
MA
Second, LSCI is susceptible to movement.
following the movement of the diaphragm creates a specific background signal in LSCI measurements.
To reduce the effect of respiratory motions, Sturesson C et al. (2011)
TE
D
measured the LBF during a state of apnea, but apnea significantly reduced the systemic and hepatic microcirculation blood flow.
In this study, we found that diaphragm-induced liver When
CE P
movement caused an approximate 150 to 200 LSPU increase in the LBF value.
comparing the LBF values of different rats, the anesthesia and respiratory motion should be
AC
controlled as much as possible. A group recently showed that movement-induced artifacts might be overcome by subtracting the back-scattered signal from an adhesive opaque surface adjacent to the ROI (Mahe et al., 2011b, 2013).
This method requires a specific signal
post-processing technique, and whether this method is suitable for LSCI measuring of microvascular LBF requires further study. Third, the backflow of the hepatic vein system might contribute to the high post-occlusion LBF value because the LBF of the no flow left liver lobe (i.e., BZ) is lower than the LBF under blood inflow occlusion (Smyrniotis et al., 2003).
The hepatic vein of the
left lobe in rats is longer than in the other lobes (Martins and Neuhaus, 2007), which might explain why the LBF of the left lobe is lower than the LBF of the median lobes during
18
ACCEPTED MANUSCRIPT ischemia and might support the contribution of hepatic vein backflow to the high LBF value after blood inflow occlusion. Multiple factors contribute to the high remaining LSCI value Some studies demonstrated that
T
when the hepatic blood inflow is completely blocked.
IP
subtracting the BZ from the raw arbitrary perfusion units did not affect the correlation
SC R
between the LSCI and LDI assessments of skin perfusion. However, subtracting the BZ introduced high variability into the baseline skin blood flux signal when recorded with LSCI
NU
or with LDI (CV=38.8% and 81.1%, respectively) (Millet et al.，2011). Because of the multiple contributing factors and high variability of the liver BZ, we do not recommend
MA
subtracting the BZ from the raw arbitrary perfusion units, particularly in ischemia and reperfusion studies.
D
Impaired hepatic microcirculation following extended hepatic resection or liver
TE
transplantation is a major cause of hepatocellular dysfunction.
Assessing the microvascular
CE P
LBF, especially intraoperatively and in real-time, has not been possible in clinical practice. In this study, we found that LSCI could accurately monitor the decrease of the LBF resulting from liver blood inflow clamping and could reflect the ischemia-reperfusion-induced liver
AC
microcirculation dysfunction after reperfusion.
These results suggest a possible application
of LSCI for monitoring the microvascular LBF intraoperatively.
The relationship between
the degree of ischemia-reperfusion-induced LBF changes and IR injury warrants further investigation.
19
ACCEPTED MANUSCRIPT Acknowledgments This work was supported by the Project of the National Natural Science Foundation of China
AC
CE P
TE
D
MA
NU
SC R
IP
T
(81271738) and the National Key Technology R&D Program of China (2012BAI06B01).
20
ACCEPTED MANUSCRIPT References 1. Basak, K., Manjunatha, M., Dutta, P.K.. 2012. Review of laser speckle-based analysis in
T
medical imaging. Med. Biol. Eng. Comput. 50(6), 547–558.
IP
2. Boas, D.A., Dunn, A.K.. 2010. Laser speckle contrast imaging in biomedical optics. J.
SC R
Biomed. Opt. 15(1), 011109.
3. Dixon E, Vollmer CM, Bathe OF, Sutherland F.. 2005. Vascular occlusion to decrease blood loss during hepatic resection. Am. J. Surg. 190(1):75-86.
NU
4. Dong, J.H., He, X.D., Li, K., Duan, H.C., Peng, Z.M., Cai, J.X.. 2002. Tolerance limit of
Pancreat. Dis. Int. 1(1), 57-62.
MA
rats to normothermic hepatic inflow occlusion under portal blood bypass. Hepatobiliary
5. Dunn, A.K.. 2012. Laser speckle contrast imaging of cerebral blood flow. Ann. Biomed.
TE
D
Eng. 40(2), 367-77.
6. Kernick, D,P,, Tooke, J.E., Shore, A.C.. 1999. The biological zero signal in laser Doppler
CE P
fluximetry-origins and practical implications. Pflugers. Arch. 437(4), 624-31. 7. Kernick, D.P., Shore, A.C.. 2000. Characteristics of laser Doppler perfusion imaging in
AC
vitro and in vivo. Physiol. Meas. 21(2), 333-40. 8. Ladurner, R., Feilitzsch, M., Steurer, W., Coerper, S., Königsrainer, A., Beckert, S.. 2009. The impact of a micro-lightguide spectrophotometer on the intraoperative assessment of hepatic microcirculation: a pilot study. Microvasc. Res. 77(3), 387-8. 9. Lesurtel, M., Lehmann, K., de Rougemont, O., Clavien, P.A.. 2009. Clamping techniques and protecting strategies in liver surgery. HPB (Oxford). 11(4),290-5. 10. Mahé, G., Haj-Yassin, F., Rousseau, P., Humeau, A., Durand, S., Leftheriotis, G., Abraham, P.. 2011. Distance between laser head and skin does not influence skin blood flow values recorded by laser speckle imaging. Microvasc. Res. 82(3), 439-42. 11. Mahé, G., Rousseau, P., Durand, S., Bricq, S., Leftheriotis, G., Abraham, P.. 2011b. Laser
21
ACCEPTED MANUSCRIPT speckle contrast imaging accurately measures blood flow over moving skin surfaces. Microvasc. Res. 81(2),183-8.
T
12. Mahé, G., Humeau-Heurtier, A., Durand, S., Leftheriotis, G, Abraham, P.. 2012.
SC R
imaging. Circ. Cardiovasc. Imaging, 5(1),155-63.
IP
Assessment of skin microvascular function and dysfunction with laser speckle contrast
13. Mahe, G., Abraham, P., Le Faucheur, A., Bruneau, A., Humeau-Heurtier, A., Durand, S.. 2013. Cutaneous microvascular functional assessment during exercise: a novel approach
NU
using laser speckle contrast imaging. Pflugers. Arch. 465(4), 451-8.
MA
14. Martins, P.N., Neuhaus, P.. 2007. Surgical anatomy of the liver, hepatic vasculature and bile ducts in the rat. Liver Int. 27(3),384-92.
15. Millet, C., Roustit, M., Blaise, S., Cracowski, J.L.. 2011. Comparison between laser
TE
D
speckle contrast imaging and laser Doppler imaging to assess skin blood flow in humans. Microvasc. Res. 82, 147–151.
CE P
16. Nishida, T., Ueshima, S., Kazuo, H., Ito, T., Seiyama, A., Matsuda, H.. 2000. Vagus nerve is involved in lack of blood reflow into sinusoids after rat hepatic ischemia. Am. J.
AC
Physiol. Heart Circ. Physiol. 278(5), H1565-70. 17. Puissant, C., Abraham, P., Durand, S., Humeau-Heurtier, A., Faure, S., 2013. Lefthériotis G, Rousseau P, Mahé G. Reproducibility of non-invasive assessment of skin endothelial function using laser Doppler flowmetry and laser speckle contrast imaging. PLoS One 8(4),e61320. 18. Pannen, B.H.. 2002. New insights into the regulation of hepatic blood flow after ischemia and reperfusion. Anesth. Analg. 94(6),1448-57. 19. Richter, S., Sperling, J., Kollmar, O., Menger, M.D., Schilling, M.K.. 2010. Laser Doppler flowmetry of hepatic microcirculation during Pringle's maneuver: determination of spatial and temporal liver tissue perfusion heterogeneity. Eur. Surg. Res.
22
ACCEPTED MANUSCRIPT 44(3-4),152-8. 20. Roustit, M., Millet, C., Blaise, S., Dufournet, B., Cracowski, J.L.. 2010. Excellent
T
reproducibility of laser speckle contrast imaging to assess skin microvascular reactivity.
IP
Microvascular. Research. 80(3), 505–511.
SC R
21. Rousseau, P., Mahé, G., Haj-Yassin, F., Dur and S., Humeau, A., Leftheriotis, G., Abraham, P.. 2011. Increasing the "region of interest" and "time of interest", both reduce the variability of blood flow measurements using laser speckle contrast imaging.
NU
Microvasc. Res. 82(1), 88-91.
MA
22. Smyrniotis, V., Kostopanagiotou, G., Lolis, E., Theodoraki, K., Farantos, C., Andreadou, I., Polymeneas, G., Genatas, C., Contis, J.. 2003. Effects of hepatovenous back flow on ischemic- reperfusion injuries in liver resections with the pringle maneuver. J. Am. Coll.
TE
D
197(6), 949-54.
23. Sturesson, C., Milstein, D.M., Post, I.C., Maas, A.M., van Gulik, T.M.. 2013. Laser
34-40.
CE P
speckle contrast imaging for assessment of liver microcirculation. Microvasc. Res. 87,
AC
24. Tawadrous, M.N., Zhang, X.Y., Wheatley, A.M.. 2001, Microvascular origin of laser-Doppler flux signal from the surface of normal and injured liver of the rat. Microvasc. Res. 62(3), 355-65. 25. Tew, G.A., Klonizakis, M., Crank, H., Briers, J.D., Hodges, G.J.. 2011, Comparison of laser speckle contrast imaging with laser Doppler for assessing microvascular function. Microvasc. Res. 82(3), 326-32. 26. Vollmar, B., Glasz, J., Leiderer, R., Post, S., Menger, M.D.. 1994, Hepatic microcirculatory perfusion failure is a determinant of liver dysfunction in warm ischemia-reperfusion. Am. J. Pathol. 145(6), 1421–1431. 27. Vollmar, B., Glasz, J., Post, S., Menger, M.D.. 1996. Role of microcirculatory 23
ACCEPTED MANUSCRIPT derangements in manifestation of portal triad cross-clamping-induced hepatic reperfusion injury. J. Surg. Res. 60, 49–54.
T
28. Wheatley, A.M., Zhao, D.. 1993. Intraoperative assessment by laser Doppler flowmetry
IP
of hepatic perfusion during orthotopic liver transplantation in the rat. Transplantation
AC
CE P
TE
D
MA
NU
SC R
56(6), 1315-8.
24
ACCEPTED MANUSCRIPT Figure Legends
A：Schematic drawing representing the ischemic model of 90% of the rat liver by
IP
study.
T
Fig. 1 Graphical illustration of the surgical operations on the rat livers and the protocol in the
SC R
blood inflow occlusion on the pedicles of the left lobe (L), left median lobe (LM), right median lobe (RM) and right lobe (R), with no occlusion of the caudate lobe (C), which B: The
NU
accounts for 8-10% of the liver weight and is kept as a passage of the portal blood.
left liver lobe is ligated to occlude the blood inflow and outflow used in liver biological zero C: The experimental procedures for LSCI measurements of ischemia and
MA
measurements.
reperfusion induced liver blood flow changes at 5 min after laparotomy (Baseline), 5 min
D
after the start of ischemia (Post-I), 5 min before the end of ischemia (Pre-R), 5 and 60 min
CE P
TE
after the start of reperfusion (Post-R, Post-R60).
Fig. 2 Example images of LSCI measurements of the rat liver marked with circular regions of interest (ROIs) on the flux image, photo image and color image.
In the flux images, the red
AC
areas correspond to the regions of high perfusion and high flux, and the blue regions indicate low perfusion and low flux.
Fig. 3 Changes in the LBF of the sham operation rats (the SO group) and the liver ischemia and reperfusion rats (the IR group) measured with LSCI at 5 min after laparotomy (Baseline), 5 min after the start of ischemia (I-5 min), 5 min before the end of ischemia (I-115 min), 5 and 60 min after the start of reperfusion (R-5 min, R-60 min). A, Representative LSCI images of the SO and IR groups at the indicated time points; B, The LBF changes in the 10 rats in the IR group; C, D, In the comparison of the LBF of the SO group and the IR group, the LBF is expressed as the raw arbitrary perfusion unit (LSPU) (C) or the percent of its 25
ACCEPTED MANUSCRIPT baseline level (D). *compared with the SO group and the Baseline of the IR group, P<0.05; #
T
compared with the I-5 min, I-115 min and R-60 min of the IR group, P<0.05.
IP
Fig. 4 The spatial heterogeneity of the LSCI measurements by comparing the variability The spatial
SC R
among different ROIs in the rats of the IR group at the indicated time points.
heterogeneity of the LBF increased following the ischemia and reperfusion. * P<0.05 vs.
NU
Baseline, I-5 min and I-115 min.
apnea and recovery conditions.
MA
Fig. 5 Representative LBF continuously monitored with LSCI of the rat under deep anesthesia, A, The flux images at the beginning of deep anesthesia (a),
start of apnea (b), end of apnea (c) and end of recovery (d).
ROIs 1-4 are located at the left,
TE
D
left median, right median and right liver lobes. B, A photograph showing the position of the rat under LSCI monitoring.
C, The representative line graph of the LBF of ROIs 1-4 under
Black arrow: the start of apnea; Red arrow: the end of apnea.
AC
ROI4-green line.
ROI1-black line, ROI2-red line, ROI3-blue line and
CE P
continuous LSCI monitoring.
Fig. 6 Histopathological changes of the rat liver after ischemia and reperfusion (200X).
A, B
from the sham operation group, showing the lobule structural integrity and normal hepatic cords and hepatocytes; C, D from the ischemia and reperfusion group, showing sinusoidal congestion and hepatocytes with karyopyknosis and necrosis. liver capsule.
26
B, D the liver tissue under the
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
27
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
28
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
29
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
30
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
31
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
32
ACCEPTED MANUSCRIPT Highlights: 1.
The first use of LSCI technology in monitoring hepatic microcirculation
Proposed a simple method for LSCI monitoring and analyzing of liver
IP
2.
T
dysfunction
First investigation of liver movement on LSCI measurement of liver
TE
D
MA
NU
microcirculation
CE P
4.
The first investigation of liver biological zero in LSCI measurement
AC
3.
SC R
microcirculation
33
S0026-2862(14)00070-3 doi: 10.1016/j.mvr.2014.04.010 YMVRE 3423
To appear in:
Microvascular Research
Received date: Revised date: Accepted date:
16 November 2013 14 April 2014 24 April 2014
Please cite this article as: Li, Chong Hui, Wang, Hong Dong, Hu, Jian Jun, Ge, Xin Lan, Pan, Ke, Zhang, Ai Qun, Dong, Jia Hong, The monitoring of microvascular liver blood flow changes during ischemia and reperfusion using laser speckle contrast imaging, Microvascular Research (2014), doi: 10.1016/j.mvr.2014.04.010
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 The monitoring of microvascular liver blood flow changes during ischemia
T
and reperfusion using laser speckle contrast imaging
IP
Chong Hui Li, Hong Dong Wang, Jian Jun Hu, Xin Lan Ge, Ke Pan, Ai Qun Zhang, Jia Hong
SC R
Dong*
MA
PLA Medical College，Beijing, China
NU
Department and Institute of Hepatobiliary Surgery, Chinese PLA General Hospital, Chinese
*Corresponding author: Department of Hepatobiliary Surgery, Chinese PLA General Hospital,
TE
AC
CE P
[email protected]
D
28 Fuxing Road, Beijing 100853 China. Tel.: 86-010-66938030; E-mail address:
1
ACCEPTED MANUSCRIPT Abstract Objective: The recovery of microvascular liver blood flow (LBF) after ischemia is an
T
important determinant of the degree of hepatocellular injury. Laser speckle contrast imaging
IP
(LSCI) was recently suggested to be a suitable instrument for monitoring the LBF. This study
SC R
was designed to evaluate LSCI in monitoring the LBF changes during liver ischemia and reperfusion (IR).
Methods: A rat model with 120-min ischemia and 60-min reperfusion to 90% of the liver
NU
(entire liver except the caudate lobe, which was kept as portal blood bypass) was used.
The
MA
LBF of the sham operation (SO) group and the IR group was measured with LSCI at the following time points: before ischemia (Baseline), 5 min after the start of ischemia (I-5 min), 5 min before the end of ischemia (I-115 min) and 5 and 60 min after the start of reperfusion
D
The reproducibility among different rats or repeated measurements,
TE
(R-5 min and R-60 min).
the liver histopathology, the liver biological zero (BZ) and the influence of liver movement on
CE P
the LSCI measurements were investigated. Results: The entire exposed liver surface after laparotomy was suitable for full-view LSCI Establishing many circular or oval regions of interest (ROIs) on the LSCI flux
AC
imaging.
image was a simple and convenient method for calculating and comparing the LBF of different ROIs and different liver lobes. There was good-to-moderate intra-individual and inter-individual reproducibility for the LSCI measurements of the LBF in the rats of the SO group. In the IR group, the total blood inflow occlusion resulted in a notable drop of the LBF from baseline (P<0.05) that remained for the 120 min of ischemia.
The LBF decreased
further after reperfusion (P<0.05), reflecting the IR-induced liver microcirculation dysfunction.
The histopathological examination revealed severe hepatic sinus congestion
and damaged hepatocytes in the IR group.
The no flow BZ and liver movement contributed
to the LBF values.
2
ACCEPTED MANUSCRIPT Conclusions: LSCI technology is a simple, convenient and accurate method for the real-time monitoring of microvascular LBF changes during ischemia and reperfusion, regardless of the This finding suggests the possible
T
contribution of biological zero and liver movement.
SC R
IP
application of LSCI for monitoring the microvascular LBF changes intraoperatively.
Keywords: microvascular liver blood flow, laser speckle contrast imaging, ischemia and
NU
reperfusion, liver microcirculation
MA
Abbreviations: laser speckle contrast imaging (LSCI); ischemia and reperfusion (IR); liver blood flow (LBF); biological zero (BZ); regions of interest (ROI)；time of interest (TOI); laser
AC
CE P
TE
D
speckle perfusion unit (LSPU); arbitrary unit (AU)
3
ACCEPTED MANUSCRIPT Introduction The Pringle maneuver (temporary occlusion of the hepatoduodenal ligament) is widely This
T
employed during hepatectomy to reduce intraoperative blood loss (Dixon et al., 2005).
IP
maneuver inevitably results in ischemia and subsequent reperfusion injury, which could cause The ischemic time and
SC R
significant postoperative complications (Lesurtel et al., 2009).
restoration of liver perfusion is an important determinant of the degree of hepatocellular injury because microcirculatory collapse corresponds to a profound reduction in tissue It would be helpful to monitor the microvascular liver
NU
oxygenation (Vollmar et al., 1994).
MA
blood flow (LBF) changes during ischemia and reperfusion (IR) intraoperatively with an appropriate and convenient instrument in a surgical environment. Laser speckle contrast imaging (LSCI) is a recently marketed technique that is based High frame rate LSCI provides
TE
D
on speckle contrast analysis (Basak et al., 2012).
non-contact full-field imaging over wide areas with excellent spatial and temporal resolutions
CE P
and theoretically combines the advantages of laser Doppler flowmetry (LDF) and laser Doppler imaging (LDI) (Roustit et al., 2010; Puissant et al., 2013). Applications of LSCI
AC
include pre-clinical studies of neurological disorders and clinical applications, including dermatological (Kernick and Shore, 2000), neurosurgical and endoscopic studies (Boas and Dunn, 2010; Dunn, 2012). LSCI was recently used to assess the LBF during sequential liver inflow occlusions, and the method was able to produce reproducible real-time blood perfusion measurements of hepatic microcirculation that correlated well with sidestream dark field imaging-derived sinusoidal blood flow velocity measurements (Sturesson et al., 2013).
Compared with the
established techniques for LBF measurements, LSCI has the advantage of non-contact measurement over a large surface with high-speed data acquisition (Richter et al., 2010).
In
Sturesson’s (2013) study, the liver blood inflow was occluded for only 3 min, and there was
4
ACCEPTED MANUSCRIPT no obvious IR injury to the liver.
The LBF measurements were obtained during a period of
apnea to minimize movement artifacts.
Apnea affects blood pressure and circulation and is It is well known that the laser signal never
T
not regularly used in small animal experiments.
IP
reaches zero in skin recordings under a situation of arrested flow (tourniquet ischemia)
cells in venules.
SC R
because of spontaneous Brownian motion of macromolecules and the remaining red blood The remaining non-zero signal is called biological zero (BZ) (Kemick et al.,
1999). The liver BZ in LSCI measurement has not been addressed.
NU
The present study was designed to evaluate the application of LSCI in monitoring the
MA
microvascular LBF changes during ischemia and reperfusion in a condition of normal anesthesia and spontaneous breathing in rats.
The experimental factors influencing the
AC
CE P
TE
D
accuracy of the LSCI measurement of LBF were analyzed.
5
ACCEPTED MANUSCRIPT Methods Animals
T
Male Wistar rats weighing 240 to 270 g were obtained from the Experimental Animal The rats were
IP
Center of the Academy of Military Medical Science (Beijing, China).
allowed food and water ad libitum.
SC R
maintained at 24°C under pathogen-free conditions with a 12/12-hour dark/light cycle and All the experiments performed in this study were
NU
approved by the Animal Research Committee of Chinese PLA General Hospital.
MA
Research design
The rats were divided into three experimental groups: the sham operation group (SO, n=10), the ischemia and reperfusion group (IR, n = 10) and the liver biological zero group
TE
D
(BZ, n=5). Ischemia and reperfusion was performed in the SO and IR groups, and the LBF was measured with LSCI at 5 time points (Fig. 1): 5 min after laparotomy, which was taken as
CE P
the baseline of LBF (Baseline); 5 min after the start of ischemia (I-5 min); 5 min before the end of 120 min of ischemia (i.e., the start of reperfusion) (I-115 min); and 5 min and 60 min The study of liver biological zero signals
AC
after the start of reperfusion (R-5 min, R-60 min).
and the effect of liver movement on LSCI measurements were performed in the BZ group. The left liver lobe was ligated at the base (Fig. 1B) to occlude the blood inflow and outflow. After the LSCI measurement, the rat was assigned to the induction of apnea under continuous LSCI monitoring.
The surgical procedures A rat model with IR to 90% of the liver was used (Dong et al., 2002) (Fig. 1A). In this rat model, the caudal lobe, which represented 10% of the liver, was not occluded but was kept as a passage of the portal blood to reduce intestinal congestion during liver ischemia (also see
6
ACCEPTED MANUSCRIPT the supplemental materials). After overnight fasting but with free access to water, the rats were anaesthetized with 1.5% isoflurane inhalation and placed on a thermostatically
T
controlled heating pad that maintained the body temperature at 37°C throughout the operation
IP
and monitoring periods.
SC R
After a midline laparotomy, the common hepatic pedicles of the left and median lobes and the right lobe were gently isolated and temporally clamped with two lengths of “0” surgical suture tied in bowknots. In the SO group, the hepatic pedicles were isolated but not
NU
clamped. A continuous suture was used to close the peritoneal skin, and the rat was returned
MA
to a cage. After 120 min of ischemia, the peritoneal skin was reopened, and reperfusion of the liver was achieved by untying the bowknots, followed by re-closure of the peritoneal skin. In LDF and LSCI research, the BZ of a specific organ or tissue is defined as the signal
TE
D
obtained in the absence of vascular flow. When the hepatic BZ for LSCI was measured in the BZ group, the pedicle of the left lobe of the liver was first ligated with 3-0 silk sutures to
CE P
occlude the blood inflow and outflow of the left lobe (Fig. 1B). The LBF of the left lobe was then measured with LSCI and taken as the BZ of the rat liver. Three of the BZ rats were given
AC
deep ether anesthesia to induce apnea. After approximately 20 seconds of apnea, the ether was removed, and the rats were given fresh air to promote recovery. This procedure was used to evaluate the influence of respiration-induced liver movement on LSCI measurements. Ether was used to induce deep anesthesia because of its rapid anesthesia and recovery.
Laser speckle contrast imaging procedures The LBF was assessed using the moorFLPI-2 Full-Field Laser Perfusion Imager (Moor Instruments, Axminster, UK). The auto focus function was used to automatically confirm the distance between the scan head and the rat liver, and the distance was 17.11.3 cm. scanning area was approximately 57 cm.
The
The sampling frequency was 25 Hz, and the
7
ACCEPTED MANUSCRIPT image acquisition rate was one frame per second in the high-resolution (752580 pixels) mode. The duration of the recording was 15 seconds.
There was no direct sunlight or
T
infrared radiation, and the room temperature was maintained at 26°C. During the LSCI
IP
measurement, the skin on both sides of the xiphoid was pulled upwards with two surgical
SC R
retractors to reduce the influence of respiratory diaphragmatic movements.
Region of interest (ROI) settings
NU
After the laparotomy, the right median lobe (RM), the left median lobe (LM) and the The LSCI imaging was undertaken by direct
MA
lower part of the left lobe (L) were exposed.
laser exposure of the three liver parts at their original positions to maintain the normal liver The surfaces of the three liver lobes were not at the same level, but the
D
blood flow (Fig. 2).
TE
majority of areas of the LSCI image were homogeneous, except for some small blue artifact spots or areas resulting from specular reflection of the liver surface. A total of 6 to 10 circular
CE P
or oval ROIs approximately 25 mm2 each were placed on the LSCI flux image of each rat liver, and as large of an area as possible for the analysis was used (Fig. 2).
The ROIs were
AC
simultaneously displayed on the photo image and the color image. The artifact spots or areas where the flux value was not accurately measured were excluded from the ROIs. This ROI setting method facilitates a comparison of the flux value of different ROIs and different liver lobes.
Data processing and analysis The manufacturer's software (moorFLPI-2 Review V4.0, Moor Instruments, Axminster, UK) was used to analyze the images. The LSCI quantifies the perfusion in the arbitrary laser speckle perfusion unit (LSPU).
The software enabled the mean LSPU to be calculated
over a defined ROI and time of interest (TOI). The flux in the LSPU of each ROI was
8
ACCEPTED MANUSCRIPT averaged over 15 seconds of recording (i.e., the TOI was 15 seconds) (Rousseau et al., 2011). Next, the flux of all the ROIs was averaged to obtain the LBF of each rat liver.
The LBF
T
was expressed as a percentage of the baseline value when needed.
IP
The offline analysis consisted of calculations of the reproducibility and heterogeneity,
SC R
which were expressed as the coefficient of variation (CV), calculated as the standard deviation (SD) divided by the mean, with ≤10%, 10-25% and ≥25% considered good,
NU
moderate and poor reproducibility, respectively (Richter et al., 2010; Rousseau et al., 2011; Tew et al., 2011). The CV for the inter-individual reproducibility was calculated as the SD for
MA
the baseline LBF of each group divided by the mean. The CV for the intra-individual reproducibility was calculated as the SD for the repeatedly measured LBF at the five time
D
points of each rat in the SO group divided by the mean. The CV for the spatial heterogeneity
CE P
liver divided by the mean.
TE
among different ROIs of each rat was calculated as the SD for the LBF of all ROIs in each rat
Histopathological assessment
AC
For the SO and IR groups, the rats were sacrificed after the final LSCI measurements, and the liver tissues were removed and fixed in formalin. The paraffin embedded liver tissues were cut into 5 μm sections and stained with hematoxylin and eosin for the histopathological assessment. Light microscopy examination was performed to evaluate the IR injury.
Statistical analysis The data were expressed as the mean ± SD. The normality of the LBF values for the SO and IR groups were tested with the Kolmogorov-Smirnov test and the Shapiro-Wilk test (P>0.05). The means of the LBF values recorded at each time point within a group were compared with the ANOVA test for repeated measurements. Comparisons of the individual 9
ACCEPTED MANUSCRIPT time points between the two groups were made by Student’s t-test. The differences were considered statistically significant at P<0.05. The statistical analyses were accomplished using
AC
CE P
TE
D
MA
NU
SC R
IP
T
SPSS version 17.0 (SPSS Inc., Chicago, IL, USA).
10
ACCEPTED MANUSCRIPT Results The baseline measurement of the LBF in rats
T
Under these experimental LSCI monitoring and analysis conditions, the LBF values of the
The baseline LBF of the two groups did not show a significant
SC R
LSPU (n=10), respectively.
IP
SO group and IR group at baseline were 713.99106.07 LSPU (n=10) and 728.0788.92
difference (P>0.05).
The CV for the baseline LBF of the SO and IR groups were 14.86%
and 12.21%, respectively, exhibiting moderately acceptable inter-individual variability.
In
NU
the SO group, each rat was repeatedly measured five times at different time points, and the
MA
CV of the repeated measurements for the 10 rats was 4.62%2.44% (1.93%-9.36%),
D
indicating good intra-individual reproducibility.
TE
Ischemia and reperfusion-induced LBF changes measured with LSCI In the IR group, the total blood inflow occlusion to 90% of the liver (except the caudal
CE P
lobe) resulted in a marked drop in the LBF from baseline (P<0.05) (Fig. 3). As illustrated in Fig.3B, the LBF changes of all the 10 rats in IR group were given. In some rats, the LBF
AC
values of I-115 min were lower than the values of I-5 min, and in other rats, these values were higher. But there was no statistically significant change between the mean LBF values of the two time points. The LBF further decreased after the start of reperfusion (P<0.05), reflecting that reperfusion induced microcirculation dysfunction.
At 60 min after the start of
reperfusion, the LBF recovered slightly but remained much lower than baseline, showing severe IR injury.
The LBF in the SO group did not show significant changes at any
monitored time points (P>0.05).
When the raw arbitrary LSCI value of each rat liver was
expressed as a percentage of its baseline value, as shown in Fig. 3D, the changes of the LBF following ischemia and reperfusion displayed a similar trend as the raw arbitrary LSCI value in the SO and IR groups.
11
ACCEPTED MANUSCRIPT
The measured spatial heterogeneity
T
The spatial heterogeneity of the LSCI measurement was analyzed by comparing the
IP
variability of different ROIs in each rat in the IR group. The CV values are shown in Fig. 4.
SC R
The measured spatial heterogeneity increased following the occlusion of liver blood inflow and the start of reperfusion, but only the CV value of the LBF at R-60 min was significantly higher than that of other time points (P<0.05 vs. baseline, I-5 min, I-115 min and R-5 min),
NU
which reflected the heterogeneity of the LBF due to IR injury.
MA
A comparison of the LBF of the different liver lobes in the IR group found that the mean LBF of the LM was lower than the mean LBF of the RM and LM at baseline, but there was
D
no significant difference between the median lobe and the left lobes (L: 663.0163.87 LSPU, After
TE
LM: 739.9091.21 LSPU, RM: 745.82119.27 LSPU, L vs. LM and RM, P>0.05).
the occlusion of the blood inflow, the LBF of the left lobe was significantly lower than the
CE P
LBF of the right median and left median lobes (L: 289.1335.77 LSPU, LM: 370.7449.30 LSPU, RM: 332.0140.36 LSPU, L vs. LM and RM, P<0.05).
The LBF of the left lobe was
AC
lower than the right median and left median lobes at I-115 min (L: 265.3429.35 LSPU, LM: 349.3551.43 LSPU, RM: 318.0545.46 LSPU, L vs. LM and RM, P<0.05). There were no statistical significances between the LBF of the left lobe and the LBF of the right median and left median lobes at 5 min and 60 min after the start of reperfusion.
The effects of biological zero and movement of the liver on LSCI measurements The BZ was 244.0959.80 LSPU for the normal rat livers, which accounts for approximately 35% of the baseline LBF.
The remaining LBF signal during ischemia
accounts for 43%-45% of the baseline LBF. When subtracting the liver BZ from the
12
ACCEPTED MANUSCRIPT remaining LBF signal during ischemia, the residual LBF signal was 8%-10% of the baseline LBF.
As illustrated in Fig. 5A, the left lobe and left median lobe
IP
anesthesia-induced apnea.
T
The influence of movement on LSCI measurements was evaluated by deep
SC R
showed lower LBF values than the right lobe and right median lobe at the beginning of monitoring because the occlusion of the blood inflow and outflow of the left lobe stopped the blood supply to the left median lobe.
Following the inhalation of a high concentration of
NU
ether, the LBF values of the four liver lobes decreased gradually (Fig. 5C).
At
MA
approximately 2:30 (2.5 min after the start of the LSCI measurement), respiratory movement stopped, and the LBF graph line became smooth; the LBF of the right lobes dropped quickly. Following this period of apnea, all of the LBF graph lines reached their lowest levels.
The decrease in
TE
D
cessation of ether inhalation, the respiration and LBF gradually recovered.
After
the LBF in the left lobe, following the induction of apnea, represented the effect of
CE P
diaphragm-induced liver movement because there was no blood flow in the left lobe. liver movement led to an approximate 150 to 200 LSPU increase in the LBF value.
This
During
AC
apnea, the LBF value of the left lobe represented the actual BZ of the rat liver, and it was approximately 150 LSPU.
Apnea resulted in a significant decrease in the systemic and
microcirculation blood flow, and the LBF of the right lobe dropped quickly.
Because of the
difficulty in avoiding diaphragmatic movement and apnea-induced blood flow decreases at the same time, the value of the liver BZ consisted of the actual BZ and the movement-induced LBF increase.
Histopathological changes of the rat liver The histopathological analysis of liver tissues from the SO group revealed a normal liver lobule structure and clear and complete hepatic cords (Fig. 6). 13
There was no sinusoidal
ACCEPTED MANUSCRIPT congestion, expansion or necrosis of the hepatocytes.
In the liver tissues from rats that
experienced ischemia and reperfusion, the lobular structure remained intact, but some hepatic
Hepatic subcapsular sinus congestion and thrombin were
IP
and swelling were observed.
T
cord structures were not clear, and sinusoidal congestion and hepatocytes with karyopyknosis
AC
CE P
TE
D
MA
NU
SC R
observed.
14
ACCEPTED MANUSCRIPT Discussion The liver blood flow is dually supplied by the hepatic artery and the portal vein, and the LDF and intravital fluorescence microscopy have
T
hepatic sinusoid is the capillary network.
IP
frequently been used to investigate the microvascular LBF during ischemia and reperfusion
SC R
(Tawadrous et al., 2001; Nishida et al., 2000). Typically, LDF measures the LBF at a single point, resulting in large inter-site and inter-individual variability (Richter et al., 2010; Wheatley et al., 1993).
Using intravital fluorescence microscopy, it was found that hepatic
NU
microcirculatory perfusion failure was a determinant of liver dysfunction in warm Intravital fluorescence microscopy requires
MA
ischemia-reperfusion (Vollmar et al., 1994).
toxic fluorescent dyes for contrast enhancement and is not suitable for monitoring the intraoperative microvascular LBF in humans or large animals because of their marked liver
D
In this study, we evaluated the use of LSCI in the assessment of IR-induced
TE
thickness.
microvascular LBF changes in rats. The results showed that LSCI is a useful instrument to
decrease
that
CE P
measure the LBF intraoperatively. resulted
from
LSCI could effectively and accurately monitor the LBF
liver
blood
inflow
occlusion
and
reflected
the
AC
ischemia/reperfusion-induced liver microcirculation dysfunction. According to a recent study, increases in the "region of interest" and "time of interest" reduce the variability of blood flow measurements using LSCI (Rousseau et al., 2011).
This
study suggested that an ROI larger than 10 mm2 and a TOI longer than 1 second were required to reduce the variability in LSCI measurements. lobes were simultaneously monitored with LSCI.
In our study, three exposed liver
A circular or oval ROI of approximately
25 mm2 was selected as the appropriate size for the rat liver LSCI analysis.
By copying the
first ROI, 6 to 10 ROIs were placed on the LSCI image of each rat liver to cover almost all of the homogenous imaging area.
The LSCI measurements and ROI setting method in our
study could increase the region of interest while avoiding the artifact areas, and are
15
ACCEPTED MANUSCRIPT convenient for the comparison of the LBF values within different ROIs or different liver lobes. This is an alternative, simple method for monitoring the microvascular LBF, and has the
T
advantage of not requiring a breathing machine and apnea, as was necessary in a study by
IP
Sturesson C et al. (2013). The surfaces of the three exposed liver lobes were not at the same
SC R
level, but the liver LSCI images were primarily homogeneous except for some small blue artifact spots that resulted from the specular reflection. This observation supports the finding that the distance between the laser head and the skin does not influence the skin blood flow
NU
values recorded by LSCI (Mahé et al., 2011a).
MA
Preliminary studies showed that the hepatic microcirculation is homogenous with no differences occurring between the single lobes, especially when assessed using a tissue depth of 2-mm (Ladurner, 2009). We found that there was no significant difference in the LBF
TE
D
between the median lobe and the left lobe at baseline and during reperfusion.
However, the
LBF of the left lobe was significantly lower than the median lobe during ischemia.
These
CE P
results might reflect the different anatomical location of the left and median lobes, which might influence the compensatory response of liver perfusion, such as back flow from the
AC
hepatic vein (Martins and Neuhaus, 2007). Similar to LDF, although the measured LSCI microvascular LBF is not the arbitrary blood flow and is a relative value (Mahé et al., 2012), our results showed that the LBF of the rat liver was relatively stable at a specific anesthesia level, environmental conditions and instrumental measuring parameters. Good-to-moderate intra-individual and inter-individual reproducibilities were observed in the LBF measurements with LSCI for the baseline and repeated measurements in the SO group. These results demonstrated the possibility to compare the arbitrary LBF values from multiple measurements and multiple experimental groups without the need to depend on relative measurements. In the SO group, the LBF did not show a significant difference at each measuring time
16
ACCEPTED MANUSCRIPT point. In the IR group, the LBF decreased significantly after total blood inflow occlusion, and there was a further decrease in the LBF after the beginning of reperfusion, reflecting the In a study by
T
microcirculation dysfunction caused by hepatic ischemia-reperfusion injury.
IP
Sturesson C et al. (2013), the total liver blood inflow was only clamped for 5 min, and the
SC R
LBF could recover to the original level. By prolonging the ischemic time, as was performed in our study, the LBF could not recover to the original level because of severe IR injury. This finding was consistent with the results of previous studies with intravital fluorescence
NU
microscopy, which revealed that reperfusion injury of the liver after warm ischemia was
et al., 1994, 1996).
MA
characterized by impaired sinusoidal perfusion with a reduction of erythrocyte flux (Vollmar Our results from the liver tissue pathological analysis showed sinusoidal
congestion within the liver and under the liver capsule.
Intracellular edema and
TE
D
hepatocellular and sinusoidal endothelial cell injury formed the structural basis for the decline in the microvascular LBF (Pannen, 2002).
CE P
After the total liver blood inflow occlusion, the LBF decreased to approximately 45% of baseline. Theoretically, when the hepatic blood inflow is completely blocked, the
result.
AC
microvascular LBF should drop to zero. We hypothesize three possible explanations for this First, the laser speckle and laser Doppler technologies have a BZ phenomenon.
For
the laser Doppler measurements, the skin blood flow does not reach the value of zero when perfusion is absent because of the Brownian motion of macromolecules arising from the interstitial space (Kernick, et al., 1999).
A portion of this signal might be attributed to the
remaining red blood cells in the venules under the condition of arrested flow (tourniquet ischemia) (Kernick and Shore, 2000).
A recent skin blood flow study showed a higher BZ
with LSCI than with LDI (Millet et al., 2011).
Regarding the liver tissue, the origin of the
BZ with LSCI remains unknown, but possible sources for the liver BZ include the Brownian motion or other passive movements of fluid, macromolecules and red blood cells.
17
Residual
ACCEPTED MANUSCRIPT vasomotion of the hepatic sinusoidal was not observed.
The blood supply to the liver is
much richer than the supply to other organs, which indicates that when the blood inflow and
T
outflow are occluded, there is more blood remaining in the hepatic sinusoid, and the
At 5 min after the start of reperfusion, the LDF further decreased to
SC R
signals during ischemia.
IP
Brownian motion of the remaining red cells might contribute largely to the remaining LSCI
approximately 30% of the baseline, indicating that the BZ of a rat liver that undergoes IR was much lower than the BZ during ischemia because of the IR injury. This might be related to the
NU
microcirculation dysfunction due to hepatic sinusoidal congestion and hemagglutination. The regular movement of the liver
MA
Second, LSCI is susceptible to movement.
following the movement of the diaphragm creates a specific background signal in LSCI measurements.
To reduce the effect of respiratory motions, Sturesson C et al. (2011)
TE
D
measured the LBF during a state of apnea, but apnea significantly reduced the systemic and hepatic microcirculation blood flow.
In this study, we found that diaphragm-induced liver When
CE P
movement caused an approximate 150 to 200 LSPU increase in the LBF value.
comparing the LBF values of different rats, the anesthesia and respiratory motion should be
AC
controlled as much as possible. A group recently showed that movement-induced artifacts might be overcome by subtracting the back-scattered signal from an adhesive opaque surface adjacent to the ROI (Mahe et al., 2011b, 2013).
This method requires a specific signal
post-processing technique, and whether this method is suitable for LSCI measuring of microvascular LBF requires further study. Third, the backflow of the hepatic vein system might contribute to the high post-occlusion LBF value because the LBF of the no flow left liver lobe (i.e., BZ) is lower than the LBF under blood inflow occlusion (Smyrniotis et al., 2003).
The hepatic vein of the
left lobe in rats is longer than in the other lobes (Martins and Neuhaus, 2007), which might explain why the LBF of the left lobe is lower than the LBF of the median lobes during
18
ACCEPTED MANUSCRIPT ischemia and might support the contribution of hepatic vein backflow to the high LBF value after blood inflow occlusion. Multiple factors contribute to the high remaining LSCI value Some studies demonstrated that
T
when the hepatic blood inflow is completely blocked.
IP
subtracting the BZ from the raw arbitrary perfusion units did not affect the correlation
SC R
between the LSCI and LDI assessments of skin perfusion. However, subtracting the BZ introduced high variability into the baseline skin blood flux signal when recorded with LSCI
NU
or with LDI (CV=38.8% and 81.1%, respectively) (Millet et al.，2011). Because of the multiple contributing factors and high variability of the liver BZ, we do not recommend
MA
subtracting the BZ from the raw arbitrary perfusion units, particularly in ischemia and reperfusion studies.
D
Impaired hepatic microcirculation following extended hepatic resection or liver
TE
transplantation is a major cause of hepatocellular dysfunction.
Assessing the microvascular
CE P
LBF, especially intraoperatively and in real-time, has not been possible in clinical practice. In this study, we found that LSCI could accurately monitor the decrease of the LBF resulting from liver blood inflow clamping and could reflect the ischemia-reperfusion-induced liver
AC
microcirculation dysfunction after reperfusion.
These results suggest a possible application
of LSCI for monitoring the microvascular LBF intraoperatively.
The relationship between
the degree of ischemia-reperfusion-induced LBF changes and IR injury warrants further investigation.
19
ACCEPTED MANUSCRIPT Acknowledgments This work was supported by the Project of the National Natural Science Foundation of China
AC
CE P
TE
D
MA
NU
SC R
IP
T
(81271738) and the National Key Technology R&D Program of China (2012BAI06B01).
20
ACCEPTED MANUSCRIPT References 1. Basak, K., Manjunatha, M., Dutta, P.K.. 2012. Review of laser speckle-based analysis in
T
medical imaging. Med. Biol. Eng. Comput. 50(6), 547–558.
IP
2. Boas, D.A., Dunn, A.K.. 2010. Laser speckle contrast imaging in biomedical optics. J.
SC R
Biomed. Opt. 15(1), 011109.
3. Dixon E, Vollmer CM, Bathe OF, Sutherland F.. 2005. Vascular occlusion to decrease blood loss during hepatic resection. Am. J. Surg. 190(1):75-86.
NU
4. Dong, J.H., He, X.D., Li, K., Duan, H.C., Peng, Z.M., Cai, J.X.. 2002. Tolerance limit of
Pancreat. Dis. Int. 1(1), 57-62.
MA
rats to normothermic hepatic inflow occlusion under portal blood bypass. Hepatobiliary
5. Dunn, A.K.. 2012. Laser speckle contrast imaging of cerebral blood flow. Ann. Biomed.
TE
D
Eng. 40(2), 367-77.
6. Kernick, D,P,, Tooke, J.E., Shore, A.C.. 1999. The biological zero signal in laser Doppler
CE P
fluximetry-origins and practical implications. Pflugers. Arch. 437(4), 624-31. 7. Kernick, D.P., Shore, A.C.. 2000. Characteristics of laser Doppler perfusion imaging in
AC
vitro and in vivo. Physiol. Meas. 21(2), 333-40. 8. Ladurner, R., Feilitzsch, M., Steurer, W., Coerper, S., Königsrainer, A., Beckert, S.. 2009. The impact of a micro-lightguide spectrophotometer on the intraoperative assessment of hepatic microcirculation: a pilot study. Microvasc. Res. 77(3), 387-8. 9. Lesurtel, M., Lehmann, K., de Rougemont, O., Clavien, P.A.. 2009. Clamping techniques and protecting strategies in liver surgery. HPB (Oxford). 11(4),290-5. 10. Mahé, G., Haj-Yassin, F., Rousseau, P., Humeau, A., Durand, S., Leftheriotis, G., Abraham, P.. 2011. Distance between laser head and skin does not influence skin blood flow values recorded by laser speckle imaging. Microvasc. Res. 82(3), 439-42. 11. Mahé, G., Rousseau, P., Durand, S., Bricq, S., Leftheriotis, G., Abraham, P.. 2011b. Laser
21
ACCEPTED MANUSCRIPT speckle contrast imaging accurately measures blood flow over moving skin surfaces. Microvasc. Res. 81(2),183-8.
T
12. Mahé, G., Humeau-Heurtier, A., Durand, S., Leftheriotis, G, Abraham, P.. 2012.
SC R
imaging. Circ. Cardiovasc. Imaging, 5(1),155-63.
IP
Assessment of skin microvascular function and dysfunction with laser speckle contrast
13. Mahe, G., Abraham, P., Le Faucheur, A., Bruneau, A., Humeau-Heurtier, A., Durand, S.. 2013. Cutaneous microvascular functional assessment during exercise: a novel approach
NU
using laser speckle contrast imaging. Pflugers. Arch. 465(4), 451-8.
MA
14. Martins, P.N., Neuhaus, P.. 2007. Surgical anatomy of the liver, hepatic vasculature and bile ducts in the rat. Liver Int. 27(3),384-92.
15. Millet, C., Roustit, M., Blaise, S., Cracowski, J.L.. 2011. Comparison between laser
TE
D
speckle contrast imaging and laser Doppler imaging to assess skin blood flow in humans. Microvasc. Res. 82, 147–151.
CE P
16. Nishida, T., Ueshima, S., Kazuo, H., Ito, T., Seiyama, A., Matsuda, H.. 2000. Vagus nerve is involved in lack of blood reflow into sinusoids after rat hepatic ischemia. Am. J.
AC
Physiol. Heart Circ. Physiol. 278(5), H1565-70. 17. Puissant, C., Abraham, P., Durand, S., Humeau-Heurtier, A., Faure, S., 2013. Lefthériotis G, Rousseau P, Mahé G. Reproducibility of non-invasive assessment of skin endothelial function using laser Doppler flowmetry and laser speckle contrast imaging. PLoS One 8(4),e61320. 18. Pannen, B.H.. 2002. New insights into the regulation of hepatic blood flow after ischemia and reperfusion. Anesth. Analg. 94(6),1448-57. 19. Richter, S., Sperling, J., Kollmar, O., Menger, M.D., Schilling, M.K.. 2010. Laser Doppler flowmetry of hepatic microcirculation during Pringle's maneuver: determination of spatial and temporal liver tissue perfusion heterogeneity. Eur. Surg. Res.
22
ACCEPTED MANUSCRIPT 44(3-4),152-8. 20. Roustit, M., Millet, C., Blaise, S., Dufournet, B., Cracowski, J.L.. 2010. Excellent
T
reproducibility of laser speckle contrast imaging to assess skin microvascular reactivity.
IP
Microvascular. Research. 80(3), 505–511.
SC R
21. Rousseau, P., Mahé, G., Haj-Yassin, F., Dur and S., Humeau, A., Leftheriotis, G., Abraham, P.. 2011. Increasing the "region of interest" and "time of interest", both reduce the variability of blood flow measurements using laser speckle contrast imaging.
NU
Microvasc. Res. 82(1), 88-91.
MA
22. Smyrniotis, V., Kostopanagiotou, G., Lolis, E., Theodoraki, K., Farantos, C., Andreadou, I., Polymeneas, G., Genatas, C., Contis, J.. 2003. Effects of hepatovenous back flow on ischemic- reperfusion injuries in liver resections with the pringle maneuver. J. Am. Coll.
TE
D
197(6), 949-54.
23. Sturesson, C., Milstein, D.M., Post, I.C., Maas, A.M., van Gulik, T.M.. 2013. Laser
34-40.
CE P
speckle contrast imaging for assessment of liver microcirculation. Microvasc. Res. 87,
AC
24. Tawadrous, M.N., Zhang, X.Y., Wheatley, A.M.. 2001, Microvascular origin of laser-Doppler flux signal from the surface of normal and injured liver of the rat. Microvasc. Res. 62(3), 355-65. 25. Tew, G.A., Klonizakis, M., Crank, H., Briers, J.D., Hodges, G.J.. 2011, Comparison of laser speckle contrast imaging with laser Doppler for assessing microvascular function. Microvasc. Res. 82(3), 326-32. 26. Vollmar, B., Glasz, J., Leiderer, R., Post, S., Menger, M.D.. 1994, Hepatic microcirculatory perfusion failure is a determinant of liver dysfunction in warm ischemia-reperfusion. Am. J. Pathol. 145(6), 1421–1431. 27. Vollmar, B., Glasz, J., Post, S., Menger, M.D.. 1996. Role of microcirculatory 23
ACCEPTED MANUSCRIPT derangements in manifestation of portal triad cross-clamping-induced hepatic reperfusion injury. J. Surg. Res. 60, 49–54.
T
28. Wheatley, A.M., Zhao, D.. 1993. Intraoperative assessment by laser Doppler flowmetry
IP
of hepatic perfusion during orthotopic liver transplantation in the rat. Transplantation
AC
CE P
TE
D
MA
NU
SC R
56(6), 1315-8.
24
ACCEPTED MANUSCRIPT Figure Legends
A：Schematic drawing representing the ischemic model of 90% of the rat liver by
IP
study.
T
Fig. 1 Graphical illustration of the surgical operations on the rat livers and the protocol in the
SC R
blood inflow occlusion on the pedicles of the left lobe (L), left median lobe (LM), right median lobe (RM) and right lobe (R), with no occlusion of the caudate lobe (C), which B: The
NU
accounts for 8-10% of the liver weight and is kept as a passage of the portal blood.
left liver lobe is ligated to occlude the blood inflow and outflow used in liver biological zero C: The experimental procedures for LSCI measurements of ischemia and
MA
measurements.
reperfusion induced liver blood flow changes at 5 min after laparotomy (Baseline), 5 min
D
after the start of ischemia (Post-I), 5 min before the end of ischemia (Pre-R), 5 and 60 min
CE P
TE
after the start of reperfusion (Post-R, Post-R60).
Fig. 2 Example images of LSCI measurements of the rat liver marked with circular regions of interest (ROIs) on the flux image, photo image and color image.
In the flux images, the red
AC
areas correspond to the regions of high perfusion and high flux, and the blue regions indicate low perfusion and low flux.
Fig. 3 Changes in the LBF of the sham operation rats (the SO group) and the liver ischemia and reperfusion rats (the IR group) measured with LSCI at 5 min after laparotomy (Baseline), 5 min after the start of ischemia (I-5 min), 5 min before the end of ischemia (I-115 min), 5 and 60 min after the start of reperfusion (R-5 min, R-60 min). A, Representative LSCI images of the SO and IR groups at the indicated time points; B, The LBF changes in the 10 rats in the IR group; C, D, In the comparison of the LBF of the SO group and the IR group, the LBF is expressed as the raw arbitrary perfusion unit (LSPU) (C) or the percent of its 25
ACCEPTED MANUSCRIPT baseline level (D). *compared with the SO group and the Baseline of the IR group, P<0.05; #
T
compared with the I-5 min, I-115 min and R-60 min of the IR group, P<0.05.
IP
Fig. 4 The spatial heterogeneity of the LSCI measurements by comparing the variability The spatial
SC R
among different ROIs in the rats of the IR group at the indicated time points.
heterogeneity of the LBF increased following the ischemia and reperfusion. * P<0.05 vs.
NU
Baseline, I-5 min and I-115 min.
apnea and recovery conditions.
MA
Fig. 5 Representative LBF continuously monitored with LSCI of the rat under deep anesthesia, A, The flux images at the beginning of deep anesthesia (a),
start of apnea (b), end of apnea (c) and end of recovery (d).
ROIs 1-4 are located at the left,
TE
D
left median, right median and right liver lobes. B, A photograph showing the position of the rat under LSCI monitoring.
C, The representative line graph of the LBF of ROIs 1-4 under
Black arrow: the start of apnea; Red arrow: the end of apnea.
AC
ROI4-green line.
ROI1-black line, ROI2-red line, ROI3-blue line and
CE P
continuous LSCI monitoring.
Fig. 6 Histopathological changes of the rat liver after ischemia and reperfusion (200X).
A, B
from the sham operation group, showing the lobule structural integrity and normal hepatic cords and hepatocytes; C, D from the ischemia and reperfusion group, showing sinusoidal congestion and hepatocytes with karyopyknosis and necrosis. liver capsule.
26
B, D the liver tissue under the
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
27
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
28
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
29
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
30
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
31
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
32
ACCEPTED MANUSCRIPT Highlights: 1.
The first use of LSCI technology in monitoring hepatic microcirculation
Proposed a simple method for LSCI monitoring and analyzing of liver
IP
2.
T
dysfunction
First investigation of liver movement on LSCI measurement of liver
TE
D
MA
NU
microcirculation
CE P
4.
The first investigation of liver biological zero in LSCI measurement
AC
3.
SC R
microcirculation
33