Lasers in Surgery and Medicine 33:226–231 (2003)
Effects of Dehydration on the Optical Properties ofIn Vitro Porcine Liver
Dan Zhu, PhD, Qingming Luo, PhD,* and Jian Cen, MS
The Key Laboratory of Biomedical Photonics of Ministry of Education, Huazhong University of Science and Technology,Wuhan, 430074, China
Background and Objectives: It is of crucial importanceto determine the effects of dehydration on the optical pro-perties of tissue so that optimization of therapy and devicefor laser applications in medicine can be made.Study Design/Materials and Methods: After beingincubated directly or indirectly in an oven of 378C for dif-ferent periods of time, the porcine liver samples wereweighed with an electronic scale and their optical proper-ties were measured by a double integrating sphere system.Results:When samples were incubated directly for 20 hours,the average weight loss was 68.5%� 1.2%, and the ab-sorption coefficient and scattering coefficient increased146.1%� 26.9% and 10.8%� 1.1%, respectively. In com-parison, there was only 21.5%� 1.0% of water loss for thecontrol samples, and the absorption coefficient increased30.7%� 7.6%, while the reduced scattering coefficientincreased 386.5%� 29.7%. The effective attenuation coeffi-cients increased 111% and 103% for dehydration group andcontrol group, respectively.Conclusions: The absorption coefficient and effective at-tenuation coefficient increase with dehydration level oftissue. However, there is no direct correlation between de-hydration and reduced scattering coefficient. Lasers Surg.Med. 33:226–231, 2003. � 2003 Wiley-Liss, Inc.
Key words: hydration level; absorption coefficient; reduc-ed scattering coefficient; effective attenuation coefficient
INTRODUCTION
Laser induced interstitial coagulation is a commonmethod employed for treating different types of tumors.During photocoagulation, laser induced rise in tempera-ture can cause immediate irreversible damage through cellnecrosis and denaturation of structural proteins in bothtargeted tissue and surrounding healthy tissue [1]. A com-mon objective in medical laser application is the coagula-tion of a desired volume of tissue with minimal or controlledthermal effects on the surrounding healthy tissue. Opti-mization of therapy or devices for laser applications inmedicine often requires knowledge of the light distributioninside the target tissue, which depends not only on laserparameters but also on optical properties of tissue [2].
During photothermal processes, the changes in appear-ance of tissues induce dynamic changes in optical proper-ties [1,3,4]. Previous studies on thermally induced opticalproperty changes in myocardium emphasized irreversible
changes between 60 and 758C, the temperature range inwhich extracellular protein and collagen denaturation areknown to occur. Derbyshire et al. [5] and Splinter et al. [6]concluded that protein macromolecules coagulation isresponsible for light scattering and has little effects onabsorption based on their observation of increase in thescatter coefficient and relatively constant in the absorptioncoefficient as a result of tissue coagulation. People thusoften only considered dynamics of scattering coefficientwhen they modeled photothermal distribution of tissueduring the process of laser induced interstitial thermo-therapy [1,3,7].
However, the change was observed not only in scatteringcoefficient of tissue, but also in the absorption coefficientduring heating process [8,9]. When the human aortasamples were heated in a constant temperature water bathat 1008C for 300� 10 seconds, Cilesiz et al. [9] observed bothincrease in scattering coefficient and the absorption coef-ficient in the visible and near IR due to the tissue weightloss. They suggested that dehydration and protein coagula-tion during photothermal treatment of tissue are importantfactors altering optical properties of tissue. Since dehydra-tion usually occurs during thermal damage of tissue [3,9],it is very difficult to investigate the relationship betweenhydration level and optical properties of tissue during thephotothermal process. There is a need to independentlystudy the relationship between tissue dehydration and itsoptical properties.
In this work, the optical properties of porcine liver wereinvestigated during the process of tissue dehydration. Toexclude the influence of dehydration from other factors,porcine liver samples are divided into two groups: dehy-dration group and control group. In the former group, thesamples were incubated in an oven at a temperature of
Contract grant sponsor: National Nautre Science Foundationof China; Contract grant number: 59836240; Contract grantsponsor: NSF of China for distinguished young scholars; Contractgrant number: 60025514; Contract grant sponsor: China Pstdoc-toral Science Foundation; Contract grant number: 2002031256.
*Correspondence to: Qingming Luo, PhD, The Key Laboratoryof Biomedical Photonics of Ministry of Education, HuazhongUniversity of Science and Technology, Wuhan, 430074, China.E-mail: [emailprotected]
Accepted 19 June 2003Published online in Wiley InterScience(www.interscience.wiley.com).DOI 10.1002/lsm.10215
� 2003 Wiley-Liss, Inc.
378C. While in the latter group, the samples were firstsandwiched between two cover slides and then wereincubated indirectly in the same oven. Due to the differentarrangements, two groups experienced different extents ofdehydration during the same periods of time. To determinethe dehydration levels as a function of incubation time, thesamples in both groups were weighed periodically with anelectronic scale during incubation. The optical propertiesincubated samples were measured by a double integratingsphere system at the same time.
MATERIALS AND METHODS
Experimental Set-Up
A double integrating sphere system was used to measurethe optical properties of porcine liver in vitro. This methodis commonly employed to measure absorption and scatter-ing properties of biological tissue [10–14]. Figure 1 showsthe schematic diagram of experiment system [15,16].
A 1.6 mW, 1mmbeamindiameter He–Nelaser (632.8nm)beam was chopped mechanically at 1 kHz (Model SR540),and then splitted two beams by a beam splitter. Thereflected beam was a small fraction (20%) of laser beam,which irradiated toareferencesphereof 70mmin diameter.The other (80%) of laser beam irradiated a sample mountedin a round holder (25 mm in diameter), which was placedin between two identical integrating spheres of 210 mm indiameter. The light fluence within reflectance sphere,transmittance sphere and reference sphere were detectedwith Si PIN photodiodes (1223-01, Hamamatsu), and theninput into a lock-in-amplifier (Model SR830) in turn.Finally, PC sampled the output signals from Lock-in-amplifier. In this system, the reference measurements oflaser power could reduce experimental error because offluctuant of laser power.
Reflectance and transmittance measurements withinthe spheres were made relative to the signal from a stan-dard plate, which was placed at sample aperture of thereflect sphere or transmission sphere, respectively. All themeasurements on reference plate (99%) are made with asingle sphere. Figure 2 shows the diagram of the referencemeasurements. Based on the measurements of referencesignal, reflection and transmission for a 99% standard
plate and tissue sample, the reflectance (R) and transmit-tance (T) were derived by following Equations [15]
R ¼ Vr=Vref
VR0=VrefR0ð1Þ
T ¼ Vt=Vref
VT0=VrefT0ð2Þ
where the values of VrefR0 and VrefT0 were from the refer-ence sphere, and the values of VR0 and VT0 were from thereflectance sphere and transmittance sphere when a 99%standard plate was placed at the sample aperture (seeFig. 2). Vref, Vr, and Vt were the measurements of referencesignal, reflected light and transmitted light for sample asdescribed in Figure 1.
Eight sets of measurements were made on each sample,i.e., 4 on each side. After each set of measurements, thesample was displaced by a few millimeters to allow theincident beam to fall on a different part of tissue. Based onthe measurements of reflectance (R), transmittance (T),and the thickness of samples, the optical properties weredetermined by the inverse adding-doubling algorithm [17].In our system, optical properties could be calculated imme-diately after attaining each set of measurements becausethe sampling program was integrated with inverse adding-doubling algorithm [16].
Tissue Sample Preparation
Fresh excised porcine liver was obtained from a localslaughterhouse within 1 hour postmortem. To remove re-sidual blood on the surface, we soaked the liver in 0.9%isotonic saline solution for 10 minutes. The in vitro hy-dration levels of tissue were not altered by this cleaningmethod. It was found to be difficult to cut non-frozen tissueinslices smaller thanmillimeterespecially forhighlyelastictissues, so the liverwaswrapped insaline-moistened plasticmembrane and cooled to �208C. Prior to any measure-ments, the tissue sample was cut to squares of 3� 3 cmSquare to cover the sample holder of 25 mm. The thicknessof samples was about 800–1,000 mm.
Soon after being cut, most of samples were immediatelyplaced on one glass slide, which were in the group ofdehydration experiments. Whereas others were carefullysandwiched between two slides which were in control groupand in fresh group. The optical properties of fresh sampleswere attained based on the thickness of samples and themeasurements from the double integration sphere system.
Fig. 1. Experimental apparatus consisting of two integrating
spheres system.
Fig. 2. Diagram showing how the reference measurements
are made.
DEHYDRATION EFFECTS ON OPTICAL PROPERTIES 227
The samples for dehydration group and control groupwere weighed with the electronic scale, and then put into378C electrothermal incubator (SKP£-01). To determinethe effects of dehydration on optical properties of tissue, thesamples after being kept in incubator for different periodsof time were weighed, and then the thickness of sampleswere measured, the optical properties were finally deter-mined based on the measurements from the double inte-gration sphere system. Different samples were used indehydration group. However, the same samples in controlgroup were subjected to several subsequent measurements.In order to keep the same heating time and determine thechange in hydration level of sample for control group anddehydration group, the samples in dehydration for followup experiment were also taken out from incubator andweighed simultaneously, and replaced into incubator withthe samples of control group after the optical propertiesof control group were measured. The measurements on asample were discontinued when no further changes in theweight of the sample in the dehydration group were obser-ved and the sample was then taken out of the sample pool.
The weight of sample was determined by weighing of theslide/sample (or /slide) arrangement and subtracting theweight of the slide (or slides). The dehydration level ofsample is determined by the equation
w ¼ ms �mt
msð3Þ
where ms was the weight for fresh sample, and mt was theweight of sample at different period of incubation time forcontrol group and dehydration group.
RESULTS
A total of 49 porcine liver samples were examined in thiswork. Five of them were fresh samples under the roomtemperature of 208C, 37 samples were in the dehydrationgroup, and 7 were in the control group. Table 1 summarizesthe results of the measurements from fresh samples,dehydration group, and control group at different periodsof incubation time. The dehydration level and the opticalproperties at 632.8 nm are expressed as the mean�SD.We first obtained the mean value from eight measurementson each sample, and then calculated the mean�SD fromdifferent samples.
Time-Dependent Dehydration Level for theDehydration and Control Groups
When samples of porcine liver were exposed indirectly ordirectly to incubator (378C), hydration level of sampleschanged with incubation time. Figure 3 shows the relation-ship between the dehydration level and incubation timefor both the groups.
Thirty-seven samples in the dehydration group were di-vided into 7 subgroups. Each subgroup had 5 or 6 samples,which were incubated for different periods of time. Thesamples weight dropped sharply during the initial 8.5hoursof incubation, and then the hydration level slowly de-creased. After 20 hours of incubation, the weight of tissuesample changed little. Also, we observed that the samplethickness decreased with the water loss. The maximumaverage weight loss reached up to 68.5%� 1.2%, whichaccompanied with an average thickness shrinkage of52.8%� 6.5%.
Since each sample in the control group was sandwichedbetween two slides, it could be measured at different timeafter it was exposed indirectly to incubator. For controlgroup, the loss of water content was relatively less. Theaverage weight loss was 21.5%� 1.0% after 20 hours ofincubation. The average thickness shrinkage was 19.8%�3.1%.
Time-Dependent Absorptionand Scattering Properties
For fresh porcine liver, the absorption coefficient is0.26� 0.005 mm�1, and the reduced scattering coefficientis 0.37� 0.03 mm�1, based on the five fresh samples. Theoptical properties would change when samples of porcineliver were exposed indirectly or directly to incubator (378C).Figure 4 shows the time dependence of the absorption andscattering properties of porcine liver from the dehydrationgroup and control group.
From Figure 4a, we observed that the absorption coeffi-cient from the dehydration group increased gradually inthe beginning, and finally remained at a constant duringdehydration process. The maximum increase in the ab-sorption coefficient was 146.1%� 26.9% for the dehydra-tion group. However, there was only 30.7%� 7.6% increasefor the control group. The change in the absorption coef-ficient for the dehydration group was much larger than the
TABLE 1. Time-Dependence of Dehydration Level and Optical Properties at 632.8 nm (Temperature 378C)
T (8C) Time (hour)
Dehydration group Control group
w% �a (mm�1) �0s (mm�1) w% �a (mm�1) �0
s (mm�1)
20 0 0 0.26� 0.005 0.37� 0.03 0 0.26� 0.005 0.37� 0.03
37 1.4 15.3� 2 0.30� 0.02 0.38� 0.06 2.1� 0.02 0.27� 0.02 0.49� 0.04
3 33.7� 1.8 0.40� 0.02 0.38� 0.02 4.84� 0.05 0.29� 0.02 0.66� 0.05
4 45.4� 4.1 0.45� 0.04 0.37� 0.06 6.36 � 0.04 0.30� 0.03 0.75� 0.06
5.5 52� 2.5 0.49� 0.05 0.38� 0.06 8.03 � 0.08 0.28� 0.04 0.92� 0.08
8.5 63� 1.5 0.55� 0.08 0.40� 0.02 12.4 � 0.09 0.31� 0.02 1.35� 0.08
16 68.0� 1.6 0.63� 0.06 0.39� 0.05 19.7 � 1.1 0.32� 0.03 1.72� 0.1
20 68.5� 1.2 0.64� 0.07 0.41� 0.04 21.5� 1.0 0.34� 0.02 1.80� 0.11
228 ZHU ET AL.
result for the control group when both types of samples hadbeen placed in incubator for the same period of time.
Figure 4b shows the reduced scattering coefficient as afunction of incubation time. For the dehydration group,reduced scattering coefficient changed a little. There was aslight increase of 10.8%� 1.1%. In contrast, the change inthe reduced scattering coefficient was much larger for thecontrol group, which was increased evidently with time.It reached a maximum of 386.5%� 29.7%.
There could be different dehydration process if differentincubator is used, and time-varying optical properties willbe different. Therefore, it is more important to investigatethe relationship between optical properties and dehydra-tion level based on measurements.
Effects of Dehydration on Absorptionand Scattering Properties
Figures 5a and b display the absorption coefficient andreduced scattering coefficient of porcine liver as the func-tion of dehydration level.
Figure 5a shows that the absorption coefficient increasedwith water loss of porcine liver for both groups. The resultsdemonstrated that the absorption coefficient was elated tothe hydration level of tissue sample. The more water
content lost, the larger absorption coefficient increased.Therefore, the data from both groups can be described witha single function. That is, absorption coefficient could bedetermined by dehydration level, no matter how the samplewas dehydrated.
In contract, the effect of dehydration on reduced scatter-ing coefficient is evident different between the dehydrationgroup and control group, as shown in Figure 4b. The re-duced scattering coefficient remained relatively constantwhen the sample was exposed directly to incubator and wasdehydrated gradually. There was a very large increase inthe reduced scattering coefficient for the control group eventhough there was a little dehydration.
Since simplex dehydration hardly impacted on the re-duced scattering coefficient of tissue sample, the resultsuggests that there is no direct relation between thereduced scattering coefficient and hydration level.
DISCUSSION
The above results indicated that the maximum averageweight loss was 68.5%� 1.2% for the dehydration group,which was very close to 69%� 1.5% water content of porcineliver [18]. That means that the sample of porcine liver isdehydrated completely. The dehydration level for the con-trol group was much less because the maximum averageweight loss was only 21.5%� 1.0% at the same period ofincubation time. The results showed that there were ob-vious differences between absorption and scattering pro-perties for both groups.
Fig. 3. Time dependence of dehydration level of porcine liver
from the dehydration and control groups (378C).
Fig. 4. Time dependence of the optical properties of porcine
liver from the dehydration and control groups. a: Absorption
coefficient; (b) reduced scattering coefficient.
Fig. 5. Dehydration dependence of the optical properties of
porcine liver from the dehydration and control groups. a:
Absorption coefficient and (b) reduced scattering coefficient.
Fig. 6. The effective attenuation coefficient for the dehy-
dration and control groups changed with (a) time and (b)
dehydration level.
DEHYDRATION EFFECTS ON OPTICAL PROPERTIES 229
The optical-thermal response of laser-irradiated tissueis governed by the heat diffusion equation with the heatsource term Q (W �mm�3) given by
Q ¼ mafðz; rÞ ð4Þ
where ma is the absorption coefficient of tissue (mm�1),which determinates the efficiency of photothermal con-version, and f (z,r) is the local fluence rate of the laser light(W �mm�2) which is determined by effective attenuationcoefficient [19,20].
Using the following equation,
meff ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi3maðma þ m0sÞ
qð5Þ
The corresponding effective attenuation coefficient (meff)could be determined from the absorption coefficient (ma)and reduced scattering coefficient (m0s). Time-dependentand dehydration-dependent effective attenuation coefficientfor both groups is shown in Figures 6a and b, respectively.
When samples were exposed to incubator at 378C, theeffective attenuation coefficient increased with time. Eventhough there are evident differences for dehydration level,but a little difference for effective attenuation coefficientbetween the dehydration group and the control group. After20 hours of incubation, the maximum average increase ineffective attenuation coefficient is 103% for the dehydrationgroup and 111% for the control group.
Even though almost the same changes were observed inthe effective attenuation coefficient, the transmittance forthe control group is lower than that for the dehydrationgroup during the process of dehydration, which could resultin the different changes in the thickness of the sample. Themaximum change in thickness for the control group is onethird of that for the dehydration group.
Cilesiz et al. [9] observed that the absorption coefficientand the reduced scattering coefficient and the effective at-tenuation coefficient increased by 42.2%, 9.1%, and 26.9%,respectively, while the average weight lost 46%� 7.6%at the wavelength of 630 nm during thermal coagulation ofhuman aorta. Similarly, we observed increase of absorptioncoefficient of 72.9%, reduced scattering coefficient of 2.0%,and effective attenuation coefficient of 47.9% at the wave-length of 632.8 nm while the weight loss resulted fromdehydration was 45� 4.1%.
Dehydration induced the increase in the absorptioncoefficient could result from shrinkage of the tissue samplesbecause of denser packing of cells, whereas the number ofchromophores remained constant.
The changes in scattering properties attributed mainlyto tissue sample putridity. There is a little change in thewater content at 378C for the control group, which helps topropagate anaerobe in porcine liver in vitro. In contrast,sample putridity was not observed for the dehydrationgroup because the water was lost quicker. Further veri-fication about this phenomenon is under investigation.
CONCLUSION
In this study, the effects of hydration level on the opticalproperties of porcine liver were analyzed in vitro. The ab-
sorption coefficient of porcine liver increased with hydra-tion levels for the dehydration group and control group.The more water content lost, the absorption coefficientincreased. However, there is no direct relationship betweenhydration level and the reduced scattering coefficient.The reduced scattering coefficient changed slightly for thedehydration group, whereas the reduced scattering coeffi-cient increased sharply for the control group even thoughthere was a little weight loss. The changes in absorptionand scattering properties resulted in the same effectiveattenuation coefficient change for both groups duringdehydration process.
ACKNOWLEDGMENTS
The authors thank Dr. Yongwu Yang (SerOptix, Inc.,Belmont), Yang Weng and Xiaohua Lv (Huazhong Univer-sity of Science and Technology, P. R. China) for their kindhelp.
REFERENCES
1. Rastegar S, Motamedi M. A theoretical analysis of dynamicvariation of temperature dependent optical properties inthe response of laser irradiated tissue. SPIE 1990;1646:228–235.
2. Matthew GS, Megumi NI, Michael CK, Michael DS. A theore-tical comparison of energy sources—microwave, ultrasoundand laser—for interstitial thermal therapy. Phys Med Biol1998;43:3535–3547.
3. Kim MB, Jacques LS, Rastegar S, Tomesen S, Motamedl M.Nonlinear finite-element analysis of the role of dynamicchanges in blood perfusion and optical properties in lasercoagulation of tissue. IEEE J Selected Top Quantum Electron1996;2:922–933.
4. Chan EK, Brown DT, Kovach IS, Welch AJ. Laser assistedsoldering: Effects of hydration on solder-tissue adhesion.J Biomed Opt 1998;3:456–461.
5. Derbyshire DJ, Bogen DK, Unger M. Thermally inducedoptical property changes in myocardium at 1064 nm. LaserSurg Med 1990;10:28–34.
6. Splinter R, Stevenson RH, Littmann L, Tuntelder RJ,Chuang CJ, Tatsis GP, Thompson M. Optical properties ofnormal, diseased, and laser photocoagulated myocardiumat the Nd:YAG wavelength. Lasers Surg Med 1991;11:117–124.
7. Zhu D, Luo QM, Zhu GM, Liu W. Kinetic thermal responseand damage in laser coagulation of tissue. Lasers Surg Med2002;22:313–321.
8. Pickering JW, Bosman S, Posthumus P, Blokland P, Beek JF,Van Gemert MJC. Changes in the optical properties (at632.8 nm) of slowly heated myocardium. Appl Opt 1993;32:367–371.
9. Cilesiz IF, Welch AJ. Light dosimetry: Effects of dehydrationand thermal damage on the optical properties of the humanaorta. Appl Opt 1993;32:477–487.
10. Pickering JW, Prahl SA, Wieringen NV, Beek JB, SterenborgHJVM, Van Gemert MJC. A double integrating spheresystem for measuring the optical properties of tissue. ApplOpt 1993;32:399–410.
11. Roggan A, Dorschel K, Minet O, Wolff D, Muller G. Theoptical properties of biological tissue in the near infraredwavelength range-review and measurements. In: Muller G,Roggan A, editors. Laser-induced interstitial thermotherapy.Bemingham, Washington: SPIE Press; 1995. pp 10–44.
12. Lin WC, Motamedi M, Welch AJ. Dynamics of tissue opticsduring laser heating of turbid media. Appl Opt 1995;35:3413–3420.
13. Beek JF, Blokland P, Posthumus P, Aalders M, Pickering JW,Sterenborg HJCM, Van Gemert MJC. In vitro double-
230 ZHU ET AL.
integrating-sphere optical properties of tissues between 630and 1064 nm. Phys Med Biol 1997;42:2255–2261.
14. Roggan A, Friebel M, Dorschel K, Hahn A, Muller G.Optical properties of circulating human blood in the wave-length range 400–2500 nm. J Biomed opt 1999;4:36–46.
15. Zhu D, Luo QM, Zeng SQ, Yin M, Ruan Y. A modified double-integrating-sphere system for measuring the optical pro-perties of tissue. Acta Photonics Sinca 2001;30:1175–1181(in Chinese).
16. Cen J, Luo QM, Zhu D, Zeng SQ. A double-integrating-spheresystem for measuring the optical properties of tissue based onsingle-chip microcomputer. Chinese J Scientific Instrument(in press) (in Chinese).
17. Prahl SS, Van Gemert MJC, Welch AJ. Determining theoptical properties of turbid media by using the adding-doubling method. Appl Opt 1993;32:559–568.
18. Roggan A, Muller G. Dosimetry and computer based irradia-tion planning for laser-induced interstitial thermotherapy(LITT). In: Muller G, Roggan A, editors. Laser-inducedinterstitial thermotherapy. Bemingham, Washington: SPIEPress; 1995. pp 114–156.
19. Welch AJ. The thermal response of laser irradiated tissue.IEEE J Quantum Electron 1984;20:1470–1481.
20. Welch AJ, Van Gemert MJC. Optical-thermal response oflaser-irradiated tissue. New York, London: Plenum Press; 1995.
DEHYDRATION EFFECTS ON OPTICAL PROPERTIES 231
