Alternative cooling and heating as a novel minimally invasive approach for treating obesity

International Journal of Thermal Sciences 64 (2013) 29e39

Contents lists available at SciVerse ScienceDirect

International Journal of Thermal Sciences journal homepage: www.elsevier.com/locate/ijts

Alternative cooling and heating as a novel minimally invasive approach for treating obesity Zi-Qiao Sun a, Yang Yang a, Jing Liu a, b, * a b

Key Laboratory of Cryogenics, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100190, China Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing 100084, China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 26 August 2011 Received in revised form 15 June 2012 Accepted 7 August 2012 Available online 7 September 2012

Patients with obesity often suffer pain and risks arising from complications in their pursuit for a trimmer figure. Therefore, an effective and safe treatment for obesity is urgently needed. In this paper, we propose a novel minimally invasive way to treat the target obesity tissues via alternative cooling and heating produced by a microprobe. The validity of the method was evaluated through both numerical simulation and differential scanning calorimetry (DSC) experiments. Theoretical prediction presents the significant effect of typical surgery and assesses the influence of various temperature boundary conditions on the microprobe to obtain the ideal therapeutic effect. An estimation standard was also established on the basis of cryolipolysis and hyperthermia. The DSC test confirms physical and chemical changes in cells during the cooling and heating process. The treatment planning will be varied with operational target, such as longer or shorter treatment time, one or multiple probes, etc. The 3D reconstruction employing MATLAB shows a simulated destruction area. This work is expected to serve as the foundation for identifying a new cure for obesity. Ó 2012 Elsevier Masson SAS. All rights reserved.

Keywords: Obesity Microprobe Surgery 3D simulation Freezing Heating Physical therapy

1. Introduction The accumulation of excess body fat in the body, not only brings inconvenience to people who suffer from it, but also becomes a major global public health problem. To date, nearly two million Chinese adults are obese, with an overweight percentage of 22.8%. In the developed countries, such as those in Europe and America, this proportion is higher by 30% [1,2]. As a result, the obesity treatment and fat removal have become popular areas of research with worldwide significance. Many efforts have been made in finding the best treatment for obesity. Surgery is generally accepted as the most effective way for fat removal. In terms of applied methods, it can be categorized into invasive and non-invasive types. As the most famous fat removal method, liposuction can directly remove the adipose tissues and produce an immediate effect after the invasive surgery [3e6]. However, some patients suffer from infection and a syndrome such as massive necrotizing fasciitis caused by subcutaneous damage

* Corresponding author. Key Laboratory of Cryogenics, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100190, PR China. Tel.: þ86 10 82543765; fax: þ86 10 82543767. E-mail address: [email protected] (J. Liu). 1290-0729/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.ijthermalsci.2012.08.003

after surgery [7]. Consequently, the body may fail to function, and fatal complications [8e10], such as blood poisoning, may occur. As an alternative, less invasive methods have been intensively investigated to prevent the side effects of liposuction. These methods include high-intensity focused ultrasound, radiofrequency method, and treatment with infrared light. Although these methods can reduce the probability of infection, they are not effective enough. On the other hand, cryolipolysis, for example, is the latest noninvasive method for fat treatment, but takes at least two months for a full treatment [11e15]. Therefore, to balance the relationship between curative effect and direct injury, we propose a distinct and minimally invasive surgical method, in which alternative cooling and heating is introduced into adipose tissues by an inserted microprobe. It is well known that the normal living cells cannot survive extreme temperatures. According to the concept of cryolipolysis, temperatures below 277 K can cause inflammatory destruction in adipose tissues. In hyperthermia, on the other hand, a temperature above 316 K would cause the target tissue to spontaneously lose activity, which is a mature principle for tumor treatment [16]. Meanwhile, cell viability is not only associated with temperature, but also related to the temperature changing rate. In cryopreservation, 1 and 40 K/min are acceptable cooling and re-warming rates, respectively [17e19]. Actually, low cooling rate (5e

30

Z.-Q. Sun et al. / International Journal of Thermal Sciences 64 (2013) 29e39

Nomenclature C D h k Ql Qm t T Ta Tb Tc Tf

heat capacity [J/m3  C] diameter of vessel [m] convective heat transfer coefficient [W/m2  C] thermal conductivity [W/m  C] latent heat [J/m3] metabolic heat generation [W/m3] time [s] temperature [ C] artery temperature [ C] mean blood temperature [ C] body core temperature [ C] surrounding air temperature [ C]

180  C/min) and very high cooling rates (>5000  C/min) would lead a high viability of cells, which, respectively, allows the cell water outflow to occur completely and thus avoids intracellular crystallization. The rapid heat flow would induce intracellular crystallization and/or vitrification before any water flows out of the cell. In other hands, a rapid cooling rate (180e5000  C/min) would lead a low viability of cells, which allows the heat flow to prevail over water outflow (in this case, cell water crystallization would occur as water was flowing out of the cell). Crystallization is a critical factor for cell death rate [20]. Therefore, if a rapid cooling rate is applied, the death rate of cells would increase. At the same time, a relatively low re-warming rate can damage cells because of intracellular re-crystallization. Therefore, if a destruction effect is required, temperature changing rate should be induced via a high cooling rate and low re-warming rate. In this paper, we introduce the method which has been successfully applied in the tumor treatment to implement both thermal ablation and cold devastation on the adipose tissue [21,22]. The work is dedicated to investigating the ablation and apoptotic behavior during the implementation process. Numerical simulation presents the treatment process and reveals that temperature can be employed as the sole assessment criterion for the target tissue. Meanwhile, the result of the parametric study documents the adaptability and validity of the proposed method. Additionally, the differential scanning calorimetry (DSC) experiment confirms the injury characteristics of alternative cooling and heating damage by thermal property analysis, and further illustrates the effectiveness of the treatment.

temperature on the microprobe wall [ C] velocity of blood flow [m/s] Cartesian coordinates [m] location

Tw v x,y,z X

Greek symbols blood perfusion [ml/s/ml] computation domain density [kg/m3] thermal conductivity [J/kg K] cb

ub U rb

Subscripts b blood u unfrozen tissue

vacant and the point depicted in Fig. 1(b) is the origin of the coordinates (0,0,0). To determine the ablation and apoptotic effects on the adipose tissue and the temperature distribution around the microprobe location, we adopt the Pennes bioheat equation [23], which describes the influence of blood flow upon temperature distribution in the tissue in terms of heat sinks or sources with variable temperatures:

rc

vTðX; tÞ ¼ VKðXÞ$V½TðX; tÞ þ rb cb ub ðXÞ½Ta  TðX; tÞ vt þ Q ðX; tÞ X ¼ U

Effective section

Skin Microprobe

Simulation area

Ice ball

Fat layer

Muscle

a hf ,T

2. Material and methods

Effective section 40mm

Skin 2mm

2.1. Heat transfer model and calculation domain In an actual surgery, to maximally utilize the effective section of the surgical probe, the microprobe is supposed to be inserted parallel to the skin surface (Fig. 1(a)), which appears as the best way. The human body is prescribed as a three-layer model consisting of muscle, fat, and skin. For simplification, a 3D rectangular geometry with dimensions of 0.04 m  0.08 m  0.07 m (Fig. 1(b)) is selected as the analyzed spatial domain. Given the alternative cooling and heating process in the treatment, the microprobe is inserted into the middle of the fat layer and can be simplified as a cylinder of 55 mm length and 2.5 mm radius. In addition, the 40 mm long black part of the microprobe (Fig. 1(b)) serves as the effective section which introduces various temperatures on the wall. For convenience, the domain inside the microprobe is deemed

(1)

Microprobe Coordinates origin

Fat layer 40mm

Muscle

Y Z

40mm

70mm

X

b Fig. 1. (a) Sketch of a prescribed actual surgery. (b) Schematic of the calculation domain.

Z.-Q. Sun et al. / International Journal of Thermal Sciences 64 (2013) 29e39

8 < Q ðX; tÞ ¼ 0 T < 273 K Q ðX; tÞ ¼ Qm ðX; tÞ þ Ql 273 K  T  277 K : Q ðX; tÞ ¼ Qm ðX; tÞ 277 K < T

(2)

where, T(X,t) is the tissue temperature; r and c are the density and specific heat of living tissue, respectively; rb and cb denote the density and specific heat of the blood, respectively; K(X) and ub(X) represent the space-dependent thermal conductivity of the tissue and blood perfusion, respectively; Qm is the metabolic heat generation rate. In the present model, blood perfusion and metabolic heat generation only exist in the muscle tissue. Ta is the arterial temperature that often treated as a constant; and U signifies the analyzed spatial domain, described as follows. X comprises the Cartesian coordinates x, y and z, where y denotes tissue depth from the body core, while x and z are coordinates along the surface; Ql is the latent heat and will only release when the temperature falls in range from 273 to 277 K. Neither the blood perfusion nor metabolic heat generation exists when the temperature of tissue becomes below 273 K. Fat tissue can be regarded as a composition of water and many kinds of lipids, thus the implementation of solidification occurred during an interval of temperature range (273e277 K). The latent heat is coupled with the heat transfer equation, presenting as a part of inner heat source in the freezing process. For the tissue freezing process, instead of tracking the liquidesolid front explicitly, an enthalpy-porosity formulation is employed to modeling the solidification/melting process. The liquidesolid mushy zone is treated as a porous zone with porosity equal to the liquid fraction. The boundary conditions for the calculation domain are prescribed as follows:

k

k

vTðx; y; z; tÞ ¼ 0 at x ¼ 20 mm; vx ¼ 20 mm vTðx; y; z; tÞ ¼ 0 at z ¼ 70 mm; vz ¼ 70 mm

k

k

vTðx; y; z; tÞ ¼ 0 at x vx ð3Þ vTðx; y; z; tÞ ¼ 0 at z vz ð4Þ

vTðx; y; z; tÞ Tðx; y; z; tÞ ¼ Tc at y ¼ 40 mm; k vy   ¼ hf Tf  T at y ¼ 40 mm

(5)

where, the skin surface is defined at y ¼ 40 mm and the body core at y ¼ 40 mm; hf is the apparent heat convection coefficient between the skin surface and the surrounding air (this coefficient is the overall contribution from natural convection and radiation under a physiological state); and Tf is the surrounding air temperature. We adopt the adiabatic conditions on the boundaries along the x and z directions on the basis of the consideration that the positions are far from the center of the domain, and the temperature field is almost unaffected by the central domain or external heating/cooling. The various temperatures on the microprobe wall can be described as follows:

Tw ¼ Tw ðX; tÞ

c ¼ cb ¼ 4200 J/kg K. The muscle tissue is regarded as having a thickness of 40 mm with thermal parameters k ¼ 0.5 W/m K, Qm ¼ 4200 W/m3, ub ¼ 0.0005 ml/s ml, and Ql ¼ 324,000 J/kg. The thickness of the skin is generally within the range from 0.5 to 4 mm, and treated in the present study as 2 mm. Its thermal conductivity is 0.3 W/m K. The thickness of the fat layer is taken as 38 mm; its thermal conductivity is 0.2 W/m K, and latent heat Ql ¼ 324,000 J/ kg. The body core temperature is equal to arterial temperature Ta ¼ Tc ¼ 310 K. The temperature of the surrounding air at the skin surface is Tf ¼ 300 K, and the heat convection coefficient is hf ¼ 10 W/m K which comes mainly from the natural convection and radiation occurred at the skin surface [24]. The 3D numerical simulations coupled with the tissue heat transfer equations are developed from the commercial software FLUENT in the calculation domain, with the parameters and boundary conditions given above. The temperature distribution in the entire calculation domain, especially in the thermal ablation and cold devastation domain introduced by the microprobe, can then be determined. In complicated operations, such as wide-area treatment, the action of one probe may not be sufficient. Thus, multiple probes may be required to facilitate efficient cure. The initial steady-state temperature distribution in the calculation domain is determined by setting the microprobe wall as an adiabatic surface. Then a typical temperature curve (Fig. 2) in the tumor treatment, is prearranged in the fat layer for the first attempt [17,26], in which the times of heating and cooling are identical. In detail, it can be described as: from 0 to 3 min, the temperature of the microprobe wall is 93 K, and then shifts to a high temperature at 353 K for 3 min. After this, the temperature goes back to 93 K as a new cycle begins. The simulation is completed after 12 min with the end of the second temperature change cycle. Three points and one monitoring line are mounted to observe instant temperature variation. As presented in Fig. 3, the distances between the microprobe axis and three points on the cross-section Z ¼ 35 mm along direction y are respectively 3, 4, and 5 mm. The monitoring line starts at Point (0,40,35) and ends at Point (0,0,35). The performance of three probes is also being considered. In Fig. 4, three parallel probes are arranged in the fat layer as the above single microprobe and the calculation domain is consistent with Fig. 1. The axis of the microprobe B is at (x ¼ 0 mm, y ¼ 20 mm) and the other two microprobes A and C are located at both sides with a 10 mm distance. In the cross-section Z ¼ 35 mm, the distance between the axis of microprobe B and the five points which are chosen to observe the temperature variation are respectively 5, 13,

400

353K 350

Temperature ( K )

where

31

300 250 200 150

(6)

100

where, Tw(X,t) is the space and time dependent temperature on the wall; X comprises the Cartesian coordinates x, y, and z on the probe wall; and t denotes time. In the numerical calculations, the typical parameters for the tissue are set as follows [24,25]: r ¼ rb ¼ 1000 kg/m3 and

50

93K temperature 0

3

6

9

Time(min) Fig. 2. Typical temperature arrangement on the microprobe.

12

32

Z.-Q. Sun et al. / International Journal of Thermal Sciences 64 (2013) 29e39

Skin Monitoring line

C B A

Fat layer Microprobe Y

Coordinates : C (0, 25, 35) B (0, 24, 35) A (0, 23, 35) Line: x=0; y∈ [0,40]; z=35.

Fig. 3. Relative position of monitoring points and monitoring line (only skin and fat layer domains are presented).

and 15 mm along direction x, and 3 and 5 mm along direction y. In the simulation, the temperature variations with the alternate rate 1:3:1:3 are arranged on the microprobe wall. In treating adiposity, the volume of fat that one operation can eliminate should be predicted. In the present study, MATLAB-3D reconstruction is employed. 2.2. Differential scanning calorimetry (DSC) experiment Although the effectiveness of alternative heating and cooling treatment has been evaluated in the numerical simulation, the mechanism of the curing effect is unclear. Thus, a DSC test on the pig adipose tissue is implemented to detect the effect of separated cooling and heating process, and the alternative cooling and heating process, respectively. Alternative cooling and heating was proposed to realize targeted ablation on fat cells. Here, differential scanning calorimetry experiments were implemented to confirm the damage property of heating and cooling effect. And DSC 200 PC (by NETZSCH, Germany) was adopted. Before experiments, the machine was calibrated under a heating and cooling rate of 10.0 K/ min by sapphire which serves as a standard sample. Combining the curve of standard sample and test sample, the curve of cp varies with the temperature could be obtained. For each test group, three repeated measurements were implemented to confirm the consistency of results. For the test on the cooling group, the temperature curve presented in Fig. 5(a) indicates that the sample was cooled from 293 to 233 K, and then re-warmed back to 293 K at a constant temperature changing rate 10 K/min for twice. For the test on the heating group, as shown in Fig. 5(b), the sample was heated from 293 to 333 K, and then returns to 293 K at a constant temperature changing rate 10 K/min for twice. The damage effect of separated cooling and heating can be evaluated as above. However, the method proposed in this paper offers the actions of freezing and heating simultaneously, therefore

Y

HG

B F

3. Results 3.1. Computational results

X

A

the combined test of these two effects should also be performed. For the test on alternative heating and cooling, as shown in Fig. 5(c), the sample was heated from 293 to 333 K, and then returns to 293 K at a constant temperature changing rate 10 K/min for twice.

D E

C

Coordinates D (0, 25, 35) E (0, 23, 35) F (5, 20, 35) G (13, 20, 35) H (15, 20, 35)

X Fig. 4. Relative position of monitoring points for the simulations of the three probes in the fat layer.

3.1.1. Typical results of the implementation process The temperature distribution in the calculation domain on each moment can be obtained. Fig. 6(a) depicts the temperature distributions on the cross-section X ¼ 0 mm of the skin and fat layer at t ¼ 3, 6, 9, and 12 min moments. Red and blue colors represent the upper and lower limits of the temperature scale. Obviously, the maximum effect is always achieved at the final moment in each temperature change cycle. Fig. 6(b) presents the solidification state at the same cross-section; 0 and 1 represent the frozen and normal states, respectively. The ice ball grows around the microprobe and reaches a maximum volume at 3 min/9 min, then almost melts in the succeeding heating process. The frozen region at the 9 min moment is larger than that in the 3 min moment (For interpretation of the references to color in this paragraph, the reader is referred to the web version of this article.). Fig. 7(a) shows that the temperatures of the monitoring points change with temperature variations in microprobe. It can be seen that the closer to the microprobe, the lower temperature can be obtained. Point A reaches the lowest and highest temperatures because of its smallest distance from the microprobe. Fig. 7(b) depicts the temperature changing rate at the monitoring points, in which both cooling and re-warming rates at the monitoring points also vary with temperature changing in the microprobe. As previously mentioned, in the heating process, the area with a temperature above 316 K can be treated as an effectively injured region; on the other hand, a real damage effect of cooling would be identified by simultaneously satisfying such three conditions: a temperature below 277 K, a cooling rate higher than 10 K/min and a re-warming rate lower than 40 K/min. The cooling residence time should not be too short, which plays an important role in cooling and lipid phase transition of the cell membrane. From Fig. 7(a) and (b), it is clear that in the cooling process, the fat cells at Points A and B have already been damaged. However, this injury is not related to Point C for the relatively high temperatures, although the cooling rate there is greater than 10 K/min and has exceeded the injury level from a cryopreservation perspective. In the consequent heating process, although the temperature in each point has not reached 316 K, the re-warming rate always stay at 40 K/min, which can conform the destruction effect in Points A and B. Therefore, the three criteria for cooling effect can be simplified into one: a temperature below 277 K. It means once the area has been exposed to temperatures below 277 K, the degree of damage can be determined. The temperature distributions in the simulation domain are captured by the monitoring line at t ¼ 3, 9, 6, and 12 min for each section, respectively. As described in Fig. 8, the points are employed to represent the temperature of the monitoring lines, while the lines are used to demonstrate the damage boundary of freezing or heating. In accordance with the new temperature criterion, red and black lines in Fig. 8(a) respectively display the cooling destruction boundary at 3 and 9 min, where the ice ball formed at 9 min is larger than 3 min. Similarly, the heating damage range at 6 min is higher than that at 12 min (Fig. 8(b)). The comparison in Fig. 8 indicates that freezing introduces a more substantial destruction area than the hyperthermia under the same treatment time. Moreover, the microprobe achieves the maximum destruction area at the end of the second freezing period (For

Z.-Q. Sun et al. / International Journal of Thermal Sciences 64 (2013) 29e39

300

273K 270 240 0

6

b

12

18

24

Temperature

330 316K 300 270 240

0

4

Time (min)

8

Time (min)

12

16

c Temperature(K)

Temperature 330

Temperature(K)

Temperature(K)

a

33

Temperature

330

316K

300 273K 270 240 0

10

20

30

40

Time (min)

Fig. 5. The temperature arrangement for the test. (a) On the cooling group; (b) on the heating group; (c) on the alternative heating and cooling group.

interpretation of the references to color in this paragraph, the reader is referred to the web version of this article.). 3.1.2. Effect of different boundary temperatures on the microprobe wall The simulation of three probes reveals that under the same treatment duration, freezing results in a larger area of destruction than heating does. Consequently, different temperature curve will be applied on the microprobe wall. First, the heat treatment time is tripled to expand the final scope of hyperthermia (Fig. 9(a)). The following periods of heating and freezing are also tripled. From the result as shown in Fig. 9(b) and (c), the extension of the heating period widens the effective area of hyperthermia, and the range of the heating effect at 12 min is only slightly smaller than that of the previous freezing range obtained at 3 min. Moreover, the performance of the following tripled cooling also results in a larger ice ball than the first cooling period. In summary, the extension of the heating period is acceptable because the ice ball provides a protective layer around the uncured area. Then, the later freezing period is doubled in the simulation (Fig. 9(d)). As an improvement, Fig. 9(e) and (f) demonstrates that the adjustment leads to a relatively smaller ice ball at the end of the

second freezing period compared with that in the previous simulation, and the later hyperthermia area is also correspondingly enlarged. The succeeding simulations further reduce the duration of the second freezing period (Fig. 9(g)); thus, the lengths of the two freezing periods are the same. The results displayed in Fig. 9(h) and (i) show that the two freezing scopes are equal, and so are the two heating processes. The result documents that the length of alternate freezing and heating will directly affect the injured region. In order to introduce a precise and effective destruction to the target region, a rate of 1:3:1:3 is appropriate. For different situations, the treatment plan should be carefully evaluated in advance to adapt to the ultimate goal of surgery. 3.1.3. Effect of multiple probes Fig. 10 illustrates that the three probes result in a lower temperature and a higher temperature changing rate, compared with those achieved by one probe in direction y. These effects not only cause greater damage, but also offer an expanded treatment range than one probe. The monitoring Points F, G, and H are located in direction x. It can be observed in Fig. 11(a) that Point G yields the lowest

Fig. 6. Results of typical temperature arrangement. (a) The temperature distribution of fat layer at the cross section X ¼ 0, at t ¼ 3, 6, 9, 12 min; (b) solidification state diagram of fat layer at the cross section X ¼ 0, at t ¼ 3, 6, 9, 12 min (liquid fraction: 0 e frozen 1 e unfrozen).

Z.-Q. Sun et al. / International Journal of Thermal Sciences 64 (2013) 29e39

a 300 Temperature(K)

C

280

277K

260

B

240 220

A 0

A-23mm B-24mm C-25mm

3

6 Time (min)

9

12

Temperature change rate(K/min)

34

b 60 30

C

0

B

-30 -60 -90 0

A-23mm B-24mm C-25mm

A 3

6 Time(min)

9

12

Fig. 7. (a) Temperature responses of monitoring points A, B, and C. (b) Temperature change rateetime at monitoring points A, B, and C.

3.1.4. 3D reconstruction of the treatment area For the above-mentioned case, the three probes are used to implement the curing process of alternative cooling and heating. The 3D reconstruction of the damaged region can be realized as follows. First, in the simulation domain, 15 XY sections are nonuniformly collected along dimension z according to the change rates of the freezing area in each cross-section. Six representative sections are shown in Fig. 13(a). The yellow and red areas represent the injured and uncured areas, respectively. Second, after the reconstruction, Fig. 13(b) shows that the damage volume is 14,527 mm3 at the end of the second cooling period. This can be considered as the final destruction area after the whole treatment (For interpretation of the references to color in this paragraph, the reader is referred to the web version of this article.).

Temperature(K)

a

300

277K

250 200

9min

150

3min 3min 9min

100 0

10

20

Y(mm)

30

40

3.2. DSC experimental results 3.2.1. Effect of the single cooling process In Fig. 14(a), a DSC jump marked by the circle means the phase change occurs at about 248 K. Compared with the value of the phase transition temperature in the two cooling processes, the temperature difference of the two is less than 1 K. Meanwhile, cp is the heat absorbed or released by a unit mass of a material per 1 K under consistent pressure, which can be adopted to distinguish one kind of substance from another. For the comparison between the two cooling process, Fig. 15(b) shows that the maximum difference of the cp value is only about 0.1 J/(g K). Minimal change is observed in both DSC and cp in the two cooling processes, which indicates that no substance change occurs during the experiment. Therefore, the destruction caused by freezing mostly involves physical and mechanical destruction. Clearly, phase transition leads to the formation of intracellular and extracellular ice. During this process, the fat cell was considered as being destroyed in the paper. But in a practical surgery, the exact percentage of the damaged cell should still be characterized by more researches in the near future. 3.2.2. Effect of the single heating process For this case, no phase transition is observed. Therefore, no DSC jump can be detected in Fig. 15(a). However, the value of DSC differs to some extent. Fig. 15(b) shows that the maximum value of cp undergoes a huge change from 4.7 to 2.7 J/(g K) during the two heating processes. It is demonstrated that the destruction caused by heating process mostly involves matter change, especially the cell components change. 3.2.3. Effect of the alternative cooling and heating process For the test on alternative heating and cooling, the temperature ranging from 233 to 333 K at 10 K/min and the DSC curve of each cooling process varies so much from each other. However, the

b

360

T e m p e r a tu r e (K )

temperature among the three points for the nearest distance to the microprobe. Meanwhile, although exhibiting the same distance from Probe A, Point F achieves a lower temperature than the Point H, which is caused by a superimposed cooling effect of Probes A and B. The curve in Fig. 11(b) illustrates that the destroyed area undergoes lipocyte destruction with the assessment factor at a cooling rate above 10 K/min and a heating rate below 40 K/min. This result also reveals that the effect of multiple probes is greater than that of one probe. The previous numerical analysis shows that the largest damage arises at 15 min, the end of the second cooling process. According to the injury standard of below 277 K, the destruction area in both situations is represented by the white region in the three dimensions in Fig. 12. Multiple probes tend to provide a larger range of effects than a single probe does, but the unique superiority of one probe should not be disregarded. Such single probe can be applied as an effective way to treat tiny mounds of flesh. Therefore, for individual situations, the application of probes should be flexible to attain effective implementation.

340

6min 12min

320

316K 6min

300 280 0

10

20

12min 30

40

Y(mm)

Fig. 8. Temperature distributions on the monitoring line: (a) Terminal of freezing process at 3 and 9 min. (b) Terminal of heating process at 6 and 12 min.

Z.-Q. Sun et al. / International Journal of Thermal Sciences 64 (2013) 29e39

35

Ratio of cooling and heating:1:3:3:3

b Temperature(K)

Temperature(K)

353K 300

200

93K

100 0

6

12

c

300

277K

250 200

3min

150

21min 3min 21min

100

18

24

30

0

10

20

30

320

316K

300

12min

280 0

30min

10

20

30

f 360

12min 27min

Temperature(K)

200

93K

100 12

18

277K

250 200

3min

150

18min

100 50

24

Temperature(K)

300

300

3min 18min

0

10

20

30

340 320

316K

300

0

40

12min

27min

280

10

20

Y(mm)

Time(min)

40

Y(mm)

e

353K

Temperature(K)

340

40

Ratio of cooling and heating:1:3:2:3

d 400

6

12mm 30mm

Y(mm)

Time(min)

0

360

Temperature(K)

a 400

30

40

Y(mm)

Ratio of cooling and heating:1:3:1:3

g 400

h 277K

250

300

Temperature(K)

Temperature(K)

Temperature(K)

i 360

300

353K

200

200

3min

150

93K

100 0

6

12

15min 3min 15min

100

18

24

0

10

20

30

12min 24min

340 320

316K

300

same range 0

40

10

30

40

Y(mm)

Y(mm)

Time(min)

20

Fig. 9. Arrangements of temperature variations on the microprobe and results; (a), (d), (g) show three types of temperature variations on the microprobe; (b, c), (e, f), (h, i) show the corresponding outcomes. For all images in the picture, the temperature variations in the monitoring line is located at the cross-section Z ¼ 35 mm; (b), (e), (h) present the range of freezing at the end of the first and second cycle; (c), (f), (i) present the range of heating at the end of the first and second cycle ends.

phase transition temperature in the second freezing process is 10 K which is lower than that in the first freezing process (Fig. 16(a)). This evident difference indicates that the heating effect provides a cooling enhancement in the freezing process and yields a relatively low phase transition temperature. Being accordant with the curve of DSC, the curve of cp also presents an evident difference between the two cooling process which reveals the enhancement of cooling once again (Fig. 16(b)).

b 315 316K

300 One probe

277K

270 Three probes

240 210

One probe Three probes

0

6

12

Time(min)

18

24

Temperature(K)

Temperature(K)

a 330

For comparison, the cooling process from 293 to 233 K and that from 333 to 233 K are arranged in Table 1. During the former cooling process, only one exothermic peak appears, whereas two are observed in the later cooling process. Differences between the freezing point temperature and the maximum DSC value are detected. All results show that in the alternative cooling and heating process, the high temperature leads to chemical changes in the cell, and consequently influences the following cooling process.

305

One probe

295 285 Three probe

277K

275 265

One probe Three probes

0

6

12

18

24

Time(min)

Fig. 10. Comparison of the effects of one probe and three probes at the same point. (a) Temperature variations at Point E. (b) Temperature variations at Point D.

Z.-Q. Sun et al. / International Journal of Thermal Sciences 64 (2013) 29e39

a Temperature(K)

325 316K

300

H 277K

275

F

250

G 225 0

6

12

F point G point H point 18 24

Time(min)

T em p eratu re C h an g e R ate(K /m in )

36

b 80

G

40

H 0 -40

F

-80 0

6

12

F point G point H point 18 24

Time(min)

Fig. 11. Temperature responses of the monitoring points for the simulations of the three probes. (a) Temperature variations in points F, G, and H. (b) Temperature change rates of points F, G, and H.

In summary, the single cooling effect mostly involves physical destruction, such as thermal stress and so on, whereas the single heating effect is significantly related to component change. Therefore, the alternative cooling and heating imposes a stronger treating effect than that of cooling or heating individually. 4. Discussion As an attempt, we introduce for the first time the alternative cooling and heating to implement both thermal ablation and cold devastation on the adipose tissue. Such method had been successfully applied in the tumor treatment. In order to verify the effect on the adipose tissue, numerical simulation and DSC experiments have been carried out in this study. Combining the concept of hyperthermia, cryolipolysis and cryopreservation, we generalize a damage effect criterion for this method. In the heating process, the area with temperature above 316 K would be considered as injured; Meanwhile, in the cooling process, a real damage effect would be identified by simultaneously satisfying such three conditions: a temperature below 277 K, a cooling rate higher than 10 K/min and a re-warming rate lower than 40 K/min. After the calculation, such complicated criterion can be simplified as: once the area has been exposed to temperatures above 316 K or below 277 K, the degree of damage can be determined. Meanwhile, we design two alternating periods in the calculation in which the cooling process is always first implemented. This is due to a more substantial destruction area could be obtained in the

cooling process under the same treatment time. The frozen area provides a protective layer surrounding the uncured area. In order to introduce a precise and effective destruction to the target region, a rate of 1:3:1:3 for the alternative cooling and heating is appropriate, in which the two freezing scopes are equal, and so are the two heating processes. Furthermore, the number of the microprobes can be flexibly selected to fulfill the conformal obesity cure. For example, the single probe can be applied as an effective way to treat tiny mounds of flesh and multiple microprobes tend to provide a larger range of destruction. In addition, the arrangement of the multiple microprobes also can be regulated to match the obesity treatment. It is easy to see that the ice ball was not melted symmetrically in the succeeding heating process. The thermal boundary conditions of fat simulated block were not the same in each surface. The fat block surface adjoin to the skin was influenced by the heat convection on the skin, with the room temperature as 300 K. At the same time, the surface next to the body core always maintained at 310 K. This difference would lead to that the ice ball formation on the body core side will be heated with more intensity and melted faster than the other side. A surgical planning should consider such phenomena and work out an appropriate scheme for the treatment. On the basis of the above-mentioned results, we adopt DSC to test the damage characteristics occurred in such method. Hysteresis effect was observed in the test. Since the present DSC test was to qualitatively analyze the damage behavior of the heating and cooling, thus the hysteresis effect did not affect the experimental conclusion. Finally, the destruction in the adipose cell caused by the

Fig. 12. Injured range depicted by the temperature distribution at the end of the second freezing process: (a) Single microprobe; (b) Three microprobes.

Z.-Q. Sun et al. / International Journal of Thermal Sciences 64 (2013) 29e39

37

Fig. 13. 3D reconstruction of the treatment range. (a) Six representative sections in the Z direction. (b) Display of the damage domain by 3D reconstruction.

280

273K

260 240

1.0

0

6

12

18

24

Time(min)

0.5

0.0

300

Temperature(K)

1.5

b 12

300

cp( J/ ( g • K ) )

Temperature(K)

DSC(mW/mg)

a 2.0

9

6

280

273K

260 240 0

6

12

18

24

Time(min)

3 c in the first cooling process

DSC in the first cooling process DSC in the second cooling process

240

250

260

270

280

0

290

c in the second cooling process

240

Temperature(K)

250

260

270

Temperature(K)

Fig. 14. DSC and cp curves for single cooling process from 293 to 233 K: (a) DSC; (b) cp.

b

1.0 Temperautre(K)

DSC(mW/mg)

0.8 0.6

Time(min)

0.4 0.2 0.0

DSC in the first heating process DSC in the second heating process

300

310

6 5

cp(J/ (g •K ))

a

skin incisions. The cannulas mechanically dislodge adipocytes, which are then aspirated with negative pressure. In this process, the skin incisions were very small. However, the trauma under the skin would be larger than it seems. The liposuction surgery could not delete the adipose tissues just as it plans, more or less. The biggest advantage of alternative cooling and heating as a novel minimally invasive approach for treating obesity is its good efficiency of adipose ablation. The target fat cells would be destroyed with no harm to the surrounding healthy tissues. Moreover, the damaged fat cells do not need to be removed from the tissues of patient which can subsequently be absorbed by human body, and thus minimize the subcutaneous trauma. When this minimally invasive alternative freezing and heating method is employed to treat obesity, the following steps have to be completed. First, all information on the target tissue and the microprobe should be

320

Temperature(K)

330

Temperature(K)

single cooling process mostly involves physical and mechanical destruction. The chemical injury, especially the cell components change, can be introduced by the single heating process. For the alternative cooling and heating process, first cooling also only provides a physical and mechanical destruction to the adipose cell. The subsequent high temperature leads to further chemical changes and then influences the followed cooling process. As a result, the alternative cooling and heating imposes a stronger effect than a single cooling or heating. The unique advantage of the proposed treatment lies in its good balance between potentially side effects and effective treatment outcomes, reducing the infection/syndrome stemming from subcutaneous damage and long-term inefficient treatment. For liposuction surgery, it involves the permanent surgical removal of subcutaneous fat by means of metal cannulas placed through small

4.7

4 2J/(g*K)

3 2.7

Time(min)

2 1 0

cp in the first heating process cp in the second heating process

300

310

320

Temperature(K)

Fig. 15. DSC and cp curves for single heating process from 293 to 333 K: (a) DSC; (b) cp.

330

Z.-Q. Sun et al. / International Journal of Thermal Sciences 64 (2013) 29e39

DSC(mg/Wg)

1.5

1.0

330

b 316K

8

300 273K 270 240 0

10

20

30

40

Time(min)

0.5

0.0

DSC in the first cooling process DSC in the second cooling process

240

Temperature(K)

Temperature(K)

a

270

300

330

c p (J/ (g •K ))

38

6

316K

300 273K 270 240 0

4

10

20

30

40

Time(min)

2 0

330

cp in the first cooling process

cp in the second cooling process

240

Temperature(K)

260

280

300

320

Temperature(K)

Fig. 16. DSC and cp curves for alternative cooling and heating process from 233 to 333 K: (a) DSC; (b) cp.

and heating method for the treatment of obesity. Specific assessment and monitoring still need to be developed in future studies.

Table 1 Curve comparison with different points.

Cooling from 293 K Cooling from 333 K

Number of peaks

Temperature of freezing points (K)

Maximum DSC value (mW/mg)

1 2

250 258

1.75 1.4

Acknowledgments This work is partially supported by the NSFC Grant 81071255. References

confirmed. This information includes the volume of adipose requiring treatment, thermal conductivity, the total action time, and temperature variations in the microprobe. Other details of surgical implementation should also be determined. Temperature variations should not be limited by the temperature range based on tumor treatment, whose typical temperatures are 93 and 353 K. Different operations require varied plans, and the duration of cooling and heating should be determined beforehand. For example, if the target is adult with abdomen fatty tissue, multiple probes are necessary to realize large treatment scope. Meanwhile, if the operation is efficiently handled at a reasonable heating or freezing action time, the damage area contributes less on the skin or body core, which means that such treatment can be regarded as an apoptotic behavior and will cause no harm to the human body. However, even as the simulation results indicate good performance in fat elimination, the effect of this method in practical treatment remains incompletely clear. More in vivo experiments are necessary in the near future. During the real treatment, monitoring equipment is needed to observe freezing or heating implementation, including the ice ball generation or melting process. Conventional monitoring equipment, such as CT, MIR, and infrared equipment, can be used. 5. Conclusion This paper proposed a new obesity treatment modality which can provide a balance between potentially side effects and effective treatment outcomes. The alternative cooling and heating introduced by the microprobes could serve as rather promising therapy candidates. Such method is firstly applied in the tumor treatment and has been confirmed as an effective method. Therefore, recurring to numerical simulation, we demonstrate the injured area caused by the microprobe. A rate of 1:3:1:3 for the alternative cooling and heating is appropriate for the precise and effective destruction. The number of the microprobe also can be flexibly selected to fulfill the conformal obesity cure. As a result of the DSC experiment, such alternative cooling and heating method could impose a larger effect than cooling or heating individually does. However, all the experiments presented in this paper only preliminarily demonstrate the feasibility of the alternative freezing

[1] M. Li, M.J. Dibley, D.W. Sibbritt, H. Yan, Dietary habits and overweight/obesity in adolescents in Xi’an City, China, Asia. Pac. J. Clin. Nutr. 19 (2010) 76e82. [2] H. Boerschmann, M. Pflüger, L. Henneberger, A. Ziegler, S. Hummel, Prevalence and predictors of overweight and insulin resistance in offspring of mothers with gestational diabetes mellitus, Diabetes Care 33 (2010) 1845e1849. [3] O. Heymans, P. Castus, F.X. Grandjean, D.V. Zele, Liposuction: review of the techniques, innovations and applications, Acta Chir. Belg. 106 (2006) 647e653. [4] K.J. Stewart, D.A. Stewart, B. Coghlan, D.H. Harrison, B.M. Jones, N. Waterhouse, Complications of 278 consecutive abdominoplasties, J. Plast. Reconstr. Aesthet. Surg. 59 (2006) 1152e1155. [5] R.A. Yoho, J.J. Romaine, D. O’Neil, Review of the liposuction, abdominoplasty, and face lift mortality and morbidity risk literature, Dermatol. Surg. 31 (2005) 733e743. [6] B.M. Coldiron, C. Healy, N.I. Bene, Office surgery incidents: what seven years of Florida data show us, Dermatol. Surg. 34 (2008) 285e291. [7]. D.J. Barillo, L.C. Cancio, S.H. Kim, K.Z. Shirani, C.W. Goodwin, Fatal and nearfatal complications of liposuction, South. Med. J. 91 (1998) 487e492. [8]. S.Z. Aasi, D.J. Leffell, Complications in dermatologic surgery e how safe is safe? Arch. Dermatol. 139 (2003) 213e214. [9]. J. Murillo, M. Torres, L. Bofill, A. Rios-Fabra, E. Irausquin, R. Isturiz, M. Guzman, J. Castro, L. Rubino, M. Cordido, V.C. Infect, Skin and wound infection by rapidly growing mycobacteria e an unexpected complication of liposuction and liposculpture, Arch. Dermatol. 136 (2000) 1347e1352. [10]. H. Meyers, B.A. Brown-Elliott, D. Moore, J. Curry, C. Truong, Y.S. Zhang, R.J. Wallace, An outbreak of Mycobacterium chelonae infection following liposuction, Clin. Infect. Dis. 34 (2002) 1500e1507. [11] A.A. Nelson, D. Wasserman, M.M. Avram, Cryolipolysis for reduction of excess adipose tissue, Semin. Cutan. Med. Surg. 28 (2009) 244e249. [12] D. Manstein, H. Laubach, K. Watanabe, W. Farinelli, D. Zurakowski, R.R. Anderson, Selective cryolysis: a novel method of non-invasive fat removal, Lasers Surg. Med. 40 (2008) 595e604. [13] M.M. Avram, R.S. Harry, Cryolipolysis (TM) for subcutaneous fat layer reduction, Lasers Surg. Med. 41 (2009) 703e708. [14] K.B. Klein, B. Zelickson, J.G. Riopelle, E. Okamoto, E.P. Bachelor, R.S. Harry, J.A. Preciado, Non-invasive cryolipolysis (TM) for subcutaneous fat reduction does not affect serum lipid levels or liver function tests, Lasers Surg. Med. 41 (2009) 785e790. [15] S.R. Coleman, K. Sachdeva, B.M. Egbert, J. Preciado, J. Allison, Clinical efficacy of noninvasive cryolipolysis and its effects on peripheral nerves, Aesthetic Plast. Surg. 33 (2009) 482e488. [16] P. Vaupel, M.R. Horsman, Tumour perfusion and associated physiology: characterization and significance for hyperthermia, Int. J. Hyperthermia 26 (2010) 209e210. [17] M.A. Henry, E.E. Noiles, D. Gao, P. Mazur, J.K. Critser, Cryopreservation of human spermatozoa. IV. The effects of cooling rate and warming rate on the maintenance of motility, plasma-membrane integrity, and mitochondrialfunction, Fertil. Steril. 60 (1993) 911e918. [18]. X.D. Cui, D.Y. Gao, B.F. Fink, H.C. Vasconez, L.L.Q. Pu, Cryopreservation of human adipose tissues, Cryobiology 55 (2007) 269e278. [19]. L.L.Q. Pu, Cryopreservation of adipose tissue, Organogenesis 5 (2009) 138e142.

Z.-Q. Sun et al. / International Journal of Thermal Sciences 64 (2013) 29e39 [20] F. Dumont, P.A. Marechal, P. Gervais, Cell size and water permeability as determining factors for cell viability after freezing at different cooling rates, Appl. Environ. Microb. 70 (2004) 268e272. [21] Y.X. Zhou, J. Liu, Y.H. Yu, L. Gui, Z.S. Deng, Y.G. Lv, New cryoprobe system with powerful heating features and its performance tests on biomaterials, in: Proceeding of ASME 2003 International Mechanical Engineering Congress and Exposition, Washington, DC, USA (2003). pp. 183e184. [22] J. Liu, Y.X. Zhou, T.H. Yu, L. Gui, Z. Deng, Y.G. Lv, Minimally invasive probe system capable of performing both cryosurgery and hyperthermia treatment on target tumor in deep tissues, Minim Invasive Ther Allied Technol. 13 (2004) 47e57.

39

[23] H.H. Pennes, Analysis of tissue and arterial blood temperatures in the resting human forearm, J. Appl. Physiol. 1 (1948) 93e122. [24] Z.S. Deng, J. Liu, Numerical simulation of selective freezing of target biological tissues following injection of solutions with specific thermal properties, Cryobiology 50 (2005) 183e192. [25] Z.S. Deng, J. Liu, Numerical study of the effects of large blood vessels on threedimensional tissue temperature profiles during cryosurgery, Numer. Heat Transfer, Part A: Appl. 49 (2006) 47e67. [26] J. Liu, Y.X. Zhou, A probe tumor treatment device with alternating high and low temperature refrigerant transport to achieve the temperature shift quickly, Patent: ZL 01268378.7, China, 2001 (in Chinese).