Original Thoracic
363
Laser Resection of Lung Tissue: Heat Accumulation from Adjacent Laser Application and How to Cool It Down Andreas Kirschbaum1
Peter Rexin2
Anika Pehl2
1 Department of Surgery, University Hospital, Marburg, Germany 2 Department of Pathology, University Hospital, Marburg, Germany 3 Department of Pathology, University Hospital, Erlangen, Germany
Detlef Bartsch1
Karl Quint3
Address for correspondence Andreas Kirschbaum, MD, Department of Surgery, University Hospital, Baldingerstrasse, Marburg 35033, Germany (e-mail: [email protected]).
Abstract
Keywords
► experimental ► metastases/ metastasectomy ► surgical equipment
received May 22, 2013 accepted September 17, 2013 published online December 2, 2013
Background Heat accumulation might induce thermal damage of the surrounding lung tissue, especially when multiple lesions are resected in one session. The present study aimed to investigate whether heat accumulates in the immediate vicinity of the resection surface and leads to thermal damage of the lung parenchyma, and what is the most effective cooling strategy in this situation. Materials and Methods In normothermic perfused paracardial swine lobes (n ¼ 6), four punctiform laser lesions forming a square were created. Each lesion was lasered at a power of 100 W for 5 seconds. Two test conditions with square sides of either 1.0 or 0.5 cm were compared. Temperatures were recorded immediately after completing the laser procedure in the square center and in the corners using a thermal camera and continued during the cooling process at 10-second intervals until normothermia (37°C). We examined two cooling methods: rinsing with ice-cold (4°C) Ringer solution during the laser procedure (group B, n ¼ 6) or submerging the lung in ice-cold water for 5 seconds immediately after laser application (group C, n ¼ 6). In the control group A (n ¼ 6), there was no cooling. Results In the 0.5 cm squares, mean temperature in the center immediately after laser application was 103.17 8.56°C, significantly higher than in the corners (76.39 2.87°C, p < 0.05). Normothermia in the quadrant corners was reached after 81 14 and after 108 29 seconds in the centers. Tissue in the square center revealed histological signs of thermic cell damage. In the 1.0 cm squares, mean temperature in the center was 64 5°C, and in the corners was 77 3.1°C (p < 0.05). Normothermia was regained after 93 22 seconds in the center and 120 21 seconds in the corners. Histological examination in the 1.0-quadrant centers revealed no signs of thermic cell damage. Submerging the lobe into ice-cold water lowered the temperature rapidly to under 40°C, and normothermia was regained after 75 1.3 seconds. Conclusion Laser application to the lung parenchyma causes considerable heat accumulation in closely related lesions. To prevent such cell damage, a distance of at least 1.0 cm between laser targets should be maintained. If no topical cooling method applied, sufficient time for spontaneous tissue cooling before additional laser application should be provided. The most effective cooling strategy against heat accumulation is submerging in ice-cold water for at least 5 seconds.
© 2014 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0033-1358780. ISSN 0171-6425.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Thorac Cardiovasc Surg 2014;62:363–368.
Laser Resection of Lung Tissue
Kirschbaum et al.
Introduction About 30% of patients with various malignancies develop lung metastases in one or both sides.1 Although solitary lung metastases are quite rare, multiple tumors are very common. Under certain circumstances, that is, the patient has adequate cardiopulmonary capacity, no mediastinal lymph node involvement, and an R0 resection is technically feasible, the resection of multiple metastases is recommended.2 Bilateral lung metastases are resected in separate sessions approaching the more complex side first. In fact, patients with multiple metastases usually undergo thoracotomy because further metastases are often detected intraoperatively by bidigital palpation of the lung tissue.3 A radical lymphadenectomy might also be performed depending on the primary tumor type.4 The goal of surgery for lung metastases is to spare as much healthy lung tissue as possible. In this regard, conventional operative methods have some disadvantages. Usually the tumors are atypically clamped off, resected, and the lung parenchyma has to sew over. Alternatively, resection with a stapler device can be employed. These techniques often lead to a considerable loss of healthy lung tissue. In deeper metastases, extra-anatomical resection by these methods might not be possible, so that an anatomic resection (i.e., segment removal or lobectomy) must be performed resulting in even more loss of healthy parenchyma. Alternatively, when the patients are not candidates for surgery owing to poor cardiopulmonary reserve or there were too much lesions, radiofrequency ablation might be a treatment option.5 Laser-guided resection of lung metastases has theoretically some key advantages compared with conventional surgical techniques. It enables the surgeon to operate in a very gentle manner on the lung parenchyma and causes very little bleeding.6 Usually the metastasis is resected with a safety margin of 5 mm around the tumor. When the laser strikes the lung tissue, its intensive heat causes vaporization of the lasered tissue and coagulation of the surrounding tissue.7,8 Thus, larger metastases can be removed with minimal blood loss. Up to 100 metastases in a single lobe can be removed by the Nd:YAG laser LIMAX 120 (Gebrüder Martin GmbH & Co KG, Tuttlingen, Germany) at a wavelength of 1,318 nm and a power of up to 120 W.6 However, as the temperature in the surrounding healthy tissue rises during every laser application, there is the theoretical risk of heat accumulation and critical cell damage in the tissue adjacent to each metastasis being resected. Since there are no data as yet regarding this issue, an experimental model was developed to analyze the heat accumulation in lung tissue and its relationship to the laser application distance. In addition, we aimed to determine by which method the lung tissue can effectively be cooled down.
Materials and Methods Laser Application Experiments were conducted on paracardial lung lobes (n ¼ 6) from freshly slaughtered pigs. After removing the Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
heart and both lungs en bloc, the paracardial lobes were inspected for any anomalies. The pulmonary arteries supplying the lobes were dissected from the surrounding tissue and cannulated with a peripheral venous catheter (12G). The lobes were then rinsed with heparin-saline solution (5.000 IU) until the drained fluid was free of blood particles. The lobes were then wrapped in moist compresses and immediately transported to the laboratory in a cold bag. In the laboratory with a constant ambient temperature, these specimens were fixed vertically and the cannulas hooked up to saline solution warmed to 42° C, which at the point of entry into the lung lobe reached a temperature of 37°C. All lobes were then perfused (perfusion pressure 20 cm H2 O) with this solution at normotonia, and the perfusion’s homogeneity documented by a thermal imaging camera (IC 120LV, Trotec, Heinsberg, Germany). The laser-lesion points were marked on the lung surface with a stencil forming a square with the four points either 0.5 or 1.0 cm apart. The Nd: YAG laser LIMAX 120 (KLS Martin) was then applied for 5 seconds at 100 W vertically on each point on the lung surface in diagonal sequence. During and after each laser application, the area was observed with the thermal imaging camera. The temperature was recorded in a freeze-frame mode every 10 seconds in the square corners and in the square center (►Fig. 1). To evaluate the temperature drop mathematically, a nonlinear curve fit based on the one-phase exponential decay algorithm of the Prism 5 software (Graphical and Statistical Software, La Jolla, California, United States) was applied. Any morphological thermal damage in the lung tissue located between the lesions was evaluated by histological examination (hematoxylin and eosin staining) after the removal of the respective lung parenchyma.
Cooling Methods These lasered lobes were our controls (group A) in our cooling strategy experiments. In group B, another set of six lobes were lasered identically to group A, but the lasered areas were rinsed with ice-cold saline solution (at 4°C) during the laser procedure. The lobes in group C (n ¼ 6) were submerged in ice-cold water (at 4°C) for 5 seconds immediately after finishing the laser application. The surface temperature in all groups was recorded continuously by the thermal imaging camera as described above. Temperature was continuously measured until normothermia was achieved (defined as the lasered area’s peak temperature of 37°C).
Statistical Analysis At the beginning of our study, we did explorative experiments, using data for a power analysis (nonparametric test with GPower software, Institute of Experimental Psychology, Heinrich Heine University, Düsseldorf, Germany). Owing to the effect size, the sample size of the groups was calculated n ¼ 6. The Mann–Whitney U test was applied to determine differences in the initial temperatures (in the 0.5 vs. 1.0 cm quadrant corners). The Wilcoxon signed-rank test was used to compare paired temperature values in the quadrant corners
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
364
Kirschbaum et al.
Fig. 1 Experiment setup. (A) Laser lesions on the lung surface at a distance of 0. 5 cm. (B) Corresponding thermographic picture with temperature zones. (C) Example of a suspended paracardial lobe with attached 1 cm square stencil.
versus their centers. We applied the Kruskal–Wallis test and Dunn test for multiple comparisons to compare the lobes’ exposure times to temperatures over 42°C in groups A to C. All values are given as the mean standard deviation, unless otherwise noted. p values < 0.05 were considered statistically significant. These procedures were discussed and approved by our biostatistician. For statistics and graphing, GraphPad Prism 5 (Graphical and Statistical Software) was used.
Results The mean temperatures immediately after laser completion in the 0.5- and 1.0-cm square corners were 76 2.9 and 77 3.0°C, revealing no significant difference between the two groups with respect to the temperature at the direct laser application site (p ¼ 0.061). However, mean temperatures in the quadrant centers did differ significantly. Mean temperature in the 0.5-cm square centers was 103.2 8.5°C compared with 64.8 5.3°C in the 1.0-cm square centers (p < 0.05, corner vs. center temperature). Temperatures also differed depending on the distance of the laser lesion. When the laser lesions were 1.0 cm apart, the temperature in the quadrant center was significantly lower than that in the corners, and significantly higher when the lesions were 0.5 cm apart (p < 0.05, centers of 0.5 cm squares vs. 1 cm squares, ►Fig. 2).
The cooling kinetics in the 0.5 cm quadrant corners followed the formula: y ¼ 41.74 e0.028x þ 33.68. The lasered corners reached normothermia (37°C) after 81 14 seconds, whereas the quadrant centers took considerably longer with 108 30 seconds (p < 0.05, ►Fig. 3). The cooling kinetics in the centers followed the formula: y ¼ 62.14 e0.028x þ 38.52. In line with their lower initial temperature, the 1-cm square centers cooled down considerably faster in 93 22 seconds compared with the 0.5-cm square centers, and the kinetics followed the formula: y ¼ 28.83 e0.025x þ 34.49. The formula for cooling in the corners was y ¼ 38.54 e0.026x þ 36.77, and normothermia was reached after 120 22 seconds (p < 0.05). In the tissue between the laser lesions, histological examination of the lung parenchyma from the 1-cm quadrants revealed no morphological anomalies compared with normal parenchyma. However, in the 0.5-cm quadrants, the tissue between the lesions displayed increased cell vacuolization and membrane destruction, which resembled thermal cell damage (►Fig. 4). In group A, which received no additional cooling after laser application, the mean temperature after laser application was 76 2.9°C. Spontaneous cooling of the lung parenchyma took place exponentially, achieving normothermia after 91 14 seconds. The lung parenchyma’s mean exposure time to temperature over 42°C was 60 10 seconds. When the lasered area was rinsed with ice-cold saline during the laser procedure (group B), the initial temperature after Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
365
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Laser Resection of Lung Tissue
Laser Resection of Lung Tissue
Kirschbaum et al.
Fig. 2 (A) Initial temperature at the end of laser application. Temperature measurement was started right after completing laser application and continued until the highest measurable temperature dropped below 37°C, referred to as normothermia. Heat accumulation at the center of the 0.5 cm squares is evident by higher temperatures in the center compared with the corners. (B) Time to normothermia. Following laser application time until normothermia was measured. In the center of the 0.5 cm squares, a longer time period was needed to achieve normothermia compared with the corners, whereas in the 1.0 cm squares, the center cools down more rapidly compared with the corners. Asterisks indicate a value of p < 0.05 (Mann–Whitney U test) between corners and center.
completion of resection was 67 2.5°C. The lung tissue cooled down exponentially attaining normothermia after 93 20 seconds (►Fig. 5). Group B tissue was exposed to temperatures over 42°C for 46 15 seconds. Regarding the laser’s effect on lung tissue, it was noticed that the refraction of the laser rays by the continuous cold rinse lead to less efficacy in terms of tissue vaporization and instead to increased coagulation (►Fig. 6). Tissue vaporization is the preferred laser effect, especially if the resection of multiple metastases is intended. In group C, the mean initial temperature was 88 4.4°C which dropped dramatically to 44 3.3°C during the 5-second cooling period (►Fig. 5). Normothermia of the lung tissue was regained after 75 20 seconds. Group C tissue was exposed to temperatures over 42°C for only 18 9 seconds. This underlines the effect of the efficacy of the cooling method in
group C, which was very effective despite higher temperatures in this group compared with group B at the onset.
Discussion Laser application provides the ability to radically resect lung metastases of various sizes with a safety margin while sparing a maximum of healthy lung parenchyma, even when removing hard-to-access tumors.8 If a lobe contains multiple metastases, laser application is the only surgical option for local resection.9,10 However, the intact lung tissue heats up with each laser application. When laser lesions lie in close proximity, excessive heat can accumulate, and persistent high temperatures can lead to thermal tissue damage. Our experimental model demonstrated that four punctiform laser lesions forming squares with edge lengths of 0.5
Fig. 3 Temperature curves at the corners and centers of squares with 0.5 and 1.0 cm sides. Square stencils with these side lengths were used to reproducibly apply laser lesions to the surface of the paracardial pig lobes. Shown are the time-dependent changes in temperature at the corners and the center of the 0.5 and 1.0 cm squares. Temperature at the center of the 0.5 cm squares are higher compared with the corners, inversely when compared with the 1.0 cm squares, where the center is cooler than the corners. The dotted lines represent a temperature of 37.0°C. Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
366
Fig. 4 Histologic examination of the area () between the laser lesions (!) in a distance of 0.5 cm (hematoxylin and eosin staining, 12.5).
and 1.0 cm between consecutive lesions heated up the quadrant’s center considerably. However, the present study showed for the first time that distance is highly relevant. If the quadrant’s lesions were 1 cm apart, their centers were much cooler than the corners, and as initial temperature was lower, the lung tissue in the center also cooled down faster. Most important with 1 cm distance histological evaluation showed no signs of thermal damage in the cells in the central parenchyma. Closely adjacent laser lesions, for example, 0.5 cm apart, as in our model, cause substantial heat accumulation in the interstitial lung parenchyma at the square centers. Heat accumulates by the concentric spread and overlapping temperatures surrounding each lesion. When heat accumulates, the tissue is subjected to prolonged heat exposure, which dramatically raises the risk of thermal damage to the cells—a risk that becomes especially clinically relevant when resecting multiple adjacent metastases in a short time. So in these cases, the surgeon should apply a
Kirschbaum et al.
cooling strategy to protect the lung tissue from damage. As the lung tissue after laser resection cools down primarily by the mechanism of local heat release, the most effective cooling method might also be a local approach. In the present study, an experimental model was designed to analyze a safe and effective cooling method after lung laser resection. The results show for the first time that cooling with 4°C saline for 5 seconds is a very effective cooling method. Tissue temperatures immediately dropped down to 41°C, and normothermia was achieved in just 75 seconds. The tissue’s mean exposure time to temperature > 42°C was effectively shortened to less than 20 seconds and the laser’s efficacy is unaffected. In contrast, the cold rinse during the laser intervention only lowered the temperature at the end of laser application by approximately 10°C and the tissue’s cooling kinetics after finishing the laser application remained unchanged compared with the noncooled controls. The tissue’s exposure to temperatures > 42°C was only 13 seconds shorter than in the control group, resulting to an exposure to temperatures > 45°C more than 45 seconds. In addition, this cooling method was associated with another drawback. Cold rinse during laser application dramatically reduced the laser’s capacity to vaporize malignant tissue. These results are in contrast with the results of Beer et al,11 who examined the influence of water/air cooling on collateral tissue damage using a diode laser in a vitro study on liver tissue. We think that especially cooling by air during laser application is an interesting point to investigate it in further studies.
Limitations to the Study Choosing six pig lobes per group could be interpreted as an underpowered study; however, our aim was to demonstrate an effect that needs to be addressed in the real-life situation. We tried to create a reproducible lesion model to examine exactly the problem of tissue heat accumulation. All animal experimental models share the inherent limitation that they do not represent the exact condition in humans (e.g., the size and mass of the lobe). As this is an important issue in the real
Fig. 5 Time course of the temperature kinetics in groups A to C. Temperature recordings were performed continuously and documented every 10 seconds. Shown are temperature curves for group A (no cooling), B (rinsing with ice-cold saline during laser application), and C (submerging in icecold water for 5 seconds after laser application). The bold dotted line indicates a temperature of 42°C, the fine dotted line the temperature of 37° C, which was defined as normothermia. Note the temperature drop in group C when submerged in ice-cold water after laser resection, which we identified as the most effective cooling method. Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
367
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Laser Resection of Lung Tissue
Laser Resection of Lung Tissue
Kirschbaum et al.
Fig. 6 Effect of rinsing the lungs with saline during laser application: (A) Example of laser markings (!) after completing laser application in group B. Of note is the lower vaporization and stronger coagulation of the lung tissue. (B) Example of the laser marks (!) after completing laser application in group A.
life setting, the data presented are of high significance to proper patient care. As a consequence of our study, the surgeon should prompt to observe this phenomenon at the operation site and address it using adequate methods. In our opinion, there is a need for further studies. First, the mechanical alteration of the lung tissue, which was injured by heat accumulation, is not understood yet. Second, the rate of air leakage in laser resected areas after reventilation compared with other methods is not known. Third, we also have to compare these results with the resection by other devices, such as the monopolar cutter or the Harmonic scalpel, which are also in use for the resection of lung metastases.
References 1 Nichols FC. Pulmonary metastasectomy. Thorac Surg Clin 2012;
22(1):91–99, vii 2 The International Registry of lung Metastases: long-term results of
3 4
5
6
Conclusion In conclusion, based on the presented data, laser lesions further than 1 cm apart do not entail the risk of heat accumulation. However, if lung metastases are closer than 1 cm to each other, the surgeon should either wait 3 to 5 minutes for spontaneous tissue cooling before resecting the next metastasis, or proceed with the resection of a lesion at a distant site. If accelerated cooling is needed, our data show that submerging the lung in ice-cold saline immediately after completing the laser application is the most effective cooling strategy.
7
8
9
10
Acknowledgment We thank PD Dr. Helmut Sitter (Biostatistician, Institute for Surgical Research, University Hospital Giessen and Marburg GmbH, Marburg) for his friendly help and support for the statistical analysis of our data.
Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
11
lung metastasectomy. J Thorac Cardiovasc Surg 1997;113(1): 37–49 Hornbech K, Ravn J, Steinbrüchel DA. Current status of pulmonary metastasectomy. Eur J Cardiothorac Surg 2011;39(6):955–962 Pfannschmidt J, Klode J, Muley T, Dienemann H, Hoffmann H. Nodal involvement at the time of pulmonary metastasectomy: experiences in 245 patients. Ann Thorac Surg 2006;81(2):448–454 Yasui K, Kanazawa S, Sano Y, et al. Thoracic tumors treated with CT-guided radiofrequency ablation: initial experience. Radiology 2004;231(3):850–857 Rolle A, Pereszlenyi A, Koch R, Richard M, Baier B. Is surgery for multiple lung metastases reasonable? A total of 328 consecutive patients with multiple-laser metastasectomies with a new 1318nm Nd:YAG laser. J Thorac Cardiovasc Surg 2006;131(6):1236–1242 Rolle A, Koch R, Alpard SK, Zwischenberger JB. Lobe-sparing resection of multiple pulmonary metastases with a new 1318nm Nd:YAG laser—first 100 patients. Ann Thorac Surg 2002;74(3): 865–869 Venuta F, Rolle A, Anile M, Martucci N, Bis B, Rocco G. Techniques used in lung metastasectomy. J Thorac Oncol 2010;5(6, Suppl 2): S145–S150 Rolle A, Pereszlenyi A, Koch R, Bis B, Baier B. Laser resection technique and results of multiple lung metastasectomies using a new 1,318. nm Nd:YAG laser system. Lasers Surg Med 2006;38(1): 26–32 Rolle A, Koch R, Alpard SK, Zwischenberger JB. Lobe-sparing resection of multiple pulmonary metastases with a new 1318nm Nd:YAG laser—first 100 patients. Ann Thorac Surg 2002;74(3): 865–869 Beer F, Körpert W, Buchmair AG, et al. The influence of water/air cooling on collateral tissue damage using a disode laser with an innovative pulse design – an in vitro study. Lasers Med Sci 2013; 28(3):96–971
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
368
Copyright of Thoracic & Cardiovascular Surgeon is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
363
Laser Resection of Lung Tissue: Heat Accumulation from Adjacent Laser Application and How to Cool It Down Andreas Kirschbaum1
Peter Rexin2
Anika Pehl2
1 Department of Surgery, University Hospital, Marburg, Germany 2 Department of Pathology, University Hospital, Marburg, Germany 3 Department of Pathology, University Hospital, Erlangen, Germany
Detlef Bartsch1
Karl Quint3
Address for correspondence Andreas Kirschbaum, MD, Department of Surgery, University Hospital, Baldingerstrasse, Marburg 35033, Germany (e-mail: [email protected]).
Abstract
Keywords
► experimental ► metastases/ metastasectomy ► surgical equipment
received May 22, 2013 accepted September 17, 2013 published online December 2, 2013
Background Heat accumulation might induce thermal damage of the surrounding lung tissue, especially when multiple lesions are resected in one session. The present study aimed to investigate whether heat accumulates in the immediate vicinity of the resection surface and leads to thermal damage of the lung parenchyma, and what is the most effective cooling strategy in this situation. Materials and Methods In normothermic perfused paracardial swine lobes (n ¼ 6), four punctiform laser lesions forming a square were created. Each lesion was lasered at a power of 100 W for 5 seconds. Two test conditions with square sides of either 1.0 or 0.5 cm were compared. Temperatures were recorded immediately after completing the laser procedure in the square center and in the corners using a thermal camera and continued during the cooling process at 10-second intervals until normothermia (37°C). We examined two cooling methods: rinsing with ice-cold (4°C) Ringer solution during the laser procedure (group B, n ¼ 6) or submerging the lung in ice-cold water for 5 seconds immediately after laser application (group C, n ¼ 6). In the control group A (n ¼ 6), there was no cooling. Results In the 0.5 cm squares, mean temperature in the center immediately after laser application was 103.17 8.56°C, significantly higher than in the corners (76.39 2.87°C, p < 0.05). Normothermia in the quadrant corners was reached after 81 14 and after 108 29 seconds in the centers. Tissue in the square center revealed histological signs of thermic cell damage. In the 1.0 cm squares, mean temperature in the center was 64 5°C, and in the corners was 77 3.1°C (p < 0.05). Normothermia was regained after 93 22 seconds in the center and 120 21 seconds in the corners. Histological examination in the 1.0-quadrant centers revealed no signs of thermic cell damage. Submerging the lobe into ice-cold water lowered the temperature rapidly to under 40°C, and normothermia was regained after 75 1.3 seconds. Conclusion Laser application to the lung parenchyma causes considerable heat accumulation in closely related lesions. To prevent such cell damage, a distance of at least 1.0 cm between laser targets should be maintained. If no topical cooling method applied, sufficient time for spontaneous tissue cooling before additional laser application should be provided. The most effective cooling strategy against heat accumulation is submerging in ice-cold water for at least 5 seconds.
© 2014 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0033-1358780. ISSN 0171-6425.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Thorac Cardiovasc Surg 2014;62:363–368.
Laser Resection of Lung Tissue
Kirschbaum et al.
Introduction About 30% of patients with various malignancies develop lung metastases in one or both sides.1 Although solitary lung metastases are quite rare, multiple tumors are very common. Under certain circumstances, that is, the patient has adequate cardiopulmonary capacity, no mediastinal lymph node involvement, and an R0 resection is technically feasible, the resection of multiple metastases is recommended.2 Bilateral lung metastases are resected in separate sessions approaching the more complex side first. In fact, patients with multiple metastases usually undergo thoracotomy because further metastases are often detected intraoperatively by bidigital palpation of the lung tissue.3 A radical lymphadenectomy might also be performed depending on the primary tumor type.4 The goal of surgery for lung metastases is to spare as much healthy lung tissue as possible. In this regard, conventional operative methods have some disadvantages. Usually the tumors are atypically clamped off, resected, and the lung parenchyma has to sew over. Alternatively, resection with a stapler device can be employed. These techniques often lead to a considerable loss of healthy lung tissue. In deeper metastases, extra-anatomical resection by these methods might not be possible, so that an anatomic resection (i.e., segment removal or lobectomy) must be performed resulting in even more loss of healthy parenchyma. Alternatively, when the patients are not candidates for surgery owing to poor cardiopulmonary reserve or there were too much lesions, radiofrequency ablation might be a treatment option.5 Laser-guided resection of lung metastases has theoretically some key advantages compared with conventional surgical techniques. It enables the surgeon to operate in a very gentle manner on the lung parenchyma and causes very little bleeding.6 Usually the metastasis is resected with a safety margin of 5 mm around the tumor. When the laser strikes the lung tissue, its intensive heat causes vaporization of the lasered tissue and coagulation of the surrounding tissue.7,8 Thus, larger metastases can be removed with minimal blood loss. Up to 100 metastases in a single lobe can be removed by the Nd:YAG laser LIMAX 120 (Gebrüder Martin GmbH & Co KG, Tuttlingen, Germany) at a wavelength of 1,318 nm and a power of up to 120 W.6 However, as the temperature in the surrounding healthy tissue rises during every laser application, there is the theoretical risk of heat accumulation and critical cell damage in the tissue adjacent to each metastasis being resected. Since there are no data as yet regarding this issue, an experimental model was developed to analyze the heat accumulation in lung tissue and its relationship to the laser application distance. In addition, we aimed to determine by which method the lung tissue can effectively be cooled down.
Materials and Methods Laser Application Experiments were conducted on paracardial lung lobes (n ¼ 6) from freshly slaughtered pigs. After removing the Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
heart and both lungs en bloc, the paracardial lobes were inspected for any anomalies. The pulmonary arteries supplying the lobes were dissected from the surrounding tissue and cannulated with a peripheral venous catheter (12G). The lobes were then rinsed with heparin-saline solution (5.000 IU) until the drained fluid was free of blood particles. The lobes were then wrapped in moist compresses and immediately transported to the laboratory in a cold bag. In the laboratory with a constant ambient temperature, these specimens were fixed vertically and the cannulas hooked up to saline solution warmed to 42° C, which at the point of entry into the lung lobe reached a temperature of 37°C. All lobes were then perfused (perfusion pressure 20 cm H2 O) with this solution at normotonia, and the perfusion’s homogeneity documented by a thermal imaging camera (IC 120LV, Trotec, Heinsberg, Germany). The laser-lesion points were marked on the lung surface with a stencil forming a square with the four points either 0.5 or 1.0 cm apart. The Nd: YAG laser LIMAX 120 (KLS Martin) was then applied for 5 seconds at 100 W vertically on each point on the lung surface in diagonal sequence. During and after each laser application, the area was observed with the thermal imaging camera. The temperature was recorded in a freeze-frame mode every 10 seconds in the square corners and in the square center (►Fig. 1). To evaluate the temperature drop mathematically, a nonlinear curve fit based on the one-phase exponential decay algorithm of the Prism 5 software (Graphical and Statistical Software, La Jolla, California, United States) was applied. Any morphological thermal damage in the lung tissue located between the lesions was evaluated by histological examination (hematoxylin and eosin staining) after the removal of the respective lung parenchyma.
Cooling Methods These lasered lobes were our controls (group A) in our cooling strategy experiments. In group B, another set of six lobes were lasered identically to group A, but the lasered areas were rinsed with ice-cold saline solution (at 4°C) during the laser procedure. The lobes in group C (n ¼ 6) were submerged in ice-cold water (at 4°C) for 5 seconds immediately after finishing the laser application. The surface temperature in all groups was recorded continuously by the thermal imaging camera as described above. Temperature was continuously measured until normothermia was achieved (defined as the lasered area’s peak temperature of 37°C).
Statistical Analysis At the beginning of our study, we did explorative experiments, using data for a power analysis (nonparametric test with GPower software, Institute of Experimental Psychology, Heinrich Heine University, Düsseldorf, Germany). Owing to the effect size, the sample size of the groups was calculated n ¼ 6. The Mann–Whitney U test was applied to determine differences in the initial temperatures (in the 0.5 vs. 1.0 cm quadrant corners). The Wilcoxon signed-rank test was used to compare paired temperature values in the quadrant corners
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
364
Kirschbaum et al.
Fig. 1 Experiment setup. (A) Laser lesions on the lung surface at a distance of 0. 5 cm. (B) Corresponding thermographic picture with temperature zones. (C) Example of a suspended paracardial lobe with attached 1 cm square stencil.
versus their centers. We applied the Kruskal–Wallis test and Dunn test for multiple comparisons to compare the lobes’ exposure times to temperatures over 42°C in groups A to C. All values are given as the mean standard deviation, unless otherwise noted. p values < 0.05 were considered statistically significant. These procedures were discussed and approved by our biostatistician. For statistics and graphing, GraphPad Prism 5 (Graphical and Statistical Software) was used.
Results The mean temperatures immediately after laser completion in the 0.5- and 1.0-cm square corners were 76 2.9 and 77 3.0°C, revealing no significant difference between the two groups with respect to the temperature at the direct laser application site (p ¼ 0.061). However, mean temperatures in the quadrant centers did differ significantly. Mean temperature in the 0.5-cm square centers was 103.2 8.5°C compared with 64.8 5.3°C in the 1.0-cm square centers (p < 0.05, corner vs. center temperature). Temperatures also differed depending on the distance of the laser lesion. When the laser lesions were 1.0 cm apart, the temperature in the quadrant center was significantly lower than that in the corners, and significantly higher when the lesions were 0.5 cm apart (p < 0.05, centers of 0.5 cm squares vs. 1 cm squares, ►Fig. 2).
The cooling kinetics in the 0.5 cm quadrant corners followed the formula: y ¼ 41.74 e0.028x þ 33.68. The lasered corners reached normothermia (37°C) after 81 14 seconds, whereas the quadrant centers took considerably longer with 108 30 seconds (p < 0.05, ►Fig. 3). The cooling kinetics in the centers followed the formula: y ¼ 62.14 e0.028x þ 38.52. In line with their lower initial temperature, the 1-cm square centers cooled down considerably faster in 93 22 seconds compared with the 0.5-cm square centers, and the kinetics followed the formula: y ¼ 28.83 e0.025x þ 34.49. The formula for cooling in the corners was y ¼ 38.54 e0.026x þ 36.77, and normothermia was reached after 120 22 seconds (p < 0.05). In the tissue between the laser lesions, histological examination of the lung parenchyma from the 1-cm quadrants revealed no morphological anomalies compared with normal parenchyma. However, in the 0.5-cm quadrants, the tissue between the lesions displayed increased cell vacuolization and membrane destruction, which resembled thermal cell damage (►Fig. 4). In group A, which received no additional cooling after laser application, the mean temperature after laser application was 76 2.9°C. Spontaneous cooling of the lung parenchyma took place exponentially, achieving normothermia after 91 14 seconds. The lung parenchyma’s mean exposure time to temperature over 42°C was 60 10 seconds. When the lasered area was rinsed with ice-cold saline during the laser procedure (group B), the initial temperature after Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
365
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Laser Resection of Lung Tissue
Laser Resection of Lung Tissue
Kirschbaum et al.
Fig. 2 (A) Initial temperature at the end of laser application. Temperature measurement was started right after completing laser application and continued until the highest measurable temperature dropped below 37°C, referred to as normothermia. Heat accumulation at the center of the 0.5 cm squares is evident by higher temperatures in the center compared with the corners. (B) Time to normothermia. Following laser application time until normothermia was measured. In the center of the 0.5 cm squares, a longer time period was needed to achieve normothermia compared with the corners, whereas in the 1.0 cm squares, the center cools down more rapidly compared with the corners. Asterisks indicate a value of p < 0.05 (Mann–Whitney U test) between corners and center.
completion of resection was 67 2.5°C. The lung tissue cooled down exponentially attaining normothermia after 93 20 seconds (►Fig. 5). Group B tissue was exposed to temperatures over 42°C for 46 15 seconds. Regarding the laser’s effect on lung tissue, it was noticed that the refraction of the laser rays by the continuous cold rinse lead to less efficacy in terms of tissue vaporization and instead to increased coagulation (►Fig. 6). Tissue vaporization is the preferred laser effect, especially if the resection of multiple metastases is intended. In group C, the mean initial temperature was 88 4.4°C which dropped dramatically to 44 3.3°C during the 5-second cooling period (►Fig. 5). Normothermia of the lung tissue was regained after 75 20 seconds. Group C tissue was exposed to temperatures over 42°C for only 18 9 seconds. This underlines the effect of the efficacy of the cooling method in
group C, which was very effective despite higher temperatures in this group compared with group B at the onset.
Discussion Laser application provides the ability to radically resect lung metastases of various sizes with a safety margin while sparing a maximum of healthy lung parenchyma, even when removing hard-to-access tumors.8 If a lobe contains multiple metastases, laser application is the only surgical option for local resection.9,10 However, the intact lung tissue heats up with each laser application. When laser lesions lie in close proximity, excessive heat can accumulate, and persistent high temperatures can lead to thermal tissue damage. Our experimental model demonstrated that four punctiform laser lesions forming squares with edge lengths of 0.5
Fig. 3 Temperature curves at the corners and centers of squares with 0.5 and 1.0 cm sides. Square stencils with these side lengths were used to reproducibly apply laser lesions to the surface of the paracardial pig lobes. Shown are the time-dependent changes in temperature at the corners and the center of the 0.5 and 1.0 cm squares. Temperature at the center of the 0.5 cm squares are higher compared with the corners, inversely when compared with the 1.0 cm squares, where the center is cooler than the corners. The dotted lines represent a temperature of 37.0°C. Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
366
Fig. 4 Histologic examination of the area () between the laser lesions (!) in a distance of 0.5 cm (hematoxylin and eosin staining, 12.5).
and 1.0 cm between consecutive lesions heated up the quadrant’s center considerably. However, the present study showed for the first time that distance is highly relevant. If the quadrant’s lesions were 1 cm apart, their centers were much cooler than the corners, and as initial temperature was lower, the lung tissue in the center also cooled down faster. Most important with 1 cm distance histological evaluation showed no signs of thermal damage in the cells in the central parenchyma. Closely adjacent laser lesions, for example, 0.5 cm apart, as in our model, cause substantial heat accumulation in the interstitial lung parenchyma at the square centers. Heat accumulates by the concentric spread and overlapping temperatures surrounding each lesion. When heat accumulates, the tissue is subjected to prolonged heat exposure, which dramatically raises the risk of thermal damage to the cells—a risk that becomes especially clinically relevant when resecting multiple adjacent metastases in a short time. So in these cases, the surgeon should apply a
Kirschbaum et al.
cooling strategy to protect the lung tissue from damage. As the lung tissue after laser resection cools down primarily by the mechanism of local heat release, the most effective cooling method might also be a local approach. In the present study, an experimental model was designed to analyze a safe and effective cooling method after lung laser resection. The results show for the first time that cooling with 4°C saline for 5 seconds is a very effective cooling method. Tissue temperatures immediately dropped down to 41°C, and normothermia was achieved in just 75 seconds. The tissue’s mean exposure time to temperature > 42°C was effectively shortened to less than 20 seconds and the laser’s efficacy is unaffected. In contrast, the cold rinse during the laser intervention only lowered the temperature at the end of laser application by approximately 10°C and the tissue’s cooling kinetics after finishing the laser application remained unchanged compared with the noncooled controls. The tissue’s exposure to temperatures > 42°C was only 13 seconds shorter than in the control group, resulting to an exposure to temperatures > 45°C more than 45 seconds. In addition, this cooling method was associated with another drawback. Cold rinse during laser application dramatically reduced the laser’s capacity to vaporize malignant tissue. These results are in contrast with the results of Beer et al,11 who examined the influence of water/air cooling on collateral tissue damage using a diode laser in a vitro study on liver tissue. We think that especially cooling by air during laser application is an interesting point to investigate it in further studies.
Limitations to the Study Choosing six pig lobes per group could be interpreted as an underpowered study; however, our aim was to demonstrate an effect that needs to be addressed in the real-life situation. We tried to create a reproducible lesion model to examine exactly the problem of tissue heat accumulation. All animal experimental models share the inherent limitation that they do not represent the exact condition in humans (e.g., the size and mass of the lobe). As this is an important issue in the real
Fig. 5 Time course of the temperature kinetics in groups A to C. Temperature recordings were performed continuously and documented every 10 seconds. Shown are temperature curves for group A (no cooling), B (rinsing with ice-cold saline during laser application), and C (submerging in icecold water for 5 seconds after laser application). The bold dotted line indicates a temperature of 42°C, the fine dotted line the temperature of 37° C, which was defined as normothermia. Note the temperature drop in group C when submerged in ice-cold water after laser resection, which we identified as the most effective cooling method. Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
367
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Laser Resection of Lung Tissue
Laser Resection of Lung Tissue
Kirschbaum et al.
Fig. 6 Effect of rinsing the lungs with saline during laser application: (A) Example of laser markings (!) after completing laser application in group B. Of note is the lower vaporization and stronger coagulation of the lung tissue. (B) Example of the laser marks (!) after completing laser application in group A.
life setting, the data presented are of high significance to proper patient care. As a consequence of our study, the surgeon should prompt to observe this phenomenon at the operation site and address it using adequate methods. In our opinion, there is a need for further studies. First, the mechanical alteration of the lung tissue, which was injured by heat accumulation, is not understood yet. Second, the rate of air leakage in laser resected areas after reventilation compared with other methods is not known. Third, we also have to compare these results with the resection by other devices, such as the monopolar cutter or the Harmonic scalpel, which are also in use for the resection of lung metastases.
References 1 Nichols FC. Pulmonary metastasectomy. Thorac Surg Clin 2012;
22(1):91–99, vii 2 The International Registry of lung Metastases: long-term results of
3 4
5
6
Conclusion In conclusion, based on the presented data, laser lesions further than 1 cm apart do not entail the risk of heat accumulation. However, if lung metastases are closer than 1 cm to each other, the surgeon should either wait 3 to 5 minutes for spontaneous tissue cooling before resecting the next metastasis, or proceed with the resection of a lesion at a distant site. If accelerated cooling is needed, our data show that submerging the lung in ice-cold saline immediately after completing the laser application is the most effective cooling strategy.
7
8
9
10
Acknowledgment We thank PD Dr. Helmut Sitter (Biostatistician, Institute for Surgical Research, University Hospital Giessen and Marburg GmbH, Marburg) for his friendly help and support for the statistical analysis of our data.
Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
11
lung metastasectomy. J Thorac Cardiovasc Surg 1997;113(1): 37–49 Hornbech K, Ravn J, Steinbrüchel DA. Current status of pulmonary metastasectomy. Eur J Cardiothorac Surg 2011;39(6):955–962 Pfannschmidt J, Klode J, Muley T, Dienemann H, Hoffmann H. Nodal involvement at the time of pulmonary metastasectomy: experiences in 245 patients. Ann Thorac Surg 2006;81(2):448–454 Yasui K, Kanazawa S, Sano Y, et al. Thoracic tumors treated with CT-guided radiofrequency ablation: initial experience. Radiology 2004;231(3):850–857 Rolle A, Pereszlenyi A, Koch R, Richard M, Baier B. Is surgery for multiple lung metastases reasonable? A total of 328 consecutive patients with multiple-laser metastasectomies with a new 1318nm Nd:YAG laser. J Thorac Cardiovasc Surg 2006;131(6):1236–1242 Rolle A, Koch R, Alpard SK, Zwischenberger JB. Lobe-sparing resection of multiple pulmonary metastases with a new 1318nm Nd:YAG laser—first 100 patients. Ann Thorac Surg 2002;74(3): 865–869 Venuta F, Rolle A, Anile M, Martucci N, Bis B, Rocco G. Techniques used in lung metastasectomy. J Thorac Oncol 2010;5(6, Suppl 2): S145–S150 Rolle A, Pereszlenyi A, Koch R, Bis B, Baier B. Laser resection technique and results of multiple lung metastasectomies using a new 1,318. nm Nd:YAG laser system. Lasers Surg Med 2006;38(1): 26–32 Rolle A, Koch R, Alpard SK, Zwischenberger JB. Lobe-sparing resection of multiple pulmonary metastases with a new 1318nm Nd:YAG laser—first 100 patients. Ann Thorac Surg 2002;74(3): 865–869 Beer F, Körpert W, Buchmair AG, et al. The influence of water/air cooling on collateral tissue damage using a disode laser with an innovative pulse design – an in vitro study. Lasers Med Sci 2013; 28(3):96–971
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
368
Copyright of Thoracic & Cardiovascular Surgeon is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
