- Email: [email protected]
Defining New Metrics in Microwave Ablation of Pulmonary Tumors: Ablation Work and Ablation Resistance Score
CLINICAL STUDY
Defining New Metrics in Microwave Ablation of Pulmonary Tumors: Ablation Work and Ablation Resistance Score Ramsey A. Al-Hakim, MD, Fereidoun G. Abtin, MD, Scott J. Genshaft, MD, Erin Kutay, MD, MPH, and Robert D. Suh, MD
ABSTRACT Purpose: To investigate pulmonary microwave ablation metrics including ablation work, ablation resistance score, and involution. Materials and Methods: Retrospective review was performed of 98 pulmonary tumor ablations using the NeuWave Certus Microwave Ablation System (NeuWave Medical, Madison, Wisconsin) in 71 patients (32 men and 39 women; mean age, 64.7 y ⫾ 11.5). Ablation work was defined as sum of (power) * (time) * (number of antennas) for all phases during an ablation procedure. Ablation zone was measured on CT at 3 time points: after procedure, 1–3 months (mean 47 d), and 3–12 months (mean 292 d). Ablation zones were scored based on location for pulmonary lobe (upper ¼ 1, middle/lingula ¼ 2, lower ¼ 3) and region (peripheral ¼ 1, parenchymal ¼ 2, central ¼ 3), and the 2 were summed for ablation resistance score. Results: Ablation zone on CT at 1–3 months was significantly smaller in regions with higher ablation resistance score (P o .05). There was a significant correlation between ablation work and ablation zone measured on CT performed after procedure (P o .001), at 1–3 months (P o .001), and at 3–12 months (P o .05). Ablation zone significantly decreased from after procedure to 1–3 months (P o .001) and from 1–3 months to 3–12 months (P o .001), with change from after procedure to 1–3 months significantly greater (P o .01). Conclusions: Pulmonary microwave ablation zone is significantly smaller in regions with higher ablation resistance score. Ablation work correlates to ablation zone with a nonlinear involution pattern in the first year and may be useful for planning before the procedure.
ABBREVIATIONS LN = Lung (antenna), PR = Precision (antenna)
Thermal ablative techniques applied to the lung include radiofrequency (RF) ablation, microwave ablation, and cryoablation. Compared with RF ablation, microwave From the Department of Radiology, Division of Thoracic Imaging, University of California, Los Angeles, 757 Westwood Plaza, Suite 1638, Los Angeles, CA 90024. Received January 24, 2016; final revision received May 20, 2016; accepted May 21, 2016. Address correspondence to R.A.A.; E-mail: [email protected] F.G.A. is a paid consultant for HealthTronics, Inc (Austin, Texas) and participated in a clinical trial for Galil Medical, Inc (Arden Hills, Minnesota). R.D.S. is a paid consultant for NeuWave Medical, Inc (Madison, Wisconsin) and HealthTronics, Inc, receives grants from HealthTronics, Inc, and performed research for Galil Medical, Inc. None of the other authors have identified a conflict of interest. & SIR, 2016 J Vasc Interv Radiol 2016; 27:1380–1386 http://dx.doi.org/10.1016/j.jvir.2016.05.026
ablation in the lungs has been shown to produce a larger ablation zone (1). Wolf et al (2) reported the clinical success of microwave ablation for treatment of pulmonary malignancies in patients deemed medically inoperable with cancer-specific mortality yielding a 1year survival of 83%. Perfusion-mediated tissue cooling (heat sink) describes the perfusion-related loss of energy deposition associated with thermal ablation, which may be a cause of local recurrence secondary to sublethal temperatures of the target lesion (3–5). Prior studies demonstrated the heat-sink effect in the lung with RF ablations adjacent to vessels 4 3 mm, with an increased rate of recurrence after ablation of tumors in direct contact with such vessels (6,7). Studies have also suggested that microwave ablation is less susceptible to the heat-sink effect compared with RF ablation (8–10). In addition to perfusion, impedance caused by ventilation is
Volume 27
’
Number 9
’
September
’
2016
a unique confounding factor associated with thermal ablation in the lungs. Decreased ventilation via bronchial occlusion with single-lung ventilation has been shown in animal studies to increase microwave and RF ablation zone size in the lungs (11,12). Both ventilation and perfusion have a meaningful impact on pulmonary microwave ablation, and each has been shown to have regional variances throughout the lung (13,14). Studies have demonstrated higher perfusion of the lower/central lung and greater ventilation in the lower lung (13,15–17). The in vivo effects of these regional perfusion and ventilation variations on microwave ablation have not been well studied to date. Multiple previous studies reviewed imaging after procedures to demonstrate characteristics predicting treatment success, including an ablation zone size 4 times greater than the tumor size before ablation (18–23). Although it is believed that ablation energy delivered is important for ablation zone size, previous studies mostly focused on ablation power (24,25). Ablation work (power * time) is a measure of the total energy delivered and, in contrast to power alone, can simultaneously account for power, ablation time, number of antennas, and differences in these parameters between ablation phases. Clinical outcomes with thermal ablation in the lungs are an important area of continued investigation. However, the purpose of this study is to investigate technical factors of in vivo microwave pulmonary ablation, including ablation work, temporal involution, and regional ablation resistance.
MATERIALS AND METHODS After institutional review board approval, a retrospective chart review of 98 noncavitary pulmonary tumor ablations using a NeuWave Certus Microwave Ablation System (NeuWave Medical, Inc, Madison, Wisconsin) was performed. All cases were performed percutaneously under computed tomography (CT) guidance by 1 of 3 operators (R.D.S., F.G.A., S.J.G.) between June 2011 and July 2015. Exclusion criteria included use of an ablation antenna other than the Precision (PR) or Lung (LN) antenna. Ablation parameters were at the discretion of the operator and with the following treatment endpoints: tissue temperature 4 601C, concentric ground-glass margin 4 5 mm, and operator security with obtained technical success. Ablation power and time were obtained from the ablation system recordings for each ablation phase (defined as continuous ablation at a single power); 45 of 98 (45.9%) pulmonary tumors were treated with 4 1 ablation phase. Ablation work was calculated as the sum of work (power * time) for each individual phase according to the equation n P Pi nti nN (WA = ablation work [J], n = number WA ¼ i¼1
of ablation phases, P = power [W], t = ablation time [s], N = number of ablation antennas).
1381
CT scans performed immediately after the procedure, 1–3 months after the procedure (earliest thoracic CT scan in 1–3 month period, n = 80), and 3–12 months after the procedure (latest thoracic CT scan in 3–12 month period, n = 66) were reviewed by a single senior radiology resident (R.A.A.) blinded to procedural ablation factors. Ablation zone refers to the area of ablation on CT and was defined as the outer edge of the peripheral margin of dense ground-glass opacity (5,23). Maximal length and maximal perpendicular measurements were obtained for each ablation zone on the image with the largest apparent ablation zone. Ablation zone was then calculated according to the equation for an ellipse (A ¼ πab). Involution of the ablation zone refers to the decrease in ablation zone over time as a result of elimination of induced coagulation and accompanying cicatrization (5). The shortest distance from the ablation zone isocenter to the closest costal pleura was measured for each ablation zone and categorized into the peripheral (r 2 cm), parenchymal (2–4 cm), or central (Z 4 cm) pulmonary region. Each ablation was assigned a score for pulmonary lobe of ablation (upper lobe ¼ 1, middle lobe/lingula ¼ 2, lower lobe ¼ 3) and pulmonary region of ablation (peripheral ¼ 1, parenchymal ¼ 2, central ¼ 3). The 2 scores were summed for an ablation resistance score ranging from 2 to 6. Ablation resistance scores were categorized into low (score of 2–3), medium (score of 4), and high (score of 5–6) ablation resistance score groups (Fig 1). Statistical analysis was performed using IBM SPSS Statistics for Windows software (IBM Corporation, Armonk, New York). Variables with a significantly skewed distribution (tumor size before ablation, ablation zone after procedure, ablation zone at 1–3 months, ablation zone at 3–12 months, ablation work, and maximum/perpendicular ablation dimensions) were compared using Kruskal-Wallis test. Otherwise, variables with an assumed normal distribution and homogeneous variance based on Levene test were comparatively analyzed using analysis of variance. Statistical significance was assessed with an α value of .05. Ordinal and nominal variables were compared using Fisher exact test. All correlations were performed with Pearson correlation method (2-tailed) with calculation of the Pearson correlation coefficient (r). Mean values are reported with ⫾ SD. Patients included 32 men and 39 women. Mean patient age was 64.7 years ⫾ 11.5. Oncologic histology included 34 primary pulmonary tumors (28 adenocarcinoma, 4 squamous cell carcinoma, 2 neuroendocrine) and 64 metastatic pulmonary tumors (21 colorectal, 19 renal, 7 uterine, 7 adenoid cystic carcinoma, 2 pancreatic, 2 hepatic, 2 ependymoma, 1 endometrial adenocarcinoma, 1 myxoid fibrosarcoma, 1 melanoma, 1 cardiac sarcoma). Mean maximal tumor diameter was 1.5 cm ⫾ 0.7 (range, 0.4–3.6 cm); 4 tumors were 4 3 cm
1382
’
Ablation Work and Ablation Resistance Score
Al-Hakim et al
’
JVIR
months (mean 3.0 cm2 ⫾ 2.9) (P o .001) (Fig 2). The ablation zone change on CT scan at 1–3 months (mean 47 d ⫾ 24) was significantly greater (P o .01) than the subsequent ablation zone change at 3–12 months (mean 292 d ⫾ 110). Similarly, the percentage ablation zone reduction from after the procedure to 1–3 months (mean 42.0% ⫾ 28.7) was significantly greater than the percentage ablation zone reduction from 1–3 months to 3–12 months relative to the original ablation zone (mean 27.2% ⫾ 20.5) (P o .001).
Ablation Resistance Score
Figure 1. Schematic demonstrating different ablation resistance score groups: low score (ablation resistance score ¼ 2– 3), medium score (ablation resistance score ¼ 4), and high score (ablation resistance score ¼ 5–6). Dashed lines indicate divisions between pulmonary regions, and solid lines indicate divisions between pulmonary lobes. Dot-dashed line represents division between left upper lobe and lingula.
There was no significant difference between the ablation resistance score groups in tumor size before ablation, antenna type, ablation work, maximum ablation power, total ablation time, or patient age. There was a significant difference in ablation zone at 1–3 months between the low (8.2 cm2 ⫾ 4.6), medium (6.4 cm2 ⫾ 4.7), and high (4.6 cm2 ⫾ 3.1) ablation resistance score groups (P o .05) (Fig 3). This statistically significant stratification of ablation zone between ablation resistance score groups was not observed after the procedure or at 3– 12 months, although there was a similar non–statistically significant trend in both groups (P ¼ .07 and P ¼ .15). Independent stratification between upper, middle, and lower lobes and peripheral, parenchymal, and central pulmonary zones did not show a significant difference in ablation zone at any time point.
in maximal diameter, and no tumors were 4 5 cm in maximal diameter.
Antenna Analysis
RESULTS Ablation Work Ablation procedures were performed using the NeuWave Certus PR (n ¼ 59) or LN (n ¼ 39) antenna, with single antenna microwave ablation (n ¼ 82) or simultaneous microwave ablation using 2 (n ¼ 15) or 3 (n ¼ 1) antennas. Mean ablation work for all ablations was 41.3 kJ ⫾ 21.3 (range, 9–117 kJ). There was a significant correlation of ablation work with ablation zone on CT after the procedure (P o .001, r ¼ .633), at 1–3 months (P o .001, r ¼ .620), and at 3–12 months (P o .05, r ¼ .249). There was no significant correlation of total ablation time (range, 164–1,380 s) with ablation zone at any time point. There was a significant correlation of maximum ablation power with ablation zone after the procedure (P o .001, r ¼ .414), at 1–3 months (P o .001, r ¼ .555), and at 3–12 months (P o .01, r ¼ .361).
Ablation Zone Involution There was a significant reduction in mean ablation zone on CT scans from after the procedure to 1–3 months (mean 4.8 cm2 ⫾ 4.6) and from 1–3 months to 3–12
All cases performed with a single ablation antenna using either the Certus LN (n ¼ 35) or PR (n ¼ 47) ablation antenna were subanalyzed. There was significantly higher ablation work with use of the LN antenna (mean 49.9 kJ ⫾ 19.5) versus the PR antenna (mean 30.8 kJ ⫾ 15.1), and the LN antenna produced a significantly larger ablation zone over the PR antenna at all time points (P o .01 for all). In addition, there was significantly higher maximum ablation power used for the LN antenna (mean 95.8 W ⫾ 23.9; range, 40–140 W) versus the PR antenna (mean 53.9 W ⫾ 16.6; range, 20– 65 W). Most PR antenna ablations were performed with a maximum power of 65 W (68.1%; 32 of 47) or 30 W (21.3%; 10 of 47), whereas most LN antenna ablations were performed with a maximum power of 90 W (65.7%; 23 of 35) or 140 W (14.3%; 5 of 35). The PR antenna demonstrated a significant correlation of ablation work with ablation zone after the procedure (P o .001, r ¼ .521) and at 1–3 months (P o .001, r ¼ .513) but not at 3–12 months. In addition, the PR antenna had a significant correlation of ablation work with the maximal and perpendicular measurements of the ablation zone after the procedure and at 1–3 months (P o .01 for all). The LN antenna ablation work did not have a significant correlation with
Volume 27
’
Number 9
’
September
’
2016
1383
Figure 2. Mean ablation zone at all time points (SE bars included), with a statistically significant involution at both time points (P o .001). The ablation zone reduction from after procedure to 1–3 months was significantly greater than ablation zone reduction at 3–12 months (P o .01). Percentage ablation zone reduction relative to ablation zone after procedure between each time point is shown for reference.
The PR single antenna ablations were stratified into 3 near-equal sampled categories based on ablation work: r 25 kJ (n ¼ 16), 25–35 kJ (n ¼ 15), and Z 35 kJ (n ¼ 16) (Table). There was a significant difference in ablation zone, maximal ablation zone length, and perpendicular ablation zone measurement after the procedure and at 1– 3 months between these 3 categories (P o .01), with the exception of the perpendicular ablation zone measurement at 1–3 months, which showed a similar trend between groups without statistical significance (P ¼ .058). The maximum ablation power was significantly lower for the r 25 kJ category compared with the 25–35 kJ and Z 35 kJ categories (P o .001).
DISCUSSION
Figure 3. Mean ablation zone for ablation resistance score groups at all time points (SE bars included), with a significant difference between the ablation resistance score groups at 1–3 months (P o .05).
ablation zone or dimensional measurements after the procedure, at 1–3 months, or at 3–12 months.
Analysis of 98 pulmonary tumor ablations with a single microwave ablation system demonstrates a correlation of pulmonary ablation work with ablation zone on imaging during the first year. Given lack of correlation between ablation times with ablation zone, the ablation work correlation is predominantly driven by the selection of ablation power. However, use of work as a reference allows simultaneous quantification and standardization of ablation power and time in the setting of multiple ablation phases and multiple ablation antennas, even when ablation phases are performed at different power levels. These results also demonstrate a disproportionately higher reduction in ablation zone at 1–3 months (mean 47 d) compared with 3–12 months (mean 292 d), indicative of a nonlinear involution pattern. Positive deviation from the expected involution curve
1384
’
Ablation Work and Ablation Resistance Score
Al-Hakim et al
’
JVIR
Table . Ablation Zone Measurements within Ablation Work Categories for NeuWave Certus Precision Microwave Antenna o 25 kJ (n ¼ 16)
25–35 kJ (n ¼ 15)
4 35 kJ (n ¼ 16)
Maximum ablation power (W)*†
36.9 ⫾ 14.9
62.7 ⫾ 9.0
62.8 ⫾ 8.8
Ablation time (s)*‡
474 ⫾ 174
578 ⫾ 169
839 ⫾ 221
6.4
7.2
10.1
3.9
4.1
4.7
2.1
2.2
2.7
Area (cm2)‡
2.8
3.7
5.2
Maximum length (cm)‡ Perpendicular measurement (cm)
2.6 1.4
2.8 1.6
3.5 1.8
Ablation Zone after Procedure Area (cm2)‡ Maximum length (cm)‡ Perpendicular measurement (cm)‡ Ablation Zone at 1–3 Months
Note–Data are means unless otherwise noted. *Data are mean ⫾ SD. † Significant difference between o 25 kJ group and other 2 groups (P o .001). ‡ Significant difference between all 3 groups (P o .01).
may potentially be used as a clinical guide for predicting early incomplete tumor ablation or tumor progression. The correlation of clinical and imaging outcomes relative to a similar involution curve should be the focus of future study. Multiple ablation antennas are available for use with the NeuWave Certus Microwave Ablation System. Initially, the LN antenna was the only available ablation antenna designed for lung ablation with the NeuWave System. In mid-2012, the PR antenna was introduced at our institution, and some operators adopted this antenna into their practice. Data using these antennas demonstrate a strong relationship of ablation work to ablation zone with use of the PR antenna but not the LN antenna. This result is likely related to the smaller sample size of lung ablation performed with the LN antenna given the similar trend without statistical significance at after the procedure (P ¼ .07, r ¼ .308) and at 3–12 months (P ¼ .10, r ¼ .363). Similarly, the relatively small sample size did not allow reliable statistical subanalysis of simultaneous ablations with 2 and 3 antennas. The correlation of PR antenna ablation work with ablation zone allowed stratification of ablation work into 3 ablation work categories that produced significantly different ablation zones (Table). These data can be used as a starting reference for pulmonary ablation procedural planning using the NeuWave Certus PR antenna, and this reference is potentially a more complete reference than what is currently available. Based on these results, an ablation power of 30 W with an ablation time of approximately 10 minutes should be used to achieve an ablation zone in the o 25 kJ category. To increase the ablation zone to the 25–35 kJ category, the power should be increased to 65 W with an ablation time of approximately 10 minutes. Finally, to reach the 4 35 kJ category, the ablation time should be increased to 4 10 minutes with a power of 65 W.
Prior studies evaluating the susceptibility of microwave ablation to perfusion-mediated tissue cooling (heat-sink effect), a perfusion-related decrease in energy deposition limiting thermal ablation, have targeted its effect on ablation based on the presence of vasculature, specifically vessel size, within and in the immediate vicinity of the ablation zone in the liver (3,4). Although a similar heat-sink effect related to a minimum threshold of vessel size would be expected with microwave ablation in the pulmonary parenchyma, the lung environment that dictates the deposition of ablation energy and eventually ablation zone is no doubt more complex, given the immense generalized perfusion of the lung through its microvasculature as well as the effect of continuous ventilation throughout respiration (7). Simply, the differential pulmonary perfusion and ventilation throughout the lung must be taken into account, perhaps more so than the anticipated influence of single or multiple vessels in the region of targeted ablation. It was our hypothesis that regions of higher perfusion and ventilation would result in a smaller ablation zone size. Based on studies of lung perfusion and ventilation, the lower lobes have higher perfusion and ventilation compared with the upper lobes and were therefore assigned a higher ablation resistance score (13,17). In addition, the central lung zones have higher perfusion and were assigned a higher ablation resistance score. Using this method, ablation resistance score correlated to a statistically significant stratification of ablation zone: smaller ablation zone in regions with a higher ablation resistance score, larger ablation zone in regions with a lower ablation resistance score, and intermediate ablation zone with midrange ablation resistance score. In physics terms, the efficiency of energy deposition by microwave energy may be referred to as effective conductivity (26). Ablation resistance score is fundamentally a qualitative inverse measurement of effective conductivity in the lung. With further validation, this proposed ablation resistance scoring
Volume 27
’
Number 9
’
September
’
2016
system may find value for microwave ablation for pulmonary malignancies not only with implications for planning before the procedure but also as a decision tool during the procedure. Tumor ablations anticipated in high ablation resistance score regions would require more ablation work to achieve the desired ablation zone. During the procedure, the decision to apply additional or more ablation work would be relatively straightforward during the treatment of tumors within high ablation resistance score regions, especially if ground-glass envelopment of the targeted tumor, an imaging surrogate used as an ablation endpoint, was less than expected. This study has several limitations. Ablations were monitored by the operators using some objective imaging parameters, such as the manufacturer ablation chart and ground-glass zone surrounding the tumor after ablation, with awareness of known phenomena such as perfusion-mediated tissue cooling. However, the retrospective aspect of this study with subjective operator treatment parameters and endpoint determination did not allow for reliable standardization and resulted in varied imaging timing after ablation. Furthermore, although ablation work was used as the primary indicator of energy deposited, it is still unclear from our analysis if all ablation work is the same. For example, are 2 5-minute ablation cycles at 65 W the same as 1 10minute ablation cycle at 65 W? The retrospective nature of this study and limited sample size did not allow reliable stratification to analyze for this difference. In addition, although our proposed ablation resistance scoring is roughly based on perfusion and ventilation variances, the regional variations of these parameters are complex, positional, and likely not fully captured by the 3 ablation resistance score groups. Furthermore, although a single trainee performed data collection, consensus regarding the definition and imaging findings of the ablation zone were established by the subspecialty trained radiologists before collection of data. Finally, the focus of this study was to describe ablation work and ablation resistance score, but future studies are required to correlate these findings with pathology and clinical outcomes, ideally using volumetric analysis (27). Specifically, future studies correlating the true histologic tumor necrosis volume and potential biomarkers over time with ablation work and the corresponding imaging ablation zone in different ablation resistance groups would validate the use of these new metrics. In conclusion, the findings in this study provide insight into microwave ablation properties in the lung and have implications for use in clinical care of patients receiving lung ablation, including planning before the procedure, decision guidance during the procedure, and imaging follow-up after the procedure. Although prior studies have demonstrated varied response to thermal ablation of lung tumors adjacent to vessels 4 3 mm, the deposition of ablation energy and subsequent ablation
1385
zone is clearly more complex. The ablation resistance score proposed in this study takes into account varied perfusion and ventilation throughout the lung when correlating with variations of the ablation zone; higher ablation resistance scores were found to correlate with smaller ablation zones, which can have important treatment implications (6,7). Similarly, ablation work accounts for differential power and time for multiple ablation phases and/or antennas and correlates with ablation zone on follow-up imaging in the first year. With validation by future clinical studies, the approach using ablation work and ablation resistance score during the planning and treatment phases of lung tumor ablation with monitoring of procedural imaging could decrease the likelihood of incomplete tumor ablation.
REFERENCES 1. Brace CL, Hinshaw JL, Laeseke PF, et al. Pulmonary thermal ablation: comparison of radiofrequency and microwave devices by using gross pathologic and CT findings in a swine model. Radiology 2009; 251: 705–711. 2. Wolf FJ, Grand DJ, Machan JT, et al. Microwave ablation of lung malignancies: effectiveness, CT findings, and safety in 50 patients. Radiology 2008; 247:871–879. 3. Yu NC, Raman SS, Kim YJ, et al. Microwave liver ablation: influence of hepatic vein size on heat-sink effect in a porcine model. J Vasc Interv Radiol 2008; 19:1087–1092. 4. Lu DS, Raman SS, Vodopich DJ. Effect of vessel size on creation of hepatic radiofrequency lesions in pigs: assessment of the “heat sink” effect. AJR Am J Roentgenol 2008; 178:47–51. 5. Ahmed M, Solbiati L, Brace C, et al. Image-guided tumor ablation: standardization of terminology and reporting criteria—a 10-year update. J Vasc Interv Radiol 2014; 25:1691–1705. 6. Gillams AR, Lees WR. Radiofrequency ablation of lung metastases: factors influencing success. Eur Radiol 2008; 18:672–677. 7. Steinke K, Haghighi KS, Wulf S, et al. Effect of vessel diameter on the creation of ovine lung radiofrequency lesions in vivo: preliminary results. J Surg Res 2005; 124:85–91. 8. Goldberg SN, Grassi CJ, Cardella JF, et al; Society of Interventional Radiology Technology Assessment Committee; International Working Group on Image-Guided Tumor Ablation. Image-guided tumor ablation: standardization of terminology and reporting criteria. Radiology 2005; 235:728–739. 9. Hinshaw JL, Lee FT Jr. Cryoablation for liver cancer. Tech Vasc Interv Radiol 2007; 10:47–57. 10. Wright AS, Sampson LA, Warner TF, et al. Radiofrequency versus microwave ablation in a hepatic porcine model. Radiology 2005; 236: 132–139. 11. Oshima F, Yamakado K, Akeboshi M, et al. Lung radiofrequency ablation with and without bronchial occlusion: experimental study in porcine lungs. J Vasc Interv Radiol 2004; 15:1451–1456. 12. Santos RS, Gan J, Ohara CJ, et al. Microwave ablation of lung tissue: impact of single-lung ventilation on ablation size. Ann Thorac Surg 2010; 90:1116–1120. 13. Levin DL, Chen Q, Zhang M, et al. Evaluation of regional pulmonary perfusion using ultrafast magnetic resonance imaging. Magn Reson Med 2001; 46:166–171. 14. Milic-Emili J. Static distribution of lung volumes. Comprehensive Physiology 2011; 561–574. 15. Grassino AE, Bake B, Martin RR, et al. Voluntary changes of thoracoabdominal shape and regional lung volumes in humans. J Appl Physiol 1975; 39:997–1003. 16. Sybrecht G, Landau L, Murphy BG, et al. Influence of posture on flow dependence of distribution of inhaled 133Xe boli. J Appl Physiol 1976; 41: 489–496. 17. Chang YH, Yu CP. A model of ventilation distribution in the human lung. Aerosol Sci Technol 1999; 30:309–319.
1386
’
Ablation Work and Ablation Resistance Score
18. Gazelle GS, Goldberg SN, Solbiati L, et al. Tumor ablation with radiofrequency energy. Radiology 2000; 217:633–646. 19. Anderson EM, Lees WR, Gillams AR. Early indicators of treatment success after percutaneous radiofrequency of pulmonary tumors. Cardiovasc Intervent Radiol 2009; 32:478–483. 20. Lee JM, Jin GY, Goldberg SN, et al. Percutaneous radiofrequency ablation for inoperable non-small cell lung cancer and metastases: preliminary report. Radiology 2004; 230:125–134. 21. Yamamoto A, Nakamura K, Matsuoka T, et al. Radiofrequency ablation in a porcine lung model: correlation between CT and histopathologic findings. AJR Am J Roentgenol 2005; 185:1299–1306. 22. de Baère T, Palussière J, Aupérin A, et al. Midterm local efficacy and survival after radiofrequency ablation of lung tumors with minimum followup of 1 year: prospective evaluation. Radiology 2006; 240:587–596.
Al-Hakim et al
’
JVIR
23. Chheang S, Abtin F, Gutierrez A, et al. Imaging features following thermal ablation of lung malignancies. Semin Intervent Radiol 2013; 30: 157–168. 24. Sonntag PD, Hinshaw JL, Lubner MG, et al. Thermal ablation of lung tumors. Surg Oncol Clin N Am 2011; 20:369. 25. Andreano A, Huang Y, Franca Meloni M, et al. Microwaves create larger ablations than radiofrequency when controlled for power in ex vivo tissue. Med Phys 2010; 37:2967–2973. 26. Brace CL. Radiofrequency and microwave ablation of the liver, lung, kidney and bone: what are the differences. Curr Probl Diagn Radiol 2009; 38:135–143. 27. Ozaki T, Mori I, Nakamura M, et al. Microwave cell death: immunohistochemical and enzyme histochemical evaluation. Pathol Int 2003; 53: 686–692.
Defining New Metrics in Microwave Ablation of Pulmonary Tumors: Ablation Work and Ablation Resistance Score Ramsey A. Al-Hakim, MD, Fereidoun G. Abtin, MD, Scott J. Genshaft, MD, Erin Kutay, MD, MPH, and Robert D. Suh, MD
ABSTRACT Purpose: To investigate pulmonary microwave ablation metrics including ablation work, ablation resistance score, and involution. Materials and Methods: Retrospective review was performed of 98 pulmonary tumor ablations using the NeuWave Certus Microwave Ablation System (NeuWave Medical, Madison, Wisconsin) in 71 patients (32 men and 39 women; mean age, 64.7 y ⫾ 11.5). Ablation work was defined as sum of (power) * (time) * (number of antennas) for all phases during an ablation procedure. Ablation zone was measured on CT at 3 time points: after procedure, 1–3 months (mean 47 d), and 3–12 months (mean 292 d). Ablation zones were scored based on location for pulmonary lobe (upper ¼ 1, middle/lingula ¼ 2, lower ¼ 3) and region (peripheral ¼ 1, parenchymal ¼ 2, central ¼ 3), and the 2 were summed for ablation resistance score. Results: Ablation zone on CT at 1–3 months was significantly smaller in regions with higher ablation resistance score (P o .05). There was a significant correlation between ablation work and ablation zone measured on CT performed after procedure (P o .001), at 1–3 months (P o .001), and at 3–12 months (P o .05). Ablation zone significantly decreased from after procedure to 1–3 months (P o .001) and from 1–3 months to 3–12 months (P o .001), with change from after procedure to 1–3 months significantly greater (P o .01). Conclusions: Pulmonary microwave ablation zone is significantly smaller in regions with higher ablation resistance score. Ablation work correlates to ablation zone with a nonlinear involution pattern in the first year and may be useful for planning before the procedure.
ABBREVIATIONS LN = Lung (antenna), PR = Precision (antenna)
Thermal ablative techniques applied to the lung include radiofrequency (RF) ablation, microwave ablation, and cryoablation. Compared with RF ablation, microwave From the Department of Radiology, Division of Thoracic Imaging, University of California, Los Angeles, 757 Westwood Plaza, Suite 1638, Los Angeles, CA 90024. Received January 24, 2016; final revision received May 20, 2016; accepted May 21, 2016. Address correspondence to R.A.A.; E-mail: [email protected] F.G.A. is a paid consultant for HealthTronics, Inc (Austin, Texas) and participated in a clinical trial for Galil Medical, Inc (Arden Hills, Minnesota). R.D.S. is a paid consultant for NeuWave Medical, Inc (Madison, Wisconsin) and HealthTronics, Inc, receives grants from HealthTronics, Inc, and performed research for Galil Medical, Inc. None of the other authors have identified a conflict of interest. & SIR, 2016 J Vasc Interv Radiol 2016; 27:1380–1386 http://dx.doi.org/10.1016/j.jvir.2016.05.026
ablation in the lungs has been shown to produce a larger ablation zone (1). Wolf et al (2) reported the clinical success of microwave ablation for treatment of pulmonary malignancies in patients deemed medically inoperable with cancer-specific mortality yielding a 1year survival of 83%. Perfusion-mediated tissue cooling (heat sink) describes the perfusion-related loss of energy deposition associated with thermal ablation, which may be a cause of local recurrence secondary to sublethal temperatures of the target lesion (3–5). Prior studies demonstrated the heat-sink effect in the lung with RF ablations adjacent to vessels 4 3 mm, with an increased rate of recurrence after ablation of tumors in direct contact with such vessels (6,7). Studies have also suggested that microwave ablation is less susceptible to the heat-sink effect compared with RF ablation (8–10). In addition to perfusion, impedance caused by ventilation is
Volume 27
’
Number 9
’
September
’
2016
a unique confounding factor associated with thermal ablation in the lungs. Decreased ventilation via bronchial occlusion with single-lung ventilation has been shown in animal studies to increase microwave and RF ablation zone size in the lungs (11,12). Both ventilation and perfusion have a meaningful impact on pulmonary microwave ablation, and each has been shown to have regional variances throughout the lung (13,14). Studies have demonstrated higher perfusion of the lower/central lung and greater ventilation in the lower lung (13,15–17). The in vivo effects of these regional perfusion and ventilation variations on microwave ablation have not been well studied to date. Multiple previous studies reviewed imaging after procedures to demonstrate characteristics predicting treatment success, including an ablation zone size 4 times greater than the tumor size before ablation (18–23). Although it is believed that ablation energy delivered is important for ablation zone size, previous studies mostly focused on ablation power (24,25). Ablation work (power * time) is a measure of the total energy delivered and, in contrast to power alone, can simultaneously account for power, ablation time, number of antennas, and differences in these parameters between ablation phases. Clinical outcomes with thermal ablation in the lungs are an important area of continued investigation. However, the purpose of this study is to investigate technical factors of in vivo microwave pulmonary ablation, including ablation work, temporal involution, and regional ablation resistance.
MATERIALS AND METHODS After institutional review board approval, a retrospective chart review of 98 noncavitary pulmonary tumor ablations using a NeuWave Certus Microwave Ablation System (NeuWave Medical, Inc, Madison, Wisconsin) was performed. All cases were performed percutaneously under computed tomography (CT) guidance by 1 of 3 operators (R.D.S., F.G.A., S.J.G.) between June 2011 and July 2015. Exclusion criteria included use of an ablation antenna other than the Precision (PR) or Lung (LN) antenna. Ablation parameters were at the discretion of the operator and with the following treatment endpoints: tissue temperature 4 601C, concentric ground-glass margin 4 5 mm, and operator security with obtained technical success. Ablation power and time were obtained from the ablation system recordings for each ablation phase (defined as continuous ablation at a single power); 45 of 98 (45.9%) pulmonary tumors were treated with 4 1 ablation phase. Ablation work was calculated as the sum of work (power * time) for each individual phase according to the equation n P Pi nti nN (WA = ablation work [J], n = number WA ¼ i¼1
of ablation phases, P = power [W], t = ablation time [s], N = number of ablation antennas).
1381
CT scans performed immediately after the procedure, 1–3 months after the procedure (earliest thoracic CT scan in 1–3 month period, n = 80), and 3–12 months after the procedure (latest thoracic CT scan in 3–12 month period, n = 66) were reviewed by a single senior radiology resident (R.A.A.) blinded to procedural ablation factors. Ablation zone refers to the area of ablation on CT and was defined as the outer edge of the peripheral margin of dense ground-glass opacity (5,23). Maximal length and maximal perpendicular measurements were obtained for each ablation zone on the image with the largest apparent ablation zone. Ablation zone was then calculated according to the equation for an ellipse (A ¼ πab). Involution of the ablation zone refers to the decrease in ablation zone over time as a result of elimination of induced coagulation and accompanying cicatrization (5). The shortest distance from the ablation zone isocenter to the closest costal pleura was measured for each ablation zone and categorized into the peripheral (r 2 cm), parenchymal (2–4 cm), or central (Z 4 cm) pulmonary region. Each ablation was assigned a score for pulmonary lobe of ablation (upper lobe ¼ 1, middle lobe/lingula ¼ 2, lower lobe ¼ 3) and pulmonary region of ablation (peripheral ¼ 1, parenchymal ¼ 2, central ¼ 3). The 2 scores were summed for an ablation resistance score ranging from 2 to 6. Ablation resistance scores were categorized into low (score of 2–3), medium (score of 4), and high (score of 5–6) ablation resistance score groups (Fig 1). Statistical analysis was performed using IBM SPSS Statistics for Windows software (IBM Corporation, Armonk, New York). Variables with a significantly skewed distribution (tumor size before ablation, ablation zone after procedure, ablation zone at 1–3 months, ablation zone at 3–12 months, ablation work, and maximum/perpendicular ablation dimensions) were compared using Kruskal-Wallis test. Otherwise, variables with an assumed normal distribution and homogeneous variance based on Levene test were comparatively analyzed using analysis of variance. Statistical significance was assessed with an α value of .05. Ordinal and nominal variables were compared using Fisher exact test. All correlations were performed with Pearson correlation method (2-tailed) with calculation of the Pearson correlation coefficient (r). Mean values are reported with ⫾ SD. Patients included 32 men and 39 women. Mean patient age was 64.7 years ⫾ 11.5. Oncologic histology included 34 primary pulmonary tumors (28 adenocarcinoma, 4 squamous cell carcinoma, 2 neuroendocrine) and 64 metastatic pulmonary tumors (21 colorectal, 19 renal, 7 uterine, 7 adenoid cystic carcinoma, 2 pancreatic, 2 hepatic, 2 ependymoma, 1 endometrial adenocarcinoma, 1 myxoid fibrosarcoma, 1 melanoma, 1 cardiac sarcoma). Mean maximal tumor diameter was 1.5 cm ⫾ 0.7 (range, 0.4–3.6 cm); 4 tumors were 4 3 cm
1382
’
Ablation Work and Ablation Resistance Score
Al-Hakim et al
’
JVIR
months (mean 3.0 cm2 ⫾ 2.9) (P o .001) (Fig 2). The ablation zone change on CT scan at 1–3 months (mean 47 d ⫾ 24) was significantly greater (P o .01) than the subsequent ablation zone change at 3–12 months (mean 292 d ⫾ 110). Similarly, the percentage ablation zone reduction from after the procedure to 1–3 months (mean 42.0% ⫾ 28.7) was significantly greater than the percentage ablation zone reduction from 1–3 months to 3–12 months relative to the original ablation zone (mean 27.2% ⫾ 20.5) (P o .001).
Ablation Resistance Score
Figure 1. Schematic demonstrating different ablation resistance score groups: low score (ablation resistance score ¼ 2– 3), medium score (ablation resistance score ¼ 4), and high score (ablation resistance score ¼ 5–6). Dashed lines indicate divisions between pulmonary regions, and solid lines indicate divisions between pulmonary lobes. Dot-dashed line represents division between left upper lobe and lingula.
There was no significant difference between the ablation resistance score groups in tumor size before ablation, antenna type, ablation work, maximum ablation power, total ablation time, or patient age. There was a significant difference in ablation zone at 1–3 months between the low (8.2 cm2 ⫾ 4.6), medium (6.4 cm2 ⫾ 4.7), and high (4.6 cm2 ⫾ 3.1) ablation resistance score groups (P o .05) (Fig 3). This statistically significant stratification of ablation zone between ablation resistance score groups was not observed after the procedure or at 3– 12 months, although there was a similar non–statistically significant trend in both groups (P ¼ .07 and P ¼ .15). Independent stratification between upper, middle, and lower lobes and peripheral, parenchymal, and central pulmonary zones did not show a significant difference in ablation zone at any time point.
in maximal diameter, and no tumors were 4 5 cm in maximal diameter.
Antenna Analysis
RESULTS Ablation Work Ablation procedures were performed using the NeuWave Certus PR (n ¼ 59) or LN (n ¼ 39) antenna, with single antenna microwave ablation (n ¼ 82) or simultaneous microwave ablation using 2 (n ¼ 15) or 3 (n ¼ 1) antennas. Mean ablation work for all ablations was 41.3 kJ ⫾ 21.3 (range, 9–117 kJ). There was a significant correlation of ablation work with ablation zone on CT after the procedure (P o .001, r ¼ .633), at 1–3 months (P o .001, r ¼ .620), and at 3–12 months (P o .05, r ¼ .249). There was no significant correlation of total ablation time (range, 164–1,380 s) with ablation zone at any time point. There was a significant correlation of maximum ablation power with ablation zone after the procedure (P o .001, r ¼ .414), at 1–3 months (P o .001, r ¼ .555), and at 3–12 months (P o .01, r ¼ .361).
Ablation Zone Involution There was a significant reduction in mean ablation zone on CT scans from after the procedure to 1–3 months (mean 4.8 cm2 ⫾ 4.6) and from 1–3 months to 3–12
All cases performed with a single ablation antenna using either the Certus LN (n ¼ 35) or PR (n ¼ 47) ablation antenna were subanalyzed. There was significantly higher ablation work with use of the LN antenna (mean 49.9 kJ ⫾ 19.5) versus the PR antenna (mean 30.8 kJ ⫾ 15.1), and the LN antenna produced a significantly larger ablation zone over the PR antenna at all time points (P o .01 for all). In addition, there was significantly higher maximum ablation power used for the LN antenna (mean 95.8 W ⫾ 23.9; range, 40–140 W) versus the PR antenna (mean 53.9 W ⫾ 16.6; range, 20– 65 W). Most PR antenna ablations were performed with a maximum power of 65 W (68.1%; 32 of 47) or 30 W (21.3%; 10 of 47), whereas most LN antenna ablations were performed with a maximum power of 90 W (65.7%; 23 of 35) or 140 W (14.3%; 5 of 35). The PR antenna demonstrated a significant correlation of ablation work with ablation zone after the procedure (P o .001, r ¼ .521) and at 1–3 months (P o .001, r ¼ .513) but not at 3–12 months. In addition, the PR antenna had a significant correlation of ablation work with the maximal and perpendicular measurements of the ablation zone after the procedure and at 1–3 months (P o .01 for all). The LN antenna ablation work did not have a significant correlation with
Volume 27
’
Number 9
’
September
’
2016
1383
Figure 2. Mean ablation zone at all time points (SE bars included), with a statistically significant involution at both time points (P o .001). The ablation zone reduction from after procedure to 1–3 months was significantly greater than ablation zone reduction at 3–12 months (P o .01). Percentage ablation zone reduction relative to ablation zone after procedure between each time point is shown for reference.
The PR single antenna ablations were stratified into 3 near-equal sampled categories based on ablation work: r 25 kJ (n ¼ 16), 25–35 kJ (n ¼ 15), and Z 35 kJ (n ¼ 16) (Table). There was a significant difference in ablation zone, maximal ablation zone length, and perpendicular ablation zone measurement after the procedure and at 1– 3 months between these 3 categories (P o .01), with the exception of the perpendicular ablation zone measurement at 1–3 months, which showed a similar trend between groups without statistical significance (P ¼ .058). The maximum ablation power was significantly lower for the r 25 kJ category compared with the 25–35 kJ and Z 35 kJ categories (P o .001).
DISCUSSION
Figure 3. Mean ablation zone for ablation resistance score groups at all time points (SE bars included), with a significant difference between the ablation resistance score groups at 1–3 months (P o .05).
ablation zone or dimensional measurements after the procedure, at 1–3 months, or at 3–12 months.
Analysis of 98 pulmonary tumor ablations with a single microwave ablation system demonstrates a correlation of pulmonary ablation work with ablation zone on imaging during the first year. Given lack of correlation between ablation times with ablation zone, the ablation work correlation is predominantly driven by the selection of ablation power. However, use of work as a reference allows simultaneous quantification and standardization of ablation power and time in the setting of multiple ablation phases and multiple ablation antennas, even when ablation phases are performed at different power levels. These results also demonstrate a disproportionately higher reduction in ablation zone at 1–3 months (mean 47 d) compared with 3–12 months (mean 292 d), indicative of a nonlinear involution pattern. Positive deviation from the expected involution curve
1384
’
Ablation Work and Ablation Resistance Score
Al-Hakim et al
’
JVIR
Table . Ablation Zone Measurements within Ablation Work Categories for NeuWave Certus Precision Microwave Antenna o 25 kJ (n ¼ 16)
25–35 kJ (n ¼ 15)
4 35 kJ (n ¼ 16)
Maximum ablation power (W)*†
36.9 ⫾ 14.9
62.7 ⫾ 9.0
62.8 ⫾ 8.8
Ablation time (s)*‡
474 ⫾ 174
578 ⫾ 169
839 ⫾ 221
6.4
7.2
10.1
3.9
4.1
4.7
2.1
2.2
2.7
Area (cm2)‡
2.8
3.7
5.2
Maximum length (cm)‡ Perpendicular measurement (cm)
2.6 1.4
2.8 1.6
3.5 1.8
Ablation Zone after Procedure Area (cm2)‡ Maximum length (cm)‡ Perpendicular measurement (cm)‡ Ablation Zone at 1–3 Months
Note–Data are means unless otherwise noted. *Data are mean ⫾ SD. † Significant difference between o 25 kJ group and other 2 groups (P o .001). ‡ Significant difference between all 3 groups (P o .01).
may potentially be used as a clinical guide for predicting early incomplete tumor ablation or tumor progression. The correlation of clinical and imaging outcomes relative to a similar involution curve should be the focus of future study. Multiple ablation antennas are available for use with the NeuWave Certus Microwave Ablation System. Initially, the LN antenna was the only available ablation antenna designed for lung ablation with the NeuWave System. In mid-2012, the PR antenna was introduced at our institution, and some operators adopted this antenna into their practice. Data using these antennas demonstrate a strong relationship of ablation work to ablation zone with use of the PR antenna but not the LN antenna. This result is likely related to the smaller sample size of lung ablation performed with the LN antenna given the similar trend without statistical significance at after the procedure (P ¼ .07, r ¼ .308) and at 3–12 months (P ¼ .10, r ¼ .363). Similarly, the relatively small sample size did not allow reliable statistical subanalysis of simultaneous ablations with 2 and 3 antennas. The correlation of PR antenna ablation work with ablation zone allowed stratification of ablation work into 3 ablation work categories that produced significantly different ablation zones (Table). These data can be used as a starting reference for pulmonary ablation procedural planning using the NeuWave Certus PR antenna, and this reference is potentially a more complete reference than what is currently available. Based on these results, an ablation power of 30 W with an ablation time of approximately 10 minutes should be used to achieve an ablation zone in the o 25 kJ category. To increase the ablation zone to the 25–35 kJ category, the power should be increased to 65 W with an ablation time of approximately 10 minutes. Finally, to reach the 4 35 kJ category, the ablation time should be increased to 4 10 minutes with a power of 65 W.
Prior studies evaluating the susceptibility of microwave ablation to perfusion-mediated tissue cooling (heat-sink effect), a perfusion-related decrease in energy deposition limiting thermal ablation, have targeted its effect on ablation based on the presence of vasculature, specifically vessel size, within and in the immediate vicinity of the ablation zone in the liver (3,4). Although a similar heat-sink effect related to a minimum threshold of vessel size would be expected with microwave ablation in the pulmonary parenchyma, the lung environment that dictates the deposition of ablation energy and eventually ablation zone is no doubt more complex, given the immense generalized perfusion of the lung through its microvasculature as well as the effect of continuous ventilation throughout respiration (7). Simply, the differential pulmonary perfusion and ventilation throughout the lung must be taken into account, perhaps more so than the anticipated influence of single or multiple vessels in the region of targeted ablation. It was our hypothesis that regions of higher perfusion and ventilation would result in a smaller ablation zone size. Based on studies of lung perfusion and ventilation, the lower lobes have higher perfusion and ventilation compared with the upper lobes and were therefore assigned a higher ablation resistance score (13,17). In addition, the central lung zones have higher perfusion and were assigned a higher ablation resistance score. Using this method, ablation resistance score correlated to a statistically significant stratification of ablation zone: smaller ablation zone in regions with a higher ablation resistance score, larger ablation zone in regions with a lower ablation resistance score, and intermediate ablation zone with midrange ablation resistance score. In physics terms, the efficiency of energy deposition by microwave energy may be referred to as effective conductivity (26). Ablation resistance score is fundamentally a qualitative inverse measurement of effective conductivity in the lung. With further validation, this proposed ablation resistance scoring
Volume 27
’
Number 9
’
September
’
2016
system may find value for microwave ablation for pulmonary malignancies not only with implications for planning before the procedure but also as a decision tool during the procedure. Tumor ablations anticipated in high ablation resistance score regions would require more ablation work to achieve the desired ablation zone. During the procedure, the decision to apply additional or more ablation work would be relatively straightforward during the treatment of tumors within high ablation resistance score regions, especially if ground-glass envelopment of the targeted tumor, an imaging surrogate used as an ablation endpoint, was less than expected. This study has several limitations. Ablations were monitored by the operators using some objective imaging parameters, such as the manufacturer ablation chart and ground-glass zone surrounding the tumor after ablation, with awareness of known phenomena such as perfusion-mediated tissue cooling. However, the retrospective aspect of this study with subjective operator treatment parameters and endpoint determination did not allow for reliable standardization and resulted in varied imaging timing after ablation. Furthermore, although ablation work was used as the primary indicator of energy deposited, it is still unclear from our analysis if all ablation work is the same. For example, are 2 5-minute ablation cycles at 65 W the same as 1 10minute ablation cycle at 65 W? The retrospective nature of this study and limited sample size did not allow reliable stratification to analyze for this difference. In addition, although our proposed ablation resistance scoring is roughly based on perfusion and ventilation variances, the regional variations of these parameters are complex, positional, and likely not fully captured by the 3 ablation resistance score groups. Furthermore, although a single trainee performed data collection, consensus regarding the definition and imaging findings of the ablation zone were established by the subspecialty trained radiologists before collection of data. Finally, the focus of this study was to describe ablation work and ablation resistance score, but future studies are required to correlate these findings with pathology and clinical outcomes, ideally using volumetric analysis (27). Specifically, future studies correlating the true histologic tumor necrosis volume and potential biomarkers over time with ablation work and the corresponding imaging ablation zone in different ablation resistance groups would validate the use of these new metrics. In conclusion, the findings in this study provide insight into microwave ablation properties in the lung and have implications for use in clinical care of patients receiving lung ablation, including planning before the procedure, decision guidance during the procedure, and imaging follow-up after the procedure. Although prior studies have demonstrated varied response to thermal ablation of lung tumors adjacent to vessels 4 3 mm, the deposition of ablation energy and subsequent ablation
1385
zone is clearly more complex. The ablation resistance score proposed in this study takes into account varied perfusion and ventilation throughout the lung when correlating with variations of the ablation zone; higher ablation resistance scores were found to correlate with smaller ablation zones, which can have important treatment implications (6,7). Similarly, ablation work accounts for differential power and time for multiple ablation phases and/or antennas and correlates with ablation zone on follow-up imaging in the first year. With validation by future clinical studies, the approach using ablation work and ablation resistance score during the planning and treatment phases of lung tumor ablation with monitoring of procedural imaging could decrease the likelihood of incomplete tumor ablation.
REFERENCES 1. Brace CL, Hinshaw JL, Laeseke PF, et al. Pulmonary thermal ablation: comparison of radiofrequency and microwave devices by using gross pathologic and CT findings in a swine model. Radiology 2009; 251: 705–711. 2. Wolf FJ, Grand DJ, Machan JT, et al. Microwave ablation of lung malignancies: effectiveness, CT findings, and safety in 50 patients. Radiology 2008; 247:871–879. 3. Yu NC, Raman SS, Kim YJ, et al. Microwave liver ablation: influence of hepatic vein size on heat-sink effect in a porcine model. J Vasc Interv Radiol 2008; 19:1087–1092. 4. Lu DS, Raman SS, Vodopich DJ. Effect of vessel size on creation of hepatic radiofrequency lesions in pigs: assessment of the “heat sink” effect. AJR Am J Roentgenol 2008; 178:47–51. 5. Ahmed M, Solbiati L, Brace C, et al. Image-guided tumor ablation: standardization of terminology and reporting criteria—a 10-year update. J Vasc Interv Radiol 2014; 25:1691–1705. 6. Gillams AR, Lees WR. Radiofrequency ablation of lung metastases: factors influencing success. Eur Radiol 2008; 18:672–677. 7. Steinke K, Haghighi KS, Wulf S, et al. Effect of vessel diameter on the creation of ovine lung radiofrequency lesions in vivo: preliminary results. J Surg Res 2005; 124:85–91. 8. Goldberg SN, Grassi CJ, Cardella JF, et al; Society of Interventional Radiology Technology Assessment Committee; International Working Group on Image-Guided Tumor Ablation. Image-guided tumor ablation: standardization of terminology and reporting criteria. Radiology 2005; 235:728–739. 9. Hinshaw JL, Lee FT Jr. Cryoablation for liver cancer. Tech Vasc Interv Radiol 2007; 10:47–57. 10. Wright AS, Sampson LA, Warner TF, et al. Radiofrequency versus microwave ablation in a hepatic porcine model. Radiology 2005; 236: 132–139. 11. Oshima F, Yamakado K, Akeboshi M, et al. Lung radiofrequency ablation with and without bronchial occlusion: experimental study in porcine lungs. J Vasc Interv Radiol 2004; 15:1451–1456. 12. Santos RS, Gan J, Ohara CJ, et al. Microwave ablation of lung tissue: impact of single-lung ventilation on ablation size. Ann Thorac Surg 2010; 90:1116–1120. 13. Levin DL, Chen Q, Zhang M, et al. Evaluation of regional pulmonary perfusion using ultrafast magnetic resonance imaging. Magn Reson Med 2001; 46:166–171. 14. Milic-Emili J. Static distribution of lung volumes. Comprehensive Physiology 2011; 561–574. 15. Grassino AE, Bake B, Martin RR, et al. Voluntary changes of thoracoabdominal shape and regional lung volumes in humans. J Appl Physiol 1975; 39:997–1003. 16. Sybrecht G, Landau L, Murphy BG, et al. Influence of posture on flow dependence of distribution of inhaled 133Xe boli. J Appl Physiol 1976; 41: 489–496. 17. Chang YH, Yu CP. A model of ventilation distribution in the human lung. Aerosol Sci Technol 1999; 30:309–319.
1386
’
Ablation Work and Ablation Resistance Score
18. Gazelle GS, Goldberg SN, Solbiati L, et al. Tumor ablation with radiofrequency energy. Radiology 2000; 217:633–646. 19. Anderson EM, Lees WR, Gillams AR. Early indicators of treatment success after percutaneous radiofrequency of pulmonary tumors. Cardiovasc Intervent Radiol 2009; 32:478–483. 20. Lee JM, Jin GY, Goldberg SN, et al. Percutaneous radiofrequency ablation for inoperable non-small cell lung cancer and metastases: preliminary report. Radiology 2004; 230:125–134. 21. Yamamoto A, Nakamura K, Matsuoka T, et al. Radiofrequency ablation in a porcine lung model: correlation between CT and histopathologic findings. AJR Am J Roentgenol 2005; 185:1299–1306. 22. de Baère T, Palussière J, Aupérin A, et al. Midterm local efficacy and survival after radiofrequency ablation of lung tumors with minimum followup of 1 year: prospective evaluation. Radiology 2006; 240:587–596.
Al-Hakim et al
’
JVIR
23. Chheang S, Abtin F, Gutierrez A, et al. Imaging features following thermal ablation of lung malignancies. Semin Intervent Radiol 2013; 30: 157–168. 24. Sonntag PD, Hinshaw JL, Lubner MG, et al. Thermal ablation of lung tumors. Surg Oncol Clin N Am 2011; 20:369. 25. Andreano A, Huang Y, Franca Meloni M, et al. Microwaves create larger ablations than radiofrequency when controlled for power in ex vivo tissue. Med Phys 2010; 37:2967–2973. 26. Brace CL. Radiofrequency and microwave ablation of the liver, lung, kidney and bone: what are the differences. Curr Probl Diagn Radiol 2009; 38:135–143. 27. Ozaki T, Mori I, Nakamura M, et al. Microwave cell death: immunohistochemical and enzyme histochemical evaluation. Pathol Int 2003; 53: 686–692.