INTRODUCTION Over the past decade, image-guided thermal ablation of primary and metastatic lung tumors has emerged as a viable therapeutic option for patients with contraindications to surgical management. Published two-year cancer-specific survival rates have been reported as high as 92% among patients with stage I non-small cell lung cancer, and a four-year local efficacy rate of 89% was observed in a large trial of 566 patients treated with ablation for pulmonary metastatic disease. In spite of these successes, however, there remain concerns regarding the safety of ablation performed in close proximity to the heart.
The safety of radiofrequency ablation performed near the heart and great vessels in humans has been previously investigated by Iguchi and colleagues in 2007 , who observed no instances of new-onset arrhythmia or cardiac ischemia. Since that time, the use of microwave ablation in lung has increased owing to its associated higher ablation zone temperatures and decreased susceptibility to heat sink and tissue heterogeneity effects. On the basis of this improved performance profile, there is a theoretical increase in the risk of injury to cardiac and peri-cardiac structures such as mediastinal nerves and pericardium when used for treatment of centrally-located tumors. In the present study, we sought to retrospectively evaluate our experience using microwave ablation for the treatment of primary and metastatic lung tumors located adjacent to the heart with an emphasis on peri-procedural and short-term patient safety.
MATERIALS AND METHODS A retrospective search was performed to identify patients who underwent microwave ablation of lung tumors located near the heart. The primary inclusion criterion was the presence of a primary or metastatic lung malignancy located 10 mm or less from one of the four cardiac chambers, as measured from the ablation antenna following placement within the tumor as well as from the closest tumor margin.
Prior to ablation, all patients were seen in consultation with an attending radiologist and a nurse practitioner. During the ablations, continuous electrocardiographic monitoring was performed prospectively by radiology nursing staff, and this was continued in a dedicated radiology recovery room for a minimum of two hours following treatment. In the absence of complications, patients were discharged home and seen in clinic for follow-up at one month. At that time, contrast-enhanced chest CT was performed to evaluate for primary treatment efficacy as well as to assess for evidence of thermal injury to cardiac or peri-cardiac structures (e.g. pericardium, mediastinal vessels and nerves, central airways). In the absence of residual unablated tumor, patients were then similarly followed at six months, 12 months, and annually thereafter.
Adverse events were recorded using the Common Terminology Criteria for Adverse Events v. 4.0. The primary outcomes of interest were the development of any new-onset arrhythmia during the ablation procedure and myocardial infarction diagnosed within 90 days of treatment. This latter variable was assessed for using both close outpatient follow-up in ablation clinic, as well as retrospective chart review for any outpatient, emergency department, or inpatient visits initiated for chest pain or other potential cardiac symptoms following ablation within our large multi-site university health care system.
To assess for any subclinical cardiac injury, all post-procedural EKGs and echocardiograms—where applicable—were reviewed for changes from baseline. Finally, the rate of pneumothorax with or without chest tube placement was calculated, and all instances of mechanical mediastinal injury from antenna placement, hospital admission, and patient death were noted.
RESULTS Ten consecutive patients (four male, six female; mean age 73.1 ± 9.5 years, range 55-87 years) with ten primary or metastatic lung tumors were identified between March, 2006 and November, 2016 who met all inclusion criteria. The mean tumor distance from the heart was 3 mm (range, 0-6 mm), and the mean distance from the microwave antenna to the heart was 7 mm (range, 2-10 mm). The mean tumor size was 3.3 ± 1.8 cm (range, 1.1-6.8 cm). Tumors were treated with an average of 1.8 microwave antennae (range, 1-4). The majority of treated tumors (7/10, 70.0%) were metastatic in origin, with colon being the most common primary site.
New-onset arrhythmia was not observed during or following any of the image-guided microwave ablation treatments, and there were no documented myocardial infarction events within 90 days of any procedure. All patients had one or more post-procedural EKGs one or more months from the ablation procedure; normal sinus rhythm was demonstrated in each case without evidence of prior myocardial ischemia (e.g. new Q waves). Four patients underwent echocardiography following ablation, all of which were normal without evidence of myocardial or valve dysfunction.
All patients received post-ablation contrast-enhanced chest CT at one month. In two of the patients, there were no demonstrable changes within the mediastinum following treatment. CTCAE Grade 1 complications occurred in the remaining eight patients, the most common of which was focal pericardial thickening and/or trace focal pericardial effusion immediately adjacent to the site of microwave ablation. In one patient with a colorectal cancer metastasis located 4 mm from the left ventricle, mild airway thickening was noted in the left upper lobe adjacent to the ablation site, as well as a small epicardial fluid collection; these findings subsequently resolved on follow-up imaging. One patient developed a small transient pleural effusion ipsilateral to the site of ablation performed for uterine leiomyosarcoma metastatic to the right middle lobe. There were no identified instances of mechanical mediastinal injury secondary to misplacement of the ablation antennae. There was no evidence of phrenic or vagus nerve injury, as determined by imaging and clinical follow-up. None of the ten treated patients required post-ablation hospitalization, and there were no ablation-related patient deaths.
During the ablation procedures, four patients developed small asymptomatic pneumothoraces. Of these patients, two were managed conservatively without chest tube placement and were discharged home on the day of treatment with subsequent resolution of pneumothorax on outpatient chest radiography (CTCAE Grade 1). Two patients required chest tube placement and were both discharged home with Heimlich valves (BD, Franklin Lakes, NJ) on the day of treatment, with the chest tubes safely removed several days later on an outpatient basis (CTCAE Grade 2).
CONCLUSIONS Herein we have described our single-institution experience performing microwave ablation of lung tumors located in close proximity to the heart in a cohort of ten patients. Though limited in scope, our results suggest the procedure can be performed with low short-term morbidity; in particular, no dysrhythmia or myocardial ischemia was observed even in patients with tumors directly abutting the pericardium. Further evaluation of oncologic outcomes among patients treated with curative intent for tumors in close proximity to the heart is necessary to prospectively replicate these findings within a larger patient population.