Endobronchial tumor debulking with a flexible cryoprobe for immediate treatment of malignant stenosis

We analyzed the data of all patients who underwent bronchoscopy in our clinic from February 2001 to September 2007. In this series we identified all patients undergoing bronchoscopic intervention with the flexible cryoprobe because of an exophytic tumor of the trachea or bronchi with subsequent symptomatic airway stenosis. In addition to the flexible cryoprobe, a laser or APC, as well as mechanical debulking with a rigid scope, were available but rarely used since the introduction of the flexible cryoprobe in our institution in 2001. The local ethics committee approved this study.

For any bronchoscopic intervention, a platelet count of at least 100 Giga/L and normal coagulation parameters (partial thromboplastin time <40 seconds and prothrombin ratio >70%) were mandatory. Medications influencing clotting time were stopped before intervention; heparin was stopped at least 12 hours before intervention.

The flexible cryoprobe (Erbokryo; ERBE, Tübingen, Germany) was used as described previously. It can be used with the rigid bronchoscope after achievement of general anesthesia but also with the flexible technique in combination with a 7.5- or 8.5-mm Bronchoflex tube (Rüsch, Kernen, Germany), which has separate oxygen tubing, after achievement of local anesthesia and sedation. The decision of which method to apply, as well as the use of additional techniques, was left to the discretion of the treating physician and based on published guidelines.^,  It should be noted that there is no ideal interventional strategy that can fully guarantee success and a safe outcome. The execution of any technique depends on the expertise of the team and the personal skills of the physician. If possible, we preassigned (in an expert discussion) the used interventional technique before each bronchoscopy. In cases with stenosis caused by an endoluminal tumor, almost all interventions were done as cryorecanalization (CR). In cases with mixed stenosis (with an additionally extraluminal tumor), we combined the techniques, such as CR and implantation of stents. Briefly, in general we used the following interventional technique. First, the cryoprobe is introduced through a flexible bronchoscope in combination with a flexible tube or a rigid bronchoscope. Second, the tip of the cryoprobe is pushed into the exophytic tumor mass, and freezing is started for 3 to 5 seconds with a footpad. Freezing is achieved with the Joules–Thomson effect: nitrous oxide is decompressed at the tip of the probe, which generates immediate cooling of the tip to −89 °C. Third, the cryoprobe and the flexible bronchoscope are abruptly mechanically removed, and thus a tumor mass of up to 10 mm is frozen at the tip of the cryoprobe and can easily be extracted. Fourth, rapid thawing of the extracted tissue is induced in a water bath for 5 seconds. Finally, the procedure is repeated until the exophytic tumor mass has been extracted and the target bronchus is reopened. After CR, subsequent stent implantation can be performed in the same session or APC can be used to stop bleeding, if needed. Additionally, some factors could influence the intervention. Among them, bleeding of the tumor limits visibility and can significantly extend the procedural duration, which normally has a range of 15 to 60 minutes. Chest radiographic analysis was not performed routinely after the intervention and remained at the discretion of the physician.

We defined our criteria to determine the success rate of CR. The bronchoscopist classified the intervention based on endoscopic criteria as (1) successful if the endobronchial tumor mass could be ablated so that either drainage of secretion or reduced airway stenosis led to a significant improvement in the patient's condition with amelioration of symptoms (eg, dyspnea, fever, or cough) and (2) not successful or a failure if the primary goal of the intervention (ie, target bronchus reopening) could not be reached.

Safety data were evaluated regarding severity of bleeding, need for a rescue operation, or prolonged mechanical ventilation for longer than 2 hours after the intervention. Bleeding was graded as mild if application of cold NaCl 0.9% solution (2 °C–4 °C) or epinephrine solution (1 mg/10 mL saline water) applied topically was sufficient to stop bleeding, moderate if additional techniques (APC or a bronchus blocker) had to be used, and severe if additional treatment was needed (transfusion of red blood cells, fresh frozen plasma, or coagulation factors; use of vasopressors, rescue operation; or prolonged mechanical ventilation) to stabilize the patient's condition.

Between February 2001 and September 2007, 5308 consecutive bronchoscopies were performed in our clinic. In this cohort bronchoscopic intervention with CR with the flexible cryoprobe for treatment of symptomatic exophytic tumor stenosis of the trachea or bronchi was used in 225 patients. The mean age of the patients was 63.9 ± 12.9 years (range, 19–83 years), and 193 (85.8%) were inpatients. One hundred fifty-six (69.3%) patients were men.

CR of obstructing tumor lesions in the central airways (trachea, main bronchus, and bronchus intermedius) was performed in 99 (44.0%) patients, of the lobe bronchi in 93 (41.3%) patients, and simultaneously in the central airways and lobe bronchi in 33 (14.7%) patients. Multifocal endobronchial lesions (in >1 bronchus) were present in 67 (29.8%) patients and were treated with the flexible cryoprobe within 1 interventional session. One hundred twenty-four (55.1%) lesions were located in the right bronchial system, 64 (28.4%) lesions were located in the left bronchial system, and 33 (14.7%) lesions were located in the trachea or on the main carina. A detailed overview of lesion localizations is shown in Table 1. In 197 (87.6%) of 225 patients, airway stenosis was malignant. One hundred forty-seven (74.6%) patients had lung cancer. Table 2 shows the histologic findings in detail.

The majority of interventions (194 [86.2%] patients) were performed solely with the flexible technique. The rigid scope, always used in combination with the flexible bronchoscope, was used in 31 (13.8%) patients. Additional techniques were rarely needed: APC (in patients with prolonged bleeding after CR) in 37 (16.4%) patients and stent implantation in 11 (4.9%) patients.

CR was successful in 205 (91.1%) of 225 interventions. In 20 (8.9%) patients the intervention was not successful and did not reopen the obstructed bronchus. The main reason for unsuccessful interventions was a longer length (>2 cm) of the complete bronchial obstruction.

Bleeding complications occurred in 27 (12.0%) patients, with mild bleeding in 9 (4.0%) patients and moderate bleeding in 18 (8.0%) patients. Severe bleeding did not occur. Thus the overall bleeding rate was low. No case of pneumothorax was detected. One manifestation of pneumomediastinum was seen in a patient with a cystadenoma, a benign tracheal tumor. In this patient injury of the pars membranacea occurred after removal of the bulky tumor. The patient was hospitalized and treated with antibiotics to prevent mediastinitis. Surgical intervention was evaluated but not necessary after spontaneous closure of the very small tracheal perforation. The patient fully recovered after 5 days without further complications.