Initial experience with a synthetic sealant PleuraSeal™ after pulmonary resections: a prospective study with retrospective case matched controls

To investigate the efficacy and postoperative outcome, PleuraSeal™ lung sealant system was prospectively evaluated as an adjunct to standard closure techniques for control of visceral pleural air leaks in patients scheduled for pulmonary resection. These data were compared with an individually case matched retrospective control cohort. Sample size calculation was based on a superiority test for two proportions using Fisher's Exact Test, estimated proportion was 72% for PSG and 15% for CG, respectively. Therefore, for an estimated power of 90% a sample size of 18 patients for each group was necessary to show a difference in the two-sided test for the primary endpoints to reach statistical significance. Data were collected at the University Hospital Freiburg, Department of Thoracic Surgery, from patient admission to discharge. Twenty consecutive patients with anatomical resection were included. Inclusion criteria for both groups were patient age 18 years or older, and a scheduled or completed (in retrospective control group) lobectomy or segmental wedge resection via an open thoracotomy approach and presence of an initial intraoperative air leak after resection. Preoperatively, groups were well balanced regarding concomitant pulmonary disease, mean FEV1, lung tissue quality, presence of diabetes and smoking habit. The study was approved by the local ethical committee (340/08, 12/02/08) and written informed consent of all patients was obtained. Only patients with intraoperative air leaks grade 1-2 based on the Macchiarini scale [20] after standard surgical suture or stapling were enrolled. All subjects had two chest tubes during the first 48 hours after thoracotomy. Routinely two silicone chest tubes (ventral 21 Ch and dorsal 24 Ch, silicone chest tubes, Redax Company, Mirandola, Italy) were placed after thoracotomy and connected to a water sealed drainage system. Suction of -20 cm H₂O was taken off once no air leaks were detected after 12 hours of observation time. The ventral chest tube (air leak chest tube) was taken out when there was no evidence of an air leak within the last 24 hours, and the dorsal tube was taken out when less than 200 ml drainage was recorded within last 24 hours. Assessment of postoperative air leakage was performed twice daily until chest tube removal and discharge. Patients were discharged and re-evaluated within 30 days after leaving the hospital, with two fixed scheduled visits to an ambulatory outpatient clinic. Primary criteria for discharge were no significant pain (VAS ≤ 2) under present pain medication and absence of chest tube.

Prior to the application of PleuraSeal™ lung sealant an intraoperative air leak test during lung inflation was carried out to evaluate the location and grade of air leaks as described above. In the event of a grade 3 air leak additional suture or stapling was performed to downgrade the pulmonary leak, so that PleuraSeal sealant was applied only in patients with grade 1 or 2 leaks. PleuraSeal™ polymer kit with a total volume of 5 ml was applied using a MicroMyst™ Applicator with a continuous flow (flow regulator) under aseptic conditions to seal intraoperative air leaks as well as prophylacticly on staple lines and other manipulated lung tissue. The lung surface was as dry as possible and the lung was inflated between 50-75% of its maximal volume. The product was uniformly distributed in a layer of 1-2 mm thickness, and on areas of extending pulmonary lesions a thicker layer was achieved of up to 3 mm. Excessive material was mechanically removed using suction or blunt dissection. Contraindications to apply PleuraSeal™ lung sealant include uncontrolled transected bronchioles >1 mm, and intrathorax infections. Additionally, the sealant was not applied to bronchial stumps or bronchial anastomoses. Following sealant application leak sites underwent another water submersion test under pressure of 25 mmHg. Sealant application could be repeated if air leakage control was not halted after the first application.

Concerning overall clinical characteristics, a total of 20 consecutive patients in the PleuraSeal™ sealant study group were matched with 20 retrospective patients that underwent pulmonary resection via an anterolateral thoracotomy as a control group. Baseline characteristics and intraoperative findings are shown in Table 1 and 2, respectively without any statistical difference between the two groups (Table 1. and 2.).

Three primary endpoints were defined by protocol: reduction of intraoperative air leakage, the presence and duration of postoperative air leak, and postoperative morbidity. Secondary parameters were duration of chest tube placement as well as time to discharge. Adverse events (AE) for the prospectively collected study cohort from the screening period prior to admission to our surgical ward until follow-up of 30-days after discharge were documented and reported if eligible. Complications were defined as occurrence of empyemas, incomplete lung inflation, pneumothorax, presence of bronchogenic fistula, chylothorax, cardiovascular complications, any other organ specific failure, severe postoperative pain, acute esophagitis, and the need for additional procedures (re-operation, pleurodesis, and replacement of chest tube).

All statistical analyses were performed using SAS^® Version 9.1. All statistical tests were two-sided at the 5% significance level. Continuous variables were summarized using mean and standard deviation, and the two sample t-test was used to test the difference between treatment groups. Categorical variables were summarized by frequencies and percentages, and the proportions were calculated and compared between treatment groups using Fisher's Exact test.

In the PleuraSeal™ sealant group two patients failed to have any intraoperative air leaks and were excluded from the study after screening. Conversely 91% of prospective lobectomy patients (20/22) demonstrated at least one intraoperative air leak. The twenty study patients received primary PleuraSeal™ sealant treatment after a water submersion test identified air leak location. The majority of air leaks occurred in the interfissure area (52%), with 23% at the staple lines, 6% at the suture lines, 6% due to adhesiolysis, and the remaining 13% due to other tissue manipulation (Table 3). For each patient the number of air leaks were counted and graded. Prior to the application of PleuraSeal™ sealant, the grade of air leak was 1.2 +/- 0.4 (mean +/- SD) in the PSG which was comparable to 1.1 +/- 0.2 (mean +/- SD) in CG. (p = 0.24). Also prior to the application of PleuraSeal™ sealant, the number of intraoperative air leaks per patient was 1.6 +/- 0.6 (mean +/- SD) in the PSG which was comparable to 1.4 +/- 0.5 (mean +/- SD) in CG (p = 0.39). In both groups additional suturing was carried out if necessary to downgrade air leaks prior to PleuraSeal™ sealant application in the PSG and closure in the CG. This was performed in 2 cases (10%) in the PSG and 6 cases (30%) in the CG without any significant difference between the two groups (p = 0.23). Neither in the PSG nor CG did we perform any additional stapling. On average 2.8 ml of PleuraSeal™ sealant was applied in less than one minute to the air leak sites and prophylacticly to all manipulated tissue. The MicroMyst™ applicator enabled multiple starts and stops during application without clogging. Only one patient needed a second application of PleuraSeal™ sealant. The mean operating time in the PSG was 2.6 hours which was comparable to the 3.1 hours in the CG (p = 0.08).

Overall morbidity was 52.5% and we could not find a statistical difference between the two study groups (p = 0.60) (Table 4). In total, 4 (10%) patients in the combined PSG and CG had severe pain postoperatively defined by a visual analogue scale (VAS) score ≥4 under present pain medication, in 5 (12.5%) patients cardiac complications were observed, and only one patient needed postoperative mechanical ventilation after onset of an adult respiratory distress syndrome. We did not experience any pleural empyema, lung emboli, incomplete lung inflation, re-thoracotomies, or re-placement of a chest tube in either of the groups. No patient died during the study period. All patients were discharged from hospital and completed the follow up period without any late complications. Hospital length of stay was 9.9 days in the PSG vs. 11.7 days in the CG, however, it did not reach statistical significance (p = 0.178). An interesting observation is that the shorter hospital length of stay of 1.8 days in the PSG corresponds exactly to the 1.8 days earlier removal of the air leak chest tube in the PSG.

At the end of surgery, 100% of the patients treated with the synthetic PleuraSeal™ sealant and 0% of the control group were air leak free (p < 0.001). One hour after surgery thru 30 day follow-up, 95% of the patients in the PSG were still air leak free compared to 15% in the CG (p < 0.001). After admission to our post operative unit all chest tubes were treated with suction as defined by protocol. The duration of postoperative air leak chest tube suction was significantly shorter in the PSG compared to the CG (Figure 1, p < 0.001), and air leak chest tube (ventral chest tube) was removed significantly earlier at 2.1 days for PSG compared to 3.9 days for CG (Figure 2, p = 0.03). Limitation of chest tube removal due to a pulmonary leak was 5% in PSG and 35% in CG (p = 0.04) without any difference for drainage time (p = 0.94). All patients in the PSG had no evidence of pneumothorax in chest X-ray before discharge and after a 30 day follow-up period compared with 4 (20%) in the CG (p = 0.16).