Patients and data collection
We reviewed all patients who underwent anatomic resection for pulmonary mycobacterial infection or symptomatic bronchiectasis between January 2010 and December 2018 at the National Cheng Kung University Hospital using data from the database of thoracic surgery and medical records. The surgical indications included a history of chronic cough, recurrent pneumonia or hemoptysis, and obvious sustaining lung lesions (i.e., bronchiectasis, infected cavity, or destroyed lobe) on chest computed tomography. We excluded patients who underwent an intentional open thoracotomy approach, among whom most had a destroyed lung and underwent pneumonectomy, and those with simultaneous lung abscess and empyema because of acute infection entity. This study, which did not require an informed consent, was approved by the National Cheng Kung University Hospital Institutional Review Board (approval no. B-ER-107-108).
We classified patients into two groups: SP-VATS and MP-VATS. The statistical analysis include the following variables: (1) demographic and clinical information (i.e., sex, age, disease etiology, comorbidity, preoperative pulmonary function test, surgical indications, sidedness, extent of resection, and procedure type) and (2) the operation characteristics (i.e., the presence of calcified LNs, adhesion score, operation time, major vessel injury, the presence of conversion, chest tube duration, intensive care unit and hospital stay, and complications).
Calcified LN and total adhesion score
Calcified LN is the calcification of LNs surrounding the bronchovascular bundle of the target lobe by the mediastinal window of the preoperative chest computed tomography (40 H level and 300 H width). When calcified LNs are present, the MP-VATS approach may be chosen intentionally depending on the surgeon’s preference.
The total adhesion score is the sum of the pleural and fissure adhesion scores. The first intraoperative step was determining the severity of pleural and fissure adhesions subjectively as follows: 0 represents no or only focal adhesion; 1 limited adhesion around the target area, requiring less time to complete adhesiolysis; and 2 diffuse adhesion or fused fissure, making adhesiolysis more time-consuming. Upon VATS exploration, we evaluated SP-VATS’ feasibility using the total adhesion score. If the score is ≧ 3 points, the surgeon could change to MP-VATS based on his judgment.
Surgical procedure
An experienced anesthesiologist intubated all patients with a double-lumen endotracheal tube placed in the lateral decubitus position. A single incision, approximately 4 cm long, was made in the fourth or fifth intercostal space at the anterior axillary line. The surgical techniques and instruments used in SP-VATS were similar to those in MP-VATS. If the operation needs to be converted, whether the initial approach is SP- or MP-VATS, minithoracotomy will be performed. Typically, we divided the pulmonary arteries first to avoid congestion of the targeted lobe. The pulmonary veins and the bronchus were then dissected separately. The bronchus was divided using an Endo-GIA stapler (Echelon Flex Endocutter, Johnson & Johnson, USA), but buttressing the bronchial stump was not routinely performed. After meticulous air leakage control, we placed a 24 Fr chest tube.
Special techniques
Because the limitations of SP-VATS, such as being a single-direction approach and limited instruments in the utility wound, we developed certain useful techniques to facilitate anatomic resection (Fig. 1).
First is the flexible hook electrocautery (Fig. 1A). Generally, blunt dissection using fingers or small peanut gauzes managed dense pleural adhesions. L-hook electrocautery and energy-based instruments were used to accelerate adhesiolysis. However, it is hard to use linear instruments for pleural adhesion in the apical or costophrenic area. So, we used flexible hook electrocautery to complete adhesiolysis.
Second is the hilum-first technique. After VATS exploration and adhesion severity evaluation, we could begin with the vascular structure dissection if hilar adhesion was relatively minor rather than a dense pleural adhesion (Fig. 1B and C). Pleural adhesion could be used as counter traction, and we could manipulate a single energy device and a suction device to avoid instrument fencing.
Third is the Satinsky vascular clamp use (Fig. 1D). Previously, when major bleeding or fear of main pulmonary artery injury during dissection occurs, we usually converted to open thoracotomy for vessel repair or proximal pulmonary artery control. Currently, we apply a Satinsky vascular clamp immediately via the utility wound to clamp the vessel and bronchus concomitantly, which could be used for bleeding control during vascular repair and massive bleeding prevention during vascular dissection when encountering dense adhesion or calcified perivascular LNs.
The final technique is the staged closure of the bronchial stump method (Fig. 1E). Calcified LNs between the pulmonary arteries and bronchus result in separation and division difficulty. Staple disruption potentially occurs if bronchovascular bundles are transected simultaneously by an Endo-GIA stapler. Therefore, we developed this technique to resolve this problem. Endoscopic scissors can cut the bronchus step by step. Suture ligation or Endo-GIA stapler can loop and sequentially divided the pulmonary arteries behind the bronchus. Finally, the opened bronchial stump can be lifted and closed using an Endo-GIA stapler (Fig. 1F).
Statistical analysis
Descriptive statistics were used to assess the patients’ demographic and perioperative characteristics. Continuous data were expressed as means ± standard deviation, whereas categorical data were expressed as frequencies and proportions. Continuous variables were analyzed using the Mann-Whitney test, and categorical variables using the chi-squared test. A probability value of less than 0.5 was considered statistically significant. All statistical analyses were performed using IBM SPSS Statistics for Windows, version 19
(IBM Corp., Armonk, N.Y).