Spontaneous Ventilation Video-Assisted Thoracoscopic Surgery for Non-small-cell Lung Cancer Patients With Poor Lung Function: Short- and Long-Term Outcomes

Objective The goal of this study was to explore the feasibility and safety of spontaneous ventilation video-assisted thoracoscopic surgery (SV-VATS) for non-small-cell lung cancer (NSCLC) patients with poor lung function. Methods NSCLC patients with poor lung function who underwent SV-VATS or mechanical ventilation VATS (MV-VATS) from 2011 to 2018 were analyzed. 1:2 Propensity score matching (PSM) was applied, and the short- and long-term outcomes between the SV-VATS group and the MV-VATS group were compared. Results Anesthesia time (226.18 ± 64.89 min vs. 248.27 ± 76.07 min; P = 0.03), operative time (140.85 ± 76.07 min vs. 163.12 ± 69.37 min; P = 0.01), days of postoperative hospitalization (7.29 ± 3.35 days vs. 8.40 ± 7.89 days; P = 0.04), and days of chest tube use (4.15 ± 2.89 days vs. 5.15 ± 3.54 days; P = 0.01), the number of N1 station lymph node dissection (2.94 ± 3.24 vs. 4.34 ± 4.15; P = 0.005) and systemic immune-inflammation index (3855.43 ± 3618.61 vs. 2908.11 ± 2933.89; P = 0.04) were lower in SV-VATS group. Overall survival and disease-free survival were not significantly different between the two groups (OS: HR 0.66, 95% CI: 0.41–1.07, P = 0.09; DFS: HR 0.78, 95% CI: 0.42–1.45, P = 0.43). Conclusions Comparable short-term and long-term outcomes indicated that SV-VATS is a feasible and safe method and might be an alternative to MV-VATS when managing NSCLC patients with poor lung function.


INTRODUCTION
Surgery is the primary approach to the treatment for the early stage of non-small cell lung cancer (NSCLC) (1). Traditional thoracotomy is controversial due to its severe complications and poor prognosis. Video-assisted thoracic surgery (VATS) has been widely used because it is minimally invasive and associated with less postoperative complications, shorter operations time and hospital stay compared with open thoracotomy (2)(3)(4).
General anesthesia with double-lumen intubation induced with mechanical ventilation during video-assisted thoracoscopic surgery (MV-VATS) has been widely recognized as a secure and effective thoracic anesthesia approach and is now considered a standard treatment option (5,6). However, some adverse effects of the MV-VATS, including intubation-related airway trauma, impaired cardiac performance, and residual neuromuscular blockade, may pose uncertain risks for the surgery and affect the prognosis of NSCLC patients (7)(8)(9)(10).
Spontaneous ventilation video-assisted thoracoscopic surgery (SV-VATS) has developed rapidly recently and achieved promising results (11). According to the reports, SV-VATS can reduce the perioperative adverse effects of tracheal intubation and general anesthesia in thoracic surgery (12)(13)(14) and provide more comfortable surgical experience for NSCLC patients. Therefore, SV-VATS has become a viable alternative to various thoracic surgeries (15)(16)(17).
Numerous studies have proved the feasibility and safety of SV-VATS in treating NSCLC (12,13,(18)(19)(20). However, since lung cancer resection in patients with poor pulmonary function is considered risky and controversial (21)(22)(23)(24), the safety and efficacy of SV-VATS in those patients remain unclear. This study aims to compare short-and long-term outcomes between the SV-VATS and the MV-VATS in treating patients with poor lung function.

Patients Inclusion
This study was reported under STROCSS 2019 guidelines (25). From January 2011 to July 2018, 5,857 NSCLC patients undergoing primary lung cancer resection at the First Affiliated Hospital of Guangzhou Medical University were consecutively identified and retrospectively collected through electronic records.
The Institutional Review Board of the hospital reviewed the study protocol and methods, IRB report ID: 2018-57.
Informed consent was obtained from every patient. Patients undergoing SV-VATS and MV-VATS were set as experimental and control groups, respectively. Patients were included based on the following criteria: (i) diagnosed with lung cancer based on pathological outcome; (ii) tumor size ≤ 5 cm; (iii) American Society of Anesthesiologists status class ≤ 3; (iv) Eastern Cooperative Oncology Group score ≤1; (v) patients underwent segmentectomy or lobectomy; (vi) no serious arrhythmia such as atrial fibrillation and frequent premature beat; and (vii) no cardiac insufficiency. Patients were excluded based on the following criteria: (i) a history of infection leading to massive pleural effusion, such as tuberculosis and pneumonia; (ii) a history of thoracic surgery. Patients and attending practitioners decided whether to receive SV-VATS or MV-VATS. Two independent radiologists performed enhanced CT scans for all included patients.

Preoperative Preparation
Electrocardiograms, heart rate, respiratory rate, pulse oxygen saturation (SpO 2 ), blood pressure, and bispectral index (BIS) were routinely monitored for patients after entering the operating room. Arterial catheters were routinely used to monitor the vital signs of the patient during the process of VATS surgery, such as blood gas analysis and the fluctuations in blood pressure. Central vein puncture were performed when necessary, so as to rescue the patient better and faster in unexpected circumstances during the operation. Atropine (0.01 mg/kg) was injected intravenously before performing anesthesia to prevent excessive airway secretions.

SV-VATS Group
The combination of local nerve block and intravenous anesthesia was employed for SV-VATS. The patient was anesthetized as follows: (i) laryngeal mask airway (Royal Fornia Medical Equipment, Guangdong, China.); (ii) intravenous anesthesia; (iii) thoracic paravertebral block (7th and 8th thoracic vertebra); (iv) visceral pleural surface anesthesia (the visceral pleural surface was anesthetized with 5 mL of 2% lidocaine); (v) operated side thoracic vagus nerve block (under direct thoracoscopic view, 2 mL of 0.25% bupivacaine was infiltrated near the vagus nerve at the level of inferior trachea for the right side of procedure and the level of aortopulmonary window for the left side of procedure). Epidural block and other nerve block methods would be used when necessary. Anesthesia was induced with target-controlled infusion (TCI) of sufentanil 0.1-0.2 mg/kg and propofol 2.0-4.0 mg/mL. No muscle relaxant was used during the procedure. Sedatives such as propofol were used when BIS was greater than 60. A laryngeal mask airway would be inserted for patients if BIS dropped below 60. Anesthesia was maintained with TCI of propofol (target plasma concentration 0.5-1.0 mg/mL), inhaled sevoflurane (± 0.8-1.5fold the minimum alveolar concentration), dexmedetomidine 0.05-0.10 µg/kg/min, remifentanil 0.05-0.15 µg/kg/min. To maintain BIS at 40-60 during operation, dexmedetomidine 0.5-1.0 mg/kg/h was added. To facilitate spontaneous ventilation, a laryngeal mask airway connected to the anesthesia machine was used. The third-generation double-tube LMA was used for ventilation management. If spontaneous ventilation cannot be used in the SV-VATS group, manual ventilation or simultaneous intermittent mandatory ventilation (SIMV) mode would be used to assist ventilation during anesthesia induction.

Surgical Process
The same surgical procedure was operated in the MV-VATS and the SV-VATS groups. Stryker 1288 HD 3-Chip Camera System (Stryker, Kalamazoo, MI, USA) was adopted as video-assisted thoracoscopic operations, and the endoscopic instruments were designed ourselves. Patients were placed in a lateral decubitus position with upper arms extended and fixed on the hand support.
According to the actual situation, surgery was carried out using a one-port or two-port approach for VATS. The thoracoscope was inserted at the 7th or 8th intercostal space on the anterior axillary line, which carried a soft incision protector to safeguard the skin, subcutaneous tissue, rib, and pleura.
Types of surgery included lobectomy and segmentectomy. N1 and N2 station lymph nodes were dissected routinely for all patients.

Data Collection and Statistical Analyses
All data were extracted by two blinded, experienced abstractors, and a third abstractor judged conflicts.
This study integrated patient demographics, preoperative laboratory examinations, anesthetic, surgical information, postoperative recovery information from electronic medical records, and survival information via telephone or interviews. The systemic immune-inflammation index (SII), neutrophilto-lymphocyte (NLR) ratio, and platelet-to-lymphocyte (PLR) ratio were calculated before and after surgery. SII was calculated as follows using formula: SII = platelet count * neutrophil count/lymphocyte count. Propensity score match (PSM) was adopted to reduce treatment selection biases. Patients who were lost to follow-up during the 90-day period were excluded from PSM. Factors including age, BMI status, smoking status, gender, pulmonary function, histological analysis, American Society of Anesthesiologists (ASA) status class, tumor position, T stage, N stage, M stage, and surgical technique were considered in PSM. Patients were matched and included in the experimental and control groups as a ratio of 1:2. The matching was performed using nearest neighbor method. Confounding variables were considered comparable when the standardized mean difference was below 0.10 and the p-value was greater than 0.05. Descriptive data were presented as means plus standard deviation, and categorical data were presented as percentages. The normality was examined using Shapiro-Wilk test. Student t-test or Wilcoxon test was performed for the comparison of inter-group continuous variables. Categorical variables were compared with Chi-squared test or Fisher's exact test. Kaplan-Meier was employed to estimate the overall survival (OS) and diseasefree survival (DFS). A log-rank test was used to calculate p-values. All tests were 2-sided, and a p < 0.05 was considered statistically significant.
The follow-up strategy followed the NCCN guidelines (32). Specifically, physical examination and non-enhanced chest CT examinations were performed every 6 months for 3 years after surgery and annually after that and included a routine history.

Patient Baseline Characteristics and Demographics Data
The flow diagram of patient recruitment was displayed in Figure 1. Overall, 5,292 NSCLC patients underwent SV-VATS or MV-VATS were included in the study. Among the enrolled patients, 396 patients were presented with poor lung function.
After performing a 1:2 PSM, 24 patients in the SV-VATS group failed to match with control group and were excluded in the study (Supplementary Table 3). Finally, 31 and 62 patients were obtained from the SV-VATS and MV-VATS group, respectively. In the SV-VATS group, 2 patients underwent segmentectomy, while 29 patients underwent lobectomy. In the MV-VATS group, 7 patients underwent segmentectomy, while 55 patients underwent lobectomy. FEV1/FVC (63 ± 7% vs. 61 ± 9%, p = 0.70) was similar between the two groups. Baseline demographic and clinical variables were well-balanced between the two groups ( Table 1). The distribution of patient characteristics and propensity scores were shown in Supplementary Figures S1, S2.

Comparison of Short-Term Outcomes
Both of the days of chest tube use (4.10 ± 2.29 days vs. 4.87 ± 2.96 days; P = 0.44) and days of postoperative hospitalization (7.74 ± 5.41 days vs. 9.97 ± 7.95 days; P = 0.20) were comparable in two groups.
There was no statistically significant difference in preoperative inflammatory indicators between the SV-VATS and MV-VATS  35; P = 0.94) after surgery was significantly lower in SV-VATS group than MV-VATS group. The incidence of complications (16 vs. 24%, P = 0.37) was comparable in the two groups. The details of postoperative outcomes were summarized in Tables 2, 3.

DISCUSSION
This retrospective study compared the short-and long-term outcomes of NSCLC patients with poor lung function treated

MV-VATS, mechanical ventilation video-assisted thoracoscopic surgery; SV-VATS, spontaneous ventilation video-assisted thoracoscopic surgery.
with SV-VATS and MV-VATS. The results indicated that SV-VATS generally exhibits comparable short-term effects, treatment prognosis, and long-term outcomes to MV-VATS, supporting its use in NSCLC patients with poor lung function. VATS is a minimally invasive thoracic surgical technique that results in significantly less postoperative pain than conventional thoracotomy. Early outcomes such as postoperative and postprocedure infection can be improved by avoiding the rib spreader, muscle tissue division, and severing of intercostal (33,34). In the traditional cognition, patients with poor lung function are unbearable to one-lung ventilation, not to mention VATS (35). Beyond this, other contraindications include morbid obesity, hemodynamic instability, extensive pleural adhesions, inability to cooperate, large centrally located tumors, a difficult airway, and others.
As the previous studies in our center on other thoracic patients undergoing SV-VATS, we concluded that non-intubated anesthesia VATS can shorten surgical operation and anesthesia time, reduce bleeding and optimize perioperative rehabilitation (12,13,(18)(19)(20)36). Studies from other centers also revealed that SV-VATS was associated with a shorter surgical time (37)(38)(39), and that it was a well-tolerated, feasible, and safe treatment option (39,40).
However, few studies have been conducted to determine the feasibility and safety of SV-VATS in treating NSCLC patients with poor lung function. The applicability of SV-VATS in such patients remains uncertain. Wang et al. (41) demonstrated that non-intubated VATS for patients with impaired lung function was technically feasible, with a sample size of only 28 patients and without control group. As a consequence, further evidence is required to fully understand the feasibility and safety of SV-VATS in patients with poor lung function.
Our study demonstrated that SV-VATS could be feasible and safe for NSCLC patients with poor lung function according to intraoperative and postoperative short-and long-term outcomes.
Anesthesia is a key area that should be considered. The differences in anesthesia between the two groups are as follows: (i) No use of relaxant in the SV-VATS group. (ii) Reduced use of paroxysms in the SV-VATS group. The difference between the two types of anesthesia cannot be considered a confounding factor in the patient's postoperative outcome, as this study systematically compares the impact of the entire surgical procedure on patients, including different anesthesia types.
Patients undergoing SV-VATS had shorter anesthesia time and faster recovery, possibly due to its simplified preoperative preparation, such as no muscle relaxants, no bronchoscopy, and no endotracheal tube. Apart from this, the SV-VATS group received a decreased dose of major amnesic agents (propofol or sevoflurane) and regional anesthesia such as paravertebral without muscle relaxants (42). With a shorter anesthesia time, the SV-VATS group exhibits a less total amount of sedative medication, reducing side effects by avoiding over-dosage. Considered to be related to the prognosis of NSCLC patients, we collected their blood routine records to calculate SII, NLR, and PLR. The SV-VATS group had significantly lower NLR and PLR after surgery than that in the MV-VATS group in this study.
Only five patient undergoing SV-VATS experienced complications after surgery. No patient developed postoperative nausea and vomiting postoperatively, which may be related to the fact that all patients did not use morphine-based analgesics postoperatively (43,44).
Patients undergoing SV-VATS have a shorter chest tube duration because of avoiding the intubation during the surgery, which can help to reduce the inflammatory responses and stress response (45,46). Numerous types of thoracic procedures have demonstrated the advantages of SV-VAT (11). First, it could effectively reduce the intubation-related complications, such as atelectasis and ventilation-induced lung injury (47). Second, without using muscle relaxants (such as cis-atracurium), residual effects, including weakness of upper airway muscles and diaphragmatic, could be avoided to some extent (7,48). Likewise, the advent of sugammadex may prevent the practice and prolonged paralysis risks. Third, SV-VATS provided better recovery for digestive and respiratory function due to the reduced use of major amnesic agents and muscle relaxants (49). Fourth, post thoracic surgery pain could be relieved for patients undergoing SV-VATS, possibly due to a better perioperative pain control by the paravertebral block (50).
Although SV-VATS is an emerging surgical technique, the surgical team should select the best surgical anesthesia method in the best interest of patients before performing SV-VATS. Besides, surgeons and anesthesiologists of the operation team should introduce the advantages and disadvantages of these two anesthesia methods to patients and their families according to the situation of patients before operation. The selection of surgical anesthesia method should be based on the willingness of the patients and their family members under the guidance of surgeon. In addition, several limitations of SV-VATS should be considered. First, chronic spontaneous breathing can result in hypercapnia (defined as PaCO 2 ≥60 mmHg) and hypoxia (defined as SpO 2 <90%). Second, since an uncontrollable ventilation situation is possible, conversion to general anesthesia with intubation may occur. Previous research indicated that the conversion rate was about 0-10% (51). Third, aspiration may occur without proper protection of the endotracheal tube, such as inadequate fasting, hiatal hernia, and gastroesophageal reflux disease.
We acknowledge several limitations in this study. First, this is a retrospective study. Although we used PSM, selection bias may exist. Second, as this is a retrospective study, we are unable to present some of the important outcome indicators, such as the patient's postoperative lung function and further indicators of lung function assessment, patient's postoperative pain, anesthesia, patient reported outcomes (PROs) or quality of recovery (QoR), data on intraoperative blood gases regarding ventilation/PaCO 2 and acid/base which we will focus on in further studies. Third, since this is a single-center analysis, external validation should be established. Two randomized, multicenter trials are currently underway across 11 centers in Asia, focusing on NSCLC and lung bullae (NCT03432637 and NCT01677442), including our center. Third, a larger SD was observed in SII, NLR and PLR, which might affect our results to some extent. For that, blood routine data were double-checked to exclude the documentation error. The recent preoperatory and postoperative blood routine data for each patient were collected to observe the changes. However, different intervals between blood routine tests exist, implying a large SD.

CONCLUSION
Our findings indicated that the SV-VATS is feasible and safe for NSCLC patients with poor lung function. Comparable short-and long-term outcomes implied that SV-VATS might be a viable alternative to MV-VATS when managing NSCLC patients with poor lung function.

CENTRAL MESSAGE
We found that SV-VATS was comparable to MV-VATS in NSCLC patients with poor lung function, implying that SV-VATS might be a feasible alternative to MV-VATS for NSCLC patients with poor lung function.

PERSPECTIVE STATEMENT
Poor lung function is a contraindication of SV-VATS. SV-VATS in such patients is controversial. We compared the short-and long-term postoperative outcomes between SV-VATS and MV-VATS in NSCLC patients with poor lung function. The results indicated the safety of SV-VATS in patients with poor lung function, implying a more extensive application of SV-VATS in such populations in the foreseeable future.

DATA AVAILABILITY STATEMENT
The datasets presented in this article are not readily available because; The patients involved in this study are from the Department of Thoracic Surgery of our hospital and are not open to anyone other than the author of this study.
Requests to access the datasets should be directed to Runchen Wang, runchen_wang@outlook.com.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Institutional Review Board of the First Affiliated Hospital of Guangzhou Medical University. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.