Outcomes in video-assisted thoracoscopic surgery lobectomies: challenging preconceived notions

Most thoracic surgical procedures in the United States are being performed by general surgeons (GSs) without any advanced training. With the recent approval of computed tomography screening for lung malignancy in high-risk populations, the number of thoracic oncologic resections is expected to rise. Previous literature has demonstrated consistently worsened outcomes for patients undergoing thoracic surgical procedure when done by nonthoracic fellowship-trained surgeons. Using the American College of Surgeons National Surgical Quality Improvement Project database, we examined short-term outcomes in patients undergoing video-assisted thoracoscopic surgery (VATS) lobectomy for malignancy.

Data were obtained from the American College of Surgeons National Surgical Quality Improvement Project from 2010-2015. We identified patients who had an International Classification of Disease 9 diagnosis of lung cancer (162) who underwent VATS lobectomy (current procedural terminology 32663). We included only adults (≥18y) and elective cases. We excluded patients who had preoperative diagnosis of sepsis, contaminated wound class, or those patients with missing American Society of Anesthesiologists classification, morbid obesity, functional status, length of stay (LOS), or sex, and race information. We identified two groups by specialty: GS versus cardiothoracic (CT) surgeon. We then performed univariate analysis. We then performed propensity score analysis using a 1:3 ratio of general surgery patients to CT patients. Outcomes of interest included 30-d postoperative mortality, 30-d postoperative morbidity, and LOS.

A total of 4105 patients were identified, 607 performed by GSs, 3508 performed by CT surgeons. The mean age for patients who underwent lobectomies by GSs was 68.6 versus 67.8 in the CT surgeon group (P < 0.05). The majority were female (58.09% GS versus 57.74% CT surgeon). There was a statistically significant difference in race between groups; patients were more likely to be African American in the CT surgeon group. Operative time was lower in the GS group as opposed to the CT surgeon group 179 min versus 196 (P < 0.01). Univariate analysis (mortality <0.1 CT surgeon and GS) and 1:3 propensity score matched analysis (0.08 GS% versus 0.08% CT surgeon) failed to demonstrate a significant difference in mortality. There was a statistically significant difference in median LOS between groups (6.2 GS versus 5.1 CT surgeon). Univariate and propensity matched analyses of pneumonia, sepsis, wound infection, deep vein thrombosis, transfusion requirement, myocardial infarction stroke, postoperative renal insufficiency, failure to wean, pulmonary embolism, reintubation, and deep organ space infection all failed to demonstrate a statistically significant difference between our groups of interest. Urinary tract infection was noted to be higher in the GS group operating room 2.29 as compared to the CT surgeon group (P value 0.02).

In this large observational study, we found that VATS lobectomies performed by GS compared to the matched CT surgeon cohort had shorter operative time, and there was no difference in major postoperative morbidity or mortality. However, LOS was higher and there was increased risk of urinary tract infection in the GS compared to matched CT surgeon cohort.