Back to Annual Meeting Program

Impact of T Status and N Status on Outcomes after Thoracoscopic Lobectomy for Lung Cancer
Nestor Villamizar, Marcus Darrabie, Jennifer Hanna, Mark Onaitis, David Harpole, Betty Tong, Thomas D'Amico, Mark Berry
Duke University, Durham, NC

A prospective, multi-institution study demonstrated the feasibility of thoracoscopic lobectomy for suspected stage I peripheral lung cancers <3cm in size. We tested the hypothesis that performing thoracoscopic lobectomy is safe and effective in patients with lung cancers that were larger, more central, or had clinically positive nodal disease.
All patients who underwent attempted thoracoscopic lobectomy for primary lung cancer between 6/1999 and 10/2010 at a single institution were reviewed. Morbidity included any perioperative complication as defined by the STS database. A model for morbidity including published preoperative risk factors as well as tumor size (>3cm vs ≤3cm), tumor location (central vs peripheral), and clinical N status (N1-N3 vs N0) was developed by multivariable logistic regression.

Of 1195 thoracoscopic lobectomies performed during the study period, 916 met the study criteria: 329 for peripheral, clinical N0 tumors ≤3cm in size and 504 for tumors that were central, clinical node positive, or >3cm in size; tumor location could not be documented for 83 patients (Table 1). Conversions to thoracotomy occurred in 36 patients (4%) and conversion rate was not higher for tumors > 3cm (p=0.89) or for central tumors (p=0.3) but was increased for patients with clinically node positive disease [11 conversions in 153 clinical N1-N3 patients (7.2%) vs 25 conversions in 763 clinical N0 patients (3.3%), p=0.03]. Overall operative mortality was 1.6% (14 patients) and morbidity was 32% (296 patients). The most common complications were atrial arrhythmia (128 patients, 14%), need for chest tube more than 5 days (107 patients, 12%) and atelectasis requiring bronchoscopy (54 patients, 6%). Although patients with larger tumors (p=0.006) and central tumors (p=0.01) had increased complications by univariate analysis, tumor size >3cm (p=0.15) and central location (p=0.5) did not significantly predict overall morbidity in multivariate analysis. Clinical node status did not predict increased complications by univariate or multivariate analysis. Significant predictors of morbidity in multivariable analysis were increasing age, decreasing FEV1, prior chemotherapy, and congestive heart failure (Table 2).

Thoracoscopic lobectomy for lung cancers that are central, clinically node positive, or larger than 3 cm does not have increased morbidity compared to peripheral, clinical N0 cancers that are smaller than 3 cm, though cancers that are clinically node positive are associated with a higher rate of conversion to thoracotomy.
Demographics and Outcomes for All Patients and stratified by location, clinical Nstatus, and size
All patients (n=916)Clinical N0Clinical N1-N3
Peripheral, ≤ 3cm (n=329)Peripheral, > 3 cm (n=81)Central, ≤ 3cm (n=165)Central, > 3 cm (n=112)Peripheral, any size (n=77)Central, any size (n=69)
Age67 (21-93)67 (38-88)69 (39-91)65 (23-93)69 (21-87)69 (40-83)69 (47-83)
Prior Chemotherapy60 (6%)9 (3%)3 (4%)13 (8%)6 (5%)14 (18%)12 (17%)
Prior Radiation48 (5%)14 (4%)3 (4%)10 (6%)5 (4%)7 (9%)6 (9%)
Perioperative Mortality14 (1.6%)4 (1%)2 (2%)2 (1%)1 (1%)1 (1%)2 (3%)
Conversion36 (4%)11 (3%)1 (1%)5 (3%)6 (5%)5 (6%)6 (9%)
Overall Morbidity296 (32%)85 (26%)32 (40%)55 (33%)50 (45%)26 (34%)25 (36%)

Multivariate risk model for complications.
Univariate p-valueMultivariate Analysis
Odds Ratio95% Confidence Intervalp-value
Age (per 1 year increase)<0.00011.0541.033-1.076<0.001
FEV1 (per 1% increase)<0.00010.9760.967-0.986<0.001
Prior Chemotherapy0.022.4531.161-5.1850.02
Congestive Heart Failure0.0032.1741.036-4.5620.04
DLCO (per 1% decrease)<0.00010.054
Tumor Size (>3cm vs ≤3 cm)0.0060.15
Smoking History0.0020.3
Tumor Location (Central vs Peripheral)0.010.5

Back to Annual Meeting Program


  Home | About WTSA | Contact Us