As a medical oncologist, my primary role is to direct general management plans for many cancer patients and to develop chemotherapy and targeted therapy regimens. These regimens are sometimes directly administered through my office, and sometimes are coordinated with oncologists closer to a patient's home. The treatment is pretty much a cookbook approach, so it's really the same no matter who administers it. On the other hand, for the approximately 1/3 of NSCLC patients who are candidates for surgery, there are major differences in clinical outcomes depending on the skill level of the surgeon. Here I'll illustrate some of the key differences that make it compelling to work with the best surgeon you can find, ideally a trained thoracic surgeon, and not just the closest surgeon or the first who has an opening on their schedule.

In 1998, Silvestri and colleagues published a summary of surgical outcomes for over 1700 patients who underwent lung surgery in South Carolina between 1991 and 1995, comparing the differences between general surgeons and specialty-trained thoracic surgeons (article here). The total number of surgeries was pretty evenly divided between general surgeons and thoracic surgeons, and the patient population was essentially identical between them. The mortality rate (death within 30 days of surgery) was significantly higher for general surgeons, however, in the patients who underwent a lobectomy (5.3% vs. 3%), and was also signficantly higher if general surgeons operated on patients over 65 or with other major medical problems. Pneumonectomies (removal of an entire lung, which is an overall easier surgery for the surgeon but much more likely to have complications or death after it) were more commonly done by general surgeons (12.2% vs. 8.8%), and had a remarkably higher mortality rate (20.2% vs. 11.8%), which did not achieve statistical significance because the number of patients undergoing pneumonectomies was so much smaller than the number who had lobectomies. The message, however, was clear: patients who underwent lung surgery from a trained thoracic surgeon had a lower likelihood of having to lose an entire lung, and a strikingly lower risk of dying from that surgery.

Other studies have corroborated these conclusions. An article several years ago in the New England Journal of Medicine (abstract here) described a review of a huge survey of patient outcomes after cancer surgery, dividing the results between low volume centers and high volume centers. As Silvestri and colleagues saw with individual surgeons, the more experienced hospitals had a lower complication rate (20% vs. 44%) and mortality rate (3% vs. 6%). Importantly, the complication and death rate immediately after surgery does not tell the whole story. Patients who receive suboptimal surgery may also not be as thoroughly staged and have complete nodal dissections that may compromise longer-term survival. In fact, the 5 year-survival was 33% after lung surgery at low-volume hospitals and was 44% in more experienced centers. And other studies have also shown improved survival after lung cancer surgery at high-volume centers (abstract here) and by individual surgeons with greater experience (abstract here).

And there are other studies with more evidence, but they all show the same results. Less specialty-trained surgeons are associated with higher risk of post-operative complications and death after surgery, are more likely to do a pneumonectomy that is more challenging for the patient but an easier surgery. Thoracic surgeons are also more likely to do a thorough lung cancer work-up that includes mediastinal staging, and an extensive nodal dissection at the time of surgery.

This affects the patients I see in the real world all the time. Today in my clinic I saw a woman who had a lobectomy for a stage IB NSCLC last year, but she never had a mediastinoscopy and had only a few lymph nodes removed at the time of her surgery by a less experienced surgeon. She recently had a major recurrence of the cancer throughout her mediastinum, compressing her airway and her esophagus, and know we're treating her very bulky disease with chemo and radiation. After her surgery, she was never offered chemo, partly because her staging showed no nodes involved, but was that really true, or was it a case of "don't ask, don't tell"? You can't find a fever if you don't take a temperature, and if you don't do proper staging and thorough surgery, you just can't know whether you're undertreating for the true stage of the cancer. Would she be cancer-free now if she had been more thoroughly staged, likely found to have mediastinal node involvement last year, and been treated aggressively then? I can't say for sure, but I know we would have a better chance of curing stage III NSCLC when it's not bulky and compressing critical structures in the middle of the chest.

At the end of the day, while there are many very talented general surgeons, the evidence clearly shows that lung cancer patients who undergo surgery by well-trained thoracic surgeons have lower complication and mortality rates, and a better overall survival years later. They also are more likely to have very thorough staging and less likely to have a pneumonectomy, only when it's truly necessary. Overall, I would recommend that lung cancer patients who are candidates for surgery should chase down the best-trained lung surgeon they can find and not the surgeon who does a lung case every few weeks between cardiac and other surgeries.