AT THE INTERNATIONAL CONFERENCE ON HEAD AND NECK CANCER
SEATTLE (FRONTLINE MEDICAL NEWS) – Oncologic transoral robotic surgery patients spend less time in the hospital if they have neck dissections at the same time, instead of later, according to a reviewof 441 patients by Stony Brook (N.Y.) University.
The average hospital length of stay (LOS) was 6 days for the 349 patients (79.1%) who had lymphadenectomy neck dissections at the same time as transoral robotic surgery (TORS). The 92 patients (20.9%) who had staged procedures – neck dissections and TORS about a month apart, with TORS usually done first – stayed in the hospital an average of 8 days (P less than .0001). After risk adjustment, LOS was 43% shorter for concurrent dissections.
Cardiac arrhythmias were also more common in staged patients, perhaps because they had general anesthesia twice in a short period or maybe because staged patients were more likely to be obese (18.5% vs. 7.5%; P less than .01).
However, there were no statistically significant outcome differences otherwise, and the investigators concluded that “concurrent and staged procedures are equally safe. It is therefore reasonable to allow operator preference and patient factors to determine surgical logistics.”
Neck dissection timing has been controversial since the advent of TORS several years ago, when surgeons and administrators realized they could fit more cases into the schedule by doing neck dissections, which can take a few hours, at a different time.
Proponents of the staged approach argue, among other things, that it reduces the risk of fistulas and tracheotomies, and allows surgeons a second go at positive margins. Advocates of concurrent procedures counter that fistulas, if found, can be repaired right away, and that same-time surgery saves money, allows for earlier adjuvant therapy, and cuts anesthesia risks.
There hasn’t been much data to settle the debate, and no one has compared LOS before, so it was “important” to look into the matter, lead investigator Catherine Frenkel, MD, a Stony Brook general surgery resident, said at the American Head and Neck Society International Conference on Head and Neck Cancer.
German investigators also recently concluded that it’s pretty much a draw between concurrent and staged dissections. In a study of 41 TORS cases, “the timing of neck dissection did not make a significant difference in the outcomes. We suggest, therefore, that aspiring and established TORS teams do not restrict their appropriate indications due to robotic slot and theatre time constraints, but perform each indicated TORS case as soon as possible within their given systems, even if the neck dissections cannot be done on the same day,” they said ( Eur J Surg Oncol. 2015 Jun;41[6]:773-8 ).
In addition to obese patients, those who had tongue or tonsil lesions were more likely to be staged in the Stony Brook analysis. About half of the surgeons in the study stuck solely to concurrent procedures, while a handful opted for the staged approach, and the rest did both. Perhaps not surprisingly, high-volume surgeons – those who did five or more TORS cases per year – were more likely to stage.
Almost two-thirds of patients had at least one complication, most commonly renal failure, heart problems, extended ventilation, and surgical errors, which included accidental punctures, postop fistulas, hemorrhages, and wound complications. A total of 13% of patients had at least one postop readmission. Apart from arrhythmias, there were no statistically significant differences in complication or 30-day readmission rates between concurrent and staged patients. High-volume surgeons were less likely to have complications.
Postop bleeding was another common problem, and more likely with staged surgeries (12% vs. 7%). Concurrent procedures had a slightly higher rate of new tracheotomies and gastrostomies, but again the differences were not statistically significant, even with pedicle and free-flap reconstruction. There was no outside funding for the work, and the investigators had no relevant conflicts of interest.
