Discussion

 

The use of routine prophylactic central lymph node dissection in PTC remains controversial. We have presented the design and feasibility of a randomized controlled trial to evaluate whether such an approach would lead to improved or impaired outcomes of PTC patients. The single most important factor in calculating sample size scenarios required to obtain sufficient statistical power is the event rate. It is well documented that PTC without pre- or intraoperative central neck nodal disease (i.e., cN0) has low recurrence rates, even in the setting of histopathological presence of lymph node metastasis, with rates ranging from 2% to 9 % (12,14–20). Since event rates are low in these PTC patients, we demonstrate that thousands of patients would need to be included in the study and followed for many years. As a point of reference, approximately 3%–7% of all U.S. patients with cN0 PTC would need to be recruited into the study with long-term surveillance to achieve sufficient statistical power (1). Similarly, morbidity rates are low after surgery for PTC, and large sample sizes would be necessary to detect a statistically significant difference between the two groups. Permanent hypoparathyroidism occurs in approximately 1%–2% (range 0.8%–5.4%) of patients after total thyroidectomy (6). The reported range of permanent hypoparathyroidism after total thyroidectomy with prophylactic central lymph node dissection is 0%–14.3% (6). The rate of unintentional permanent recurrent laryngeal nerve injury after total thyroidectomy ranges between 0% and 5.5%, whereas the rate is 0%–5.7% after total thyroidectomy with central lymph node dissection (6). Similar to studies analyzing the rate of permanent hypoparathyroidism, these studies are retrospective and self-reported and laryngoscopy was used selectively. Thus, the true incidence of nerve injury is largely unknown.

The success of a multicenter prospective randomized controlled trial of prophylactic central lymph node dissection in cN0 PTC would depend largely on as much standardization as possible of the diagnosis, treatment, and surveillance of the study subjects. We realize that the proposed protocol is not uncontroversial, for instance, the attempt to standardize the operative management and postoperative radioactive iodine treatment independent of the stage of PTC at final pathology. This could potentially hamper the enthusiasm for participation of treating physicians and impede study subject enrollment.

The current investigation did not specifically address the costs of this hypothetical study. However, a large simple trial over 7 years (4 years enrollment, 3 years minimum follow-up, 7 years maximum follow-up, and 5 years average follow-up) with enrollment of 5840 patients would be expensive. Approximately 30 centers would be required to enroll 50 subjects per year over 4 years to achieve the target sample size. At a minimum, one full-time study coordinator per study site would be needed to enroll and follow 200 patients for study outcomes for an average of 5 years at a total cost of approximately $15 million over a 7-year study period. In addition, funding would be required for startup and closeout costs, a national study coordinator, statistical and data coordinating center, site monitoring, investigator meetings, data and safety monitoring board, etc. No funding would be requested for participating patients, study surgeons, endocrinologists, radiologists, pathologists, and nuclear medicine physician. No funds would be requested for the medical care of the subjects as it would adhere to recommended state-of-the-art practice. We estimate that the total cost of the study would be approximately $20 million or $3425 per enrolled study subject.

In summary, our analysis of a hypothetical randomized controlled trial to evaluate the role of routine prophylactic central lymph node dissection for cN0 PTC indicates that given the low rates of newly identified structural disease and morbidity, prohibitively large samples sizes would be required to detect statistically and clinically relevant differences between the groups. This means that no single institution could perform this study alone. Therefore, a multi-institutional collaborative trial would need to be done. This would introduce significant variation in surgical techniques, as well as local institutional variability in inclusion and follow-up of patients. In addition, a large number of surgeons, endocrinologists, radiologists, nuclear medicine physicians, and pathologists would need to participate. These issues would introduce additional variables and challenges. We thus conclude that a randomized controlled trial of prophylactic central lymph node dissection in PTC is not readily feasible. Studies reporting the absence of an outcome difference based on small sample sizes may have inadequate statistical power from which to draw such a conclusion.