Proponents for elective neck dissection argue that it can be completed with a low incidence of complications and reoperation in the central compartment, for recurrence is associated with a higher incidence of complications. Evidence supporting prophylactic neck dissection of the central compartment has been extrapolated from experience with medullary thyroid carcinoma (MTC). Tisell et al. developed the technique for microdissection for thyroid cancer using magnifying loops and bipolar cautery [9]. Dralle et al. described compartment-orientated microdissection for MTC with a low frequency of complications with respect to recurrent nerve injury and hypoparathyroidism (3.1 %) [10]. On the basis of these observations, the technique was extrapolated to patients of WDTC. Tisell’s group reported on 195 patients with WDTC treated with total thyroidectomy and central node dissection [11]. The results were compared with contemporary series in other Scandinavian populations, including Norway and Finland. Death caused by thyroid cancer in their series was 1.6 % versus 8.4 % and 11.1 %, respectively, in the comparison groups. Scheumann et al. reported on 342 patients without making specific comparison between therapeutic versus elective neck dissection [12]. These authors stressed that for a tumour confined to the thyroid (no extrathyroidal extension), survival was increased by systematic compartment-orientated dissection. Sywak and co-workers showed the thyroglobulin levels to be lower in the group of 56 patients undergoing total thyroidectomy and central node dissection compared with 391 patients with total thyroidectomy alone [13]. The groups were comparable in terms of MACIS scores and radioactive iodine treatment. An undetectable thyroglobulin level was more likely with a central node dissection [13]. Total thyroidectomy resulted in 0.5 % hypoparathyroidism with no nerve injury, whereas for total thyroidectomy and ipsilateral central node dissection the rates were 1.8 % and 1.0 %, respectively. Gemsenjäger et al. noted hypocalcaemia in 2.4 % and recurrent nerve injury in 7.1 % of 42 patients undergoing therapeutic node dissections [14]. No hypocalcaemia or nerve injury was noted in the central node dissection or prophylactic node dissection group.

Many surgeons feel that a prophylactic neck dissection is not indicated, as the nodal metastases can be removed when they become clinically evident in the majority of patients, and that the survival rates of patients treated prophylactically and therapeutically are comparable. The other argument against routine central node dissection is the widespread use of radioactive iodine, which may obscure the benefits of neck dissection. Whereas the results from Tisell’s group were excellent, over a quarter of their patients had T1 disease [11]. It is not clear whether these patients would have done just as well with total thyroidectomy alone and no central neck dissection. Assessing the risk : benefit ratio of prophylactic neck dissections, Henry et al. evaluated complications after 50 total thyroidectomies for benign disease, and compared them to those in 50 patients with PTC (cN0) who were treated with total thyroidectomy and central node dissection [15]. The first group had three cases of transient nerve palsy (6 %) and 2 (4 %) in the second group. The multinodular group had no cases of permanent hypothyroidism, but 4 of transient hypoparathyroidism (8 %). In the thyroid cancer group, 7 patients developed transient hypoparathyroidism (14 %) and 2 patients (4 %) remained with definitive or permanent hypoparathyroidism. The authors concluded that after total thyroidectomy for PTC, central node dissection did not increase recurrent laryngeal nerve morbidity, but was responsible for a high rate of hypoparathyroidism, especially in the early postoperative course. They were of the opinion that even after taking into account the possible benefits, it is difficult to advocate routine central node dissection [15].

Roh et al. did not use a microdissection technique and reported hypocalcaemia in 30.5 % of patients after central compartment lymph node dissection [16]. In total thyroidectomy-alone group, the incidence of hypocalcaemia was 9.6 %. The mean number of parathyroid glands was 1.2 and was similar in the two groups. Four recurrences were observed: 3 of 73 of the total thyroidectomy alone, and 1 of 82 in the node dissection group. The central nodal group was involved in 62.2 % and the lateral group in 25.6 % recurrences. Considering overall survival, recurrence patterns and morbidity, these workers could not recommend routine use of central node dissection.

Pereira et al. reported a 60 % prevalence of positive lymphadenopathy in 43 clinically node-negative patients undergoing central node dissection and 58.1 % patients developed transient hypocalcaemia that was associated with incidental parathyroidectomy, number of nodes removed and thymectomy [17]. Permanent hypoparathyroidism was observed in 4.6 % of patients and transient vocal cord paralysis in 7 %. Central neck recurrences were not seen in any patient, but 5/43 patients developed lateral recurrences.

Frasoldati et al. described a series of patients in which there was no emphasis on central node dissection [18]. Central neck recurrence was observed in 5.8 % and it represented 60 % of all neck recurrences at 44 months, despite the use of radioactive iodine. Although central compartment node dissection is thought to reduce the risk of central neck failure, the same cannot be claimed about the lateral neck recurrences. This technique is also associated with a higher complication rate, especially of hypoparathyroidism.

The ATA guidelines recommend a therapeutic central compartment (level VI) neck dissection for patients with clinically involved central or lateral neck lymph nodes, in addition to total thyroidectomy [6]. For small (T1 or T2), non-invasive, clinically node-negative PTCs and most FTCs, a neartotal or total thyroidectomy without prophylactic central neck dissection might be appropriate and a prophylactic central-compartment neck dissection (ipsilateral or bilateral) may be performed in patients with PTC with clinically uninvolved central neck lymph nodes, especially for advanced primary tumours (T3 or T4). However, caution should be exercised when interpreting these recommendations in view of available surgical expertise. For patients with small, non-invasive, apparently node-negative tumours, the balance of risk and benefit may favour simple near-total thyroidectomy with close intraoperative inspection of the central compartment with compartmental dissection, only in the presence of obviously involved lymph nodes. This approach may increase the chance of future locoregional recurrence, but overall this approach may be safer in less experienced surgical hands.