Total Thyroidectomy Versus Lobectomy for Thyroid Cancer: Single-Center Data and Literature Review

We report data from an Italian series of DTCs divided into ATA low- and intermediate-risk classes and treated with LT or TT. In both classes, after a mean follow up period of 7 years, the patients treated with LT were found to require more additional therapies after initial surgery, to harbor a worse disease response at the dynamic risk stratification, and to have a lower event-free survival according to Kaplan-Meyer curves than the patients treated with TT. In particular, at the last follow-up visit, the prevalence rates for event-free cases after TT and LT were respectively 95% and 87.5% in the low-risk group versus 89% and 50% in the intermediate-risk group. The risk of persistence/relapse for the intermediate-risk patients treated with LT was higher than previously reported,24 although scanty and controversial data exist on this topic, mainly because the ATA risk classification is not included in most series.1 Discrepancies also could be due to the different proportions of micro- and macrocarcinomas included in the different cohorts. In this context, we found that by dividing our cases into micro- and macrocarcinomas, only the latter were significantly more frequently event-free when treated with TT. However, microcarcinomas did not have a significantly different final outcome. Additional treatments were needed only for the patients treated with LT, especially if belonging to the intermediate-risk category.

Most of the studies available on this topic indicate a lack of differences in outcome between the two surgical procedures, whereas in five large series, including the largest series published to date (4813 patients), TT was associated with a better outcome.3‐7 The differences could be due to several factors including sample size, diagnostic tools (particularly the sensitivity of Tg measurement), the criteria used to define disease persistence, and possibly ethnicity. In this context, the two largest Italian studies found TT associated with a better outcome than LT.3,5

Surprisingly, no published data on this topic from other European Countries have been published, likely due to the limited indication for LT as a therapeutic option for suspicious cytology or thyroid cancer. Indeed, in Europe, TT is a preferred procedure due to either the higher complexity required for outcome evaluation of patients treated with LT or the relatively huge availability of high-volume endocrine surgeons.

A Tg cutoff of 30 μg/L for patients not treated with levothyroxine after LT has been proposed recently for the definition of persistent or cured disease, but we could not validate this in our series. Indeed, although lower levels are expected in our patients, all receiving thyroxine treatment, the Tg levels were below 10 μg/L in four of six patients with structural disease, indicating that the 30-μg/L threshold would have led to a misdiagnosis in the majority of cases.

On the other hand, the trend of Tg, TgAb, or both was always consistent with US and radiologic findings, allowing correct identification of persistence or confirmation of an event-free outcome for the 40 patients treated with LT. Different data were obtained by Park et al.,28 who in a cohort of low-risk patients treated with LT found a progressive increase in Tg/TSH, without differences between cured and recurrent patients, indicating a limited value of Tg trend evaluation for predicting disease recurrence. In contrast to our study, none of the patients in the Korean study was treated with levothyroxine after LT, and a possible effect on Tg reliability can be hypothesized to explain the different results obtained.

The major limitation of the current study was the retrospective nature, as for all the studies published to date on this topic. Although data from a prospective study would certainly be highly significant, our study had the advantage of including only patients followed up in a single tertiary care center, with the result that the clinical record was thus extremely accurate, and information was very rarely missed. Moreover, our study is one of only a few studies and the only study from a European center that included micro- and macro-PTCs as well as a large number of FTCs, all of which had undergone a careful risk stratification and an accurate assessment of the response to therapy in terms of dynamic risk stratification and final disease outcome. Other potential limitations were some statistically significant differences in the clinicopathologic features of the low-risk cases.

Nevertheless, after consideration of these limitations in detail, it can be argued that they had no impact on the final results. In particular, node dissection was more frequently performed for the patients treated with TT, but no differences were found between the two groups at histology because all the removed nodes were negative. Moreover, the tumors treated with LT were larger and more frequently FTC at histology. This could have had an impact on the greater need of additional therapies for the low-risk patients treated with LT versus those treated with TT, but the final outcomes did not differ, supporting the indication for a more conservative treatment in this group of patients, as suggested by the ATA 2015 guidelines,1 even in cases with a potentially more aggressive presentation.

Finally, the selection between LT and TT possibly could have been biased by a US presurgical staging. Nevertheless, it appears that this drawback did not occur because the tumors treated with LT were significantly larger (low-risk category) or comparable in size (intermediate-risk category), and no differences were found between the tumors treated with LT and those treated with TT in terms of nodal metastases at histology.

In conclusion, our data indicate that LT can be proposed for low-risk microcarcinomas, although in a minority of cases, additional treatments are needed. Because the evaluation of an event-free outcome for patients treated with LT relies on the evidence of a Tg and/or TgAb trend stably declining, and a negative neck ultrasound, a longer follow-up period is required for a reliable definition of the response to surgery than for patients treated with TT. On the other hand, 50% of the intermediate-risk microcarcinomas treated by LT in this study needed additional treatments after initial surgery. Thus, despite the higher frequency of post-surgical complications, TT should be favored for intermediate-risk tumors.

Due to the high variability of Tg secretion, from either normal or metastatic tissue, even if normalized for TSH levels, a Tg threshold to define outcome in LT cases likely will be difficult to establish. Nevertheless, the Tg trend has been found to correlate strongly with ultrasonographic findings and with the clinical status of the patient, and thus should always be included in the follow-up evaluation of patients treated by either TT without radioiodine residue ablation or by LT.