Benign thyroid nodules treatment using percutaneous laser ablation (PLA) and radiofrequency ablation (RFA)

Sir,

We read with interest the paper by Mauri et al. entitled “Benign thyroid nodules treatment using percutaneous laser ablation (PLA) and radiofrequency ablation (RFA)” published in the International Journal of Hyperthermia [1]. The authors compared two ablation techniques, namely, PLA (n = 59) and RFA (n = 31) (the number of RFA is 59 nodules in your abstract and 31 in the tables), for treating benign thyroid nodules. The results of this retrospective study showed that RFA and PLA are similarly feasible, safe and effective in treating benign thyroid nodules, while treatment duration required for RFA is lower than that for PLA but requires significantly higher energy. Despite its retrospective nature, this study is valuable owing to its originality, as it is the first study to compare RFA and PLA performed at the same institution.

We congratulate the authors for their interesting and excellent results as well as for trying to maintain the treatment balance between the two groups. First, they used standard devices and two techniques namely, the moving-shot technique for RFA and the pull-back technique for PLA. The moving-shot technique is a standard RFA technique recommended by current guidelines [2]. We agree with the authors’ opinions that performing RFA via the free hand moving-shot technique enables more personal tailoring of the treatment on a case-by-case basis, but requires higher technical expertise. Second, these image-guided ablations were performed by interventional radiologists with 5 to 30 years of experience, which could enhance the efficacy of this study. Third, the inclusion of patients was decided by a multidisciplinary team. Four, contrast-enhanced ultrasound (CEUS) performed at the end of the ablation was used to immediately assess the result of the ablation, and additional treatment was decided when CEUS demonstrated zones of residual enhancement [1]. Finally, we were able to fully appreciate their efforts by communicating with the authors through e-mail. Although not mentioned in their manuscript, the authors treated only solid or predominantly solid tumours, and first performed an ethanol ablation for cystic or predominantly cystic nodules. Recent randomised trials [3,4] suggest that ethanol ablation should be a first-line treatment modality for cystic and predominantly cystic thyroid nodules. Furthermore, all the cases were treated via single-session ablation (e-mail communication), therefore, number of treatment was balanced in two groups. It is well-known that solidity and number of treatment sessions are important factors of volume reduction after ablation [5,6]. However, initial nodule volume is also an important factor affecting volume reduction after ablation [5]. Although this study achieved similar volume reduction at 12-month follow-up, the initial volumes of two groups are significantly different (RFA, 32.7 mL vs. LA, 20.3 mL). This is an important limitation of this study.

The authors have also described several limitations of their study. First, the study involved a small number of patients (n = 59 in the PLA group and n = 31 in the RFA group; the number of patients were different in abstract and tables) and retrospective design. Second, this result may not reflect the general practice, as the authors are highly experienced in this field. Finally, the 12-month follow-up period may not accurately reflect the long-term results. Therefore, we concur with the authors’ opinion that a multicenter prospective study and randomised trials are necessary in the future.

Finally, we would like to suggest several important points that should be considered in future prospective randomised trials. First, we would like to address the experience of the operators. Randomization of operators in one centre for both RFA and PLA can be justified when the operators have sufficient experience in handling both modalities. However, there can be significant bias despite the randomisation if the operator has sufficient experience in one modality but not in the other. Therefore Ha et al. [5] performed a Bayesian network meta-analysis, which is an indirect comparison method, for comparing the two modalities. As no centre at that time had accumulated sufficient clinical experience using both ablation techniques, data on direct comparison of the two techniques in a prospective randomised controlled trial were not available. Recently, the Korean Society of Thyroid Radiology has made some suggestions regarding the operators’ experience and learning curve with respect to thyroid RFA. Approximately 50–100 cases of thyroid RFA for benign nodules are required for the efficacy [7,8] and safety [9] evaluation. Second, the relationship between nodule size and cosmetic problems should be considered. Although, the first randomised controlled international collaborative trial conducted in Korean and Italian centres enrolled patients with similar-sized thyroid nodules, cosmetic problems were observed to be significantly higher in Korean patients [2]. Korean patients with medium-sized thyroid nodules show more significant cosmetic problems than European patients, which is explained by the fact that Korean patients have thinner necks. Therefore, such differences among patients from different countries need to be understood. Finally, we would like to address the size of the thyroid nodules. The two treatment modalities did not show any volume difference during the single-session treatment of nodules that were too large or too small. Therefore, medium-sized thyroid nodules may be appropriate candidates for such comparative studies. The first randomised controlled international collaborative trial also used medium-sized (10–20 mL) thyroid nodules to verify the efficacy of single-session ablation [7].

In conclusion, this study is valuable for comparing the application of RFA and LA for treating benign thyroid nodules. Further, several measures have been taken in this retrospective study to maintain a balance between the two modalities in order to overcome possible bias. Therefore, we agree with the authors’ opinions and suggest several points that should be considered in future randomised trials.