Modified supraclavicular approach for hemithyroidectomy with central neck dissection in unilateral papillary thyroid microcarcinoma: a feasibility-focused retrospective cohort study Authors: Ziyu Wang, Kangcong Liang, Xidi Wang, Zhiqiang Zhang, Jingyu Xun, Heng Cao, Kefeng Lei Published: 2026 ABSTRACT BackgroundConventional open thyroidectomy provides reliable exposure for hemithyroidectomy and central neck dissection but leaves a visible anterior cervical scar. Remote-access approaches may improve cosmetic outcomes but often require specialized equipment, longer operative time, and greater technical complexity. This study evaluated the feasibility and perioperative outcomes of a modified supraclavicular approach in selected patients with unilateral papillary thyroid microcarcinoma (PTMC).MethodsThis single-center retrospective cohort study included 130 patients with unilateral PTMC who underwent hemithyroidectomy with isthmusectomy and central neck dissection between June 2023 and January 2026. Patients were categorized into three surgical groups: modified supraclavicular approach, infraclavicular endoscopic approach, and conventional midline open approach. Perioperative outcomes, central lymph node yield, postoperative day 1 parathyroid hormone levels, hospitalization costs, complications, incision-related outcomes, and the learning curve were assessed. To address baseline imbalance, the primary comparison between the modified supraclavicular and conventional midline open approaches was analyzed using propensity score overlap weighting based on age, sex, body mass index (BMI), comorbidity status, Hashimoto’s thyroiditis, tumor size, tumor side, and tumor location.ResultsAmong the 130 patients included, 29 underwent the modified supraclavicular approach, 65 underwent conventional midline open thyroidectomy, and 36 underwent the infraclavicular endoscopic approach, which was retained as an institutional contextual cohort. In the primary comparison with the midline open group, no statistically significant difference was detected in operative time or central lymph node yield after adjustment for relevant covariates. No intraoperative conversion occurred in the modified supraclavicular group. No postoperative hematoma requiring reoperation, clinically suspected permanent recurrent laryngeal nerve dysfunction, or permanent hypoparathyroidism was documented. The modified supraclavicular group had lower postoperative pain scores and less frequent neck discomfort, although these subjective outcomes should be interpreted cautiously because assessment was retrospective and unblinded. Cosmetic superiority over conventional open or remote-access thyroidectomy was not assessed using a validated scar scale. Cumulative sum (CUSUM) analysis identified an approximate transition in operative performance at case 14.ConclusionsIn selected patients with unilateral PTMC for whom hemithyroidectomy with ipsilateral central neck dissection had already been planned, the modified supraclavicular approach appeared technically feasible as an alternative cervical access route. In the primary adjusted comparison with conventional midline open thyroidectomy, no statistically significant differences were detected in operative time or central lymph node yield, and no intraoperative conversion occurred. These findings should be considered exploratory and do not establish equivalence, non-inferiority, cosmetic superiority, or long-term oncological effectiveness. Highlight box Key findings • In the overlap-weighted primary comparison, operative time and central lymph node yield did not differ significantly between the modified supraclavicular and conventional midline open approaches; no conversion occurred. • Pain scores and neck discomfort were lower after modified supraclavicular surgery, but these retrospective, unblinded findings are exploratory. • CUSUM suggested a transition at case 14, not a validated proficiency threshold. What is known and what is new? • Open thyroidectomy provides reliable exposure but leaves an anterior cervical scar. Remote-access techniques avoid or relocate the neck scar but may require longer operative routes and specialized resources. • Controlled sternocleidomastoid retraction expanded the supraclavicular corridor, permitting direct exposure of the thyroid bed and ipsilateral central compartment. In selected patients with PTMC, no statistically significant differences were detected in operative time or central lymph node yield compared with midline open surgery. What is the implication, and what should change now? • This approach may be considered for carefully selected patients who already require hemithyroidectomy with ipsilateral central neck dissection and prefer a lower lateral cervical incision. • Prospective multicenter validation with standardized laryngeal, scar, patient-reported, and oncological outcomes is required before wider adoption. Introduction The incidence of papillary thyroid carcinoma (PTC) has increased steadily over recent decades, largely owing to the increased detection of small, low-risk tumors, particularly papillary thyroid microcarcinoma (PTMC) (1,2). Although PTMC is associated with an excellent overall prognosis, central lymph node metastasis remains relatively common in a subset of patients (3,4). Accordingly, for patients with unilateral PTMC requiring surgical intervention, the ideal operative approach should not only minimize surgical morbidity and improve cosmetic outcomes but also ensure adequate exposure for recurrent laryngeal nerve identification and safe central neck dissection (5). For thyroid cancer surgery, the clinical value of a new access route should therefore be judged not only by incision-related outcomes but also by its ability to maintain adequate central compartment exposure, recurrent laryngeal nerve identification, parathyroid preservation, and lymph node dissection adequacy. Conventional open thyroidectomy via a low-collar cervical incision remains the standard approach because it provides excellent surgical exposure, familiar anatomical orientation, and a favorable safety profile (5). However, this approach leaves a visible anterior neck scar that is difficult to hide, potentially impairing cosmetic satisfaction and quality of life, especially among young patients and female individuals (6,7). To overcome these limitations, several minimally invasive and remote-access techniques have been developed, including transoral vestibular, breast or areolar, axillary, infraclavicular endoscopic, and robotic approach (8,9). Although these approaches can effectively eliminate or relocate visible cervical scars, they typically require longer dissection pathways, more extensive flap mobilization, more complex creation of the working space, and specialized equipment, which may prolong operative time, increase healthcare costs, and steepen the learning curve (10). Supraclavicular and lateral cervical small-incision approaches have emerged as practical alternatives that bridge the gap between conventional open thyroidectomy and remote-access techniques. By placing the incision in the lower lateral neck, these approaches improve scar concealment while preserving a short dissection corridor and the familiar anatomic orientation of standard open thyroid surgery. Previous studies have reported favorable perioperative safety and improved cosmetic outcomes with lateral cervical small-incision or supraclavicular oblique-incision thyroidectomy in selected patients with early-stage unilateral thyroid cancer. More recently, the refined supraclavicular approach has further highlighted the potential of this surgical corridor to reduce surgical trauma without compromising oncologic adequacy (11). Nevertheless, in previously described supraclavicular or lateral cervical approaches, the working space is typically established through conventional intermuscular planes adjacent to the sternocleidomastoid or strap muscles (11-13). In a substantial proportion of patients, this surgical corridor may be limited by anatomic variation, increased muscle tension, local tissue adhesions, or the technical demands of central compartment dissection (14,15). Limited surgical exposure may hinder reliable identification of the recurrent laryngeal nerve, jeopardize preservation of the parathyroid glands, and compromise the adequacy of central lymph node dissection. Therefore, improving operative exposure while preserving incision concealability remains an important technical consideration (16). In the present study, we evaluated a modified supraclavicular approach using controlled retraction of the sternocleidomastoid muscle to expand the lateral operative corridor and improve exposure of the thyroid bed, recurrent laryngeal nerve entry zone, and ipsilateral central compartment. The primary objective was to assess the feasibility and perioperative outcomes of this approach in comparison with conventional midline open thyroidectomy in patients with unilateral PTMC undergoing hemithyroidectomy with isthmusectomy and central neck dissection. The infraclavicular endoscopic cohort was retained as institutional contextual information because remote-access surgery serves a distinct cosmetic purpose. We present this article in accordance with the STROBE reporting checklist (available at https://gs.amegroups.com/article/view/10.21037/gs-2026-0388/rc). Methods Study design and patient selection This single-center retrospective cohort study was conducted in the Department of Thyroid and Breast Surgery at The Seventh Affiliated Hospital of Sun Yat-sen University. Consecutive patients who underwent thyroid surgery between June 2023 and January 2026 and had a pathological diagnosis of unilateral PTMC were retrospectively reviewed. PTMC was defined as PTC with a maximum tumor diameter of 1 cm or less. All patients underwent preoperative cervical ultrasonography. Fine-needle aspiration cytology (FNA) was selectively performed for nodules with sonographic features suspicious for malignancy according to institutional practice, rather than routinely for all subcentimeter nodules. No patient had preoperative radiologic evidence of lateral neck lymph node metastasis. Eligible patients were categorized according to the surgical approach used: modified supraclavicular approach, infraclavicular endoscopic approach, or conventional midline open thyroidectomy. All included patients underwent unilateral hemithyroidectomy with isthmusectomy and central neck dissection. During the study period, ipsilateral central neck dissection was routinely performed in surgically treated unilateral PTMC patients at our institution after preoperative evaluation and shared decision-making. This institutional practice was not intended to suggest that prophylactic central neck dissection should be routinely applied to all clinically node-negative PTMC patients, but rather defined the population in which the present surgical-access comparison was performed. The choice of surgical approach was determined through shared decision-making between the surgeon and patient after consideration of tumor characteristics, patient preference regarding incision concealment, economic factors, equipment availability, and surgeon assessment of technical feasibility. Because of the retrospective design, the final surgical approach was not randomized. In response to the distinct clinical purposes of different surgical access routes, the primary comparison of interest was defined as the modified supraclavicular approach versus conventional midline open thyroidectomy. These two approaches both use a direct cervical operative field and leave a cervical incision, allowing assessment of whether the incision can be relocated laterally without compromising operative exposure or perioperative safety. The infraclavicular endoscopic group was retained as an institutional contextual cohort rather than as a comparator for cosmetic or overall superiority over remote-access thyroidectomy. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of The Seventh Affiliated Hospital, Sun Yat-sen University (Approval No. KY-2026-197). The requirement for individual informed consent was waived because of the retrospective design and use of de-identified data. Written informed consent was obtained from the patient for publication of the surgical images. Patients were eligible for inclusion if they met the following criteria: age 18 years or older; pathologically confirmed unilateral PTMC; no preoperative radiologic evidence of lateral neck lymph node metastasis; treatment with hemithyroidectomy, isthmusectomy, and central neck dissection; surgery performed using one of the three predefined approaches; and complete core clinical, operative, pathological, and perioperative records available. Patients were excluded if they had bilateral thyroid carcinoma or extensive multifocal disease requiring total thyroidectomy, concomitant lateral neck dissection, previous neck surgery or reoperative thyroid surgery, other head and neck malignancies, or incomplete core clinical or perioperative data. Intraoperative conversion was not treated as an exclusion criterion. Instead, conversion was predefined as a feasibility endpoint and was defined as failure to complete the initially planned surgical approach with completion of the operation through another surgical access route. The eight excluded reoperative cases were not reoperations for complications after the index procedures. These patients had undergone previous thyroid surgery at outside hospitals and were admitted for contralateral newly developed thyroid disease or completion surgery after thyroid cancer was diagnosed following partial thyroidectomy. They were excluded because prior neck surgery could substantially alter cervical anatomy and confound operative difficulty, safety outcomes, and learning-curve assessment. Potential sources of bias were addressed by applying predefined inclusion and exclusion criteria, standardizing the extent of surgery across the three groups, extracting data from institutional records using consistent definitions, and adjusting multivariable models for clinically relevant covariates. These covariates included age, sex, body mass index (BMI), American Society of Anesthesiologists (ASA) physical status classification, comorbidities, Hashimoto’s thyroiditis, tumor size, tumor laterality, and tumor location within the thyroid lobe, as available. Because this was a retrospective non-randomized study, residual confounding and selection bias could not be completely eliminated. The study size was determined by the number of eligible patients who met the inclusion and exclusion criteria during the study period. No formal sample size calculation was performed because of the retrospective design and the early clinical introduction of the modified supraclavicular approach. A flowchart detailing the patient selection procedure is presented in Figure 1. Patient selection flowchart. Surgical techniques In the modified supraclavicular group, patients were placed in the supine position under general anesthesia, with the neck slightly extended and rotated to the contralateral side. A 5 cm oblique incision was made along a natural skin crease approximately 2 cm above the ipsilateral clavicle. After sequential division of the skin, subcutaneous tissue, and platysma, subplatysmal flaps were elevated superiorly to the thyroid cartilage and inferiorly to the suprasternal notch. The surgical plane was developed between the sternocleidomastoid muscle and the strap muscles. Controlled lateral retraction of the sternocleidomastoid muscle was then performed to expand the lateral working space and improve exposure of the thyroid bed. All procedures in the three study groups were performed by the same experienced thyroid surgery team. Operations were led by the same senior thyroid surgeon, with one of two experienced attending surgeons serving as the first assistant. The extent of thyroid resection, principles of central neck dissection, perioperative analgesia, drainage management, and discharge criteria were standardized across the three surgical approaches. Carbon nanoparticles or mitoxantrone-based tracer was used to assist parathyroid identification according to patient choice and institutional practice. The thyroid lobe was dissected along the true capsular plane. The middle thyroid vein was identified, divided, and ligated. Superior pole vessels were carefully controlled, with deliberate preservation of the external branch of the superior laryngeal nerve and the cricothyroid muscle. Dissection of the inferior and posterolateral thyroid parenchyma was performed close to the capsule to preserve the parathyroid glands and their vascular pedicles. The thyroid lobe was retracted medially to expose the recurrent laryngeal nerve at its laryngeal entry point and along its cervical course. The isthmus was then transected, completing hemithyroidectomy with isthmusectomy. Prophylactic ipsilateral central neck dissection was subsequently performed, including removal of lymphatic and fibrofatty tissue from the prelaryngeal, pretracheal, and ipsilateral paratracheal compartments, corresponding to ipsilateral level VI dissection. The principal technical advantage of this modified technique is controlled retraction of the sternocleidomastoid muscle, which expands the lateral operative corridor and improves access to the thyroid bed, recurrent laryngeal nerve entry zone, and central compartment while preserving a relatively concealed supraclavicular incision. Representative operative steps of the modified supraclavicular approach are shown in Figure 2. Surgical technique of the modified supraclavicular approach. (A) Preoperative incision planning. A 5 cm oblique incision was designed along a natural skin crease, approximately 2 cm superior to the ipsilateral clavicle. (B) Controlled retraction of the sternocleidomastoid muscle was used to expand the lateral operative corridor and expose the thyroid bed; the asterisk indicates the thyroid lobe. (C) Thyroid lobectomy was performed after nanocarbon injection; the asterisk indicates the nanocarbon-stained thyroid lobe, and the arrow indicates the parathyroid gland. (D) Critical anatomic structures were identified and preserved during lateral dissection; the open triangle indicates the recurrent laryngeal nerve. In the infraclavicular endoscopic group, an endoscopic working space was established through the infraclavicular region according to standard institutional practice. Under magnified endoscopic visualization, hemithyroidectomy with isthmusectomy and central neck dissection was completed. In the midline open group, surgery was performed through a standard low transverse cervical collar incision. After elevation of subplatysmal flaps and separation of the strap muscles, hemithyroidectomy with isthmusectomy and central neck dissection was performed under direct vision, with routine identification and preservation of the recurrent laryngeal nerve and parathyroid glands. Perioperative management A closed-suction drain was routinely placed in the thyroid bed after hemithyroidectomy with central neck dissection according to institutional practice during the study period. The drain was generally removed on postoperative day 2–3 when the drainage fluid was clear and the 24-hour drainage volume was <30 mL, without evidence of active bleeding or chyle leakage. Postoperative analgesia was standardized across groups and was primarily based on non-steroidal anti-inflammatory drugs. Opioid analgesics or patient-controlled analgesia were used selectively in patients with inadequate pain control or greater pain sensitivity. Patients were considered eligible for discharge when vital signs were stable, there was no active bleeding, respiratory distress, or clinically significant voice abnormality, serum calcium status was stable without symptomatic hypocalcemia, the incision showed no evidence of infection, the drain had been removed or institutional drainage criteria had been met, and the patient could tolerate oral intake and ambulate independently. Data collection and outcomes Data were collected from electronic medical records, operative and anesthesia records, laboratory databases, pathological reports, and billing records. Baseline variables included age, sex, BMI, ASA physical status classification, comorbidity status, Hashimoto’s thyroiditis, tumor size, tumor side, and tumor location within the thyroid lobe. Tumor size was obtained from the final pathological report; for multifocal tumors, the largest lesion was recorded. Tumor location was categorized as upper, middle, or lower pole. Secondary outcomes included intraoperative blood loss, postoperative drainage volume, length of postoperative hospital stay, postoperative day 1 parathyroid hormone level, total hospitalization cost, central lymph node positivity, postoperative complications, incision-related outcomes, and learning-curve characteristics. Perioperative endpoints included operative time, estimated intraoperative blood loss, postoperative drainage volume, length of postoperative hospital stay, and total hospitalization costs. Operative time was defined as the interval from skin incision to completion of skin closure. Postoperative drainage volume was defined as the cumulative drainage volume recorded while the drain remained in place. Length of postoperative hospital stay was calculated from the day of surgery to discharge. Total hospitalization costs included all expenses incurred during hospitalization. Pathologic lymph node outcomes included the number of central lymph nodes dissected, the number of metastatic central lymph nodes, and central lymph node positivity, defined as the presence of at least one pathologically confirmed metastatic lymph node in the central compartment. Parathyroid hormone levels measured on postoperative day 1 were assessed as an indicator of early preservation of parathyroid function. Safety and patient-reported outcomes Safety and patient-reported outcomes were primarily compared between the modified supraclavicular and conventional midline open groups. Safety outcomes included postoperative cervical hematoma, reoperation for hematoma, clinically documented postoperative hoarseness, biochemically confirmed postoperative hypocalcemia, and inadvertent parathyroidectomy identified on final pathological examination. All patients underwent preoperative videolaryngoscopy. Postoperative laryngoscopy was not routinely performed and was reserved for patients with clinically suspected vocal cord dysfunction. Postoperative pain was assessed using a numeric rating scale at approximately postoperative day 3. Subjective neck discomfort was defined as patient-reported neck tightness, pulling sensation, foreign-body sensation, or discomfort during neck movement. Information regarding neck discomfort and the need for scar concealment was obtained from postoperative outpatient records, telephone interviews, and structured follow-up. No validated scar assessment instrument was used. Learning curve analysis All modified supraclavicular procedures were arranged chronologically according to the date of surgery. Operative time was used as the primary metric for learning-curve assessment. A scatter plot with a smoothed trend line was used to visualize changes in operative time with accumulating surgical experience. In addition, cumulative sum (CUSUM) analysis was performed by cumulatively summing the deviation of each individual operative time from the overall mean operative time. The maximum point of the CUSUM curve was used to identify an approximate transition point. Operative times before and after the transition point were compared using the Mann-Whitney U test. The identified transition point was considered exploratory rather than a validated proficiency threshold. Statistical analysis All statistical analyses were performed using R software, version 4.5.1 (R Foundation for Statistical Computing, Vienna, Austria). Continuous variables were summarized as means with standard deviations or medians with interquartile ranges (IQRs), as appropriate, and categorical variables were summarized as frequencies and percentages. Overall comparisons among the three surgical groups were performed using analysis of variance or the Kruskal-Wallis test for continuous variables and the Pearson Chi-squared test or Fisher exact test for categorical variables, as appropriate. Because the primary clinically relevant comparison was between the modified supraclavicular and conventional midline open approaches, pairwise comparisons between these two groups were additionally performed for the prespecified perioperative and pathological outcomes. To account for multiple testing across nine outcomes, pairwise P values were adjusted using the Holm method. To address baseline imbalance between the modified supraclavicular and conventional midline open groups, propensity score overlap weighting was applied. Propensity scores were estimated using age, sex, BMI, comorbidity status, Hashimoto’s thyroiditis, tumor size, tumor side, and tumor location. Covariate balance before and after weighting was assessed using standardized mean differences, with an absolute standardized mean difference <0.10 considered acceptable. For the overlap-weighted outcome analyses, positive continuous outcomes were evaluated using generalized linear models with a gamma distribution and log-link function. Central lymph node yield was analyzed using a weighted Poisson regression model, and central lymph node positivity was analyzed using weighted logistic regression. Robust standard errors were used, and effect estimates were reported as adjusted ratios or odds ratios with 95% confidence intervals (CIs). Exact 95% CIs were calculated for selected uncommon safety events. CUSUM analysis was used to explore the learning curve of the modified supraclavicular approach, and operative times before and after the estimated transition point were compared using the Mann-Whitney U test. All statistical tests were two-sided, and P<0.05 was considered statistically significant. Given the exploratory retrospective design and limited sample size, the adjusted analyses were not interpreted as demonstrating equivalence or non-inferiority. Results Patient characteristics, operative feasibility, and pathological findings A total of 130 patients with unilateral PTMC were included in the final analysis. Of these, 29 underwent the modified supraclavicular approach, 36 underwent the infraclavicular endoscopic approach, and 65 underwent conventional midline open thyroidectomy. All patients underwent hemithyroidectomy with isthmusectomy and ipsilateral level VI central neck dissection, including removal of lymphatic and fibrofatty tissue from the prelaryngeal, pretracheal, and ipsilateral paratracheal compartments. No intraoperative conversion occurred in either the modified supraclavicular group or the infraclavicular endoscopic group, and all procedures were completed through the initially planned surgical access route. The patient selection process and conversion status are presented in Figure 1. Baseline characteristics are summarized in Table 1. Significant between-group differences were observed in age and the prevalence of comorbidities. Patients in the conventional midline open group were older than those in the modified supraclavicular and infraclavicular endoscopic groups, with a median age of 51.00 (IQR: 38.00–58.00) years, compared with 37.00 (32.00–44.00) years and 36.00 (33.75–40.25) years, respectively (P<0.001). At least one comorbidity was present in 12 patients (41.4%) in the modified supraclavicular group, 27 patients (75.0%) in the infraclavicular endoscopic group, and 27 patients (41.5%) in the conventional midline open group (P=0.003). Table 1Baseline demographic and tumor characteristics by surgical approachVariableModified supraclavicular (n=29)Infraclavicular endoscopic (n=36)Midline open (n=65)P valueAge, years37.00 [32.00–44.00]36.00 [33.75–40.25]51.00 [38.00–58.00]<0.001Sex0.47 Male8 (27.6)7 (19.4)20 (30.8) Female21 (72.4)29 (80.6)45 (69.2)BMI, kg/m²23.72 [22.85–25.39]23.58 [22.37–27.62]25.04 [20.87–27.86]0.99ASA physical status0.71 I2 (6.9)3 (8.3)3 (4.6) II27 (93.1)33 (91.7)62 (95.4)Any comorbidity12 (41.4)27 (75.0)27 (41.5)0.003Hashimoto’s thyroiditis7 (24.1)13 (36.1)14 (21.5)0.27Tumor size, cm0.50 [0.50–0.70]0.60 [0.50–0.90]0.65 [0.50–1.00]0.38Tumor side0.32 Left15 (51.7)12 (33.3)28 (43.1) Right14 (48.3)24 (66.7)37 (56.9)Tumor location0.89 Upper pole7 (24.1)6 (16.7)16 (24.6) Middle pole16 (55.2)20 (55.6)34 (52.3) Lower pole6 (20.7) 10 (27.8)15 (23.1)Values are presented as median [IQR] or n (%). P values were calculated using the Kruskal-Wallis test for continuous variables and the Pearson Chi-squared test for categorical variables; the Fisher-Freeman-Halton exact test was used for ASA physical status because of sparse cells. ASA, American Society of Anesthesiologists; BMI, body mass index; IQR, interquartile range. Baseline demographic and tumor characteristics by surgical approach Values are presented as median [IQR] or n (%). P values were calculated using the Kruskal-Wallis test for continuous variables and the Pearson Chi-squared test for categorical variables; the Fisher-Freeman-Halton exact test was used for ASA physical status because of sparse cells. ASA, American Society of Anesthesiologists; BMI, body mass index; IQR, interquartile range. No statistically significant between-group differences were detected in sex distribution (P=0.47), BMI (P=0.99), ASA physical status classification (P=0.71), Hashimoto’s thyroiditis (P=0.27), tumor size (P=0.38), tumor side (P=0.32), or tumor location within the thyroid lobe (P=0.89). In particular, the distributions of upper-, middle-, and lower-pole tumors were similar across the three surgical groups. Regarding central nodal outcomes, the median central lymph node yield was 7.00 (4.00–11.00) nodes in the modified supraclavicular group, 4.50 (3.00–6.00) nodes in the infraclavicular endoscopic group, and 8.00 (4.00–12.00) nodes in the conventional midline open group, with a statistically significant overall difference among the three groups (P<0.001). The number of metastatic central lymph nodes did not differ significantly among the groups (P=0.06), with a median of 0 metastatic nodes in each group. Central lymph node specimens from all three groups were processed according to the same institutional pathological protocol. Perioperative and pathological outcomes Unadjusted perioperative and pathological outcomes are summarized in Table 2. Significant overall differences among the three groups were observed in operative time (P<0.001), intraoperative blood loss (P=0.006), total hospitalization cost (P<0.001), and central lymph node yield (P<0.001). These overall differences should be interpreted in the context of the distinct clinical purposes of the three surgical approaches, with the infraclavicular endoscopic cohort retained primarily as institutional contextual information. Table 2Unadjusted perioperative and pathological outcomes by surgical approachOutcomeModified supraclavicular (n=29)Infraclavicular endoscopic (n=36)Midline open (n=65)Overall P valueHolm-adjusted P value†Operative time, min90.00 [80.00–100.00]150.50 [125.00–182.50]82.00 [75.00–105.00]<0.001>0.99Intraoperative blood loss, mL20.00 [20.00–30.00]10.00 [10.00–20.00]20.00 [10.00–20.00]0.0060.32Postoperative drainage volume, mL117.00 [85.00–146.00]111.00 [72.25–136.50]110.00 [86.00–156.00]0.58>0.99Length of postoperative hospital stay, days3.00 [3.00–4.00]3.00 [2.00–3.25]3.00 [3.00–4.00]0.06>0.99Postoperative day 1 PTH, pg/mL44.70 [38.50–56.00]50.95 [38.35–64.77]49.10 [35.80–58.80]0.72>0.99Total hospitalization cost, CNY19,785.74 [18,742.10–21,340.60]24,012.82 [22,660.58–25,502.82]21,332.34 [19,810.69–23,845.62]<0.0010.17Central lymph node yield, n7.00 [4.00–11.00]4.50 [3.00–6.00]8.00 [4.00–12.00]<0.001>0.99Metastatic central lymph nodes, n0.00 [0.00–1.00]0.00 [0.00–0.00]0.00 [0.00–1.00]0.060.32Central lymph node positivity14 (48.3)8 (22.2)17 (26.2)0.0470.34Values are presented as median [IQR] or n (%). Overall, three-group comparisons were performed using the Kruskal-Wallis test or Pearson Chi-squared test, as appropriate. The prespecified primary pairwise comparison was between the modified supraclavicular and conventional midline open groups; pairwise P values were obtained using the Mann-Whitney U test or Fisher's exact test, as appropriate, and adjusted across nine outcomes using the Holm method. The infraclavicular endoscopic cohort is presented as institutional contextual information. †, Holm-adjusted P value refers to the modified supraclavicular versus midline open comparison. CNY, Chinese yuan; IQR, interquartile range; PTH, parathyroid hormone. Unadjusted perioperative and pathological outcomes by surgical approach Values are presented as median [IQR] or n (%). Overall, three-group comparisons were performed using the Kruskal-Wallis test or Pearson Chi-squared test, as appropriate. The prespecified primary pairwise comparison was between the modified supraclavicular and conventional midline open groups; pairwise P values were obtained using the Mann-Whitney U test or Fisher's exact test, as appropriate, and adjusted across nine outcomes using the Holm method. The infraclavicular endoscopic cohort is presented as institutional contextual information. †, Holm-adjusted P value refers to the modified supraclavicular versus midline open comparison. CNY, Chinese yuan; IQR, interquartile range; PTH, parathyroid hormone. The infraclavicular endoscopic group had the longest operative time, with a median of 150.50 (125.00–182.50) minutes. Median operative time was 90.00 (80.00–100.00) minutes in the modified supraclavicular group and 82.00 (75.00–105.00) minutes in the conventional midline open group. No statistically significant difference was detected between the modified supraclavicular and midline open groups in the prespecified pairwise comparison after correction for multiple testing. Median intraoperative blood loss was 20.00 (20.00–30.00) mL in the modified supraclavicular group, 10.00 (10.00–20.00) mL in the infraclavicular endoscopic group, and 20.00 (10.00–20.00) mL in the midline open group. Although the unadjusted pairwise comparison between the modified supraclavicular and midline open groups yielded a nominal P value below 0.05, this difference was no longer statistically significant after Holm correction. No significant overall differences were observed in postoperative drainage volume (P=0.58), length of postoperative hospital stay (P=0.06), or postoperative day 1 parathyroid hormone (PTH) level (P=0.72). Total hospitalization cost differed among the three groups (P<0.001); however, the difference between the modified supraclavicular and midline open groups did not remain statistically significant after correction for multiple testing. Therefore, hospitalization cost was interpreted as an exploratory resource-use outcome rather than evidence of an independent economic advantage of the modified approach. The median central lymph node yield was 7.00 (4.00–11.00) nodes in the modified supraclavicular group, 4.50 (3.00–6.00) nodes in the infraclavicular endoscopic group, and 8.00 (4.00–12.00) nodes in the conventional midline open group. Although the overall three-group difference was statistically significant (P<0.001), no statistically significant difference was detected between the modified supraclavicular and midline open groups in the primary pairwise comparison after Holm correction. The median number of metastatic central lymph nodes was 0 in all three groups. Central lymph node positivity was identified in 14 of 29 patients (48.3%) in the modified supraclavicular group, 8 of 36 patients (22.2%) in the infraclavicular endoscopic group, and 17 of 65 patients (26.2%) in the midline open group. Although the overall three-group comparison reached nominal statistical significance, the primary pairwise comparison between the modified supraclavicular and midline open groups was not statistically significant after correction for multiple testing. Primary adjusted analysis To address the marked baseline imbalance between the modified supraclavicular and conventional midline open groups, propensity score overlap weighting was applied for the primary adjusted analysis. Propensity scores were estimated using age, sex, BMI, comorbidity status, Hashimoto’s thyroiditis, tumor size, tumor side, and tumor location. After weighting, the measured baseline covariates were well balanced between the two groups, with all absolute standardized mean differences below 0.10 (Table S1 and Figure S1). In the overlap-weighted analysis, the modified supraclavicular approach was not significantly associated with operative time compared with the conventional midline open approach (adjusted ratio =1.04, 95% CI: 0.90–1.22, P=0.58). No statistically significant associations were detected for intraoperative blood loss (adjusted ratio =1.28, 95% CI: 0.86–1.91, P=0.22), postoperative drainage volume (adjusted ratio =1.16, 95% CI: 0.85–1.58, P=0.35), length of postoperative hospital stay (adjusted ratio =0.96, 95% CI: 0.79–1.16, P=0.68), postoperative day 1 PTH level (adjusted ratio =1.02, 95% CI: 0.76–1.36, P=0.91), or total hospitalization cost (adjusted ratio =0.92, 95% CI: 0.78–1.08, P=0.32). Central lymph node yield was also not significantly associated with surgical approach after weighting (adjusted ratio =0.94, 95% CI: 0.61–1.46, P=0.80). For central lymph node positivity, the modified supraclavicular group showed a numerically higher odds ratio than the conventional midline open group, but the association was not statistically significant (adjusted OR =3.01, 95% CI: 0.64–14.15, P=0.16). The wide CI reflects the limited sample size and number of events and should therefore be interpreted cautiously. These adjusted results are summarized in Figure 3 and Table 3. Given the limited sample size, the estimates should be regarded as exploratory and should not be interpreted as evidence of equivalence or non-inferiority. Propensity score overlap-weighted associations between the modified supraclavicular and conventional midline open approaches. The conventional midline open approach was used as the reference category. Propensity scores were estimated using age, sex, BMI, comorbidity status, Hashimoto’s thyroiditis, tumor size, tumor side, and tumor location. Overlap-weighted gamma-log models were used for positive continuous outcomes, a weighted Poisson model was used for central lymph node yield, and weighted logistic regression was used for central lymph node positivity, with robust standard errors. Error bars indicate 95% CIs. The vertical reference line represents the null value of 1.0. BMI, body mass index; CI, confidence interval; PTH, parathyroid hormone. Table 3Primary adjusted comparison between the modified supraclavicular and conventional midline open approachesOutcomeEffect measureAdjusted estimate (95% CI)P valueOperative timeRatio1.04 (0.90–1.22)0.58Intraoperative blood lossRatio1.28 (0.86–1.91)0.22Postoperative drainage volumeRatio1.16 (0.85–1.58)0.35Length of postoperative hospital stayRatio0.96 (0.79–1.16)0.68Postoperative day 1 PTHRatio1.02 (0.76–1.36)0.91Total hospitalization costRatio0.92 (0.78–1.08)0.32Central lymph node yieldRatio0.94 (0.61–1.46)0.80Central lymph node positivityOdds ratio3.01 (0.64–14.15)0.16The midline open approach was the reference. Propensity scores included age, sex, body mass index, comorbidity status, Hashimoto’s thyroiditis, tumor size, tumor side, and tumor location. Overlap weighting was used to address the marked baseline age imbalance and limited common support. Effective sample sizes after weighting were 24.5 and 33.8. Gamma-log models were used for positive continuous outcomes, a weighted Poisson model for nodal yield, and weighted logistic regression for nodal positivity, all with robust standard errors. Estimates are exploratory and do not establish equivalence or non-inferiority. CI, confidence interval; PTH, parathyroid hormone. Primary adjusted comparison between the modified supraclavicular and conventional midline open approaches The midline open approach was the reference. Propensity scores included age, sex, body mass index, comorbidity status, Hashimoto’s thyroiditis, tumor size, tumor side, and tumor location. Overlap weighting was used to address the marked baseline age imbalance and limited common support. Effective sample sizes after weighting were 24.5 and 33.8. Gamma-log models were used for positive continuous outcomes, a weighted Poisson model for nodal yield, and weighted logistic regression for nodal positivity, all with robust standard errors. Estimates are exploratory and do not establish equivalence or non-inferiority. CI, confidence interval; PTH, parathyroid hormone. Learning curve The 29 modified supraclavicular procedures were arranged chronologically according to the date of surgery. Operative time showed substantial variability during the early cases, followed by a numerical decrease and reduced dispersion as surgical experience accumulated. The CUSUM analysis of operative time increased during the initial phase, reached its maximum at case 14, and subsequently declined, identifying case 14 as the approximate transition point between the early learning phase and the later phase (Figure 4). Learning curve of the modified supraclavicular approach. (A) Chronological operative time across 29 consecutive modified supraclavicular procedures with a smoothed trend line. (B) CUSUM analysis of operative time. The CUSUM curve reached its maximum at case 14, which was considered an approximate transition point. The vertical reference line indicates case 14. This transition point should be interpreted as exploratory rather than as a validated proficiency threshold. CUSUM, cumulative sum. Median operative time was 97.50 [82.50–122.50] minutes for cases 1–14 and 89.00 (77.50–95.00) minutes for cases 15–29. Although operative time was numerically lower and less variable in the later phase, the difference between phases did not reach statistical significance (P=0.13). No intraoperative conversion occurred during either phase. For descriptive purposes, the median operative time in the post-transition modified supraclavicular cases was 89.00 (77.50–95.00) minutes, compared with 82.00 (75.00–105.00) minutes in the conventional midline open group, with no statistically significant difference detected (P=0.86). These findings suggest stabilization of operative performance after approximately 14 cases but should not be interpreted as a definitive proficiency threshold because of the limited sample size. Safety and patient-reported outcomes Safety and patient-reported outcomes were assessed primarily between the modified supraclavicular and conventional midline open groups and are summarized in Table 4. All patients underwent preoperative videolaryngoscopy. Postoperative laryngoscopy was not routinely performed and was reserved for patients with clinically suspected vocal cord dysfunction. Table 4Safety and patient-reported outcomes in the modified supraclavicular and conventional midline open groupsOutcomeModified supraclavicular (n=29)Midline open (n=65)P valueSafety outcomes Preoperative videolaryngoscopy performed29 (100.0)65 (100.0)– Postoperative cervical hematoma0 (0.0%; 95% CI: 0.0–11.9%)0 (0.0%; 95% CI: 0.0–5.5%)– Reoperation for postoperative hematoma0 (0.0%; 95% CI: 0.0–11.9%)0 (0.0%; 95% CI: 0.0–5.5%)– Clinically documented postoperative hoarseness2 (6.9%; 95% CI: 0.8–22.8%)5 (7.7%; 95% CI: 2.5–17.0%)>0.99 Biochemically confirmed postoperative hypocalcemia0 (0.0%; 95% CI: 0.0–11.9%)0 (0.0%; 95% CI: 0.0–5.5%)– Inadvertent parathyroidectomy on final pathology0 (0.0%; 95% CI: 0.0–11.9%)0 (0.0%; 95% CI: 0.0–5.5%)– Permanent hypoparathyroidism0 (0.0)0 (0.0)–Patient-reported outcomes Postoperative pain score2.12±0.843.21±1.03<0.001 Neck discomfort2 (6.9)22 (33.8)0.005 Need for scar concealment9 (31.0)30 (46.2)0.18Data are presented as n (%) or mean ± standard deviation unless otherwise indicated. Exact 95% CIs are shown for selected safety outcomes. All patients underwent preoperative videolaryngoscopy; routine postoperative laryngoscopy was not performed, so documented hoarseness is not equivalent to objectively confirmed recurrent laryngeal nerve palsy. Transient paresthesia treated empirically with calcium in the absence of biochemical hypocalcemia was not classified as confirmed hypocalcemia. Pain and neck discomfort were assessed at approximately postoperative day 3 under the same analgesic protocol. No validated scar scale was used. “–” indicates that a P value was not estimable because both groups had zero events. CI, confidence interval; PTH, parathyroid hormone. Safety and patient-reported outcomes in the modified supraclavicular and conventional midline open groups Data are presented as n (%) or mean ± standard deviation unless otherwise indicated. Exact 95% CIs are shown for selected safety outcomes. All patients underwent preoperative videolaryngoscopy; routine postoperative laryngoscopy was not performed, so documented hoarseness is not equivalent to objectively confirmed recurrent laryngeal nerve palsy. Transient paresthesia treated empirically with calcium in the absence of biochemical hypocalcemia was not classified as confirmed hypocalcemia. Pain and neck discomfort were assessed at approximately postoperative day 3 under the same analgesic protocol. No validated scar scale was used. “–” indicates that a P value was not estimable because both groups had zero events. CI, confidence interval; PTH, parathyroid hormone. Clinically documented postoperative hoarseness occurred in 2 of 29 patients (6.9%) in the modified supraclavicular group and 5 of 65 patients (7.7%) in the conventional midline open group (P>0.99). Because routine postoperative laryngoscopy was not performed, these events could not be classified uniformly as laryngoscopically confirmed transient recurrent laryngeal nerve palsy. No clinically suspected permanent recurrent laryngeal nerve dysfunction was documented in the available follow-up records; however, asymptomatic vocal cord dysfunction could not be objectively excluded. No postoperative cervical hematoma, reoperation for hematoma, biochemically confirmed postoperative hypocalcemia, inadvertent parathyroidectomy on final pathological examination, or permanent hypoparathyroidism was documented in either group. The absence of observed major complications should be interpreted cautiously because the study was not powered to exclude clinically important differences in uncommon safety events. The modified supraclavicular group had a lower postoperative pain score than the conventional midline open group (2.12±0.84 vs. 3.21±1.03, P<0.001). Neck discomfort was reported by 2 patients (6.9%) in the modified supraclavicular group and 22 patients (33.8%) in the conventional midline open group (P=0.005). These subjective outcomes were assessed at approximately postoperative day 3 under a standardized institutional analgesic protocol but were collected retrospectively and without blinding. The need for scar concealment was reported by 9 patients (31.0%) in the modified supraclavicular group and 30 patients (46.2%) in the conventional midline open group, with no statistically significant difference detected (P=0.18). Because no validated scar or cosmetic satisfaction scale was used, these findings were not interpreted as evidence of cosmetic superiority. Representative postoperative scar photographs are provided in Figure S1 for illustrative purposes only. These photographs were not used for quantitative or comparative assessment of cosmetic outcomes. Short-term follow-up Patients were routinely scheduled for postoperative assessment at 1 month and every 3 months during the first 2 postoperative years. Follow-up evaluations included thyroid function tests, serum thyroglobulin, thyroglobulin antibody, and cervical ultrasonography. As of July 2026, no structural recurrence or recurrence-related reoperation had been documented among patients with available follow-up records. Because individual follow-up duration and completeness were not uniform, these observations were considered descriptive and were not used to establish long-term oncological equivalence. Discussion In this feasibility-focused retrospective cohort study, the primary clinically relevant comparison was between the modified supraclavicular approach and conventional midline open thyroidectomy, whereas the infraclavicular endoscopic cohort was retained as institutional contextual information. After accounting for baseline imbalance and multiple comparisons, no statistically significant differences were detected between the modified supraclavicular and midline open groups in operative time, central lymph node yield, postoperative drainage volume, length of postoperative hospital stay, postoperative day 1 parathyroid hormone level, or total hospitalization cost. No intraoperative conversion occurred in the modified supraclavicular group. Postoperative pain scores and neck discomfort were lower in the modified supraclavicular group, although these subjective findings should be interpreted cautiously because of the retrospective and unblinded assessment. CUSUM analysis identified an approximate transition in operative performance at case 14. Collectively, these findings support the technical feasibility of relocating the cervical incision to the supraclavicular region in selected patients, but they should not be interpreted as evidence of equivalence, non-inferiority, or superiority over either conventional open or remote-access thyroidectomy. For selected patients with unilateral low-risk differentiated thyroid cancer, thyroid lobectomy represents an established surgical option (5). In the present study, all patients had unilateral PTMC and underwent the same extent of surgery, namely hemithyroidectomy with isthmusectomy and routine central neck dissection. Therefore, the between-group comparisons primarily reflect differences in surgical access rather than differences in the extent of resection. This distinction is clinically relevant because an ideal approach for unilateral PTMC should improve incision-related outcomes while preserving operative efficiency, recurrent laryngeal nerve exposure, parathyroid gland preservation, and the adequacy of central compartment dissection (17). The clinical context of surgical selection should also be considered when interpreting this cohort. Active surveillance is an increasingly accepted management strategy for selected patients with low-risk cT1aN0M0 PTC. The present study was not designed to compare surgery with active surveillance. Rather, all patients included in this cohort had already undergone surgical management after individualized clinical evaluation and shared decision-making. The decision to proceed with surgery was based on the overall clinical context, including tumor characteristics and location, sonographic findings, and patient preference regarding definitive treatment and continued surveillance. Therefore, our findings should be interpreted as addressing the choice of surgical access among patients who had already selected surgery, rather than as supporting immediate surgery for all patients with PTMC. Conventional midline open thyroidectomy remains the most established approach because it provides direct exposure, familiar anatomical orientation, and a reliable safety profile. Remote-access thyroidectomy, in contrast, is primarily selected by patients who wish to avoid a visible cervical scar. The modified supraclavicular approach does not achieve a scarless neck and should therefore not be regarded as a substitute for remote-access thyroidectomy (6,7). Rather, it relocates the cervical incision to a lower lateral position while retaining a short direct operative route and an anatomical orientation similar to conventional open surgery (9,10). Its potential target population may therefore include selected patients who prefer a more laterally located and potentially more concealable cervical incision but do not wish to undergo, or are not suitable for, a remote-access procedure. Supraclavicular and lateral cervical small-incision approaches represent practical intermediate options between conventional cervical incision thyroidectomy and more complex remote-access procedures (18-20). By relocating the incision to the lower lateral neck, these approaches may facilitate scar concealment while preserving a short operative route to the thyroid bed and a familiar anatomical orientation (19,21). Clinical studies of lateral cervical small-incision and supraclavicular oblique-incision thyroidectomy have reported acceptable perioperative safety and favorable cosmetic outcomes in selected patients with early-stage or unilateral thyroid cancer (13,19). The recently described refined supraclavicular approach further frames this access route as a way to reduce surgical trauma while maintaining direct access to the thyroid bed and central compartment (11). The key technical modification of our approach was the use of controlled sternocleidomastoid retraction to create a wider lateral operative corridor, rather than relying solely on natural intermuscular planes. Previously described supraclavicular or lateral approaches commonly access the thyroid bed through intermuscular corridors related to the sternocleidomastoid or strap muscles (14). In some patients, this corridor may be limited by muscle tension, anatomic variation, local tissue adhesion, or the technical requirements of central neck dissection. Controlled retraction of the sternocleidomastoid muscle can enlarge the lateral exposure and provide access to the thyroid bed, recurrent laryngeal nerve entry zone, and central compartment in a manner that more closely resembles the open approach. This wider corridor may facilitate recurrent laryngeal nerve identification, parathyroid preservation, and central compartment dissection while maintaining a lower lateral cervical incision that may be easier to conceal in selected patients (22,23). An important technical question for a lateral cervical access route is whether adequate exposure of the central compartment can be maintained. In the present cohort, central neck dissection included the prelaryngeal, pretracheal, and ipsilateral paratracheal compartments, corresponding to ipsilateral level VI dissection (5,24). Although the overall central lymph node yield differed among the three groups, no statistically significant difference was detected between the modified supraclavicular and conventional midline open groups in the primary pairwise comparison after correction for multiple testing. Importantly, lymph node yield should not be interpreted as evidence of oncological equivalence (25). The median number of metastatic central lymph nodes was zero in all three groups, indicating a generally low metastatic burden. Therefore, the present nodal findings are better interpreted as evidence that central compartment exposure and lymph node retrieval were technically achievable through the modified supraclavicular corridor rather than proof of equivalent long-term oncological efficacy. The routine use of ipsilateral central neck dissection in this cohort also requires contextualization. Prophylactic central neck dissection for small clinically node-negative PTMC is not universally recommended, and practice varies across institutions and regions. In the present study, ipsilateral level VI dissection reflected the institutional surgical strategy during the study period and was performed consistently across all three groups. Accordingly, our findings should not be interpreted as supporting routine prophylactic central neck dissection for all patients with cN0 PTMC. Rather, they are applicable to selected patients in whom hemithyroidectomy with ipsilateral central neck dissection had already been planned. Regarding perioperative safety, no postoperative cervical hematoma requiring reoperation, biochemically confirmed hypocalcemia, or inadvertent parathyroidectomy was documented in the revised cohort. All patients underwent preoperative videolaryngoscopy; however, postoperative laryngoscopy was not routinely performed and was reserved for patients with clinically suspected vocal cord dysfunction. Therefore, clinically documented postoperative hoarseness cannot be considered equivalent to objectively confirmed recurrent laryngeal nerve palsy, and asymptomatic vocal cord dysfunction could not be excluded. Moreover, because only 29 patients underwent the modified supraclavicular approach, the absence of uncommon major complications should not be interpreted as evidence of equivalent safety. The present study was not powered to exclude clinically meaningful differences in rare recurrent laryngeal nerve, parathyroid, or bleeding complications. Closed-suction drainage was routinely used after hemithyroidectomy with central neck dissection at our institution during the study period and was generally removed on postoperative day 2–3 when the drainage fluid was clear and the 24-hour output was less than 30 mL. We acknowledge that routine drainage after uncomplicated thyroid lobectomy is not universally recommended. Therefore, postoperative drainage volume and length of stay in this cohort should be interpreted within the context of this institutional perioperative protocol and may not be directly generalizable to centers adopting a no-drain strategy. Lower postoperative pain scores and less frequent neck discomfort were observed in the modified supraclavicular group than in the conventional midline open group. These findings are clinically interesting but should be considered exploratory. The surgical groups differed substantially in baseline age and comorbidity, and pain and neck discomfort were subjective outcomes assessed retrospectively without blinding (7,14). Although assessments were generally performed at approximately postoperative day 3 under a standardized institutional analgesic protocol, residual confounding, recall bias, and assessor bias cannot be excluded. The lower lateral location of the supraclavicular incision provides a plausible rationale for improved scar concealability; however, cosmetic superiority was not demonstrated. The need for scar concealment did not differ significantly between the modified supraclavicular and conventional open groups, and no validated scar assessment or cosmetic satisfaction instrument was used. Therefore, postoperative photographs, if presented, should be regarded as illustrative rather than comparative evidence (18,26). Although the unadjusted total hospitalization cost was lower in the modified supraclavicular group, this finding should be interpreted cautiously. After the revised primary analysis and correction for multiple testing, the difference between the modified supraclavicular and conventional midline open groups was not statistically significant (8,27,28). In addition, only total hospitalization cost was available, and detailed procedure-specific cost components were not analyzed. Therefore, the present study cannot establish an independent economic advantage of the modified supraclavicular approach (29,30). Learning-curve assessment provided an additional feasibility signal (31,32). CUSUM analysis of operative time identified an approximate transition point at case 14, after which operative time showed lower numerical values and reduced variability. Median operative time was 97.50 (82.50–122.50) minutes for cases 1–14 and 89.00 (77.50–95.00) minutes for cases 15–29; however, this phase-wise difference did not reach statistical significance. Moreover, all modified supraclavicular procedures were performed by the same experienced thyroid surgery team (33,34). Accordingly, case 14 should be regarded as an exploratory transition point in this series rather than a validated proficiency threshold, and the learning process may differ in surgeons with other backgrounds or levels of experience. The modified supraclavicular corridor also has approach-specific limitations. Excessive inferior-lateral dissection may place cutaneous branches of the supraclavicular nerve at risk and may lead to sensory disturbance over the clavicular region. Wide or excessively thin flap elevation may compromise skin-flap perfusion. In addition, the lower lateral trajectory may make exposure of the superior pole, the external branch of the superior laryngeal nerve, or the contralateral prelaryngeal region more demanding in selected patients. Tumor location may therefore influence technical suitability. In the present cohort, the distributions of upper-, middle-, and lower-pole tumors did not differ significantly among groups, but the sample size was insufficient for robust location-specific subgroup analyses. These anatomical considerations support careful patient selection rather than indiscriminate application of the technique. No structural recurrence or recurrence-related reoperation was documented during the available short-term follow-up. However, follow-up duration was heterogeneous and insufficient to evaluate disease-free survival (DFS) or long-term oncological effectiveness. These observations should therefore be regarded as descriptive short-term findings rather than evidence of oncological equivalence. Several limitations should be acknowledged. First, this was a single-center retrospective study with nonrandom selection of surgical approach, resulting in substantial potential for selection bias and residual confounding. Although the revised analysis accounted for measured baseline differences, including age, BMI, comorbidity, Hashimoto’s thyroiditis, tumor size, tumor side, and tumor location, unmeasured confounding cannot be eliminated. Second, only 29 patients underwent the modified supraclavicular approach, and the study was underpowered to exclude clinically meaningful differences in uncommon complications or central lymph node outcomes. Therefore, nonsignificant findings should not be interpreted as evidence of equivalence or non-inferiority. Third, postoperative laryngoscopy was not routinely performed, limiting objective ascertainment of transient or asymptomatic recurrent laryngeal nerve dysfunction. Fourth, pain, neck discomfort, and scar-related outcomes were retrospectively assessed without blinding, and no validated scar or patient-reported outcome instrument was used, introducing potential recall and assessor bias. Fifth, only total hospitalization cost was available, precluding detailed assessment of procedure-specific economic effects. Sixth, all modified supraclavicular procedures were performed by the same experienced thyroid surgery team, and the CUSUM-derived transition point may not be generalizable to other surgeons or institutions. Seventh, routine ipsilateral central neck dissection and routine drain placement reflected institutional practice during the study period and may limit applicability to centers following different management pathways for cN0 PTMC. Finally, oncological follow-up was relatively short and heterogeneous, precluding conclusions regarding long-term recurrence, DFS, or oncological equivalence. In selected patients with unilateral PTMC for whom hemithyroidectomy with ipsilateral central neck dissection has already been planned, the modified supraclavicular approach appears technically feasible as an alternative cervical access route. In the revised primary comparison with conventional midline open thyroidectomy, no statistically significant differences were detected in operative time or central lymph node yield, and no intraoperative conversion occurred in this initial series. However, the present study was not designed or powered to establish equivalence, non-inferiority, cosmetic superiority, or long-term oncological effectiveness. Further prospective multicenter studies incorporating standardized recurrent laryngeal nerve assessment, validated patient-reported and scar outcomes, and longer oncological follow-up are required before broader adoption. Conclusions In selected patients with unilateral PTMC for whom hemithyroidectomy with ipsilateral central neck dissection had already been planned, the modified supraclavicular approach appeared technically feasible, and no intraoperative conversion occurred. In the overlap-weighted comparison with conventional midline open thyroidectomy, no statistically significant differences were detected in operative time or central lymph node yield. These exploratory findings do not establish equivalence, non-inferiority, cosmetic superiority, or long-term oncological effectiveness; prospective multicenter validation is warranted. Supplementary The article’s supplementary files as Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-0388/prf