Risk stratification for symptom-based hypocalcemia after total thyroidectomy using POD1 total calcium: a temporal validation study

Risk stratification for symptom-based hypocalcemia after total thyroidectomy using POD1 total calcium: a temporal validation study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Risk stratification for symptom-based hypocalcemia after total thyroidectomy using POD1 total calcium: a temporal validation study Yating Chen, Jiaxin Pan, Yanyang Zhao, Gang Miao This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9298761/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Background Symptom-based hypocalcemia after total thyroidectomy may occur after discharge and is not always fully reflected by early postoperative biochemical testing. Early risk stratification may help guide discharge supplementation and follow-up intensity. Methods This single-center retrospective cohort study included consecutive patients with papillary thyroid carcinoma who underwent total thyroidectomy between May 2021 and November 2025. The primary outcome was a symptom-based hypocalcemia event within 3 months after surgery, defined by telephone follow-up as paresthesia, tremor, or muscle cramps relieved by oral calcium. Patients operated on before 2025 were assigned to the development cohort, and those operated on in 2025 to the temporal validation cohort. A multivariable logistic regression model was developed using routinely available perioperative variables, including sex, age, operative duration, surgical extent, parathyroid autotransplantation number, body mass index category, postoperative day 1 (POD1) total calcium, and POD1 phosphate. Model performance was assessed by temporal validation and 10-fold cross-validation, and a supplementary same-center cross-surgeon analysis was performed. Results Among 411 patients, outcome data were available for 402 (97.8%), and 132 patients (32.8%) developed symptom-based hypocalcemia. The development cohort included 329 evaluable patients, and the validation cohort included 73. POD1 total calcium was the strongest predictor in both the expanded and main models. In temporal validation, the main model showed modest discrimination (AUC 0.629, 95% CI 0.488–0.770) with a Brier score of 0.254 and an increasing observed event rate across higher predicted-risk groups. Using a development-derived threshold, patients classified as high risk in the validation cohort had a higher event rate than those not classified as high risk (64.3% vs 38.0%). Results were broadly consistent in sensitivity analyses. A POD1 total calcium-only model showed similar discrimination to the multivariable model (AUC 0.625 vs 0.629). Conclusions A POD1 total calcium-anchored model provided modest temporal discrimination and may support pragmatic discharge risk stratification for symptom-based hypocalcemia after total thyroidectomy. A supplementary same-center cross-surgeon analysis showed a similar predictor pattern, but formal external validation in independent multi-center cohorts is still required before clinical implementation. total thyroidectomy papillary thyroid carcinoma hypocalcemia prediction model temporal validation discharge management Figures Figure 1 Figure 2 Figure 3 Background Postoperative hypocalcemia remains the most common early complication after total thyroidectomy [ 1 , 2 ] , leading to paresthesia, muscle cramps, tetany, emergency visits, prolonged hospital stay, and reduced patient satisfaction [ 3 – 6 ] . Although biochemical monitoring is routine during hospitalization, clinically relevant symptoms often develop after discharge and may not coincide with contemporaneous calcium measurements [ 1 , 7 ] . Consequently, surgeons must decide discharge supplementation regimens and follow-up intensity under uncertainty. Existing approaches to postoperative calcium management include routine prophylactic supplementation, laboratory-guided supplementation, and symptom-based algorithms [ 1 , 8 ] . However, a tool that predicts post-discharge symptom-based hypocalcemia events using data available at the time of discharge could support practical risk stratification and guide individualized follow-up [ 9 , 10 ] . This study aimed to develop and temporally validate a prediction model for symptom-based hypocalcemia within 3 months after total thyroidectomy for papillary thyroid carcinoma, using routinely available perioperative variables including postoperative day 1 (POD1) total calcium and phosphate, and to propose an actionable discharge risk stratification strategy. Methods Study design and setting This was a single-center retrospective cohort study of consecutive patients who underwent total thyroidectomy for papillary thyroid carcinoma between May 2021 and November 2025 at Beijing Hospital. Clinical variables were extracted from electronic medical records, perioperative laboratory systems, and operative notes. Post-discharge outcome status was assessed by standardized telephone follow-up at three months. This study was reported in accordance with the TRIPOD statement for the development and validation of multivariable prediction models. Participants Eligible patients were adults with pathologically confirmed papillary thyroid carcinoma who underwent total thyroidectomy with or without lymph node dissection. Surgical extent was classified in two ways: a binary classification distinguished total thyroidectomy with no lateral neck dissection from total thyroidectomy with any lateral neck dissection; a three-level classification categorized extent as A (total thyroidectomy with unilateral central neck dissection or no central neck dissection), B (total thyroidectomy with bilateral central neck dissection), and C (total thyroidectomy including lateral neck dissection). Patients were included if postoperative day 1 (POD1) morning fasting total calcium and phosphate were available. For the primary analysis, patients with missing outcome status were excluded. Supplementary same-center cross-surgeon cohort As a supplementary same-center cross-surgeon analysis, an additional cohort of consecutive patients who underwent total thyroidectomy for papillary thyroid carcinoma performed by another attending surgeon at the same institution was also analyzed, with the same inclusion and exclusion criteria, outcome definition, and predictor variables applied. Because this analysis involved refitting within the additional cohort, it was not considered formal external validation. Outcome The primary outcome was a symptom-based hypocalcemia event occurring within three months after surgery, assessed by telephone follow-up. An event was defined as the presence of typical hypocalcemia-related symptoms (scalp paresthesia, limb paresthesia, tremor, and/or muscle cramps) that were reported to be relieved by oral calcium supplementation. The outcome was recorded as 1 (event) or 0 (no event). Cases in which outcome status could not be ascertained were recorded as missing. Candidate predictors Candidate predictors were prespecified based on clinical relevance and routine availability: sex, age, operative duration (hours), surgical extent, parathyroid autotransplantation (yes/no and number), body mass index (BMI) category (≥24 kg/m² vs <24 kg/m²), preoperative parathyroid hormone (PTH), preoperative calcium and phosphate, and postoperative day 1 (POD1) fasting total calcium and phosphate. Preoperative calcium was albumin-corrected using available preoperative albumin values according to the formula: corrected calcium (mmol/L) = total calcium (mmol/L) + 0.02 × [40 − albumin (g/L)]. Because POD1 albumin was not routinely available, POD1 calcium was analyzed as measured total calcium. For the primary model, variables that could be obtained reliably before discharge were prioritized, and derived variables functionally dependent on predictors (e.g., biochemical hypocalcemia derived from POD1 calcium) were avoided to reduce collinearity. Temporal split and model development A prespecified temporal split was applied to emulate prospective deployment: surgeries before January 1, 2025 comprised the development cohort, and surgeries on or after January 1, 2025 comprised the validation cohort. A multivariable logistic regression model was fitted in the development cohort. The main model included sex, age, operative duration, binary surgical extent, parathyroid autotransplantation number, body mass index (BMI) category, postoperative day 1 (POD1) total calcium, and POD1 phosphate. Complete-case data for predictors were used among patients with non-missing outcomes. To explore the contribution of additional baseline biochemical variables, an expanded model was additionally fitted, including preoperative corrected calcium, preoperative phosphate, and preoperative parathyroid hormone (PTH). Model performance and validation Temporal validation in the 2025 cohort assessed model performance by discrimination, calibration, and overall prediction error. Discrimination was quantified using the area under the receiver operating characteristic curve (AUC) with DeLong 95% confidence intervals. Calibration was evaluated using a grouped calibration plot by quartiles, comparing observed event rates to mean predicted risk. Overall prediction error was assessed via the Brier score. Risk stratification To derive a practical discharge tool, predicted probabilities from the development model were used to define prespecified risk thresholds. The 66.7th percentile predicted risk (0.360) was used as a high-risk threshold. In the validation cohort, event rates and POD1 total calcium levels were compared between non-high-risk and high-risk groups. A three-level stratification using development tertiles was also generated as a supplementary analysis. Sensitivity analyses Sensitivity analyses were performed to assess robustness. The three-level surgical extent classification was substituted for the binary classification. Missing outcomes were handled under extreme assumptions (all missing outcomes imputed as no event vs as event). Discrimination of a POD1 total calcium-only model was compared with that of the main multivariable model. Single-predictor discrimination was evaluated for key perioperative variables. Derived calcium/phosphate variables, spline modeling, and LASSO-based variable selection were explored to assess whether model discrimination could be meaningfully improved. Ten-fold cross-validation was additionally performed as an internal validation analysis for the main model. Statistical analysis All analyses were performed in R (version 4.x). A two-sided p < 0.05 was considered statistically significant for descriptive comparisons; model evaluation focused on effect sizes and performance metrics. Results Cohort characteristics The patient selection process is shown in Figure 1. All surgical cases performed by a single surgeon (Prof. Gang Miao) between May 2021 and November 2025 were screened. Adult patients undergoing total thyroidectomy for thyroid malignancy were considered eligible. Patients were excluded if they had severe renal insufficiency, a relevant medication history affecting calcium metabolism, non-papillary thyroid carcinoma histology, preoperative calcium supplementation, missing preoperative calcium results, or preoperative hypocalcemia. After applying these criteria, 411 patients were included in the final analysis. Outcome status was available for 402 patients (97.8%); nine patients (2.2%) had missing outcome data. The overall symptom-based event rate among patients with non-missing outcomes was 32.8%. Using the prespecified temporal split, 333 patients were assigned to the development cohort and 78 to the validation cohort. Among those with non-missing outcomes, the development cohort included 329 patients with an event rate of 30.7%, whereas the validation cohort included 73 patients with an event rate of 42.5%. Baseline characteristics by outcome are shown in Table 1. Patients experiencing symptom-based events had lower postoperative day 1 (POD1) total calcium ( p < 0.001) and a higher frequency of biochemical hypocalcemia on POD1 ( p < 0.001). Preoperative albumin-corrected calcium was numerically lower in patients with events but did not differ significantly between groups ( p = 0.130). Other characteristics did not differ significantly. Model development The expanded multivariable model was fitted in the development cohort with complete predictors (n = 321). In this model, postoperative day 1 (POD1) total calcium remained the strongest predictor of symptom-based events (adjusted odds ratio [OR] per 1 mmol/L increase 0.011, 95% confidence interval [CI] 0.0014–0.090), whereas preoperative corrected calcium, preoperative phosphate, and preoperative parathyroid hormone (PTH) were not statistically significant. The primary main model was then fitted using routinely available discharge-oriented variables. In this model, POD1 total calcium was again the dominant predictor (adjusted OR per 1 mmol/L increase 0.0079, 95% CI 0.0010–0.062). Other predictors were not statistically significant. Model coefficients are reported in Table 2. Temporal validation and model performance Temporal validation of the main model was performed in the 2025 validation cohort (n = 73 with non-missing outcomes). The model achieved an area under the receiver operating characteristic curve (AUC) of 0.629 (95% confidence interval [CI] 0.488–0.770) (Fig. 2). Calibration was assessed using a grouped calibration plot, with patients divided into quartiles based on predicted risk (Fig. 3). Observed event rates increased with higher predicted risk across the quartiles. The Brier score was 0.254. Risk stratification for discharge management Using the 66.7th percentile predicted-risk threshold from the development cohort (0.360), patients in the validation cohort were stratified into non-high-risk (n = 50) and high-risk (n = 14) groups. The high-risk group had a higher event rate (64.3%) compared with the non-high-risk group (38.0%) and lower mean postoperative day 1 (POD1) total calcium (2.01 vs 2.30 mmol/L) (Table 3). The three-level tertile stratification results are provided in Table S1. Sensitivity analyses Sensitivity analyses were performed to assess the robustness of the main model. Substituting the three-level surgical extent classification for the binary classification yielded similar discrimination in the validation cohort (area under the curve [AUC] 0.624, 95% confidence interval [CI] 0.482–0.766) (Fig. S1; Table S2). Under extreme assumptions for missing outcomes, validation discrimination remained consistent (missing imputed as no event: AUC 0.618, 95% CI 0.477–0.760; missing imputed as event: AUC 0.625, 95% CI 0.491–0.758) (Table S2). A postoperative day 1 (POD1) total calcium-only model showed discrimination nearly identical to that of the main multivariable model in temporal validation (AUC 0.625 vs 0.629; Table S3 and Fig. S2), indicating that POD1 total calcium captured most of the predictive signal in this cohort. Additional exploratory analyses were performed to assess whether discrimination could be improved beyond the prespecified main model. Inclusion of derived perioperative variables, including calcium and phosphate change metrics and calcium–phosphate composite measures, did not improve validation discrimination. Similarly, spline-based modeling of POD1 total calcium and LASSO-based variable selection did not outperform the simpler main model. In LASSO analyses, POD1 total calcium remained the dominant retained predictor, whereas other perioperative variables contributed limited additional signal. Single-predictor analyses further supported this pattern. Among individual perioperative variables, POD1 total calcium demonstrated the highest discrimination in the validation cohort (AUC 0.625), outperforming parathyroid autotransplantation number (AUC 0.548), operative duration (AUC 0.543), and binary surgical extent (AUC 0.531) (Fig. S3). As an additional internal validation analysis, 10‑fold cross‑validation of the main model in the complete‑case cohort yielded a similar discrimination estimate (AUC 0.617, 95% CI 0.555–0.679; Brier score 0.210) (Table S4), supporting the internal consistency of the model while confirming that overall discrimination remained modest. Supplementary same-center cross-surgeon analysis An additional same-center cohort comprising 101 patients treated by other surgeons was analyzed using the same eligibility criteria and endpoint definition. In this supplementary cohort, 40 patients (39.6%) developed symptom-based hypocalcemia. Among individual perioperative variables, operative duration and postoperative day 1 (POD1) total calcium demonstrated the strongest discrimination, with areas under the curve (AUC) of 0.739 (95% CI 0.636-0.843) and 0.722 (95% CI 0.619-0.826), respectively. A multivariable model refitted using the same predictor structure as the main model showed an apparent AUC of 0.825 (95% CI 0.742-0.908) with a Brier score of 0.161. Substituting the three-level surgical extent classification for the binary extent variable yielded a similar apparent AUC of 0.825 (95% CI 0.742-0.907) (Fig. S4). These performance estimates are based on training (apparent) performance without internal validation and should therefore be interpreted as optimistic rather than as evidence of formal external validation. Discussion Principal findings In this single-center retrospective cohort study, we developed and temporally validated a prediction model for 3-month symptom-based hypocalcemia events after total thyroidectomy for papillary thyroid carcinoma. The model demonstrated modest discrimination in the temporal validation cohort (AUC 0.629) and supported pragmatic risk stratification at discharge, identifying a high-risk subgroup with a substantially elevated event rate (64.3% vs 38.0%). Mechanistic interpretation The strong association between postoperative day 1 (POD1) total calcium and subsequent symptom events is biologically plausible [ 11 – 13 ] . Post-thyroidectomy hypocalcemia is most commonly caused by transient parathyroid dysfunction due to gland devascularization and surgical manipulation or injury, resulting in reduced parathyroid hormone (PTH) secretion and a delayed postoperative calcium nadir [ 1 , 10 , 14 , 15 ] . In classical hypoparathyroidism, serum phosphate typically rises because reduced PTH activity decreases renal phosphate excretion [ 16 , 17 ] . Although phosphate did not independently predict symptom events in our multivariable model, its combined assessment with POD1 calcium may still provide mechanistic context. Specifically, lower calcium accompanied by relatively higher phosphate is biologically more consistent with transient postoperative parathyroid insufficiency, whereas alternative biochemical patterns may reflect other contributors to postoperative calcium decline rather than isolated parathyroid dysfunction [ 18 , 19 ] . In classical hypoparathyroid physiology, phosphate tends to rise because reduced PTH activity decreases renal phosphate excretion, whereas hungry bone physiology is more often characterized by concurrent hypocalcemia and hypophosphatemia, frequently with hypomagnesemia and increased alkaline phosphatase reflecting high bone turnover [ 20 – 22 ] . Our dataset did not include magnesium or alkaline phosphatase, which limits direct phenotyping of hungry bone syndrome; nevertheless, the clear separation in POD1 total calcium and the higher frequency of biochemical hypocalcemia among symptomatic patients support parathyroid dysfunction as the dominant pathway in this cohort. A methodological consideration is that postoperative calcium was measured as total calcium rather than ionized calcium [ 23 , 24 ] . While preoperative calcium could be corrected using available albumin values, POD1 albumin was not routinely available, precluding postoperative correction [ 12 , 25 ] . Therefore, the predictive role of POD1 calcium in this study should be interpreted as that of measured total calcium rather than ionized or albumin-corrected calcium. POD1 total calcium-only model versus the main multivariable model A notable finding was the near-identical discrimination of the POD1 total calcium-only model and the main multivariable model in temporal validation. This suggests that POD1 total calcium captured most of the predictive information in this cohort, whereas additional perioperative variables contributed limited incremental value. Consequently, the primary value of the multivariable model may lie less in substantially improving discrimination and more in providing a structured and clinically interpretable framework for discharge risk stratification. Additional exploratory analyses reinforced this conclusion. Derived calcium-phosphate variables, spline-based modeling of POD1 total calcium, and LASSO-guided model reduction did not materially improve discrimination beyond the prespecified main model. In single-predictor comparisons, POD1 total calcium outperformed parathyroid autotransplantation number, operative duration, and binary surgical extent. Moreover, 10-fold cross-validation yielded a discrimination estimate similar to that observed in temporal validation, indicating that the modest AUC was not primarily driven by the temporal split itself but rather reflects the limited predictive information available from routinely collected perioperative variables in this dataset. Taken together, these findings suggest that the current model is likely close to the best performance achievable with routinely available perioperative variables in this dataset. An additional same-center cohort from other surgeons provided supportive evidence that the dominant predictive pattern was not restricted to the original surgeon's case series. In that supplementary cohort, operative duration and POD1 total calcium again emerged as the strongest discriminators, and a refitted model using the same predictor structure demonstrated good apparent discrimination. However, because this analysis was performed within the additional cohort itself, it should be interpreted as a supplementary transportability check across surgeons within the same institution rather than as formal external validation. Clinical implications: discharge stratification and follow-up intensity Despite modest discrimination, the model’s most practical contribution is actionable risk stratification at discharge. A single high-risk threshold derived from the development cohort identified a subgroup with markedly higher symptom event rates. In practice, such patients may benefit from intensified supplementation strategies—including higher-dose calcium and consideration of active vitamin D—early laboratory reassessment (e.g., within 48–72 hours), and closer symptom surveillance. In contrast, patients in the non-high-risk group may be managed with standardized education and routine follow-up under the standard postoperative regimen (levothyroxine 50 µg once daily, calcium carbonate 1500 mg twice daily, and calcitriol 0.25 µg once daily), potentially reducing unnecessary interventions. Given that the model’s probability estimates showed calibration variability in low-risk ranges and the validation cohort was small, the model should be positioned as a pragmatic risk stratifier rather than a definitive probability calculator. Both discrimination and calibration performance should be interpreted cautiously; the modest AUC and variability observed in calibration plots suggest that the model is more suitable for risk stratification than for precise individual probability estimation. In particular, instability in lower predicted-risk ranges reflects the limited validation sample size and should not be overinterpreted. Strengths Key strengths of this study include: (1) a clinically relevant symptom-based endpoint that reflects post-discharge morbidity; (2) the use of routinely available POD1 biochemical data, facilitating implementation; (3) a prespecified temporal validation design that mimics real-world deployment; (4) comprehensive sensitivity analyses demonstrating robustness to alternative surgical extent coding, outcome missingness, and model simplification. Limitations This study has several limitations. First, the outcome relied on patient-reported symptoms assessed by telephone, introducing recall bias and subjective variability; calcium values at symptom onset were not available. Second, the primary temporal validation cohort remained relatively small (n = 73 with complete outcomes), resulting in wide confidence intervals and unstable calibration at low predicted risks. Although we additionally analyzed a same-center cohort from other surgeons, that supplementary analysis was based on refitting within the additional cohort and therefore should not be interpreted as formal external validation. Third, relevant biochemical covariates such as magnesium, alkaline phosphatase, and vitamin D status were unavailable, limiting evaluation of hungry bone physiology and potentially reducing predictive performance. However, the uniform protocol likely reduced between-surgeon and between-patient heterogeneity in treatment intensity. Finally, all cohorts were derived from a single institution, and external validation in independent multi-center settings is still required before clinical implementation. Declarations Ethics approval and consent to participate This retrospective study was approved by the Ethics Committee of Beijing Hospital (No. 2021BJYYEC-044-03). The requirement for individual informed consent was waived because of the retrospective nature of the study and the use of de-identified data. Clinical trial number: not applicable. Consent for publication Not applicable. This manuscript does not contain any individual person’s data in any form. Availability of data and materials The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request. Competing interests The authors declare that they have no competing interests. Funding This work was supported by Beijing Hospital Clinical Research Project 121 Project Task Document (BJ-2020-169) and the Special Project on the Integration of Medicine and Engineering in Beijing Hospital (BJ-2023-093). The funders had no role in the study design, data collection, analysis, interpretation, or writing of the manuscript. Authors’ contributions YC conceived the study, curated the data, performed the formal analysis, and wrote the first draft of the manuscript. JP contributed to data curation and investigation and revised the manuscript. YZ contributed to the study design and methodology, supervised the study, and critically revised the manuscript. GM conceived and supervised the study, contributed to the methodology, administered the project, and critically revised the manuscript. All authors read and approved the final manuscript. Acknowledgements The authors thank the medical and nursing staff of the Department of General Surgery, Beijing Hospital, for their support in patient management and data collection. References ORLOFF L A, WISEMAN S M, BERNET V J, et al. American Thyroid Association Statement on Postoperative Hypoparathyroidism: Diagnosis, Prevention, and Management in Adults [J]. Thyroid, 2018, 28(7): 830-841. CHO J N, PARK W S, MIN S Y. Predictors and risk factors of hypoparathyroidism after total thyroidectomy [J]. Int J Surg, 2016, 34: 47-52. GáLVEZ-PASTOR S, TORREGROSA N M, RíOS A, et al. Prediction of hypocalcemia after total thyroidectomy using indocyanine green angiography of parathyroid glands: A simple quantitative scoring system [J]. Am J Surg, 2019, 218(5): 993-999. MIN H, HAN LEE C. Electromyographic changes in a patient with hypocalcemia after thyroidectomy: A case report [J]. Turk J Phys Med Rehabil, 2022, 68(3): 422-425. YOUNGWIRTH L, BENAVIDEZ J, SIPPEL R, et al. Postoperative parathyroid hormone testing decreases symptomatic hypocalcemia and associated emergency room visits after total thyroidectomy [J]. Surgery, 2010, 148(4): 841-844; discussion 844-846. GRAINGER J, AHMED M, GAMA R, et al. Post-thyroidectomy hypocalcemia: Impact on length of stay [J]. Ear Nose Throat J, 2015, 94(7): 276-281. VAN KINSCHOT C M J, LONČAR I, VAN GINHOVEN T M, et al. A symptom-based algorithm for calcium management after thyroid surgery: a prospective multicenter study [J]. Eur Thyroid J, 2023, 12(6). SINGER M C, BHAKTA D, SEYBT M W, et al. Calcium management after thyroidectomy: a simple and cost-effective method [J]. Otolaryngol Head Neck Surg, 2012, 146(3): 362-365. PALADINO N C, GUéRIN C, GRAZIANI J, et al. Predicting risk factors of postoperative hypocalcemia after total thyroidectomy: is safe discharge without supplementation possible? A large cohort study [J]. Langenbecks Arch Surg, 2021, 406(7): 2425-2431. METERE A, BIANCUCCI A, NATILI A, et al. PTH after Thyroidectomy as a Predictor of Post-Operative Hypocalcemia [J]. Diagnostics (Basel), 2021, 11(9). PATTOU F, COMBEMALE F, FABRE S, et al. Hypocalcemia following thyroid surgery: incidence and prediction of outcome [J]. World J Surg, 1998, 22(7): 718-724. KOŠEC A, GAŠIĆ A, HERGEŠIĆ F, et al. Assessing Symptomatic Hypocalcemia Risk After Total Thyroidectomy: A Prospective Study [J]. Int Arch Otorhinolaryngol, 2024, 28(1): e12-e21. KOŠEC A, HERGEŠIĆ F, MATOVINOVIĆ F, et al. Identifying early postoperative serum parathyroid hormone levels as predictors of hypocalcaemia after total thyroidectomy: A prospective non-randomized study [J]. Am J Otolaryngol, 2020, 41(3): 102416. WANG Y H, BHANDARI A, YANG F, et al. Risk factors for hypocalcemia and hypoparathyroidism following thyroidectomy: a retrospective Chinese population study [J]. Cancer Manag Res, 2017, 9: 627-635. SITGES-SERRA A. Etiology and Diagnosis of Permanent Hypoparathyroidism after Total Thyroidectomy [J]. J Clin Med, 2021, 10(3). WEINMAN E J, LEDERER E D. PTH-mediated inhibition of the renal transport of phosphate [J]. Exp Cell Res, 2012, 318(9): 1027-1032. BRANDI M L, BILEZIKIAN J P, SHOBACK D, et al. Management of Hypoparathyroidism: Summary Statement and Guidelines [J]. The Journal of Clinical Endocrinology & Metabolism, 2016, 101(6): 2273-2283. PASIEKA J L, WENTWORTH K, YEO C T, et al. Etiology and Pathophysiology of Hypoparathyroidism: A Narrative Review [J]. J Bone Miner Res, 2022, 37(12): 2586-2601. LORENTE-POCH L, SANCHO J J, MUñOZ-NOVA J L, et al. Defining the syndromes of parathyroid failure after total thyroidectomy [J]. Gland Surg, 2015, 4(1): 82-90. BRANDI M L, BILEZIKIAN J P, SHOBACK D, et al. Management of Hypoparathyroidism: Summary Statement and Guidelines [J]. J Clin Endocrinol Metab, 2016, 101(6): 2273-2283. FLORAKIS D, KARAKOZIS S, TSELENI-BALAFOUTA S, et al. Lessons learned from the management of Hungry Bone Syndrome following the removal of an Atypical Parathyroid Adenoma [J]. J Musculoskelet Neuronal Interact, 2019, 19(3): 379-384. GUO Z, ZHAO L, XIE Y, et al. Hungry Bone Syndrome Secondary to Subtotal Thyroidectomy in A Patient With Thyrotoxicosis [J]. The American Journal of the Medical Sciences, 2021, 362(3): 314-320. EMMANOUILIDOU A, STAMATI A, AVRAMIDOU E, et al. Predicting Hypocalcemia and Identifying Supplementation Needs After Total Thyroidectomy: The Role of Perioperative PTH Measurements [J]. Biomedicines, 2025, 14(1). ROBERTSON W G, MARSHALL R W. Ionized calcium in body fluids [J]. Crit Rev Clin Lab Sci, 1981, 15(2): 85-125. DESGAGNéS N, KING J A, KLINE G A, et al. Use of Albumin-Adjusted Calcium Measurements in Clinical Practice [J]. JAMA Netw Open, 2025, 8(1): e2455251. Tables Table1. Baseline characteristics by symptom-based hypocalcemia event status Variable No event (n=270) Event (n=132) P value Continuous variables Age, years 44.01 ± 11.60 42.79 ± 10.73 0.311 Operative duration, h 2.05 ± 0.76 2.27 ± 1.61 0.066 Parathyroid autotransplantation number 0.18 ± 0.45 0.25 ± 0.50 0.144 Preoperative corrected calcium, mmol/L 2.27 ± 0.11 2.25 ± 0.08 0.130 Preoperative phosphate, mmol/L 1.17 ± 0.18 1.16 ± 0.17 0.538 Thyroglobulin, ng/mL 43.15 ± 70.45 44.29 ± 67.92 0.879 Albumin, g/L 44.64 ± 5.10 44.30 ± 3.28 0.493 Preoperative PTH, pg/mL 44.36 ± 20.93 48.93 ± 36.48 0.116 Preoperative TSH, uIU/mL 2.05 ± 1.41 2.11 ± 1.34 0.699 Preoperative FT4 1.35 ± 0.53 1.34 ± 0.32 0.837 Preoperative FT3 3.41 ± 0.68 3.27 ± 0.53 0.063 POD1 total calcium, mmol/L 2.23 ± 0.13 2.15 ± 0.15 <0.001 POD1 phosphate, mmol/L 1.42 ± 0.23 1.42 ± 0.23 0.918 Length of stay, days 6.15 ± 1.48 6.29 ± 2.16 0.448 Categorical variables Sex 0.668 Female 183 (67.8%) 93 (70.5%) Male 87 (32.2%) 39 (29.5%) Surgical extent (binary) 0.431 No lateral neck dissection 180 (66.7%) 82 (62.1%) Any lateral neck dissection 90 (33.3%) 50 (37.9%) Surgical extent (3-level) 0.662 A 23 (8.5%) 10 (7.6%) B 157 (58.1%) 72 (54.5%) C 90 (33.3%) 50 (37.9%) Parathyroid autotransplantation 0.122 No 229 (84.8%) 103 (78.0%) Yes 41 (15.2%) 29 (22.0%) BMI category 0.432 BMI <24 100 (37.0%) 55 (41.7%) BMI ≥24 170 (63.0%) 77 (58.3%) Blood type 0.982 A 72 (27.2%) 37 (28.2%) AB 29 (10.9%) 15 (11.5%) B 80 (30.2%) 37 (28.2%) O 84 (31.7%) 42 (32.1%) Biochemical hypocalcemia <0.001 No 232 (85.9%) 81 (61.4%) Yes 38 (14.1%) 51 (38.6%) Data are presented as mean ± standard deviation (SD) for continuous variables and as n (%) for categorical variables. P values were calculated using the independent t-test for continuous variables and the chi-square test or Fisher’s exact test for categorical variables, as appropriate. POD1 = postoperative day 1; PTH = parathyroid hormone; TSH = thyroid-stimulating hormone; FT3 = free triiodothyronine; FT4 = free thyroxine; BMI = body mass index. Table2. Multivariable logistic regression model for symptom-based hypocalcemia events in the development cohort (main model) Variable OR (95% CI) Sex (male vs female) 0.78 (0.43–1.43) Age (per year) 0.99 (0.96–1.01) Operative duration (per hour) 1.23 (0.90–1.70) Surgical extent (with lateral neck dissection vs without) 0.72 (0.36–1.44) Parathyroid autotransplantation number 1.11 (0.63–1.97) BMI category (≥24 kg/m² vs <24 kg/m²) 1.10 (0.64–1.90) POD1 total calcium (per 1 mmol/L increase) 0.0079 (0.0010–0.062) POD1 phosphate (per 1 mmol/L increase) 1.06 (0.31–3.61) Data are presented as odds ratios (OR) with 95% confidence intervals (CI). The model was adjusted for sex, age, operative duration, binary surgical extent, parathyroid autotransplantation number, body mass index (BMI) category, postoperative day 1 (POD1) total calcium, and POD1 phosphate. POD1 = postoperative day 1; CI = confidence interval; OR = odds ratio. Table3. Discharge risk stratification using the 66.7th percentile predicted-risk threshold (0.360) Risk group n Event rate (%) Mean POD1 total calcium (mmol/L) Mean POD1 phosphate (mmol/L) Mean length of stay (days) Non-high risk 50 38 2.3 1.49 6.12 High risk 14 64.3 2.01 1.49 6.64 Data are presented as n, event rate (%), and mean values. POD1 = postoperative day 1. Additional Declarations No competing interests reported. Supplementary Files SupplementaryInformation.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 08 May, 2026 Reviewers agreed at journal 08 May, 2026 Reviewers invited by journal 22 Apr, 2026 Editor invited by journal 09 Apr, 2026 Editor assigned by journal 08 Apr, 2026 Submission checks completed at journal 08 Apr, 2026 First submitted to journal 02 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9298761","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":629073075,"identity":"78a24225-3298-4348-b84a-fd06d190a7da","order_by":0,"name":"Yating Chen","email":"","orcid":"","institution":"Beijing Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yating","middleName":"","lastName":"Chen","suffix":""},{"id":629073076,"identity":"67946ce3-489d-40b6-9b40-6129ff433ff0","order_by":1,"name":"Jiaxin Pan","email":"","orcid":"","institution":"Peking Union Medical College","correspondingAuthor":false,"prefix":"","firstName":"Jiaxin","middleName":"","lastName":"Pan","suffix":""},{"id":629073077,"identity":"1a52a82b-9683-4b27-b59b-f56407db609f","order_by":2,"name":"Yanyang Zhao","email":"","orcid":"","institution":"Beijing Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yanyang","middleName":"","lastName":"Zhao","suffix":""},{"id":629073078,"identity":"ae1868a8-4f65-4a27-ba61-004b31037253","order_by":3,"name":"Gang Miao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+UlEQVRIiWNgGAWjYDACZijNz95++MHHBrYEBgYeIrVI9pxJM5xJlBYYMLjhYCDN28BAWItuO/PDh19q7tg13GBIMLbdwZen2957gOHnDtxazA6zGRvLHHuW3Di78cDj3DNsxWZnziUw9p7Bp4XBTFqC7XAys8yBBOPcNrbEbTdyDJgZ2/BpYf8mLfHvcDKbRIKBtCVxWnjMJD+2HbbjAWlhJFJLsTFj3+EECR5gIPeeAWo5c8bgYC8+LeePb3z449the/vjwKj8ueNY4rbjPYYPfuLRAgLMwIhIbICwj4HJA/g1MDAw/mBgsIeyawgpHgWjYBSMghEIAKUZW7AzR+5MAAAAAElFTkSuQmCC","orcid":"","institution":"Beijing Hospital","correspondingAuthor":true,"prefix":"","firstName":"Gang","middleName":"","lastName":"Miao","suffix":""}],"badges":[],"createdAt":"2026-04-02 06:11:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9298761/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9298761/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108407755,"identity":"ce0726c7-712b-45f1-b844-933af7dc7304","added_by":"auto","created_at":"2026-05-04 09:54:06","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":115927,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart of patient selection for the study. A total of 2,457 surgical cases performed between May 2021and November 2025 were screened. After applying predefined inclusion and exclusion criteria - including non-adult age, non-thyroid malignancy, severe renal insufficiency, relevant medication history, non-papillary histology, preoperative calcium supplementation, missing preoperative calcium results, and preoperative hypocalcemia - 411 patients undergoing total thyroidectomy for papillary thyroid carcinoma were included in the final cohort.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9298761/v1/b8f0fb531c0b0470e82bf7cf.png"},{"id":108408072,"identity":"3cf1a84c-5f36-4e66-9ac7-d5b9e5aa39ae","added_by":"auto","created_at":"2026-05-04 09:55:11","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":68821,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eReceiver operating characteristic (ROC) curve for the main model in the validation cohort. \u003c/strong\u003eTemporal validation of the main model was performed in the 2025 validation cohort (n = 73 with non-missing outcomes). The model demonstrated modest discrimination with an area under the curve (AUC) of 0.629 (95% confidence interval [CI] 0.488–0.770).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9298761/v1/a6da0369d3fa47a61786daa2.png"},{"id":108407663,"identity":"da40ff54-43e2-4473-ae45-5d163b2502c0","added_by":"auto","created_at":"2026-05-04 09:53:28","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":52897,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCalibration plot for the main model in the validation cohort. \u003c/strong\u003eCalibration was assessed using a grouped calibration plot, with patients divided into quartiles based on predicted risk. The dashed line indicates perfect calibration. Points represent observed event rates within each quartile, with error bars denoting 95% confidence intervals. Observed event rates increased across risk quartiles. The Brier score was 0.254.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-9298761/v1/0d921b7a78a6a13aec1fc545.png"},{"id":108408559,"identity":"c279fb5e-3dae-438a-920b-9aa25daf6ff7","added_by":"auto","created_at":"2026-05-04 09:55:43","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":534835,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9298761/v1/102d6fb4-71d2-4cb6-98ec-1c2c9d7fc47f.pdf"},{"id":108407749,"identity":"626fbe01-5e86-4a0a-aa2b-9edf53ee6616","added_by":"auto","created_at":"2026-05-04 09:54:05","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":461505,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-9298761/v1/b82cc27b2fb53288972b6413.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Risk stratification for symptom-based hypocalcemia after total thyroidectomy using POD1 total calcium: a temporal validation study","fulltext":[{"header":"Background","content":"\u003cp\u003ePostoperative hypocalcemia remains the most common early complication after total thyroidectomy\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/sup\u003e, leading to paresthesia, muscle cramps, tetany, emergency visits, prolonged hospital stay, and reduced patient satisfaction\u003csup\u003e[\u003cspan additionalcitationids=\"CR4 CR5\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]\u003c/sup\u003e. Although biochemical monitoring is routine during hospitalization, clinically relevant symptoms often develop after discharge and may not coincide with contemporaneous calcium measurements\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/sup\u003e. Consequently, surgeons must decide discharge supplementation regimens and follow-up intensity under uncertainty.\u003c/p\u003e \u003cp\u003eExisting approaches to postoperative calcium management include routine prophylactic supplementation, laboratory-guided supplementation, and symptom-based algorithms\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/sup\u003e. However, a tool that predicts post-discharge symptom-based hypocalcemia events using data available at the time of discharge could support practical risk stratification and guide individualized follow-up\u003csup\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThis study aimed to develop and temporally validate a prediction model for symptom-based hypocalcemia within 3 months after total thyroidectomy for papillary thyroid carcinoma, using routinely available perioperative variables including postoperative day 1 (POD1) total calcium and phosphate, and to propose an actionable discharge risk stratification strategy.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cem\u003eStudy design and setting\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis was a single-center retrospective cohort study of consecutive patients who underwent total thyroidectomy for papillary thyroid carcinoma between May 2021 and November 2025 at Beijing Hospital. Clinical variables were extracted from electronic medical records, perioperative laboratory systems, and operative notes. Post-discharge outcome status was assessed by standardized telephone follow-up at three months. This study was reported in accordance with the TRIPOD statement for the development and validation of multivariable prediction models.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eParticipants\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eEligible patients were adults with pathologically confirmed papillary thyroid carcinoma who underwent total thyroidectomy with or without lymph node dissection. Surgical extent was classified in two ways: a binary classification distinguished total thyroidectomy with no lateral neck dissection from total thyroidectomy with any lateral neck dissection; a three-level classification categorized extent as A (total thyroidectomy with unilateral central neck dissection or no central neck dissection), B (total thyroidectomy with bilateral central neck dissection), and C (total thyroidectomy including lateral neck dissection). Patients were included if postoperative day 1 (POD1) morning fasting total calcium and phosphate were available. For the primary analysis, patients with missing outcome status were excluded.\u003c/p\u003e\n\n\u003cp\u003eSupplementary same-center cross-surgeon cohort\u003c/p\u003e\n\u003cp\u003eAs a supplementary same-center cross-surgeon analysis, an additional cohort of consecutive patients who underwent total thyroidectomy for papillary thyroid carcinoma performed by another attending surgeon at the same institution was also analyzed, with the same inclusion and exclusion criteria, outcome definition, and predictor variables applied. Because this analysis involved refitting within the additional cohort, it was not considered formal external validation.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eOutcome\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe primary outcome was a symptom-based hypocalcemia event occurring within three months after surgery, assessed by telephone follow-up. An event was defined as the presence of typical hypocalcemia-related symptoms (scalp paresthesia, limb paresthesia, tremor, and/or muscle cramps) that were reported to be relieved by oral calcium supplementation. The outcome was recorded as 1 (event) or 0 (no event). Cases in which outcome status could not be ascertained were recorded as missing.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eCandidate predictors\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eCandidate predictors were prespecified based on clinical relevance and routine availability: sex, age, operative duration (hours), surgical extent, parathyroid autotransplantation (yes/no and number), body mass index (BMI) category (\u0026ge;24 kg/m\u0026sup2; vs \u0026lt;24 kg/m\u0026sup2;), preoperative parathyroid hormone (PTH), preoperative calcium and phosphate, and postoperative day 1 (POD1) fasting total calcium and phosphate. Preoperative calcium was albumin-corrected using available preoperative albumin values according to the formula: corrected calcium (mmol/L) = total calcium (mmol/L) + 0.02 \u0026times; [40 \u0026minus; albumin (g/L)]. Because POD1 albumin was not routinely available, POD1 calcium was analyzed as measured total calcium. For the primary model, variables that could be obtained reliably before discharge were prioritized, and derived variables functionally dependent on predictors (e.g., biochemical hypocalcemia derived from POD1 calcium) were avoided to reduce collinearity.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eTemporal split and model development\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eA prespecified temporal split was applied to emulate prospective deployment: surgeries before January 1, 2025 comprised the development cohort, and surgeries on or after January 1, 2025 comprised the validation cohort. A multivariable logistic regression model was fitted in the development cohort. The main model included sex, age, operative duration, binary surgical extent, parathyroid autotransplantation number, body mass index (BMI) category, postoperative day 1 (POD1) total calcium, and POD1 phosphate. Complete-case data for predictors were used among patients with non-missing outcomes.\u003c/p\u003e\n\u003cp\u003eTo explore the contribution of additional baseline biochemical variables, an expanded model was additionally fitted, including preoperative corrected calcium, preoperative phosphate, and preoperative parathyroid hormone (PTH).\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eModel performance and validation\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTemporal validation in the 2025 cohort assessed model performance by discrimination, calibration, and overall prediction error. Discrimination was quantified using the area under the receiver operating characteristic curve (AUC) with DeLong 95% confidence intervals. Calibration was evaluated using a grouped calibration plot by quartiles, comparing observed event rates to mean predicted risk. Overall prediction error was assessed via the Brier score.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eRisk stratification\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo derive a practical discharge tool, predicted probabilities from the development model were used to define prespecified risk thresholds. The 66.7th percentile predicted risk (0.360) was used as a high-risk threshold. In the validation cohort, event rates and POD1 total calcium levels were compared between non-high-risk and high-risk groups. A three-level stratification using development tertiles was also generated as a supplementary analysis.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eSensitivity analyses\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSensitivity analyses were performed to assess robustness. The three-level surgical extent classification was substituted for the binary classification. Missing outcomes were handled under extreme assumptions (all missing outcomes imputed as no event vs as event). Discrimination of a POD1 total calcium-only model was compared with that of the main multivariable model. Single-predictor discrimination was evaluated for key perioperative variables. Derived calcium/phosphate variables, spline modeling, and LASSO-based variable selection were explored to assess whether model discrimination could be meaningfully improved. Ten-fold cross-validation was additionally performed as an internal validation analysis for the main model.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eStatistical analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAll analyses were performed in R (version 4.x). A two-sided \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05 was considered statistically significant for descriptive comparisons; model evaluation focused on effect sizes and performance metrics.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cem\u003eCohort characteristics\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe patient selection process is shown in Figure 1. All surgical cases performed by a single surgeon (Prof. Gang Miao) between May 2021 and November 2025 were screened. Adult patients undergoing total thyroidectomy for thyroid malignancy were considered eligible. Patients were excluded if they had severe renal insufficiency, a relevant medication history affecting calcium metabolism, non-papillary thyroid carcinoma histology, preoperative calcium supplementation, missing preoperative calcium results, or preoperative hypocalcemia. After applying these criteria, 411 patients were included in the final analysis.\u003c/p\u003e\n\u003cp\u003eOutcome status was available for 402 patients (97.8%); nine patients (2.2%) had missing outcome data. The overall symptom-based event rate among patients with non-missing outcomes was 32.8%.\u003c/p\u003e\n\u003cp\u003eUsing the prespecified temporal split, 333 patients were assigned to the development cohort and 78 to the validation cohort. Among those with non-missing outcomes, the development cohort included 329 patients with an event rate of 30.7%, whereas the validation cohort included 73 patients with an event rate of 42.5%.\u003c/p\u003e\n\u003cp\u003eBaseline characteristics by outcome are shown in Table 1. Patients experiencing symptom-based events had lower postoperative day 1 (POD1) total calcium (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001) and a higher frequency of biochemical hypocalcemia on POD1 (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001). Preoperative albumin-corrected calcium was numerically lower in patients with events but did not differ significantly between groups (\u003cem\u003ep\u003c/em\u003e = 0.130). Other characteristics did not differ significantly.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eModel development\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe expanded multivariable model was fitted in the development cohort with complete predictors (n = 321). In this model, postoperative day 1 (POD1) total calcium remained the strongest predictor of symptom-based events (adjusted odds ratio [OR] per 1 mmol/L increase 0.011, 95% confidence interval [CI] 0.0014\u0026ndash;0.090), whereas preoperative corrected calcium, preoperative phosphate, and preoperative parathyroid hormone (PTH) were not statistically significant.\u003c/p\u003e\n\u003cp\u003eThe primary main model was then fitted using routinely available discharge-oriented variables. In this model, POD1 total calcium was again the dominant predictor (adjusted OR per 1 mmol/L increase 0.0079, 95% CI 0.0010\u0026ndash;0.062). Other predictors were not statistically significant. Model coefficients are reported in Table 2.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eTemporal validation and model performance\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTemporal validation of the main model was performed in the 2025 validation cohort (n = 73 with non-missing outcomes). The model achieved an area under the receiver operating characteristic curve (AUC) of 0.629 (95% confidence interval [CI] 0.488\u0026ndash;0.770) (Fig. 2).\u003c/p\u003e\n\u003cp\u003eCalibration was assessed using a grouped calibration plot, with patients divided into quartiles based on predicted risk (Fig. 3). Observed event rates increased with higher predicted risk across the quartiles. The Brier score was 0.254.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eRisk stratification for discharge management\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eUsing the 66.7th percentile predicted-risk threshold from the development cohort (0.360), patients in the validation cohort were stratified into non-high-risk (n = 50) and high-risk (n = 14) groups. The high-risk group had a higher event rate (64.3%) compared with the non-high-risk group (38.0%) and lower mean postoperative day 1 (POD1) total calcium (2.01 vs 2.30 mmol/L) (Table 3). The three-level tertile stratification results are provided in Table S1.\u003c/p\u003e\n\n\u003cp\u003e\u003cem\u003eSensitivity analyses\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSensitivity analyses were performed to assess the robustness of the main model. Substituting the three-level surgical extent classification for the binary classification yielded similar discrimination in the validation cohort (area under the curve [AUC] 0.624, 95% confidence interval [CI] 0.482\u0026ndash;0.766) (Fig. S1; Table S2). Under extreme assumptions for missing outcomes, validation discrimination remained consistent (missing imputed as no event: AUC 0.618, 95% CI 0.477\u0026ndash;0.760; missing imputed as event: AUC 0.625, 95% CI 0.491\u0026ndash;0.758) (Table S2).\u003c/p\u003e\n\u003cp\u003eA postoperative day 1 (POD1) total calcium-only model showed discrimination nearly identical to that of the main multivariable model in temporal validation (AUC 0.625 vs 0.629; Table S3 and Fig. S2), indicating that POD1 total calcium captured most of the predictive signal in this cohort.\u003c/p\u003e\n\u003cp\u003eAdditional exploratory analyses were performed to assess whether discrimination could be improved beyond the prespecified main model. Inclusion of derived perioperative variables, including calcium and phosphate change metrics and calcium\u0026ndash;phosphate composite measures, did not improve validation discrimination. Similarly, spline-based modeling of POD1 total calcium and LASSO-based variable selection did not outperform the simpler main model. In LASSO analyses, POD1 total calcium remained the dominant retained predictor, whereas other perioperative variables contributed limited additional signal.\u003c/p\u003e\n\u003cp\u003eSingle-predictor analyses further supported this pattern. Among individual perioperative variables, POD1 total calcium demonstrated the highest discrimination in the validation cohort (AUC 0.625), outperforming parathyroid autotransplantation number (AUC 0.548), operative duration (AUC 0.543), and binary surgical extent (AUC 0.531) (Fig. S3).\u003c/p\u003e\n\u003cp\u003eAs an additional internal validation analysis, 10‑fold cross‑validation of the main model in the complete‑case cohort yielded a similar discrimination estimate (AUC 0.617, 95% CI 0.555\u0026ndash;0.679; Brier score 0.210) (Table S4), supporting the internal consistency of the model while confirming that overall discrimination remained modest.\u003c/p\u003e\n\n\u003cp\u003eSupplementary same-center cross-surgeon analysis\u003c/p\u003e\n\u003cp\u003eAn additional same-center cohort comprising 101 patients treated by other surgeons was analyzed using the same eligibility criteria and endpoint definition. In this supplementary cohort, 40 patients (39.6%) developed symptom-based hypocalcemia. Among individual perioperative variables, operative duration and postoperative day 1 (POD1) total calcium demonstrated the strongest discrimination, with areas under the curve (AUC) of 0.739 (95% CI 0.636-0.843) and 0.722 (95% CI 0.619-0.826), respectively. A multivariable model refitted using the same predictor structure as the main model showed an apparent AUC of 0.825 (95% CI 0.742-0.908) with a Brier score of 0.161. Substituting the three-level surgical extent classification for the binary extent variable yielded a similar apparent AUC of 0.825 (95% CI 0.742-0.907) (Fig. S4). These performance estimates are based on training (apparent) performance without internal validation and should therefore be interpreted as optimistic rather than as evidence of formal external validation.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003ePrincipal findings\u003c/h2\u003e \u003cp\u003eIn this single-center retrospective cohort study, we developed and temporally validated a prediction model for 3-month symptom-based hypocalcemia events after total thyroidectomy for papillary thyroid carcinoma. The model demonstrated modest discrimination in the temporal validation cohort (AUC 0.629) and supported pragmatic risk stratification at discharge, identifying a high-risk subgroup with a substantially elevated event rate (64.3% vs 38.0%).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eMechanistic interpretation\u003c/h2\u003e \u003cp\u003eThe strong association between postoperative day 1 (POD1) total calcium and subsequent symptom events is biologically plausible\u003csup\u003e[\u003cspan additionalcitationids=\"CR12\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/sup\u003e. Post-thyroidectomy hypocalcemia is most commonly caused by transient parathyroid dysfunction due to gland devascularization and surgical manipulation or injury, resulting in reduced parathyroid hormone (PTH) secretion and a delayed postoperative calcium nadir\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e. In classical hypoparathyroidism, serum phosphate typically rises because reduced PTH activity decreases renal phosphate excretion\u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e. Although phosphate did not independently predict symptom events in our multivariable model, its combined assessment with POD1 calcium may still provide mechanistic context. Specifically, lower calcium accompanied by relatively higher phosphate is biologically more consistent with transient postoperative parathyroid insufficiency, whereas alternative biochemical patterns may reflect other contributors to postoperative calcium decline rather than isolated parathyroid dysfunction\u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]\u003c/sup\u003e. In classical hypoparathyroid physiology, phosphate tends to rise because reduced PTH activity decreases renal phosphate excretion, whereas hungry bone physiology is more often characterized by concurrent hypocalcemia and hypophosphatemia, frequently with hypomagnesemia and increased alkaline phosphatase reflecting high bone turnover\u003csup\u003e[\u003cspan additionalcitationids=\"CR21\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/sup\u003e. Our dataset did not include magnesium or alkaline phosphatase, which limits direct phenotyping of hungry bone syndrome; nevertheless, the clear separation in POD1 total calcium and the higher frequency of biochemical hypocalcemia among symptomatic patients support parathyroid dysfunction as the dominant pathway in this cohort.\u003c/p\u003e \u003cp\u003eA methodological consideration is that postoperative calcium was measured as total calcium rather than ionized calcium\u003csup\u003e[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/sup\u003e. While preoperative calcium could be corrected using available albumin values, POD1 albumin was not routinely available, precluding postoperative correction\u003csup\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e. Therefore, the predictive role of POD1 calcium in this study should be interpreted as that of measured total calcium rather than ionized or albumin-corrected calcium.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003ePOD1 total calcium-only model versus the main multivariable model\u003c/h2\u003e \u003cp\u003eA notable finding was the near-identical discrimination of the POD1 total calcium-only model and the main multivariable model in temporal validation. This suggests that POD1 total calcium captured most of the predictive information in this cohort, whereas additional perioperative variables contributed limited incremental value. Consequently, the primary value of the multivariable model may lie less in substantially improving discrimination and more in providing a structured and clinically interpretable framework for discharge risk stratification.\u003c/p\u003e \u003cp\u003eAdditional exploratory analyses reinforced this conclusion. Derived calcium-phosphate variables, spline-based modeling of POD1 total calcium, and LASSO-guided model reduction did not materially improve discrimination beyond the prespecified main model. In single-predictor comparisons, POD1 total calcium outperformed parathyroid autotransplantation number, operative duration, and binary surgical extent. Moreover, 10-fold cross-validation yielded a discrimination estimate similar to that observed in temporal validation, indicating that the modest AUC was not primarily driven by the temporal split itself but rather reflects the limited predictive information available from routinely collected perioperative variables in this dataset. Taken together, these findings suggest that the current model is likely close to the best performance achievable with routinely available perioperative variables in this dataset.\u003c/p\u003e \u003cp\u003eAn additional same-center cohort from other surgeons provided supportive evidence that the dominant predictive pattern was not restricted to the original surgeon's case series. In that supplementary cohort, operative duration and POD1 total calcium again emerged as the strongest discriminators, and a refitted model using the same predictor structure demonstrated good apparent discrimination. However, because this analysis was performed within the additional cohort itself, it should be interpreted as a supplementary transportability check across surgeons within the same institution rather than as formal external validation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eClinical implications: discharge stratification and follow-up intensity\u003c/h2\u003e \u003cp\u003eDespite modest discrimination, the model\u0026rsquo;s most practical contribution is actionable risk stratification at discharge. A single high-risk threshold derived from the development cohort identified a subgroup with markedly higher symptom event rates. In practice, such patients may benefit from intensified supplementation strategies\u0026mdash;including higher-dose calcium and consideration of active vitamin D\u0026mdash;early laboratory reassessment (e.g., within 48\u0026ndash;72 hours), and closer symptom surveillance. In contrast, patients in the non-high-risk group may be managed with standardized education and routine follow-up under the standard postoperative regimen (levothyroxine 50 \u0026micro;g once daily, calcium carbonate 1500 mg twice daily, and calcitriol 0.25 \u0026micro;g once daily), potentially reducing unnecessary interventions.\u003c/p\u003e \u003cp\u003eGiven that the model\u0026rsquo;s probability estimates showed calibration variability in low-risk ranges and the validation cohort was small, the model should be positioned as a pragmatic risk stratifier rather than a definitive probability calculator. Both discrimination and calibration performance should be interpreted cautiously; the modest AUC and variability observed in calibration plots suggest that the model is more suitable for risk stratification than for precise individual probability estimation. In particular, instability in lower predicted-risk ranges reflects the limited validation sample size and should not be overinterpreted.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eStrengths\u003c/h2\u003e \u003cp\u003eKey strengths of this study include: (1) a clinically relevant symptom-based endpoint that reflects post-discharge morbidity; (2) the use of routinely available POD1 biochemical data, facilitating implementation; (3) a prespecified temporal validation design that mimics real-world deployment; (4) comprehensive sensitivity analyses demonstrating robustness to alternative surgical extent coding, outcome missingness, and model simplification.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eLimitations\u003c/h2\u003e \u003cp\u003eThis study has several limitations. First, the outcome relied on patient-reported symptoms assessed by telephone, introducing recall bias and subjective variability; calcium values at symptom onset were not available. Second, the primary temporal validation cohort remained relatively small (n\u0026thinsp;=\u0026thinsp;73 with complete outcomes), resulting in wide confidence intervals and unstable calibration at low predicted risks. Although we additionally analyzed a same-center cohort from other surgeons, that supplementary analysis was based on refitting within the additional cohort and therefore should not be interpreted as formal external validation. Third, relevant biochemical covariates such as magnesium, alkaline phosphatase, and vitamin D status were unavailable, limiting evaluation of hungry bone physiology and potentially reducing predictive performance. However, the uniform protocol likely reduced between-surgeon and between-patient heterogeneity in treatment intensity. Finally, all cohorts were derived from a single institution, and external validation in independent multi-center settings is still required before clinical implementation.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cem\u003eEthics approval and consent to participate\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis retrospective study was approved by the Ethics Committee of Beijing Hospital (No. 2021BJYYEC-044-03). The requirement for individual informed consent was waived because of the retrospective nature of the study and the use of de-identified data.\u003c/p\u003e\n\u003cp\u003eClinical trial number: not applicable.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eConsent for publication\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable. This manuscript does not contain any individual person\u0026rsquo;s data in any form.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAvailability of data and materials\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCompeting interests\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eFunding\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by Beijing Hospital Clinical Research Project 121 Project Task Document (BJ-2020-169) and the Special Project on the Integration of Medicine and Engineering in Beijing Hospital (BJ-2023-093). The funders had no role in the study design, data collection, analysis, interpretation, or writing of the manuscript.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAuthors\u0026rsquo; contributions\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eYC conceived the study, curated the data, performed the formal analysis, and wrote the first draft of the manuscript. JP contributed to data curation and investigation and revised the manuscript. YZ contributed to the study design and methodology, supervised the study, and critically revised the manuscript. GM conceived and supervised the study, contributed to the methodology, administered the project, and critically revised the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAcknowledgements\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe authors thank the medical and nursing staff of the Department of General Surgery, Beijing Hospital, for their support in patient management and data collection.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eORLOFF L A, WISEMAN S M, BERNET V J, et al. American Thyroid Association Statement on Postoperative Hypoparathyroidism: Diagnosis, Prevention, and Management in Adults [J]. Thyroid, 2018, 28(7): 830-841.\u003c/li\u003e\n\u003cli\u003eCHO J N, PARK W S, MIN S Y. Predictors and risk factors of hypoparathyroidism after total thyroidectomy [J]. Int J Surg, 2016, 34: 47-52.\u003c/li\u003e\n\u003cli\u003eG\u0026aacute;LVEZ-PASTOR S, TORREGROSA N M, R\u0026iacute;OS A, et al. Prediction of hypocalcemia after total thyroidectomy using indocyanine green angiography of parathyroid glands: A simple quantitative scoring system [J]. Am J Surg, 2019, 218(5): 993-999.\u003c/li\u003e\n\u003cli\u003eMIN H, HAN LEE C. Electromyographic changes in a patient with hypocalcemia after thyroidectomy: A case report [J]. Turk J Phys Med Rehabil, 2022, 68(3): 422-425.\u003c/li\u003e\n\u003cli\u003eYOUNGWIRTH L, BENAVIDEZ J, SIPPEL R, et al. Postoperative parathyroid hormone testing decreases symptomatic hypocalcemia and associated emergency room visits after total thyroidectomy [J]. Surgery, 2010, 148(4): 841-844; discussion 844-846.\u003c/li\u003e\n\u003cli\u003eGRAINGER J, AHMED M, GAMA R, et al. Post-thyroidectomy hypocalcemia: Impact on length of stay [J]. Ear Nose Throat J, 2015, 94(7): 276-281.\u003c/li\u003e\n\u003cli\u003eVAN KINSCHOT C M J, LONČAR I, VAN GINHOVEN T M, et al. A symptom-based algorithm for calcium management after thyroid surgery: a prospective multicenter study [J]. Eur Thyroid J, 2023, 12(6).\u003c/li\u003e\n\u003cli\u003eSINGER M C, BHAKTA D, SEYBT M W, et al. Calcium management after thyroidectomy: a simple and cost-effective method [J]. Otolaryngol Head Neck Surg, 2012, 146(3): 362-365.\u003c/li\u003e\n\u003cli\u003ePALADINO N C, GU\u0026eacute;RIN C, GRAZIANI J, et al. Predicting risk factors of postoperative hypocalcemia after total thyroidectomy: is safe discharge without supplementation possible? A large cohort study [J]. Langenbecks Arch Surg, 2021, 406(7): 2425-2431.\u003c/li\u003e\n\u003cli\u003eMETERE A, BIANCUCCI A, NATILI A, et al. PTH after Thyroidectomy as a Predictor of Post-Operative Hypocalcemia [J]. Diagnostics (Basel), 2021, 11(9).\u003c/li\u003e\n\u003cli\u003ePATTOU F, COMBEMALE F, FABRE S, et al. Hypocalcemia following thyroid surgery: incidence and prediction of outcome [J]. World J Surg, 1998, 22(7): 718-724.\u003c/li\u003e\n\u003cli\u003eKO\u0026Scaron;EC A, GA\u0026Scaron;IĆ A, HERGE\u0026Scaron;IĆ F, et al. Assessing Symptomatic Hypocalcemia Risk After Total Thyroidectomy: A Prospective Study [J]. Int Arch Otorhinolaryngol, 2024, 28(1): e12-e21.\u003c/li\u003e\n\u003cli\u003eKO\u0026Scaron;EC A, HERGE\u0026Scaron;IĆ F, MATOVINOVIĆ F, et al. Identifying early postoperative serum parathyroid hormone levels as predictors of hypocalcaemia after total thyroidectomy: A prospective non-randomized study [J]. Am J Otolaryngol, 2020, 41(3): 102416.\u003c/li\u003e\n\u003cli\u003eWANG Y H, BHANDARI A, YANG F, et al. Risk factors for hypocalcemia and hypoparathyroidism following thyroidectomy: a retrospective Chinese population study [J]. Cancer Manag Res, 2017, 9: 627-635.\u003c/li\u003e\n\u003cli\u003eSITGES-SERRA A. Etiology and Diagnosis of Permanent Hypoparathyroidism after Total Thyroidectomy [J]. J Clin Med, 2021, 10(3).\u003c/li\u003e\n\u003cli\u003eWEINMAN E J, LEDERER E D. PTH-mediated inhibition of the renal transport of phosphate [J]. Exp Cell Res, 2012, 318(9): 1027-1032.\u003c/li\u003e\n\u003cli\u003eBRANDI M L, BILEZIKIAN J P, SHOBACK D, et al. Management of Hypoparathyroidism: Summary Statement and Guidelines [J]. The Journal of Clinical Endocrinology \u0026amp; Metabolism, 2016, 101(6): 2273-2283.\u003c/li\u003e\n\u003cli\u003ePASIEKA J L, WENTWORTH K, YEO C T, et al. Etiology and Pathophysiology of Hypoparathyroidism: A Narrative Review [J]. J Bone Miner Res, 2022, 37(12): 2586-2601.\u003c/li\u003e\n\u003cli\u003eLORENTE-POCH L, SANCHO J J, MU\u0026ntilde;OZ-NOVA J L, et al. Defining the syndromes of parathyroid failure after total thyroidectomy [J]. Gland Surg, 2015, 4(1): 82-90.\u003c/li\u003e\n\u003cli\u003eBRANDI M L, BILEZIKIAN J P, SHOBACK D, et al. Management of Hypoparathyroidism: Summary Statement and Guidelines [J]. J Clin Endocrinol Metab, 2016, 101(6): 2273-2283.\u003c/li\u003e\n\u003cli\u003eFLORAKIS D, KARAKOZIS S, TSELENI-BALAFOUTA S, et al. Lessons learned from the management of Hungry Bone Syndrome following the removal of an Atypical Parathyroid Adenoma [J]. J Musculoskelet Neuronal Interact, 2019, 19(3): 379-384.\u003c/li\u003e\n\u003cli\u003eGUO Z, ZHAO L, XIE Y, et al. Hungry Bone Syndrome Secondary to Subtotal Thyroidectomy in A Patient With Thyrotoxicosis [J]. The American Journal of the Medical Sciences, 2021, 362(3): 314-320.\u003c/li\u003e\n\u003cli\u003eEMMANOUILIDOU A, STAMATI A, AVRAMIDOU E, et al. Predicting Hypocalcemia and Identifying Supplementation Needs After Total Thyroidectomy: The Role of Perioperative PTH Measurements [J]. Biomedicines, 2025, 14(1).\u003c/li\u003e\n\u003cli\u003eROBERTSON W G, MARSHALL R W. Ionized calcium in body fluids [J]. Crit Rev Clin Lab Sci, 1981, 15(2): 85-125.\u003c/li\u003e\n\u003cli\u003eDESGAGN\u0026eacute;S N, KING J A, KLINE G A, et al. Use of Albumin-Adjusted Calcium Measurements in Clinical Practice [J]. JAMA Netw Open, 2025, 8(1): e2455251.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable1.\u003c/strong\u003e \u003cstrong\u003eBaseline characteristics by symptom-based hypocalcemia event status\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"\"\u003e\n \u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"665\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariable\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo event\u003cbr\u003e\u0026nbsp;(n=270)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEvent\u003cbr\u003e\u0026nbsp;(n=132)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eP value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eContinuous variables\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003eAge, years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e44.01 \u0026plusmn; 11.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e42.79 \u0026plusmn; 10.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.311\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003eOperative duration, h\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e2.05 \u0026plusmn; 0.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e2.27 \u0026plusmn; 1.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.066\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003eParathyroid autotransplantation number\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e0.18 \u0026plusmn; 0.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e0.25 \u0026plusmn; 0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.144\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003ePreoperative corrected calcium, mmol/L\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e2.27 \u0026plusmn; 0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e2.25 \u0026plusmn; 0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.130\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003ePreoperative phosphate, mmol/L\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e1.17 \u0026plusmn; 0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e1.16 \u0026plusmn; 0.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.538\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003eThyroglobulin, ng/mL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e43.15 \u0026plusmn; 70.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e44.29 \u0026plusmn; 67.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.879\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003eAlbumin, g/L\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e44.64 \u0026plusmn; 5.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e44.30 \u0026plusmn; 3.28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.493\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003ePreoperative PTH, pg/mL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e44.36 \u0026plusmn; 20.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e48.93 \u0026plusmn; 36.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.116\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003ePreoperative TSH, uIU/mL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e2.05 \u0026plusmn; 1.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e2.11 \u0026plusmn; 1.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.699\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003ePreoperative FT4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e1.35 \u0026plusmn; 0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e1.34 \u0026plusmn; 0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.837\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003ePreoperative FT3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e3.41 \u0026plusmn; 0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e3.27 \u0026plusmn; 0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.063\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003ePOD1 total calcium, mmol/L\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e2.23 \u0026plusmn; 0.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e2.15 \u0026plusmn; 0.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003ePOD1 phosphate, mmol/L\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e1.42 \u0026plusmn; 0.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e1.42 \u0026plusmn; 0.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.918\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003eLength of stay, days\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e6.15 \u0026plusmn; 1.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e6.29 \u0026plusmn; 2.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.448\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCategorical variables\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSex\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.668\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e183 (67.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e93 (70.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; Male\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e87 (32.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e39 (29.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSurgical extent (binary)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.431\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; No lateral neck dissection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e180 (66.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e82 (62.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; Any lateral neck dissection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e90 (33.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e50 (37.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSurgical extent (3-level)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.662\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; A\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e23 (8.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e10 (7.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; B\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e157 (58.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e72 (54.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e90 (33.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e50 (37.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eParathyroid autotransplantation\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.122\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e229 (84.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e103 (78.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e41 (15.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e29 (22.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBMI category\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.432\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; BMI \u0026lt;24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e100 (37.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e55 (41.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; BMI \u0026ge;24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e170 (63.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e77 (58.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBlood type\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.982\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; A\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e72 (27.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e37 (28.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; AB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e29 (10.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e15 (11.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; B\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e80 (30.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e37 (28.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; O\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e84 (31.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e42 (32.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBiochemical hypocalcemia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e232 (85.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e81 (61.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 336px;\"\u003e\n \u003cp\u003e\u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e38 (14.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e51 (38.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eData are presented as mean \u0026plusmn; standard deviation (SD) for continuous variables and as n (%) for categorical variables. P values were calculated using the independent t-test for continuous variables and the chi-square test or Fisher\u0026rsquo;s exact test for categorical variables, as appropriate. POD1 = postoperative day 1; PTH = parathyroid hormone; TSH = thyroid-stimulating hormone; FT3 = free triiodothyronine; FT4 = free thyroxine; BMI = body mass index.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable2.\u003c/strong\u003e \u003cstrong\u003eMultivariable logistic regression model for symptom-based hypocalcemia events in the development cohort (main model)\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"595\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 406px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariable\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 189px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOR (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 406px;\"\u003e\n \u003cp\u003eSex (male vs female)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 189px;\"\u003e\n \u003cp\u003e0.78 (0.43\u0026ndash;1.43)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 406px;\"\u003e\n \u003cp\u003eAge (per year)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 189px;\"\u003e\n \u003cp\u003e0.99 (0.96\u0026ndash;1.01)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 406px;\"\u003e\n \u003cp\u003eOperative duration (per hour)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 189px;\"\u003e\n \u003cp\u003e1.23 (0.90\u0026ndash;1.70)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 406px;\"\u003e\n \u003cp\u003eSurgical extent (with lateral neck dissection vs without)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 189px;\"\u003e\n \u003cp\u003e0.72 (0.36\u0026ndash;1.44)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 406px;\"\u003e\n \u003cp\u003eParathyroid autotransplantation number\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 189px;\"\u003e\n \u003cp\u003e1.11 (0.63\u0026ndash;1.97)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 406px;\"\u003e\n \u003cp\u003eBMI category (\u0026ge;24 kg/m\u0026sup2; vs \u0026lt;24 kg/m\u0026sup2;)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 189px;\"\u003e\n \u003cp\u003e1.10 (0.64\u0026ndash;1.90)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 406px;\"\u003e\n \u003cp\u003ePOD1 total calcium (per 1 mmol/L increase)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 189px;\"\u003e\n \u003cp\u003e0.0079 (0.0010\u0026ndash;0.062)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 406px;\"\u003e\n \u003cp\u003ePOD1 phosphate (per 1 mmol/L increase)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 189px;\"\u003e\n \u003cp\u003e1.06 (0.31\u0026ndash;3.61)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eData are presented as odds ratios (OR) with 95% confidence intervals (CI). The model was adjusted for sex, age, operative duration, binary surgical extent, parathyroid autotransplantation number, body mass index (BMI) category, postoperative day 1 (POD1) total calcium, and POD1 phosphate. POD1 = postoperative day 1; CI = confidence interval; OR = odds ratio.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable3.\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eDischarge risk stratification using the 66.7th percentile predicted-risk threshold (0.360)\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd class=\"fr-cell-fixed \"\u003e\n \u003cp\u003e\u003cstrong\u003eRisk group\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003en\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eEvent rate (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eMean POD1 total calcium (mmol/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eMean POD1 phosphate (mmol/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd class=\"fr-cell-handler \"\u003e\n \u003cp\u003e\u003cstrong\u003eMean length of stay (days)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eNon-high risk\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e2.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e6.12\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHigh risk\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e64.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e2.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e6.64\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eData are presented as n, event rate (%), and mean values. POD1 = postoperative day 1.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-surgery","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bsur","sideBox":"Learn more about [BMC Surgery](http://bmcsurg.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bsur/default.aspx","title":"BMC Surgery","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"total thyroidectomy, papillary thyroid carcinoma, hypocalcemia, prediction model, temporal validation, discharge management","lastPublishedDoi":"10.21203/rs.3.rs-9298761/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9298761/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eSymptom-based hypocalcemia after total thyroidectomy may occur after discharge and is not always fully reflected by early postoperative biochemical testing. Early risk stratification may help guide discharge supplementation and follow-up intensity.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eThis single-center retrospective cohort study included consecutive patients with papillary thyroid carcinoma who underwent total thyroidectomy between May 2021 and November 2025. The primary outcome was a symptom-based hypocalcemia event within 3 months after surgery, defined by telephone follow-up as paresthesia, tremor, or muscle cramps relieved by oral calcium. Patients operated on before 2025 were assigned to the development cohort, and those operated on in 2025 to the temporal validation cohort. A multivariable logistic regression model was developed using routinely available perioperative variables, including sex, age, operative duration, surgical extent, parathyroid autotransplantation number, body mass index category, postoperative day 1 (POD1) total calcium, and POD1 phosphate. Model performance was assessed by temporal validation and 10-fold cross-validation, and a supplementary same-center cross-surgeon analysis was performed.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eAmong 411 patients, outcome data were available for 402 (97.8%), and 132 patients (32.8%) developed symptom-based hypocalcemia. The development cohort included 329 evaluable patients, and the validation cohort included 73. POD1 total calcium was the strongest predictor in both the expanded and main models. In temporal validation, the main model showed modest discrimination (AUC 0.629, 95% CI 0.488\u0026ndash;0.770) with a Brier score of 0.254 and an increasing observed event rate across higher predicted-risk groups. Using a development-derived threshold, patients classified as high risk in the validation cohort had a higher event rate than those not classified as high risk (64.3% vs 38.0%). Results were broadly consistent in sensitivity analyses. A POD1 total calcium-only model showed similar discrimination to the multivariable model (AUC 0.625 vs 0.629).\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eA POD1 total calcium-anchored model provided modest temporal discrimination and may support pragmatic discharge risk stratification for symptom-based hypocalcemia after total thyroidectomy. A supplementary same-center cross-surgeon analysis showed a similar predictor pattern, but formal external validation in independent multi-center cohorts is still required before clinical implementation.\u003c/p\u003e","manuscriptTitle":"Risk stratification for symptom-based hypocalcemia after total thyroidectomy using POD1 total calcium: a temporal validation study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-04 09:53:17","doi":"10.21203/rs.3.rs-9298761/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-08T14:21:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"74814884608605317752571591035294395784","date":"2026-05-08T14:03:41+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-22T14:22:32+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-04-09T18:15:05+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-08T04:18:22+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-08T04:17:34+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Surgery","date":"2026-04-02T05:54:48+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-surgery","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bsur","sideBox":"Learn more about [BMC Surgery](http://bmcsurg.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bsur/default.aspx","title":"BMC Surgery","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"a25963ff-d50d-4e5b-bfe4-4d45e2d30ad9","owner":[],"postedDate":"May 4th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-08T14:21:11+00:00","index":110,"fulltext":""},{"type":"reviewerAgreed","content":"74814884608605317752571591035294395784","date":"2026-05-08T14:03:41+00:00","index":109,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-04T09:53:17+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-04 09:53:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9298761","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9298761","identity":"rs-9298761","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}