Parathyroid Hormone Reduction Predicts Transient Hypocalcemia after Total Thyroidectomy: A Single-Center Prospective Study

Objective We performed this study to investigate the risk factors for postoperative hypocalcemia after total thyroidectomy with central lymph node dissection (CLND). Study Design. This was a single-center prospective study based on 176 consecutive patients who underwent total thyroidectomy for papillary thyroid carcinoma. Setting. Patients were recruited between January 2016 and June 2018. Subjects and Methods. Patients who underwent bilateral (n = 155, bilateral group) and ipsilateral CLND (n = 21) after total thyroidectomy were included. The preoperative and postoperative parathyroid hormone (PTH) and calcium levels were detected. The risk factors for transient hypocalcemia were identified using logistic regression analysis and receiver operating characteristic (ROC) curve analysis. Results Fifty-one (28.98%) patients developed transient hypocalcemia, and 2 patients (1.14%) developed permanent hypoparathyroidism. There was no difference in the gender ratio or the morbidity of hypocalcemia between the patients who underwent bilateral and ipsilateral CLND. On postoperative day 1, PTH decrease was a risk factor for transient hypocalcemia in the whole cohort (β = 0.043, OR = 1.044, 95% CI 1.023–1.065, p < 0.001), bilateral group (β = 0.042, OR = 1.043, 95% CI 1.022–1.064, p < 0.001), and female patients (β = 0.049, OR = 1.050, 95% CI 1.026–1.075, p < 0.001). Tumor diameter was a risk factor for transient hypocalcemia in female patients (β = 0.499, OR = 1.647, 95% CI 1.003–2.704, p=0.049). The ROC curve analysis illustrated that 65.58%, 71.00%, and 71.00% PTH level reduction had high accuracy in predicting transient hypocalcemia in the whole cohort, bilateral group, and female patients, respectively (AUC = 0.986, 0.987, and 0.987). Conclusion Asymptomatic female patients with bilateral CLND and a 71.00% PTH level reduction were at a high risk of transient hypocalcemia.


Introduction
Total thyroidectomy or radical thyroidectomy is the main surgical management for papillary thyroid carcinoma (PTC). Transient postoperative hypocalcemia is a common complication following thyroidectomy with a worldwide incidence of 3∼30% [1][2][3]. Female gender is a risk factor for transient hypocalcemia. Approximately 25∼40% of female patients developed temporary postoperative hypocalcemia, which is higher than the ∼10% in male patients [1,3]. Although the recurrence of secondary hypoparathyroidism, especially in chronic kidney disease could be significantly reduced by autotransplantation [4,5], the high incidence of transient hypocalcemia after total thyroidectomy is still a dilemma. e treatment strategy for PTC remains controversial due to the incidence of postoperative hypoparathyroidism and hypocalcemia. Total thyroidectomy plus central lymph node dissection (CLND) potentially reduces the recurrence of PTC compared to total thyroidectomy alone [6,7]. However, several trials demonstrated that the combination of bilateral or ipsilateral CLND could increase the incidence of transient or permanent hypocalcemia and prolong the length of hospital stay [7,8]. In addition, the malignant pathology, age, female gender, and number of parathyroid glands are indicated as risk factors for asymptomatic and transient hypocalcemia [8,9]. e postoperative serum calcium (Ca) level decreased up to five days postthyroidectomy, and the late measurement of Ca levels lacked the sensitivity to predict hypocalcemia. Many surgeons are interested in predicting early hypocalcemia posttotal thyroidectomy and, therefore, use the intraoperative parathyroid hormone (PTH) and postoperative 1-hour or 4-hour PTH level test [2,10,11]. Surgeons reported that the postoperative 1-hour PTH level is reliable for predicting the development of postoperative symptomatic hypocalcemia [2,11,12]. Reddy et al. reported that the PTH level at 20 minutes after total thyroidectomy was reliable for predicting clinical hypocalcemia [12]. Surgeons discovered that the postoperative 1-hour PTH level had equivalent reliability to the postoperative day 1 PTH for predicting the development of symptomatic hypocalcemia [11]. Seo et al. evaluated the postoperative 1-hour PTH and postoperative days 1, 3, 5, and 7 PTH levels and found that the predictive power of the postoperative 1-hour PTH was equivalent to that of postoperative day 1 [13].
Other surgeons suggested that the decrease in PTH levels was more precise than the postoperative PTH [14]. Lecerf et al. suggested the decrease in the PTH level (68.5%) was more precise than the postoperative PTH level (19.4 ng/l) for predicting hypocalcemia after total thyroidectomy, and the patients with a reduction in PTH of less than 68.5% could be discharged on postoperative day 1 [14]. Updated reports on the postoperative PTH levels for predicting hypocalcemia increased pressure for early discharge of patients and freed patients from unnecessary Ca supplements.
We performed this study to investigate the difference in risk factors between patients with dissimilar surgical strategies and in female patients. e association of age, gender, preoperative PTH level, Ca level, and PTH level reduction on postoperative day 1 with the incidence of postoperative hypocalcemia after total thyroidectomy with CLND was identified. We aimed to determine whether these variables were risk factors for hypocalcemia in patients after total thyroidectomy.

Setting and Subjects.
is was a single-center prospective study carried out at the Department of Head and Neck Surgery of Zhejiang Cancer Hospital between January 2016 and June 2018. e ethics protocol was obtained from the Ethics Committee of our hospital. Patients received thyroidectomies for malignant PTC. Bilateral or ipsilateral CLND was carried out for the treatment of PTC with suspicious enlarged lymph nodes on the bilateral or ipsilateral sides. Patients were excluded if they met one of the following exclusion criteria: (1) known hyperparathyroidism; (2) pregnancy and/or breast-feeding; (3) concomitant serious illnesses like chronic renal failure and hepatic insufficiency; (4) immunological diseases like arthritis and systemic lupus erythematosus; (5) diabetes mellitus; and (6) infectious diseases like HBV-hepatitis, human immunodeficiency virus, infectious atypical pneumonia, pulmonary tuberculosis, and syphilis. Written informed consents were obtained from all participants before surgery and blood sample collection.

Grouping Strategy.
Patients were divided into comparative groups according to gender (male/female), calcemia (hypocalcemia/normocalcemia), and surgical strategies. Patients who underwent thyroidectomy with bilateral CLND (bilateral or ipsilateral cervical lymph node dissection) were assigned to the bilateral group. Patients who underwent thyroidectomy with ipsilateral CLND (with or without ipsilateral cervical lymph node dissection) were assigned to the ipsilateral group.

Hypocalcemia and Treatment.
Normocalcemia was defined as serum Ca ≥ 2.0 mmol/l without numbness and convulsion during hospitalization. Supplementation of Ca was not needed for patients with normocalcemia. Transient hypocalcemia was defined as serum Ca <2.00 mmol/l with numbness and/or convulsion on postoperative day 1. Postoperative calcium prophylaxis was required for patients with transient hypocalcemia. Convulsive patients were treated with an intravenous administration of calcium gluconate (1 g/10 ml/day) for two days. Patients with numbness were encouraged to take routine calcium carbonate D3 granules or alfacalcidol soft capsules. Temporary hypoparathyroidism was defined as symptomatic hypocalcemia (occasional numbness and/or convulsion, with <2.0 mmol/l serum Ca) during hospitalization and at the postoperative 3-month follow-up. Asymptomatic subjects with normal Ca levels on follow-up were asked to phase out daily Ca and vitamin D supplements. Asymptomatic or symptomatic patients that had low serum PTH levels (<15.0 mmol/l) at the postoperative 6-month follow-up were diagnosed with permanent hypoparathyroidism. Daily Ca and vitamin D supplements and routine hospital follow-up were required for subjects with hypoparathyroidism.

Biochemical Analysis.
Preoperative blood samples were collected on the morning of the operation to determine the preoperative Ca and PTH levels. e postoperative serum Ca and PTH levels were monitored on the morning of the 1 st (within 24 hours postsurgery), 3 rd , 5 th , and 7 th postoperative days. A low PTH level was defined as a PTH measurement <15 pg/mL within 24 hours of surgery. Patients were discharged on the 3 rd ∼7 th day postsurgery and were requested to return for clinical follow-up on the 1 st , 3 rd , 6 th , and 12 th postoperative months. e serum Ca and PTH levels were measured using a Hitachi 7600 series automatic biochemical analyzer (Hitachi, Tokyo, Japan) and electrochemiluminescence system (Roche Diagnostics GmbH, Mannheim, Germany).

Data Collection.
Biochemical characteristics including the preoperative and postoperative serum PTH and Ca levels were recorded. Demographic characteristics including age and gender; clinical characteristics such as duration of illness, tumor stage (FIGO classification), tumor diameter, node number, and lymphatic metastasis; postoperative complications (numbness and convulsion); and hypocalcemia frequency were recorded.

Statistical Analysis.
Statistical analyses were conducted using SPSS 22.0. Dichotomous variables were described as number and percentage. Differences in dichotomous variables were analyzed using the χ [2] test. Quantitative data with abnormal distribution are described as median and range (Q1∼Q3). Differences in continuous variables were analyzed using the nonparametric Mann-Whitney U test. Logistics regression analysis was performed to identify the independent risk factors for transient hypocalcemia, with the identification of odds ratio (OR) and 95% confidence interval (CI). e area under a receiver characteristic (ROC) curve (AUC) was used to measure the relative predictability of independent factors. e sensitivity and specificity of risk factors were calculated using the Clinical Calculator 1 (http://vassarstats.net/clin1.html) with the identification of 95% CI. A p < 0.05 was considered to be statistically significant.

Demographic, Clinical, and Biochemical Characteristics of Patients Included.
A total of 176 patients who underwent total thyroidectomy for PTC were included (

Postoperative Recovery of Serum PTH and Ca Levels.
e postoperative serum PTH and Ca levels in patients with normocalcemia returned to the preoperative levels within the 1 st month postsurgery (Figures 1(a) and 1(b)). In patients with transient hypocalcemia, the postoperative day 1 serum PTH and Ca levels were significantly reduced (PTH from 51.89 to 10.30 pg/ml (80.15% reduction)) and Ca was decreased from 2.34 to 1.88 mmol/l (19.51%, reduction; p < 0.01; Figures 1(c) and 1(d)). e postoperative day 1 serum PTH level was low during the first 7 days and was at its lowest level on postoperative day 5 (Figure 1(c)). e postoperative 3-month PTH level was lower than the preoperative level. e serum Ca was at its lowest level on postoperative day 1 (Figure 1(d)) and was gradually returned to a normal level within the 1 st month postsurgery. e slight decrease in serum Ca level after 1 month postsurgery may be due to the phasing-out of Ca supplementation. e female gender and bilateral or ipsilateral cervical lymph node dissection did not influence the recovery of postoperative serum PTH and Ca levels.

Comparison between Groups.
Comparison between patients with hypocalcemia (n � 51) and normocalcemia (n � 125) on postoperative day 1 demonstrated that there were significant differences in male ratio (

Discussion
Frequent monitoring of serum Ca and PTH levels after surgery is reliable for predicting postoperative hypocalcemia. Our present study illustrated that PTH level reduction, female gender, and tumor diameter were risk factors for postoperative hypocalcemia in patients who underwent total thyroidectomy with CLND. However, age, gender, and preoperative PTH and Ca levels did not show predictive features for postoperative hypocalcemia.
ere were differences in the thresholds of PTH level reduction for predicting the incidence of transient hypocalcemia in the whole cohort, patients with bilateral CLND, and female patients (65.58%, 71.00%, and 71.00%, respectively).
Most surgeons are interested in predicting early hypocalcemia after thyroidectomy. e early prediction of hypocalcemia may be necessary for the timely treatment and prevention of it and reduce hospital stay [15]. Some surgeons showed that patients with a <80% PTH level reduction could be safely discharged on the day of surgery [16]. Others expected but failed to predict the early discharge at 24 h posttotal thyroidectomy based on the postoperative day 1 PTH and Ca levels or delta PTH and Ca levels without considering the complications [15]. However, the morbidity of postoperative hypocalcemia is the most common cause of prolonged hospital stay after thyroidectomy [16]. e reliability of postoperative 20-min, 1-hour, 2-hour, and 3-hour PTH levels for predicting postoperative hypocalcemia has been confirmed [2,11,12,16]. For instance, Reddy et al. found that the PTH level at 20 minutes after total thyroidectomy was reliable for predicting clinical hypocalcemia [12]. Early prediction of postoperative hypocalcemia is certainly desirable; however, certain surgeons have indicated that the postoperative 1-hour PTH level has equivalent reliability to the postoperative day 1 PTH levels for predicting the development of symptomatic hypocalcemia [11,13]. As an example, Seo et al. indicated that the postoperative 1-hour and 1-day PTH levels had equivalent predictive power regarding the development of hypocalcemia [13]. We found that the serum PTH remained at stable levels within the first five days postsurgery. e low PTH level observed during hospitalization may be due to data missing from patients with normal PTH and Ca levels on postoperative day 1. e number of patients with recorded postoperative Ca levels decreased from 176 on day 1 to 101  on day 3 and 60 on day 7. However, the equivalent levels of postoperative serum PTH and Ca during the first 3 days postsurgery in patients exhibiting hypocalcemia may prove that the postoperative day 1 PTH levels are accurate enough to direct the initiation of Ca supplementation. In our study, 28.98% (51/176) of patients displaying transient hypocalcemia on postoperative day 1 and 2 patients (1.14%) developed permanent hypoparathyroidism at the 1year follow-up. All of these incidences were within the range of the incidence of transient and permanent hypocalcemia as previously reported [1-3, 13, 17, 18]. e differences in hypoparathyroidism incidence were attributed to the various thresholds used by the surgeons. e duration for defining permanent hypoparathyroidism ranges from 3 to 12 months postsurgery, the PTH levels range from 8.0 pg/ml to 15.0 pg/ ml [13,[19][20][21], and serum Ca levels range from 1.9 mmol/L to 2.10 mmol/L [1,22]. For instance, Eismontas et al. defined postoperative hypocalcemia as serum Ca level <2.10 mmol/L [22], while Sands et al. set the threshold at <1.90 mmol/L [1]. e inclusion of symptomatic or asymptomatic hypocalcemia also influences the reported incidence of hypoparathyroidism [1,12]. We defined permanent hypoparathyroidism as symptomatic or asymptomatic in patients with a persistent low level of Ca (<2.0 mmol/l) within 6 months of surgery and transient hypoparathyroidism as symptomatic patients with a serum Ca level <2.0 mmol/l. e unification of the criteria for defining transient or permanent hypoparathyroidism may be beneficial for surgeons and for surgical management.
Our logistics analysis revealed that the PTH level reduction on postoperative day 1 was an independent risk factor for transient hypocalcemia, however not preoperative Ca or PTH levels. e significant findings in our study were that the various PTH level reduction thresholds predicted transient hypocalcemia in different cohorts. e threshold of PTH level reduction in the whole cohort, bilateral group, and female patients was 65.58%, 71.00%, and 71.00%, respectively, with moderate-high sensitivity and specificity. e 65.58% PTH level reduction threshold was equivalent to the 68.5% reported by Lecerf et al. [14]. Lecerf et al. suggested that a 68.5% PTH level reduction in patients was a more precise indicator than the postoperative PTH (19.4 ng/l) in predicting transient hypocalcemia [14]. Consistent with the reports of Seo et al., the mean relative decline (70%) in postoperative 1-hour PTH levels showed comparable reliability with the mean postoperative PTH level (10.42 pg/mL) for predicting postoperative hypocalcemia, with higher sensitivity and specificity [13]. Our present study confirmed high accuracy, sensitivity, and specificity of the PTH level reduction as well. However, the various thresholds of the PTH level reduction in the whole cohort, bilateral group, and female patients (65.58%, 71.00%, and 71.00%, respectively) illustrate that bilateral CLND and female gender are factors that should be considered when predicting transient hypocalcemia in patients who undergo total thyroidectomy for PTC. As documented in selected literature, total thyroidectomy without CLND induced lower rates of postoperative hypocalcemia [13,18,23]. Various surgeons confirmed that bilateral CLND was an independent risk factor for the development of transient hypoparathyroidism [21,23]. Zhao et al. performed a systematic review and meta-analysis of 17 studies that included 4437 patients who underwent total thyroidectomy with and without prophylactic central neck   dissection for PTC [24]. e authors reported that central neck dissection certainly increased the morbidity of temporary and permanent hypocalcemia and the necessity for postoperative radioiodine ablation [24]. Due to the metastasis status of PTCs, all patients included in this study received total thyroidectomy with bilateral or ipsilateral CLND and bilateral or ipsilateral cervical lymph node dissection. Patients who underwent total thyroidectomy with bilateral and ipsilateral CLND did not show statistical differences in the morbidity of postoperative hypocalcemia. ese results were consistent with that from Seo et al. [13]. e clinical and demographic parameters including diabetes, thyrotoxicosis, malignant pathology, age, female gender, and parathyroid gland number are all indicated as risk factors for asymptomatic and transient hypocalcemia [8,9,23]. e probability of age as a risk factor for postoperative hypocalcemia has been controversial until now [8,15,18,25]. e mechanism for the association of age and female gender with postoperative hypoparathyroidism has not been investigated. Several research studies indicated the prevalence of vitamin D deficiency [8], and some specified there was not a significant association between vitamin D and hypocalcemia [26]. e association of tumor size with the incidence of postoperative hypoparathyroidism has been reported [17,27]. e risk factor of tumor diameter in predicting postoperative transient hypocalcemia was confirmed in female patients by univariate and multivariate regression analyses, while that in the whole cohort and bilateral group was confirmed only by univariate regression analyses. ese findings may indicate that additional risk factors for transient hypocalcemia should be considered in female patients receiving total thyroidectomy with bilateral or ipsilateral CLND.

Conclusion
Our present study documented that the threshold of postoperative day 1 PTH reduction was a reliable independent risk factor for predicting transient hypocalcemia in patients who underwent a total thyroidectomy with bilateral and/or ipsilateral CLND. ere was no difference in the morbidity of transient hypocalcemia between patients who underwent bilateral and ipsilateral CLND. However, female patients and patients with bilateral CLND had a greater PTH reduction threshold of 71.00% than the 65.58% observed in the whole cohort. is study confirmed that female gender and tumor diameter were existing independent risk factors for postoperative transient hypocalcemia in total thyroidectomy, while no risk factors were identified in male patients. Female patients with large tumor size and >71.00% PTH reduction on postoperative day 1 posttotal thyroidectomy with CLND had a high risk of transient hypocalcemia.

Data Availability
All data generated or analyzed during this study are included in this published article. e original data are available from the authors upon request.