Risk Factor Analysis and Prediction of Severe Hypocalcemia after Total Parathyroidectomy without Auto-Transplantation in Patients with Secondary Hyperparathyroidism

Objective Our study aimed to develop and validate a nomogram to predict severe hypocalcemia (SH) before total parathyroidectomy (TPTX) without auto-transplantation in patients with secondary hyperparathyroidism. Methods A total of 299 consecutive patients who underwent TPTX without transplantation for secondary hyperparathyroidism were selected from the General Hospital of Northern Theater Command between January 2013 and December 2021. Of these, patients who underwent surgery between January 2013 and December 2020 formed the training cohort (n = 208) to develop a nomogram, and those who underwent surgery thereafter formed the validation cohort (n = 91) to validate the performance of this nomogram. Univariate and multivariate logistic regression analyses were used to identify the risk factors associated with SH, and then, a nomogram was constructed. Results The incidence of postoperative SH was 27.9% and 35.2% in the training and validation cohorts, respectively. The preoperative factors associated with SH were younger age, lower serum calcium (Ca) level, higher intact parathyroid hormone (iPTH) level, and higher serum alkaline phosphatase (ALP) level. Incorporating these 4 factors, the nomogram achieved good concordance indexes of 0.866 (95%CI, 0.816–0.916) and 0.867 (95% CI, 0.793–0.941) in predicting SH in the training and validation cohorts, respectively, and had well-fitted calibration curves. The positive predictive values of the nomogram were 64.7% (54.1%–78.4%) and 75.0% (58.6%–88.5%), and negative predictive values of the nomogram were 90.0% (82.9%–93.6%) and 86.4% (73.5%–94.0%) for the training and validation cohorts, respectively. Conclusions We developed and validated a nomogram for the prediction of SH in patients who underwent TPTX without auto-transplantation for secondary hyperparathyroidism. Our nomogram may facilitate the identification of high-risk SH in patients after TPTX and optimization of preoperative decision-making.

Surgical parathyroidectomy (PTX) is necessary for those patients with severe and progressive SHPT refractory to medical treatment [12]. Previous studies showed that PTX can relieve symptoms, improve quality of life, and reduce the risk of all-cause and cardiovascular mortality in patients with severe SHPT [13][14][15][16][17]. Postoperative hypocalcemia is the most common complication of surgery, and the incidence was up to 97% in a previous study [18]. Importantly, severe hypocalcemia (SH) can lead to life-threatening sequelae, such as respiratory muscle weakness, laryngeal stridor, seizures, cardiac arrhythmias, congestive heart failure, tetany, and even sudden death [19]. Terefore, an accurate preoperative prediction of SH can help doctors make efective response and avoid the occurrence of adverse events. Te aim of this study was to investigate the risk factors of postoperative SH following total PTX (TPTX), and a nomogram was constructed to predict the development of postoperative SH using preoperative clinical characteristics in patients with SHPT who had undergone TPTX.

Patients.
Between January 2013 and December 2021, data on consecutive patients with SHPT who had undergone TPTX in the General Hospital of Northern Teater Command were obtained. Te study was approved by the Institutional Ethics Committee of the General Hospital of Northern Teater Command. All patients were informed of the risks and procedures of the surgery and signed informed consent.
Te inclusion criteria were as follows: (1) in accordance with the Kidney Disease Outcomes Quality Initiative guidelines, patients with persistently elevated serum intact PTH (iPTH) levels >800 pg/ml, uncontrolled hypercalcemia with hyperphosphatemia, severe clinical symptoms such as bone and joint pain, muscle weakness, or refractory pruritus, or refractory to medical treatment, need to undergo PTX; (2) patients underwent total PTX without auto-transplantation; (3) the surgery is technically successful with the pathological confrmation of at least 4 parathyroid glands, accompanied by an intact parathormone (iPTH) value of <60 pg/mL on postoperative day 1 (POD1). Patients who underwent TPTX with auto-transplantation or subtotal PTX (SPTX), underwent second PTX due to recurrent SHPT following the initial PTX, underwent a failure operation, had a history of liver, biliary, or pancreatic diseases, and had incomplete clinical data were excluded. Patients who underwent surgery between January 2013 and December 2020 were included in the training cohort for development of the nomogram, and those who underwent surgery between January 2021 between December 2021 were included in the validation cohort.

Clinical Variables.
We collected preoperative information on clinical variables, including gender, age, body mass index (BMI), underlying diseases, dialysis duration, dialysis modality, and preoperative laboratory tests (serum intact parathyroid hormone, serum alkaline phosphatase, serum calcium, serum phosphate, serum kalium, hemoglobin, albumin, serum creatinine, urea, prothrombin, and fbrinogen). Moreover, the postoperative serum calcium within 72 hours had been collected in this study. Te postoperative serum calcium below 2.20 mmol/L was used to diagnose hypocalcemia, and severe hypocalcemia was defned as serum calcium below 1.80 mmol/L after TPTX.

Perioperative Management and Surgical Procedures
Routine preoperative examination included serum intact PTH level, concentrations of calcium, blood phosphorus, liver and renal function tests, and ultrasonography of the thyroid and parathyroid glands. Te preoperative diagnosis was based on criteria of the KDIGO 2009 clinical practice guideline. TPTX with/without auto-transplantation of parathyroid tissue and SPTX are currently considered as standard surgical procedures in the treatment of SHPT [20,21]. Previous studies have noted that the TPTX without auto-transplantation approach has been associated with lower rates of recurrence [22]. Terefore, total parathyroidectomy without auto-transplantation is used as a surgical option in our center. TPTX with autotransplantation (TPTX + AT) and SPTX had not been selected since January 2016 in our center. All surgical procedures were performed by Dr. Guangming Cheng and his surgical team. A successful operation was defned as previously described [23].  (Table 3).

Nomogram for Predicting Postoperative Severe
Hypocalcemia. Based on the previous analyses, the independently associated risk factors were used to construct a nomogram ( Figure 1). Te resulting model was internally validated using the bootstrap validation method. Te nomogram for predicting postoperative SH in the training cohort had an unadjusted C index of 0.866 (95%CI, 0.816-0.916) and a bootstrap-corrected C index of 0.866, indicating that the nomogram has good accuracy in estimating the risk of SH. In the validation cohort, the nomogram displayed a C index of 0.867 (0.793-0.941) for the estimation of SH risk. Tere was also a good calibration curve for risk estimation (Figure 2). In addition, the calibration plots overlapped with the ideal line in the training and validation cohorts, showing adequate agreement of the predictive nomogram with actual observations (Figure 3).

Risk of Postoperative Severe Hypocalcemia Based on the
Nomogram. Te optimal cutof value of the total nomogram scores was determined to be 100. Te sensitivity, specifcity,

Discussion
Postoperative SH can increase mortality and hospitalization [25][26][27][28]. In the present study, SH was found to be present in 90 of 299 cases. And, we uncovered that the preoperative factors, including younger age, higher serum iPTH level, higher serum ALP level, and lower serum calcium level, are signifcantly associated with SH in patients who underwent TPTX without auto-transplantation. Importantly, we developed a nomogram which achieved an optimal preoperative prediction of SH in those patients who underwent TPTX without auto-transplantation for secondary hyperparathyroidism.   Figure 1: Nomogram for preoperative prediction of server hypocalcemia (SH) following total parathyroidectomy (TPTX) without auto-transplantation in patients with secondary hyperparathyroidism (SHPT). Points are signed for age, preoperative serum calcium level, preoperative serum iPTH level, and preoperative serum ALP level. Te score for each value was assigned by drawing a line upward to the "points" line, and the sum of the four scores was plotted on the "total points" line (probability of SH).

International Journal of Endocrinology
Considering the high incidence of SH after parathyroidectomy and the development of life-threatening sequelae, eforts on the risk estimation of SH have been made over the past decade [29][30][31][32][33][34][35][36][37][38][39][40][41][42][43][44][45][46]. Various risk factors have been suggested for the development of postoperative hypocalcemia in previous studies, including younger age, pruritus, higher preoperative iPTH and ALP levels, and lower preoperative serum calcium level. Moreover, the diagnostic value of these factors was evaluated by using receiver operator characteristic (ROC) analyses [31,33,40,41]. For example, one study reported a diagnostic model incorporated 3 factors (preoperative serum calcium, iPTH,   and ALP levels) in the risk estimation of SH [40]. However, the algorithm of this diagnostic model is complex, and further clinical validation is required. As is well known, the nomogram, identifed as an easy-to-use prediction tool, has high accuracy and good discrimination characteristics in predicting outcomes. In the current study, the proposed nomogram, which incorporated 4 easily available preoperative variables, performed well as supported by the C index values 0.866 (95%CI, 0.816-0.916) and 0.867 (95% CI, 0.793-0.941) in the training and validation cohorts, respectively, and the optimal calibration curves demonstrated the agreement between prediction and actual observation. Younger age was found to be a preoperative risk factor for SH in this study, which is largely consistent with the previous reports [30, 34-36, 43, 44, 46]. Explanation as to why younger patients are at a higher risk of hypocalcemia mainly includes stronger osteoblast function and greater calcium utilization efciency of bone tissue. However, the infuence of age on the postoperative hypocalcemia is still controversial. Gong et al. reported that advanced age is a risk factor of postoperative hypocalcemia [39]. In addition, some reports also revealed no association between age and postoperative hypocalcemia. It should be noticed that nearly all of these studies had been performed in a single center with a small sample size. Terefore, some researchers have appealed for more studies with larger sample sizes to verify this conclusion. Recently, in a retrospective study that included 1500 patients, Zhao et al. reported that age at the time of surgery was negatively associated with postoperative hypocalcemia [43]. Te sample size of this retrospective study is larger than that in other previous studies and seems to confrm the association between younger age and postoperative hypocalcemia.
Te explanation of preoperative serum calcium level as a risk factor of hypocalcemia after parathyroid surgery is limited in the relevant studies [11,33,35,36,46]. Te lower preoperative serum calcium in SHPT patients may indicate a higher baseline bone-remodeling status [47]. And, the remineralization of the skeleton would be stronger in those patients after PTX. Terefore, the frequency and the severity of postoperative hypocalcemia would be higher in those patients. Some studies use preoperative corrected serum calcium for research [35]. However, corrected serum calcium cannot refect the accurate serum calcium concentration in patients [48][49][50]. For patients undergoing longterm dialysis or patients with hypoproteinemia, the blood calcium level may be overestimated when using corrected serum calcium and the collinearity among variables is often neglected in these studies when screening variables. In the present study, total serum calcium level was used to predict SH after TPTX, and we identifed that calcium defciency before operation was a risk factor for SH. Importantly, the results of this study strongly suggest that appropriate calcium supplement therapy should be provided in patients with hypocalcemia before PTX to alleviate the postoperative complications of hypocalcemia.
In previous studies, preoperative iPTH as one of the risk factors of postoperative hypocalcemia has been frequently found in patients who underwent PTX [29, 31-34, 36, 37, 39, 40, 43, 44]. It is well accepted that preoperative iPTH concentration can be used to predict postoperative hypocalcemia, because it is consistent with the physiological functions of iPTH. However, some studies also found no association between preoperative iPTH and postoperative hypocalcemia. It should be noticed that the parathyroid procedures performed are rather variable in these studies [30,33,38,41,42,46], including TPTX, TPTX + AT, or SPTX. Te selected operative method defnitely afects the postoperative serum calcium content. For example, SPTX preserves a remnant parathyroid gland with its original blood supply, and then, it has a higher iPTH value and a lower risk of postoperative hypocalcemia [51,52]. Terefore, the results of these studies could not truly refect the association between preoperative iPTH and postoperative hypocalcemia. In addition, there are other explanations including skeletal resistance to iPTH, the relationship between the serum iPTH level and degree of bone remodeling being not always maintained, and the current iPTH assay being not accurate [53][54][55].
ALP as a bone formation biomarker can refect the activity of osteoblasts [56]. Te serum ALP level will be increased when the osteoblastic activity is stimulated by high level of iPTH in patients with SHPT. Teoretically, higher ALP levels before surgery indicates a more active bone remodeling state in patients, resulting in a higher incidence of hypocalcemia after PTX. In fact, preoperative ALP level as a risk factor of hypocalcemia after PTX has been well identifed in previous studies [29-32, 34, 36-45]. For example, a previous study conducted by Tan et al. reported that preoperative ALP level was a risk factor of hypocalcemia, and it suggested that preoperative ALP level should be used to instruct the management of postoperative hypocalcemia [45]. Ho et al. also identifed that preoperative ALP level was a risk factor for postoperative hypocalcemia, and they considered that the increase in ALP level was closely related to the decrease in serum calcium level and the amount of calcium supplement by observing the clinical index two weeks after operation [30]. Bone-ALP, which is directly related to bone turn-over, had been used in several previous studies [35]. But, bone-ALP is not routinely measured in  6 International Journal of Endocrinology most of the the institutions especially in the basic medical institutions, thus limiting its wide use. In the present study, we selected total ALP but not bone-ALP as a variable, and we found that total preoperative ALP level is an independent predictive factor of hypocalcemia after TPTX. Moreover, the accuracy of this nomogram was estimated using as the cutof value in the present study. Patients with a score 100 or more are considered a high-risk subgroup of SH after TPTX. Our results identifed that this nomogram allows physicians to accurately identify dialysis patients who are at a greater risk of hypocalcemia after PTX and to aggressively monitor and treat those patients with a score or more.
Tere are some limitations in our study. First and foremost, we took the frst measurement of blood level of calcium at 6-8 hours after surgery and then once daily in the morning until 72 hours after surgery. We choose the lowest value of them to distinguish the sever hypocalcemia. Hypocalcemia could start as early as 5-6 hours, and patients were treated with intravenous infusion of calcium gluconate if they were diagnosed as hypocalcemia regardless of the severity. However, the use of calcium supplements can lead to decreased incidence and severity of hypocalcemia. Second, the nomogram was constructed based on data from a single institution. It is necessary to validate the predictive value of this nomogram in the other institutions. Tird, some other factors might be correlated with postoperative hypocalcemia, such as uremic toxins and bone mineral density. Finally, this study is retrospective in nature. Te reliability of the nomogram model needs to be confrmed in further prospective studies.

Conclusion
In the present study, we identifed younger age, higher serum iPTH level, higher serum ALP level, and lower serum calcium level are the preoperative risk factors for SH after TPTX. By combining these 4 preoperative risk factors, a nomogram was constructed. Te nomogram provides an optimal preoperative estimation of SH risk in patients with SHPT underwent TPTX.

Data Availability
Te data that support the fndings of this study are available from the corresponding authors upon reasonable request.

Conflicts of Interest
Te authors declare that there are no conficts of interest.

Authors' Contributions
Chenchen He and Yibing Zhang designed and performed the research and wrote the paper. Longfei Li, Guangming Cheng, and Wei Zhang collected data and performed the statistical analysis. Yufu Tang and Chunhui Wang designed the study and revised the paper. All authors approved the submitted fnal version. Chenchen He and Yibing Zhang contributed equally to this work.