Post total thyroidectomy hypocalcemia: A novel multi-factorial scoring system to enable its prediction to facilitate an early dischargePV Pradeep1, K Ramalingam2, B Jayashree1
1 Department of Endocrine Surgery, Narayana Medical College & Superspeciality Hospital, Chinthareddypalem, Nellore, Andhra Pradesh, India
2 Department of Biochemistry, Narayana Medical College & Superspeciality Hospital, Chinthareddypalem, Nellore, Andhra Pradesh, India
Correspondence Address: Source of Support: Indian Council of Medical Research grant no: 3/2/TG-21/HRD-2011, Conflict of Interest: None DOI: 10.4103/0022-3859.109479
Source of Support: Indian Council of Medical Research grant no: 3/2/TG-21/HRD-2011, Conflict of Interest: None
Context: No single factor can predict the occurrence of post total thyroidectomy (TT) hypocalcemia. Aims: This study was conducted to look at various factors usually implicated in post TT clinically significant hypocalcemia (CSH) and to develop a scoring system using a combination of these factors to predict CSH. Settings and Design: Prospective study, tertiary care center. Materials and Methods: 145 patients, who underwent total thyroidectomy for benign goiters and early carcinoma thyroid ( < T2/N0/M0), were included. Age of the patient, presence, or absence of hyperthyroidism, pre-operative levels of serum calcium and 25 OH vitamin D, post-operative iPTH at 8 hours and calcium at 12 hours, intra-operative parathyroid preservation status, and nodule size were studied. CSH prediction score (0 to 8) was designed based on these 8 factors. Statistical Analysis: SPSS 13 software was used. For comparison between groups' independent samples T-test and Chi-square test was used. Statistical significance was set at P<0.05. A logistic regression analysis model was built to assess the significant predictors. Results: There were 22 males and 123 females. 64.82% had euthyroid multinodular goiters, 24.82% had toxic MNG, and 10.34% had an early carcinoma of thyroid. 30.34% developed CSH. CSH was observed in patients with low pre-operative calcium (P=0.008), low 25 OH vitamin D (P=0.001), low post-operative iPTH at 8 hours (P=0.001), low serum calcium at 12 hours after surgery (P=0.001) and lesser number of parathyroid identification at surgery (P=0.001). Patient age (P=0.2) and nodule size (P - 0.17) was not significant. Hypocalcemia risk score of > 3 had 91% sensitivity, 84% specificity with a PPV of 71% and NPV of 95%, whereas score of ≥ 4 had 100% specificity and PPV in predicting CSH. Conclusions: CSH after TT is multi-factorial, and a combination of factors (Hypocalcemia prediction score > 3) can be used to predict it so as to discharge patients within 24 hours after surgery.
Keywords: 25 OH vitamin D3, clinically significant hypocalcemia, iPTH, prediction score, total thyroidectomy
After total thyroidectomy (TT), temporary symptomatic hypocalcemia occurs in some patients. This is a limiting factor in an early discharge of the patients from hospital. The incidence has been reported to vary from 0.5% to 75%.  It is important to predict which patient is likely to develop CSH so that an early treatment can be initiated in these patients, and the others, in whom the occurrence of hypocalcemia is unlikely, can be discharged as a day case surgery. Post-operative serum calcium, post-operative intact PTH (iPTH), serum 25 OH vitamin D levels, pre-operative serum calcium, presence of hyperthyroidism, advanced age of the patient, parathyroid preservation, and size of goiter at surgery have all been implicated in the development of post TT hypocalcemia. ,,,,,,,,,,,,,,,,,,,,, It has been realized from various studies published in literature that no single factor can predict its occurrence. ,,,,,,,,,,,,,,,,,,,, This study was conducted to look at all these factors and to develop a scoring system using a combination of these factors. This will enable us to predict occurrence of post TT clinically significant hypocalcemia in Indian patients, which in turn will enable us to discharge the patients within 24 hours after surgery.
Ours is a prospective study conducted at a tertiary care medical center. All patients of TT operated for benign goiters and early carcinoma thyroid (< T2/N0/M0) during the period January 2010 to January 2011 are included in the final analysis. Carcinoma thyroid stage T3 and above were excluded since our policy is to do central compartment dissection for such cases, which invariably involves parathyroid autotransplantation. Patients who developed CSH at or before 12 hours of surgery (n=9) were excluded from the final analysis to avoid confounding.
An ethical clearance for the present study was obtained from the Institutional Ethical committee. An informed consent was obtained from the patients prior to the study.
Age of the patient, presence of hyperthyroidism, pre-operative levels of serum calcium and 25 OH vitamin D, post-operative iPTH at 8 hours and serum calcium at 12 hours, intra-operative parathyroid preservation status, and nodule size were studied to predict development of CSH 24 hours after surgery. Hypocalcemia prediction score (0 to 8) was designed based on these factors [Table 1]. Post-operative biochemical hypocalcemia (BH) was defined as serum calcium < 7.5 mg% occurring anytime after total thyroidectomy. Clinically significant hypocalcemia (CSH) was defined as patients who have low serum calcium (< 7.5 mg %) and carpopedal spasm (induced within 2 minutes on eliciting the Trousseau sign). Temporary hypoparathyroidism was defined as one which recovers within 6 months of thyroidectomy. Since this study was to design a scoring system, none of the patients included in this study received prophylactic calcium tablets. A maximum of 8 points and minimum of 0 points was allotted to each case.
Serum calcium is estimated by Cresolphthalein complex method using commercial kit (GmbH Germany) with HUMASTAR 600 automated chemistry analyzer (GmbH Germany). Serum iPTH is estimated by sandwich chemiluminescent immunoassay by commercial kits from (BECKMAN, USA) using ACCESS-2 automated chemiluminescent immunoassay system (BECKMAN, USA). Serum 25 OH vitamin D3 is estimated by HPLC method using commercial column and reagents from (RECIPE, GmbH, Germany) using (Younglin) HPLC, Korea.
SPSS 13 statistical software package was used. (SPSS Inc., Chicago, IL, USA) Continuous variables were described as mean/standard deviation, categorical variables as actual numbers/percentages. For comparison between groups', independent samples T-Test and Chi-square test was used. Statistical significance was set at P<0.05 A logistic regression analysis model was built to assess the influence of the independent variables on occurrence of hypocalcemia.
Among the 145 patients included in the final analysis; there were 22 males and 123 females (15.2% and 84.8%). The mean duration of goiter was 38.09±56.05 months (Range 1-300). 64.82% (n=94) were euthyroid multinodular goiters, 24.82% (n=36) were toxic MNG, and 10.34% (n=15) were early carcinoma thyroid. 30.34% (n=44) developed clinically significant hypocalcemia (CSH) after 24 hours of surgery. Comparison of the patients who developed CSH (group B) with those who did not develop (group A) revealed that CSH was observed in patients with low pre-operative calcium (P=0.008), low S Vit D (P=0.001), low post-operative iPTH at 8 hours (P=0.001), and low serum calcium at 12 hours after surgery (P=0.001). Clinically significant hypocalcemia (CSH) also occurred with lesser number of parathyroid being preserved at surgery (P=0.001). Age of the patient (P=0.2) and nodule size (P-0.17) was not significantly different between the groups. [Table 2] shows the comparison of the demographic data between groups A vs. group B. [Table 3] shows the predictive power of each of the independent variables in predicting CSH. Logistic regression analysis revealed that in our patient population, post-operative calcium at 12 hours after surgery, number of parathyroid preserved, pre-operative 25 OH vitamin D deficiency, pre-operative serum calcium levels, and operated nodule size were the important predictors of post-operative symptomatic hypocalcemia. [Table 4] depicts the logistic regression data. [Table 5] depicts the predictive power of the different hypocalcemia scores, which will predict clinically significant hypocalcemia. Hypocalcemia risk score of >3 had 91% sensitivity, 84% specificity with a PPV of 71% and NPV of 95%, whereas a score =>4 had 100% specificity and PPV in predicting CSH.
The most common problem encountered after TT is hypocalcemia, which can either be temporary or be permanent. Temporary hypocalcemia can be biochemical (BH) or symptomatic hypocalcemia (SH), which usually develops 24 to 48 hours after the thyroidectomy.  Hence, patients have to be observed for this period before they can be discharged in order to prevent development of CSH at home. This is especially true in India where the rural health care system is not well-developed.
Various factors have been implicated in the development of post TT hypocalcemia. One of them has been post-operative serum calcium levels. ,, Pfleiderer et al. demonstrated that serum calcium less than 7.6 mg% on day 1 after surgery had 95% specificity in predicting SH. We used serum calcium at 12 hours after the TT as suggested by Lombardi et al. Lo CY  et al. used serum calcium of 7.2 mg% as cut off to define hypocalcemia. We used a cut off of 7.5 mg% as one of the predictive factors, which had a sensitivity of 73%, specificity of 84% in predicting CSH [Table 3]. However, since the serum calcium had to included in the prediction model, we excluded those cases who developed CSH at or before 12 hours after surgery (n=9).
Post-operative as well as intra-operative serum iPTH has been used in various studies to predict post TT hypocalcemia. ,,,, An analysis of 9 observational studies by Noordzij et al. revealed that the PTH 6 hours after TT had 96.4% sensitivity and 91.4% specificity in detecting post-operative hypocalcemia. Payne et al. used iPTH at 6, 12, and 20 hours post-operatively and found the 12 hour value as most sensitive in prediction. Lombardi et al. used 2, 4, 6, 24, 48 hours post-operatively and found that the 4 hour and 6 hour value had the best predictive value. Since there are no definite guidelines for the time at which the iPTH sample has to be collected, we used iPTH at 8 hours after surgery as predictor in this study. The cut off values of iPTH has been different in different studies. The mean PTH was 13.52 pg/mL in patients who developed hypocalcemia.  In our study, the post-operative iPTH of ≤10 pg/mL at 8 hours after surgery had a sensitivity of 68%, specificity of 73% [Table 3]. Literature review has shown that intra-operative PTH was less sensitive and specific than when it was checked post-operatively;  therefore, we used only one post-operative iPTH value.
Serum vitamin D has been found to be low in Indian patients in various studies. , Since the serum vitamin D directly influences the calcium kinetics, especially in the post-thyroidectomy scenario, we included pre-operative vitamin D also as a predictive factor. , One study has shown that the low vitamin D levels are associated with 28-fold increase in chances of post TT clinically significant hypocalcemia.  Vitamin D levels less than 20 ng/mL was found to have sensitivity and specificity of 68% and 80% respectively in our study in predicting CSH.
Hyperthyroidism has been shown to have significant impact over bone turn over even after euthyroid status is restored. , Longer duration of Grave's disease and low vitamin D were shown as factors contributing to hypocalcemia.  In our study, we observed that presence of hyperthyroidism had a sensitivity of 46% and specificity of 86% in predicting clinically significant hypocalcemia.
Pre-operative calcium levels have been used to predict hypocalcemia post-operatively.  Yamashita et al. have reported lower pre-operative calcium levels in patients developing hypocalcemia.  In our study, low pre-operative calcium was having low sensitivity of 14% but a high specificity of 88%. This could have been due to the cut off being 8.4 mg%; a lower serum calcium limit would have increased the sensitivity.
Erbil et al. and Dedivitis et al. observed that elderly patients have significant risk factor for post-operative hypocalcemia.  Ageing is said to be associated with decreased intestinal calcium absorption, decreased renal 1α hydroxylase, decreased dermal synthesis of vitamin D. These contribute to susceptibility to hypocalcemia after TT.  In this study, we considered post-menopausal women and men > 60 years as susceptible to hypocalcemia. We observed that the age was not very sensitive in predicting the hypocalcemia [Table 3].
The lesser the number of parathyroid saved the more the chances of developing post TT hypocalcemia. Some authors have suggested that at least 3 parathyroids are to be saved while others opined that 2 functional glands are enough to prevent post TT hypocalcemia. ,, In our study, we considered 2 or <2 parathyroid identification as a risk factor for hypocalcemia. This had a sensitivity and specificity of 27 % and 96% respectively in predicting hypocalcemia [Table 3].
It has been shown that the post-operative hypocalcemia is also dependant on the surgeons experience and the operative technique used. , In this study, all surgeries were performed by a single surgeon, well trained in thyroid surgery. The surgical technique, therefore, has been uniform in all cases.
In our series, the CSH after TT was 30.3%. The incidence of permanent hypocalcemia has been 0.69% (n=1). We have studied 8 factors, which are usually implicated in post TT temporary hypoparathyroidism and applied them to develop a scoring system, which will predict the occurrence of CSH. Individually, these factors have been shown to significantly influence the occurrence of CSH by various authors. ,,,,,,,,,,,,, In our study, we observed that even though individually these factors could be used as predictors, the sensitivity was low (14-68%, [Table 3]) with PPV of 20-75% and NPV of 66-85%. However, a combination of factors and hypocalcemia score of ≥3 had a sensitivity of 91%, specificity of 84% and NPV of 95% [Table 5]. Results of the logistic regression model reveals that in our patients, out of the 8 predictors used, post-operative serum calcium at 12 hours, number of parathyroids preserved, serum 25 OH vitamin D, operative nodule size, and the pre-operative serum calcium levels are the strongest predictors of the post TT symptomatic hypocalcemia [Table 4].
The possible limitation is such a model can be the higher number of tests required, which are not done during the routine thyroidectomy (S 25 OH Vit D, Serum iPTH etc.) and also the concern of additional costs. In our center, these additional tests cost 56 US dollars. However, the 2 extra day's hospitalization costs the same i.e. 55 US dollars. The tests, therefore, saves 2 extra days of hospitalization, moreover, the vacated beds can be allotted to other patients waiting for surgery. However, we feel the strength of this study has been the prospective study design and the prediction model using multiple factors, which has not been reported in literature so far.
To conclude, development of hypocalcemia after TT is multi-factorial, and a combination of factors can be used to predict it and selectively practice early discharge of patients. In future, further studies are needed to validate the scoring system prospectively in multiple centers to bring our scoring system into clinical practice.
The authors would like to acknowledge the contributions of Sidharth Muralidharan PhD, School of Mass Communication and Journalism, The University of Southern Mississippi, USA for his valuable contributions in the statistical analysis of the data. This paper was presented as poster at the 81 st Annual meeting of the American Thyroid Association at Indian Wells, California Oct 26 th to 30 th 2011. The paper presentation was supported by the Indian Council of Medical Research grant no: 3/2/TG-21/HRD-2011.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]