Definition and diagnosis of postsurgical hypoparathyroidism after thyroid surgery: meta-analysis

Abstract

Background: Postsurgical hypoparathyroidism (PH) is the most frequent complication after thyroid surgery. The aim of this systematic review and meta-analysis is to summarize a unifying definition of PH and to elucidate the best possible approach for early detection of PH.

Methods: A systematic review of the literature according to the PICO framework using Embase, PUBMED and the Cochrane library was carried out on 1 December 2021 followed by analysis for risk of bias, data extraction and meta-analysis. All studies addressing the definition of postoperative hypoparathyroidism and/or diagnostic approaches for early detection and diagnosis were included. Case reports, commentaries, non-English articles, book chapters and pilot studies and reviews were excluded.

Results: From 13 704 articles, 188 articles were eligible for inclusion and further analysis. These articles provided heterogeneous definitions of PH. Meta-analysis revealed that postoperative measurements of parathormone (PTH) levels have a higher sensitivity and specificity than intraoperative PTH measurements to predict PH after thyroid surgery. None of the timeframes analysed after surgery within the first postoperative day (POD1) was superior to predict the onset of PH. PTH levels of less than 15 pg/ml and less than 10 pg/ml are both reliable threshold levels to predict the postoperative onset of PH. A relative reduction of mean(s.d.) PTH levels from pre- to postoperative values of 73 (standard deviation 11) per cent may also be predictive for the development of PH. The estimation of calcium levels on POD1 are recommended.

Conclusion: PH is best defined as an undetectable or inappropriately low postoperative PTH level in the context of hypocalcaemia with or without hypocalcaemic symptoms. PTH levels should be measured after surgery within 24 h. Both threshold levels below 10 and 15 pg/ml or relative loss of PTH before/after thyroid surgery are reliable to predict the onset of PH.

Fig. 1

PRISMA flowchart for the literature search

Fig. 2

Forest plot (left) and summary receiver operating characteristic curves (right) Studies analysing the value of a intraoperative parathyroid hormone measurements and b postoperative parathyroid hormone measurements to identify postsurgical hypoparathyroidism. The outcome parameter is hypocalcaemia in the presence or absence of symptoms as indicated by the study protocol of the studies included. Bivariate analysis summarized the sensitivity and specificity for each condition. TP, true positive; FP, false positive; FN, false negative; TN, true negative.

Fig. 3

Forest plot (left) and summary receiver operating characteristic curves (right) Studies analysing a predictive value of early postoperative (1–6 h) and b later postoperative (24 h/POD1) parathyroid hormone measurements to identify postsurgical hypoparathyroidism. The definition of hypoparathyroidism varied between studies. Only studies were included in the meta-analysis that investigated both time points in the same cohort of patients. Bivariate analysis summarized the sensitivity and specificity for each condition. TP, true positive; FP, false positive; FN, false negative; TN, true negative.

Fig. 4

Forest plot (left) and summary receiver operating characteristic curves (right) Studies of parathyroid hormone threshold levels of a less than 15 pg/ml and b less than 10 pg/ml to identify postsurgical hypoparathyroidism (development of hypocalcaemia). Bivariate analysis summarized the sensitivity and specificity for each condition. TP, true positive; FP, false positive; FN, false negative; TN, true negative.

Fig. 5

Forest plot (left) and summary receiver operating characteristic curves (right) Studies of parathyroid hormone threshold levels of less than10 pg/ml to identify postsurgical hypoparathyroidism (development of symptoms) are shown. Bivariate analysis summarized the sensitivity and specificity for each condition. TP, true positive; FP, false positive; FN, false negative; TN, true negative.

Fig. 6

Median of mean and 95 per cent confidence intervals from values of relative reduction of parathyroid hormone levels from eight comparable studies These were extracted to assess whether these values can be used to predict hypocalcaemia in patients after thyroid surgery. All studies together represent a cohort of 1358 patients. Unpaired non-parametric Kruskal–Wallis test was used to test for significant differences. PTH, parathyroid hormone.