Postresection Parathyroid Hormone and Parathyroid Hormone Decline Accurately Predict Hypocalcemia After Thyroidectomy

Transcription

1 Clinical Chemistry / PTH PREDICTS HYPOCALCEMIA AFTER THYROIDECTOMY Postresection Parathyroid Hormone and Parathyroid Hormone Decline Accurately Predict Hypocalcemia After Thyroidectomy Pedro Alía, PhD, 1 Pablo Moreno, MD, PhD, 2 Raúl Rigo, 1 José-Manuel Francos, MD, 2 and Miguel-Ángel Navarro, MD, PhD 1 Key Words: Thyroidectomy; Hypocalcemia; Hypoparathyroidism; Parathyroid hormone Abstract Hypocalcemia is the most frequent complication after total thyroidectomy. Parathyroid hormone (PTH) measurement has been proposed as an early predictor of this condition. Total thyroidectomy was performed in 39 patients. Hypocalcemia was present in 15 cases (38%). Patients undergoing hemithyroidectomy (n = 13) were considered control subjects not developing hypocalcemia. PTH was measured before surgery and 10 minutes after resection of the gland using a rapid (15 minutes) chemiluminescent immunometric assay. Patients developing hypocalcemia had lower calcium and postresection PTH levels and higher PTH decline than patients not developing hypocalcemia (P <.0001). PTH decline (cutoff value, 62.5%) had the better sensitivity (93.3%) for predicting hypocalcemia, allowing for a fairly safe early discharge. However, the best overall results corresponded to the combination of postresection PTH level ( 18 pg/ml [ 1.9 pmol/l]) and PTH decline (>62.5%), with a sensitivity of 90% and a specificity of 97.9%. Perioperative PTH measures can accurately predict hypocalcemia after thyroidectomy, granting the laboratory a key role in the immediate decision about calcium supplementation for patients at risk. Thyroidectomy can cause hypoparathyroidism from removal of parathyroid glands or damage to their blood supply. This is the most frequent postoperative complication, and the subsequent onset of hypocalcemia can pose severe problems, especially because most patients are discharged early but symptoms may appear later on. The laboratory can have a crucial role in identifying the most reliable markers that allow early diagnosis of hypocalcemia, helping the surgeon to establish replacement therapy, if necessary, after a short monitoring protocol. Approaches to prediction of hypocalcemia based on calcium serum values at different times after surgery 1-3 or combined assessment of calcium and parathyroid hormone (PTH) levels 6 hours after surgery 4,5 have proven useful. However, because of a search for earlier predictors, the short half-life of PTH (that had already let it emerge as an indicator of completeness of parathyroidectomy during the past decade 6 ) has led to increased interest in intraoperative PTH monitoring as an early marker of hypocalcemia Our study aimed to confirm whether the PTH value immediately after excision (postresection PTH) and the relative difference between PTH values before and after excision (PTH decline) are reliable markers for the need to establish replacement therapy with calcium following total or completion thyroidectomy. We also looked for the possibility of establishing an immediate postsurgical decision algorithm. Materials and Methods Patients From June 2003 to August 2005, 52 patients (12 males and 40 females; median age, 44.5 years; range, years) 592 Am J Clin Pathol 2007;127: Downloaded 592 from

2 Clinical Chemistry / ORIGINAL ARTICLE undergoing thyroidectomy were enrolled in a prospective study. Informed consent was obtained from patients after approval by the institutional review board. Thirty-nine patients, most with papillary or medullary thyroid cancer (n = 34) and the rest with multinodular goiter, Graves disease, or thyroid nodules, underwent total thyroidectomy (TT; including unilateral or bilateral central neck node dissection and excision of central neck recurrence), and 13 patients with multinodular goiter or thyroid nodules who underwent hemithyroidectomy (total lobectomy plus isthmusectomy) were considered control subjects (theoretically they were not at risk of hypoparathyroidism). Surgery was performed by the same surgeon. Patients in whom clinical hypocalcemia developed were treated with intravenous calcium, oral calcium, oral vitamin D, or a combination. Sampling Protocol After induction of general anesthesia, a blood sample was drawn from a peripheral intravenous line (preoperative PTH). Ten minutes after removal of the thyroid gland, while the patient was still in the operating room, a second sample was obtained (postresection PTH). The samples were handled according to the protocol already routinely used in our center for parathyroid surgery. Briefly, both samples were put in ice and carried together by an orderly from the operating room to the programmed laboratory. There they were centrifuged at 3,500 rpm for 3 minutes at 4 C and analyzed (15 minutes of instrument assay). The whole process took about 30 minutes. However, during the study, the surgeon followed his customary protocol for patient management and eventual discharge based on daily calcium level measurement and observation for clinical symptoms of hypocalcemia. Analytic Methods The intact plasma PTH concentration was measured with a chemiluminescent immunometric assay in the Immulite- Turbo analyzer (Diagnostic Products, Los Angeles, CA). The reference interval for PTH in our laboratory is 14 to 74 pg/ml ( pmol/l). The method imprecision had a coefficient of variation of 8% for concentrations in the reference interval, and the functional sensitivity (defined as the concentration for which the method imprecision has a coefficient of variation of 20%) was 7 pg/ml (0.75 pmol/l). The relative decrement of PTH or percentage of PTH decline was calculated as follows: [(Preoperative PTH Postresection PTH)/Preoperative PTH] 100. The serum calcium level was measured using a colorimetric assay with endpoint determination and sample blank in the Modular Hitachi analyzer (Roche Diagnostics, Mannheim, Germany). The reference interval for calcium in our laboratory is mg/dl ( mmol/l). Serum calcium levels were measured daily at 7:00 AM until discharge. Biochemical hypocalcemia was defined by a concentration less than 8.6 mg/dl (2.15 mmol/l), but calcium supplementation was given only for a value of 7.2 mg/dl or less ( 1.8 mmol/l) or for values between 7.2 and 8.6 mg/dl (1.8 and 2.15 mmol/l) accompanied by symptoms of clinical hypocalcemia (positive Trousseau sign, tetany, paresthesia). Statistical Methods Comparisons between groups were made with the Kruskal-Wallis and Mann-Whitney U tests. Diagnostic characteristics (receiver operating characteristic curves, sensitivity, specificity, positive and negative predictive values, and likelihood ratios) were assessed to compare the diagnostic performance of the biochemical tests. Calculations were made with the Analyse-it statistical program (Analyse-it Software, Leeds, England), using a significance level of.05. Results All patients who underwent a hemithyroidectomy remained normocalcemic, whereas 15 (38%) of 39 patients who underwent TT developed clinical signs of hypocalcemia in the first 24 to 36 hours after surgery. Thus, we had 3 groups of patients Table 1 : control hemithyroidectomy, normocalcemic TT, and hypocalcemic TT. No differences in basal PTH concentrations were observed. After surgery, a significant decline in PTH was detected for all 3 groups (29.6%, P =.048; 41.5%, P =.0003; and 80.5%, P <.0001, respectively). Almost all patients (36 of 39 TT cases and 10 of 13 hemithyroidectomy cases) had serum calcium concentrations less than 8.6 mg/dl (<2.15 mmol/l; our laboratory s lower reference limit) at 12 to 18 hours after surgery (ie, at 7:00 AM the day after surgery). However, there were significant differences in median serum calcium levels among the groups (P <.0001). In normocalcemic TT vs hemithyroidectomy cases, significant differences were observed (P =.05), but patients in whom hypocalcemia eventually developed had even lower calcium concentrations (P =.0007). As with calcium, the differences among groups observed when comparing PTH concentrations after excision of the gland and PTH decline were attributable mostly to the hypocalcemia group. In fact, no significance was found when comparing hemithyroidectomy and normocalcemic TT groups for postresection PTH (P =.14) and PTH decline (P =.12) (because value distributions in these groups were widely overlapping, as shown by the 95% confidence intervals in Table 1), whereas postresection PTH was clearly lower (P <.0001) and PTH decline was clearly higher (P <.0001) in hypocalcemic TT patients. The receiver operating characteristic curves (including only TT patients) for PTH decline, postresection PTH, and Downloaded from Am J Clin Pathol 2007;127:

3 Alía et al / PTH PREDICTS HYPOCALCEMIA AFTER THYROIDECTOMY Table 1 PTH and Calcium Levels for Hemithyroidectomy and Total Thyroidectomy Patients * Total Thyroidectomy (n = 39) Hemithyroidectomy (Normocalcemia; n = 13) Normocalcemia (n = 24) Hypocalcemia (n = 15) P Preoperative PTH (pmol/l) 7.0 ( ) 6.0 ( ) 5.4 ( ) NS Calcium, 7:00 AM (mmol/l) 2.12 ( ) 2.05 ( ) 1.80 ( ) <.0001 PTH decline (%) 29.6 ( ) 41.5 ( ) 80.5 ( ) <.0001 Postresection PTH (pmol/l) 5.3 ( ) 2.9 ( ) 1.0 ( ) <.0001 NS, not significant; PTH, parathyroid hormone. * Data are given as median (95% confidence interval). Laboratory values are given in Système International units; conversions to conventional units are as follows: calcium (mg/dl), divide by 0.25; PTH (pg/ml), divide by Kruskal-Wallis test P values for comparisons of the 3 groups are included. This calcium measure corresponds to the day after surgery. P <.05 for postresection PTH vs preoperative PTH. calcium levels on the day after surgery are shown in Figure 1. Differences between areas under the curve were not significant. A summary of diagnostic characteristics for the best cutoff point (corresponding to the maximum sensitivity-specificity pair) is given in Table 2, in which each biochemical test is considered independently. Postresection PTH and PTH decline can also be considered together to improve the diagnostic characteristics. Two different combinations are shown in Table 3 : one includes the same cutoff values as in Table 2, the other (postresection PTH, 18 pg/ml [ 1.9 pmol/l] and PTH decline, >62.5%) gives an even better outcome. The election of the second postresection PTH value is based on the dispersion data for postresection PTH and PTH decline shown in Figure 2, where 2 lines mark the cutoff points for the latter combination. Sensitivity (True Positives) No discrimination PTH decline Postresection PTH 1Ca Specificity (False Positives) Discussion Hypocalcemia is the most usual complication after TT, although the reported frequency is variable, 2,15 probably because the usual practice of administering calcium supplements to all patients to allow for early discharge may mask the Figure 1 Receiver operating characteristic curves for comparison. 1Ca7 refers to the calcium level on the next day. No significant differences were found between areas under the curve. Table 2 Comparison of Diagnostic Characteristics for Hypocalcemia * Cutoff Values for Hypocalcemia Sensitivity (%) Specificity (%) Positive LR Negative LR PPV (%) NPV (%) PTH decline 93.3 ( ) 83.3 ( ) 5.6 ( ) 0.08 ( ) 77.8 ( ) 95.2 ( ) (>62.5%) Postresection 80.0 ( ) 91.7 ( ) 9.6 ( ) 0.22 ( ) 85.7 ( ) 88.0 ( ) PTH ( 1.5 pmol/l) Calcium, 7:00 AM 80.0 ( ) 91.3 ( ) 9.2 ( ) 0.22 ( ) 85.5 ( ) 87.5 ( ) ( 1.96 mmol/l) LR, likelihood ratio; NPV, negative predictive value; PPV, positive predictive value; PTH, parathyroid hormone. * Values are given as percentage (95% confidence interval) or LR (95% confidence interval). These values correspond to the cutoff values showing the maximum sum for sensitivity and specificity in predicting clinical hypocalcemia. Calcium measures correspond to the day after surgery. Cutoff values are given in Système International units; conversions to conventional units are as follows: calcium (mg/dl), divide by 0.25; PTH (pg/ml), divide by Am J Clin Pathol 2007;127: Downloaded 594 from

4 Clinical Chemistry / ORIGINAL ARTICLE Table 3 Comparison of Diagnostic Characteristics of Combined Magnitudes for Hypocalcemia * Cutoff Values for Hypocalcemia Sensitivity (%) Specificity (%) Positive LR Negative LR PPV (%) NPV (%) PTH decline (>62.5%) 76.7 ( ) 97.9 ( ) 36.8 ( ) 0.24 ( ) 95.8 ( ) 87.0 ( ) and postresection PTH ( 1.5 pmol/l) PTH decline (>62.5%) 90.0 ( ) 97.9 ( ) 43.2 ( ) 0.10 ( ) 96.4 ( ) 94.0 ( ) and postresection PTH ( 1.9 pmol/l) LR, likelihood ratio; NPV, negative predictive value; PPV, positive predictive value; PTH, parathyroid hormone. * Values are given as percentage (95% confidence interval) or LR (95% confidence interval). The first line of values corresponds to cutoff values previously tested separately. Cutoff values are given in Système International units; conversions to conventional units are as follows: calcium (mg/dl), divide by 0.25; PTH (pg/ml), divide by PTH Decline (%) Postresection PTH (pmol/l) Figure 2 Dispersion graphic of postresection parathyroid hormone (PTH) and PTH decline data. Lines mark the cutoff values (corresponding to the second row in Table 3). Hypocalcemic status is indicated by black. 1.8 mmol/l Serum calcium 1.8 <x 2.15 mmol/l Signs and symptoms of hypocalcemia + Calcium supplementation >2.15 mmol/l Discharge PTH decline 62.5% Perioperative PTH Postresection PTH 1.9 pmol/l and PTH decline >62.5% Calcium supplementation Figure 3 Proposal for patient management for total thyroidectomy. Usual practice (left arm) may last up to 3 days. Intraoperative prediction of hypocalcemia (right arm) implies an almost immediate ability to determine the need for calcium supplementation, allowing for early discharge of patients with or without treatment. real frequency of hypocalcemia. In addition, this practice means that a high percentage of patients take unnecessary calcium supplements, which are unpleasant and could, in some cases, increase the risk of constipation, promote the development of calcium kidney stones, and inhibit iron and zinc absorption from food. In our study, supplement was given only with a calcium value of 7.2 mg/dl or less ( 1.8 mmol/l) or with values between 7.2 and 8.6 mg/dl (1.8 and 2.15 mmol/l) accompanied by symptoms of clinical hypocalcemia (positive Trousseau sign, tetany, and paresthesia) Figure 3 (left arm). Several causes have been reported in the literature (postoperative hemodilution, hungry bone syndrome, cervical lymphadenectomy, surgical manipulation of the thyroid gland) 16 that explain a transient hypocalcemic status, but permanent hypocalcemia is mainly explained by hypoparathyroidism from incidental removal of parathyroid glands or damage to their blood supply. Given that patients who undergo this kind of surgery are usually discharged early (most often the day after surgery, unless complications occur), reliable markers allowing early diagnosis of hypocalcemia have been thoroughly sought. 1-5,7-15 The aim of our study was to assess perioperative PTH measures as very early predictors of hypocalcemia after TT to attempt making this kind of surgery an ambulatory procedure in our hospital so that patients can be discharged the day of surgery, provided that no other risks exist. We included patients undergoing hemithyroidectomy as control subjects because they do not have hypocalcemia secondary to hypoparathyroidism. They seemed to show just the transient biochemical hypocalcemia already reported after many surgeries. However, we found that their median postresection PTH concentration was significantly lower than that before surgery, a possible indication of a certain level of impairment in parathyroid metabolism even when no translation into a hypocalcemic status was later observed. A much more significant difference between median preoperative and postresection PTH levels was found for patients Downloaded from Am J Clin Pathol 2007;127:

5 Alía et al / PTH PREDICTS HYPOCALCEMIA AFTER THYROIDECTOMY undergoing TT, but hypocalcemia was present in just 38% of cases those with a lower postresection PTH level and a deeper decrease in PTH. Although statistically significant, differences in calcium concentrations between patients who remained normocalcemic and patients who underwent hemithyroidectomy were much lower than differences between the 2 groups and hypocalcemic TT patients. Thus, the search for a threshold in postresection PTH values or in PTH decline (or both) seems warranted to explain the development of hypocalcemia. Different cutoff values for postoperative PTH have been proposed in different works. 4,11,13,14 This is likely due to the differences in measurement protocol and performance, and the evident need for establishing discriminant values in each center has already been stated. 14 In our work, using a postresection PTH concentration of 14 pg/ml (1.5 pmol/l) as the cutoff value gives a specificity and a positive likelihood ratio similar to those of serum calcium levels on the next day (a criterion used by surgeons, along with clinical signs and symptoms, to diagnose hypocalcemia) and works well for confirming hypocalcemia. A postresection PTH value equal to or lower than 14 pg/ml ( 1.5 pmol/l) is 9 times more likely in patients in whom hypocalcemia develops. On the other hand, the measurement of PTH decline in our work proved to be a good marker to rule out hypocalcemia. Thus, a PTH decline of less than 62.5% is 12 times more likely in patients unlikely to develop hypocalcemia (negative likelihood ratio of 0.08; Table 2). Payne et al 4 recently proposed combining postoperative PTH and serum calcium levels at 6 hours after surgery to assess the risk of hypocalcemia: levels more than 28 ng/l (>28 pg/ml [2.95 pmol/l]) and 8.56 mg/dl (2.14 mmol/l), respectively, almost rule out the risk and permit early discharge. However, perioperative PTH measurements are proving to be as useful and even earlier predictors. Higgins et al 10 and Scurry et al 11 also reported good diagnostic performance of PTH decline, with 75% as their cutoff value (best sensitivity-specificity pair). In our work, substantial improvement can be achieved in positive likelihood ratios (ie, for confirming hypocalcemia) by combining information from postresection PTH levels and PTH decline. As shown in Table 3, concurrent postresection PTH levels of 14 pg/ml or less ( 1.5 pmol/l) and PTH declines greater than 62.5% are 36.8 times more likely in patients likely to develop hypocalcemia. An even higher positive likelihood ratio is obtained when a 62.5% decline is combined with a cutoff of 18 pg/ml or less ( 1.9 pmol/l) for the postresection PTH level. This combination increases the positive likelihood ratio to As shown in Figure 2, it excludes any patient that is not at risk for hypocalcemia. PTH concentration studies after thyroidectomy should focus on maximizing sensitivity to safely discharge patients, 17 but unfortunately tests with a high (near 100%) sensitivity usually do not have an equally good specificity. 4,13,17,18 In our work, to achieve 100% sensitivity for postresection PTH levels and PTH decline independently, specificity would have fallen to less than 65%. PTH decline analysis alone (with a sensitivity of 93.3%) allows good performance for ruling out the hypocalcemia. On the other hand, the criteria proposed in Table 3 combine good sensitivity (up to 90%) and specificity (97.9%). Based on these results, we propose to use postresection PTH levels and PTH decline in combination to manage patients at risk for hypocalcemia after undergoing TT. Figure 3 shows the decision algorithm used by surgeons to date in comparison with our proposal. It is strongly recommended that every center find its own cutoff values because the values would depend on the specific performance characteristics. Compared with usual management involving serum calcium measures along with clinical signs and symptoms of hypocalcemia, perioperative PTH measurements give the laboratory a key role in providing valuable information with the advantage of almost immediate knowledge of the patient s condition. Therefore, perioperative PTH measurement in thyroid surgery may become another useful example of point-of-care testing. On the other hand, this short protocol allows surgeons to quite safely discharge patients because of a nonrisk situation or very early calcium treatment, significantly reducing the length of stay, which may have a positive effect on resource conservation. From the 1 Hormone and Genetics Unit, Department of Clinical Chemistry, and 2 Department of General and Digestive Surgery, IDIBELL-Hospital Universitari de Bellvitge, Barcelona, Spain. Address reprint requests to Dr Alía: Servicio de Bioquímica Clínica, Hospital Universitario de Bellvitge, Feixa Llarga s/n, L Hospitalet de Llobregat, Barcelona, Spain. References 1. Husein M, Hier MP, Al-Abdulhadi K, et al. Predicting calcium status post thyroidectomy with early calcium levels. Otolaryngol Head Neck Surg. 2002;127: Marohn M, LaCivita KA. Evaluation of total/near-total thyroidectomy in a short-stay hospitalization: safe and costeffective. Surgery. 1995;118: Gulluoglu B, Manukyan MN, Cingi A, et al. Early prediction of normocalcemia after thyroid surgery. World J Surg. 2005;29: Payne RJ, Hier MP, Tamilia M, et al. Same-day discharge after total thyroidectomy: the value of 6-hour serum parathyroid hormone and calcium levels. Head Neck. 2005;27: Payne R, Hier MP, Cote V, et al. Postoperative parathyroid hormone levels in conjunction with corrected calcium values as a predictor of post-thyroidectomy hypocalcemia: review of outcomes 1 year after the implementation of a new protocol. J Otolaryngol. 2005;34: Am J Clin Pathol 2007;127: Downloaded 596 from

6 Clinical Chemistry / ORIGINAL ARTICLE 6. Sokoll L, Wians FH Jr, Remaley AT. Rapid intraoperative immunoassay of parathyroid hormone and other hormones: a new paradigm for point-of-care testing. Clin Chem. 2004;50: Lindblom P, Westerdahl J, Bergenfelz A. Low parathyroid hormone levels after thyroid surgery: a feasible predictor of hypocalcemia. Surgery. 2002;131: Richards ML, Bingener-Casey J, Pierce D, et al. Intraoperative parathyroid hormone assay: an accurate predictor of symptomatic hypocalcemia following thyroidectomy. Arch Surg. 2003;138: Lombardi C, Raffaelli M, Princi P, et al. Early prediction of postthyroidectomy hypocalcemia by one single ipth measurement. Surgery. 2004;136: Higgins KM, Mandell DL, Govindaraj S, et al. The role of intraoperative rapid parathyroid hormone monitoring for predicting thyroidectomy-related hypocalcemia. Arch Otolaryngol Head Neck Surg. 2004;130: Scurry WJ, Beus KS, Hollenbeak CS, et al. Perioperative parathyroid hormone assay for diagnosis and management of postthyroidectomy hypocalcemia. Laryngoscope. 2005;115: Quiros R, Pesce CE, Wilhelm SM, et al. Intraoperative parathyroid hormone levels in thyroid surgery are predictive of postoperative hypoparathyroidism and need for vitamin D supplementation. Am J Surg. 2005;189: Soon PS, Magarey CJ, Campbell P, et al. Serum intact parathyroid hormone as a predictor of hypocalcaemia after total thyroidectomy. ANZ J Surg. 2005;75: Vescan A, Witterick I, Freeman J. Parathyroid hormone as a predictor of hypocalcemia after thyroidectomy. Laryngoscope. 2005;115: Bergamaschi R, Becouarn G, Ronceray J, et al. Morbidity of thyroid surgery. Am J Surg. 1998;176: Falk SA. Thyroid Disease. 2nd ed. Philadelphia, PA: Lippincott-Raven; Ghaheri BA, Liebler SL, Andersen PE, et al. Perioperative parathyroid hormone levels in thyroid surgery. Laryngoscope. 2006;116: Wong C, Price S, Scott-Coombes D. Hypocalcaemia and parathyroid hormone assay following total thyroidectomy: predicting the future. World J Surg. 2006;30: Downloaded from Am J Clin Pathol 2007;127: