- Email: [email protected]
Normocalcemia with elevated parathyroid hormone levels after surgical treatment of primary hyperparathyroidism
The American Journal of Surgery 182 (2001) 15–19
Scientific paper
Normocalcemia with elevated parathyroid hormone levels after surgical treatment of primary hyperparathyroidism Anne Denizot, M.D.a,*, Marco Pucini, M.D.a, Christophe Chagnaud, M.D.b, Genevie`ve Botti, M.D.c, Jean-Franc¸ois Henry, M.D.a a
General Surgery and Endocrinology Service, CHU Timone, Boulevard Jean Moulin, 13385 Marseille Cedex 5, France b Radiology Service, CHU Timone, Boulevard Jean Moulin, 13385 Marseille Cedex 5, France c Department of Medical Information, CHU Timone, Boulevard Jean Moulin, 13385 Marseille Cedex 5, France Manuscript received September 27, 2000; revised manuscript February 12, 2001
Abstract Background: Thirty percent of patients who undergo successful parathyroidectomy for primary hyperparathyroidism show unexplained elevated postoperative serum parathyroid hormone (PTH) levels despite normocalcemia. Methods: PTH levels were measured monthly in 97 patients for 6 months after parathyroidectomy. Renal function, 25-OH-vitamin D levels, serum alkaline phosphatase levels, osteocalcin, and bone densitometry were evaluated before and 6 months after surgery. PTH reactivity to calcium loading was tested at the sixth month. Results: Thirty patients had elevated PTH levels despite normocalcemia after parathyroidectomy. Before surgery, these 30 patients had higher PTH and creatinine levels, lower vitamin D levels, and more extensive bone involvement than those with normal postoperative PTH levels. In patients with normal renal function and normal vitamin D levels, postoperative PTH values correlated with preoperative PTH levels but not with bone disease. Conclusion: In most cases, elevated PTH levels after surgery is an adaptive reaction to renal dysfunction or vitamin D deficiency. If no adaptive cause can be found, persistent hyperparathyroidism must be suspected. © 2001 Excerpta Medica, Inc. All rights reserved. Keywords: Parathyroid hormone; Parathyroidectomy; Primary hyperparathyroidism; Postoperative period
Elevated serum parathyroid hormone (PTH) levels are observed after surgery in 30% of patients who undergo parathyroidectomy for primary hyperparathyroidism (pHPT) despite normalization of serum calcium levels [1-5]. The mechanisms underlying elevated serum PTH levels after parathyroidectomy are unclear. Two explanations are possible. The first involves a pathologic process and attributes the elevated serum PTH levels to persistent or recurrent pHPT. Indeed, cases of pHPT with normal calcium levels have been reported [6]. The second explanation involves an adaptative process and attributes elevated serum PTH to a response to physiological changes. In this regard increased PTH levels have been observed in patients with hypocalcemia, renal insufficiency, and vitamin D deficiency. * Corresponding author. Tel.: ⫹33(0)491-968626, ⫹33(0)491-964932; fax: ⫹33(0)491-968170. Present address: General Surgery Service, CHU Nord, Chemin des Bourrellys, 13915 Marseille Cedex 20, France. E-mail address: [email protected]
In the present prospective study we monitored serum PTH levels during the first 6 months after parathyroidectomy for pHPT and attempted to determine if any increase correlated with preoperative renal function, urinary calcium level, serum vitamin D level, bone status, or postoperative PTH reactivity.
Patients and methods Study population Of 120 patients who underwent surgery for suspected pHPT between July 20, 1993, and December 31, 1994, 97 were included in this study. The remaining 23 patients were excluded for the following reasons: misdiagnosis (hypocalciuria-hypercalcemia) in 1 case, association of pHPT with another cause of hypercalcemia in 1 case, loss from follow-up in 12 cases, preoperative PTH level lower than 40
0002-9610/01/$ – see front matter © 2001 Excerpta Medica, Inc. All rights reserved. PII: S 0 0 0 2 - 9 6 1 0 ( 0 1 ) 0 0 6 6 4 - X
16
A. Denizot et al. / The American Journal of Surgery 182 (2001) 15–19
Table 1 Outcome of parathyroidectomy for primary hyperparathyroidism in 97 patients Preoperative mean ⫾ SEM Age (years) Weight of resection specimen (g) Calcium (mmol/L) Phosphate (mmol/L) Alkaline phosphatase (UI/L) Osteocalcin (ng/mL) Bone Density (g/cm3) Z Score Urinary calcium (mmol/day) Creatinine (mol/L) 25-OH-Vitamin D (ng/mL) Parathyroid hormone (pg/mL)
59 ⫾ 10 1973 ⫾ 2156 2.91 ⫾ 0.21 0.77 ⫾ 0.13 200 ⫾ 71 17 ⫾ 8 0.837 ⫾ 0.152 ⫺0.77 ⫾ 1.27 6.4 ⫾ 3.1 95 ⫾ 23 14 ⫾ 7 155 ⫾ 99
Postoperative (6 months) mean ⫾ SEM
Normal range
Comments
2.31 ⫾ 0.09 1.06 ⫾ 0.16 120 ⫾ 32 6⫾2 0.869 ⫾ 0.145 ⫺0.35 ⫾ 1.17 3.7 ⫾ 1.8 92 ⫾ 21 16 ⫾ 7 42 ⫾ 18
2.15–2.60 0.90–1.45 70–210 2–13
Normalization* Normalization* 40% decrease* Normalization* 5% improvement* 50% improvement* * NS NS *
⬍6.75 50–110 10–40 10–65
* P ⬍0.05 (paired t test). NS ⫽ change not significant.
pg/mL in 4 cases, and PTH level higher than 40 pg/mL the day after the procedure in 5 cases. Patients with low preoperative PTH were excluded to avoid biasing the statistical analysis, and patients with high PTH immediately after the operation were excluded on the assumption that surgery had probably not been curative. This assumption was based on a previous study of PTH levels the day after surgery in patients with documented recurrence after surgery, study showing that although they were often clearly lower than preoperative levels, postoperative PTH levels were never lower than 40 pg/mL [7]. Our study population included 79 women and 18 men with a mean age of 59 years (Table 1). No patient was taking diuretics. The mean preoperative calcium level was 2.91 mmol/L, mean phosphate was 0.77 mmol/L, and mean PTH was 155 pg/mL. No renal leak hypercalciuria was observed. The operative finding was an isolated adenoma in 85 cases, cancer in 1 case, and multiple gland disease in 11 cases, including double adenoma in 1 case, diffuse hyperplasia in 4 cases, and mixed disease involving various degrees of associated adenoma and hyperplasia in 6 cases. All procedures were carried out with the patients under general anesthesia. All four glands were looked for in all cases but 8; these 8 were reoperations for which limited exploration was performed on the basis of the previous surgical findings or topographic examination. The indication for reoperation was persistent pHPT following incomplete surgical exploration in 4 cases and previous thyroid surgery in 4 cases. Measurements Calcium, phosphate, and PTH levels were measured the first day after surgery (D1), the second day after surgery (D2), 1 week after surgery (D8), and then monthly for 6
months (M1 to M6). Intact PTH was measured using a sandwich type or immunoradiometric assay, (ELSA, PTHCIS Bio International), which quantifies only intact PTH 1-84. The intra-assay and interassay coefficients of variation were 4% and 5%, respectively. The lower limit of sensitivity was 5 pg/mL. Normal values ranged between 10 and 65 pg/mL. Serum creatinine, alkaline phosphatase, osteocalcin, and vitamin D (25-OH cholecalciferol) levels were determined before surgery and 6 months after, as were urinary calcium levels. In 46 patients, bone density was evaluated by biphotonic absorptiometry using radiography before surgery and 7 months after with a Hologic QDR-1000 densitometer (Hologic Inc., Waltham, Massachusetts). These measurements were made from the anteroposterior angle on the L1 to L4 segment of the lumbar spine. Crushed vertebrae were not studied. To detect a normocalcemic persistent pHPT, oral calcium loading, which has been proved a valuable adjunct in confirming the diagnosis of pHPT in patients with normocalcemia [8,9], was tested at the sixth month in 71 patients. Calcium, phosphate, and PTH levels were measured every 30 minutes for 3 hours after oral ingestion of 20 mg/kg of calcium gluconolactate. Regulation was considered to be normal if PTH reached either 20 pg/mL before the end of testing or 50% of baseline. Statistical analysis Values are given as mean ⫾ SEM. Statistical analysis was performed using the paired t test for temporal changes within individuals, Student’s t test for two groups comparison, Newman-Keuls test for multiple groups comparison, and the chi-square test for categorical data. Correlation was calculated as Spearman’s coefficient. A P value of ⬍0.05 was considered statistically significant.
A. Denizot et al. / The American Journal of Surgery 182 (2001) 15–19
17
Fig. 1. Evolution of parathyroid parathormone (PTH) levels after surgery. Circles represent mean PTH in all 97 patients; squares represent mean PTH in 67 patients whose highest PTH value during follow-up was normal; and triangles represent mean PTH in 30 patients whose highest PTH value during follow-up exceeded normal range (normal PTH range 10 to 65 pg/mL). D ⫽ day; M ⫽ month.
Results
Comparison of patients with high and normal PTH maximum
Outcome of surgery The outcome of surgery is summarized in Table 1. Hypercalcemia was normalized within a few days after surgery in all 97 patients. Hypocalcemia requiring treatment was observed in 3 patients, including 1 in whom it persisted until the sixth month. Improvement in bone status at 6 months after surgery was documented by normalization of osteocalcin and alkaline phosphatase, increase in bone density, and decrease in z score. Urinary calcium levels significantly decreased. Renal function and 25-OH-vitamin D levels did not differ significantly before surgery and 6 months after. PTH kinetics were the same in all cases (Fig. 1). After surgery, PTH levels decreased; they were almost undetectable on D1. PTH levels began to increase on D2 and continued to rise until M2. PTH levels remained stable between M1 and M4 and then decreased slightly in M5 and M6. Elevated serum PTH levels were observed at least once in 30 of the 97 patients. The highest postoperative serum PTH level was 250 pg/mL, which is fourfold normal. In some patients elevated PTH level was continuous while in others it was intermittent. In all patients with intermittent elevation, however, serum PTH levels were always in the high normal range. In 2 patients elevated serum PTH levels occurred on the second day after surgery. Regulation of PTH was normal in 48 of the 71 patients (68%) who underwent the calcium loading test 6 months after surgery. No patient developed recurrent pHPT until June 1998, the end of the follow-up period.
We defined “PTH maximum” as the individual highest postoperative value. We compared data before surgery and PTH reactivity at M6 in two groups according to whether PTH maximum exceeded the normal range (high PTH maximum group; n ⫽ 30) or remained within the normal range (normal PTH maximum group; n ⫽ 67; Table 2). In these Table 2 Preoperative data and postoperative reactivity of parathyroid hormone (PTH) in patients whose highest value (PTH max) during follow-up was normal range and in patients whose highest value exceeded normal
Age (years) Male/female Weight of resection specimen (g) Calcium (mmol/L) Urinary calcium (mmol/day) Parathyroid hormone (pg/mL) Alkaline phosphatase (UI/L) Osteocalcin (ng/mL) Bone Density (g/cm3) Z Score Creatinine (mol/L) 25-OH-Vitamin D (ng/mL) Normal PTH regulation (% of group) * P ⬍0.05 (Student’s t test). † P ⫽ 0.05 (Student’s t test). ‡ P ⬍0.05 (chi-square test).
Normal PTH max (n ⫽ 67) mean ⫾ SEM
High PTH max (n ⫽ 30) mean ⫾ SEM
59 ⫾ 11 12/55 1771 ⫾ 1957 2.90 ⫾ 0.2 6.55 ⫾ 3.11 120 ⫾ 65 188 ⫾ 68 15 ⫾ 6 0.872 ⫾ 0.154 ⫺0.49 ⫾ 1.23 92 ⫾ 18 15 ⫾ 7 78%
59 ⫾ 9 6/24 2400 ⫾ 2541* 2.93 ⫾ 0.2 6.35 ⫾ 3.02 230 ⫾ 158* 223 ⫾ 74* 22 ⫾ 13* 0.783 ⫾ 0.135* ⫺1.35 ⫾ 1.07* 102 ⫾ 32* 12 ⫾ 6† 48%‡
18
A. Denizot et al. / The American Journal of Surgery 182 (2001) 15–19
two groups, age, gender, and serum and urinary calcium levels at the time of surgery did not significantly differ. Preoperative serum PTH levels and weight of the resection specimen were greater in the high PTH maximum group than in the normal PTH maximum group. The high PTH maximum group also demonstrated more extensive bone involvement (elevated alkaline phosphatases and osteocalcin and lower bone density), higher creatinine levels, and lower vitamin D levels in the preoperative period. Osteocalcin at M6 did not differ in both groups. Regulation of PTH ascertained by oral calcium loading test at M6 was normal in 12 of 25 patients (48%) in the high PTH maximum group and in 36 of 46 patients (78%) in the normal PTH maximum group. Creatinine and vitamin D levels Statistical analysis showed that preoperative PTH level correlated with creatinine levels and with bone status, reflected by osteocalcin and alkaline phosphatase levels (Spearman test, r ⫽ 0.47, 0.73, and 0.65 respectively, P ⬍0.05). For this reason another method was necessary to determine whether any of these preoperative factors plays a determinant role in the occurrence of postoperative elevated PTH levels. We studied the role of creatinine and vitamin D alone, because these two parameters did not change after surgery. Patients were divided into three groups. The first group included 14 patients who presented renal failure with creatinine values greater than 110 mol/L (high C group). The second group included 38 patients in whom renal function was normal but 25-OH-vitamin D levels were lower than 12 ng/mL (low D group). The third group included 33 patients in whom both renal function and vitamin D levels were normal (normal C and D group). Mean PTH maximum was 47 pg/mL in the normal C and D group. In comparison mean PTH maximum was 83 pg/mL in the high C group and 67 pg/mL in the low D group (P ⫽ 0.0009 and 0.004, respectively, Newman-Keuls test). The proportion of patients presenting serum PTH levels higher than 65 pg/mL was 50% (7 of 14) in the high C group, 42% (16 of 38) in the low D group, and 15% (5 of 32) in the normal C and D group (chi-square test, P ⬍0.01). Regulation of PTH during the calcium loading test was abnormal in 72% (8 of 11) of patients in the high C group, 37% (8 of 30) in the low D group, and 23% (5 of 22) in the normal C and D group (chi-square test, P ⬍0.01). Correlation of PTH maximum with preoperative data in the normal C and D group showed that postoperative PTH maximum depended on preoperative PTH values because mean PTH maximum was 39 pg/mL in 18 patients having preoperative PTH lower than 100 pg/mL as compared with 57 pg/mL in 15 patients having preoperative PTH higher than 100 pg/mL (Student’s t test, P ⫽ 0.03). In this group, no correlation (Spearman coefficient) was found between
PTH maximum and bone status as assessed by osteocalcin, alkaline phosphatases, or bone density.
Comments As in previous reports [1– 4], our series documents elevated serum PTH levels after surgery for pHPT in 30% of patients. We demonstrate that PTH maximum (ie, highest PTH value observed after surgery for each patient) correlates with three preoperative variables: highest PTH maximum values were observed in patients with high preoperative creatinine levels, low preoperative vitamin D levels, and high preoperative PTH levels. Several conditions may account for an increased PTH secretion: primary hyperparathyroidism, renal insufficiency, osteomalacia, and hypocalcemia. For the last three situations, increased PTH levels may be considered a normal adaptative response to supranormal stimuli, which should be identified prior to establishing the diagnosis of pHPT. Renal insufficiency even at early stages with glomerular filtration rate of 80 mL/mn/1.73 m2 has been associated with increased PTH levels [10]. The role of this factor in postoperative elevated PTH levels was confirmed by Lundgren [2] in a study with a mean follow-up of 15 years in which elevated PTH levels were associated with high creatinine levels several years after surgery for pHPT. Osteomalacia due to vitamin D deficiency is another documented cause of elevated PTH levels [11]. Bergenzfeld et al [1] emphasized the importance of low 25-OH-vitamin D levels in inducing elevated serum PTH levels after surgery. Indeed, we also found that elevated postoperative serum PTH levels correlated with low 25-OH-vitamin D levels in both the preoperative and postoperative period. Unlike them, however, we did not find an improvement in 25-OH-vitamin D levels after surgery for pHPT. We have no explanation for this discrepancy. Once believed to act simply as a precursor for more active products, 25-OHvitamin D is now thought to play an important role in absorption of calcium by the intestine [12]. Note also that pHPT is a possible cause of 25-OHvitamin D deficiency and renal dysfunction. In such cases pHPT could be concomitant with secondary hyperparathyroidism, which is revealed after surgical cure of pHPT, and which may be evidence of serious and neglected pHPT that should have been treated sooner. A third possible cause of adaptive elevated PTH level is hypocalcemia. Unlike Tisell [4] who studied ionized calcium levels, we found no evidence that hypocalcemia plays a role in postoperative elevated PTH levels. Nevertheless two mechanisms could lead to hypocalcemia after surgery for pHPT. The first is injury or functionally diminished activity of the residual parathyroid tissue, which had been suppressed by chronic hypercalcemia [13], but in this case PTH should also be low. The second possible cause for hypocalcemia is bone repair. Indeed, there is after surgery
A. Denizot et al. / The American Journal of Surgery 182 (2001) 15–19
for pHPT improvement in bone density with intense remineralization and a relative depletion of serum calcium [14]. These repair processes, called the “hungry bone syndrome,” can lead to hyperstimulation of residual parathyroid tissue. Although it would provide a convenient explanation for transient elevated PTH level after surgery, this hypothesis is not supported by our findings, as we observed no correlation between PTH maximum and indicators of preoperative bone status in 33 patients with normal renal function and normal vitamin D levels. Larger studies should be conducted to evaluate this hypothesis. In the absence of any adaptive cause for postoperative elevated PTH levels, a pathological process such as spontaneous secretion of PTH should be suspected. Failure of surgery to cure pHPT is more frequent in patients with multiple gland disease (15% failure versus up to 30% in cases involving familial polyendocrine disease) than in patients with adenoma (less than 5% failure) [15,16]. Persistent pHPT due to residual disease tissue can be suspected in patients presenting hypercalcemia immediately after surgery. Recurrent pHPT due to development of tumor in a gland that was normal at the end of the procedure is characterized by reappearance of hypercalcemia more than 6 months after surgery. In some cases procedures may have been mistakenly considered curative [17] either because some diseased tissue was overlooked or because hyperplasia was not adequately treated. In such cases, it has been suggested that the first manifestation of inadequate resection is elevation of serum PTH levels with normal calcemia [17]. Current follow-up, however, is insufficient for this hypothesis to be established, as recurrence may take many years [18]. Results of the oral calcium loading test 6 months after surgery demonstrated poor regulation of PTH in 52% of patients with high postoperative serum PTH levels. Most of these patients, however, also had high creatinine levels or low 25-OH-vitamin D levels, which both make this test inapplicable [19]. Intravenous calcium loading test would probably have been a more suitable method for evaluating patients with vitamin D deficiency. However, because it is difficult to interpret results, studying postoperative PTH reactivity in patients with elevated serum PTH levels seems of little value. Conclusion In the postoperative follow-up of patients undergoing parathyroidectomy for pHPT, serum PTH levels and calcemia must both be measured. Normal PTH and calcium levels in the first month after surgery is a reliable indicator that surgery was successful and that further surveillance is unnecessary at least for patients who had an isolated adenoma. In patients with elevated serum PTH but normal calcium levels, one should determine creatinine and 25-OHvitamin D levels. In most cases elevated PTH levels are an adaptive reaction to renal dysfunction or vitamin D deficiency. If, however, no adaptive cause is found, persistent or
19
recurrent hyperparathyroidism should be suspected and long-term surveillance is warranted. References [1] Bergenzeld A, Valdemarsson S, Tibblin S. Persistent serum elevated levels of intact parathyroid hormone after operation for sporadic adenoma: evidence of detrimental effect of severe parathyroid disease. Surgery 1996;119:624 –33. [2] Lundgren E, Rastad J, Ridefelt P, et al. Long-term effects of parathyroid operation on serum calcium and parathyroid hormone values in sporadic primary hyperparathyroidism. Surgery 1992;112:1123– 8. [3] Mulder H, Hackeng WHJ, Koster J, Van der Shaar H. Normocalcemia with persistent increase of parathyroid hormone: a prospective study. Calcif Tissue Int 1992;51:27–9. [4] Tisell LE, Jansson S, Nilsson B, et al. Transient rise in intact parathyroid hormone concentration after surgery for primary hyperparathyroidism. Br J Surg 1996;83:665–9. [5] Denizot A, Brue T, Henry JF. Hyperparathormone´mie et normocalce´mie apre`s traitement chirurgical d’une hyperparathyroı¨die primaire. Presse Med 1993;22:1930. [6] Clark OH, editor. Endocrine surgery of the thyroı¨d and parathyroid glands. Toronto: Mosby, 1985, pp 202–3. [7] Denizot A, de Boissezon C, Henry JF. Dosage per ope´ratoire de parathormone. Intereˆt dans la chirurgie de l’hyperparathyroı¨disme primaire. J Chir (Paris) 1995;132:346 –52. [8] Monchik JM, Lamberton RP, Roth U. Role of the oral calciumloading test with measurement of intact parathyroid hormone in the diagnoses of symptomatic subtle primary hyperparathyroidism. Surgery 1992;112:1103–10. [9] Thome JF, Bilezikian JP, Clemens TL, et al. Suppression of parathyroid hormone secretion with oral calcium in normal subjects and in patients with primary hyperparathyroidism. J Clin Endocrinol Metab 1990;70:951– 6. [10] Ritz E, Merke J, Mehls O. Secondary (renal) hyperparathyroidism— diagnosis and medical management. In: Rothmund M, Wells SA, editors. Progress in surgery, parathyroid surgery. Basel: Karger, 1986, pp 165– 85. [11] Aurbach GD, Marx SJ, Spiegel AM. Parathyroid hormone, calcitonin and the cholecalciferols. In: Wilson JD, Foster DW, editors. Williams textbook of endocrinology. 8th ed. Philadelphia: WB Saunders, 1992, p 1445. [12] Barger-Lux JM, Heaney RP, Lanspa SJ, et al. An investigation of sources of variation in calcium absorption efficiency. J Clin Endocrinol Metab 1995;80:406 –11. [13] Mandal AK, Udelsman R. Secondary hyperparathyroidism is an expected consequence of parathyroidectomy for primary hyperparathyroidism: a prospective study. Surgery 1998;124:1021–7. [14] Silverberg SJ, Gartenberg F, Jacobs TP et al. Increased bone mineral density after parathyroidectomy in primary hyperparathyroidism. J Clin Endocrinol Metab 1995;80:729 –34. [15] Granberg P, Cedermark B, Farnbo L, Hamberger B. Surgery for primary hyperparathyroidism: indications, intra operative decisionmaking and results. In: Rothmund M, Wells SA, editors. Progress in surgery, parathyroid surgery. Basel: Karger, 1986, pp 93–105. [16] Henry JF, Denizot A. Cervicotomie premie`re pour hyperparathyroı¨disme primaire. In: Barbier J, Henry JF, editors. L’hyperparathyroı¨disme primaire, monographies de l’association franc¸aise de chirurgie. Paris: Spinger-Verlag, 1991, pp 75–98. [17] Bruining HA, Ong GL. Causes des echecs en chirurgie parathyroı¨dienne. Chirurgie 1987;113:549 –55. [18] Bruining HA, Bierckenha¨ger JC, Ong GL, Lamberts SW. Causes of failure in operations for hyperparathyroidism. Surgery 1987;101:562– 65. [19] Valette S, Feraud C, Conte-Devolx B. Calcium charge test in secondary hyperparathyroidism [letter]. Presse Med 1997;26:164.
Scientific paper
Normocalcemia with elevated parathyroid hormone levels after surgical treatment of primary hyperparathyroidism Anne Denizot, M.D.a,*, Marco Pucini, M.D.a, Christophe Chagnaud, M.D.b, Genevie`ve Botti, M.D.c, Jean-Franc¸ois Henry, M.D.a a
General Surgery and Endocrinology Service, CHU Timone, Boulevard Jean Moulin, 13385 Marseille Cedex 5, France b Radiology Service, CHU Timone, Boulevard Jean Moulin, 13385 Marseille Cedex 5, France c Department of Medical Information, CHU Timone, Boulevard Jean Moulin, 13385 Marseille Cedex 5, France Manuscript received September 27, 2000; revised manuscript February 12, 2001
Abstract Background: Thirty percent of patients who undergo successful parathyroidectomy for primary hyperparathyroidism show unexplained elevated postoperative serum parathyroid hormone (PTH) levels despite normocalcemia. Methods: PTH levels were measured monthly in 97 patients for 6 months after parathyroidectomy. Renal function, 25-OH-vitamin D levels, serum alkaline phosphatase levels, osteocalcin, and bone densitometry were evaluated before and 6 months after surgery. PTH reactivity to calcium loading was tested at the sixth month. Results: Thirty patients had elevated PTH levels despite normocalcemia after parathyroidectomy. Before surgery, these 30 patients had higher PTH and creatinine levels, lower vitamin D levels, and more extensive bone involvement than those with normal postoperative PTH levels. In patients with normal renal function and normal vitamin D levels, postoperative PTH values correlated with preoperative PTH levels but not with bone disease. Conclusion: In most cases, elevated PTH levels after surgery is an adaptive reaction to renal dysfunction or vitamin D deficiency. If no adaptive cause can be found, persistent hyperparathyroidism must be suspected. © 2001 Excerpta Medica, Inc. All rights reserved. Keywords: Parathyroid hormone; Parathyroidectomy; Primary hyperparathyroidism; Postoperative period
Elevated serum parathyroid hormone (PTH) levels are observed after surgery in 30% of patients who undergo parathyroidectomy for primary hyperparathyroidism (pHPT) despite normalization of serum calcium levels [1-5]. The mechanisms underlying elevated serum PTH levels after parathyroidectomy are unclear. Two explanations are possible. The first involves a pathologic process and attributes the elevated serum PTH levels to persistent or recurrent pHPT. Indeed, cases of pHPT with normal calcium levels have been reported [6]. The second explanation involves an adaptative process and attributes elevated serum PTH to a response to physiological changes. In this regard increased PTH levels have been observed in patients with hypocalcemia, renal insufficiency, and vitamin D deficiency. * Corresponding author. Tel.: ⫹33(0)491-968626, ⫹33(0)491-964932; fax: ⫹33(0)491-968170. Present address: General Surgery Service, CHU Nord, Chemin des Bourrellys, 13915 Marseille Cedex 20, France. E-mail address: [email protected]
In the present prospective study we monitored serum PTH levels during the first 6 months after parathyroidectomy for pHPT and attempted to determine if any increase correlated with preoperative renal function, urinary calcium level, serum vitamin D level, bone status, or postoperative PTH reactivity.
Patients and methods Study population Of 120 patients who underwent surgery for suspected pHPT between July 20, 1993, and December 31, 1994, 97 were included in this study. The remaining 23 patients were excluded for the following reasons: misdiagnosis (hypocalciuria-hypercalcemia) in 1 case, association of pHPT with another cause of hypercalcemia in 1 case, loss from follow-up in 12 cases, preoperative PTH level lower than 40
0002-9610/01/$ – see front matter © 2001 Excerpta Medica, Inc. All rights reserved. PII: S 0 0 0 2 - 9 6 1 0 ( 0 1 ) 0 0 6 6 4 - X
16
A. Denizot et al. / The American Journal of Surgery 182 (2001) 15–19
Table 1 Outcome of parathyroidectomy for primary hyperparathyroidism in 97 patients Preoperative mean ⫾ SEM Age (years) Weight of resection specimen (g) Calcium (mmol/L) Phosphate (mmol/L) Alkaline phosphatase (UI/L) Osteocalcin (ng/mL) Bone Density (g/cm3) Z Score Urinary calcium (mmol/day) Creatinine (mol/L) 25-OH-Vitamin D (ng/mL) Parathyroid hormone (pg/mL)
59 ⫾ 10 1973 ⫾ 2156 2.91 ⫾ 0.21 0.77 ⫾ 0.13 200 ⫾ 71 17 ⫾ 8 0.837 ⫾ 0.152 ⫺0.77 ⫾ 1.27 6.4 ⫾ 3.1 95 ⫾ 23 14 ⫾ 7 155 ⫾ 99
Postoperative (6 months) mean ⫾ SEM
Normal range
Comments
2.31 ⫾ 0.09 1.06 ⫾ 0.16 120 ⫾ 32 6⫾2 0.869 ⫾ 0.145 ⫺0.35 ⫾ 1.17 3.7 ⫾ 1.8 92 ⫾ 21 16 ⫾ 7 42 ⫾ 18
2.15–2.60 0.90–1.45 70–210 2–13
Normalization* Normalization* 40% decrease* Normalization* 5% improvement* 50% improvement* * NS NS *
⬍6.75 50–110 10–40 10–65
* P ⬍0.05 (paired t test). NS ⫽ change not significant.
pg/mL in 4 cases, and PTH level higher than 40 pg/mL the day after the procedure in 5 cases. Patients with low preoperative PTH were excluded to avoid biasing the statistical analysis, and patients with high PTH immediately after the operation were excluded on the assumption that surgery had probably not been curative. This assumption was based on a previous study of PTH levels the day after surgery in patients with documented recurrence after surgery, study showing that although they were often clearly lower than preoperative levels, postoperative PTH levels were never lower than 40 pg/mL [7]. Our study population included 79 women and 18 men with a mean age of 59 years (Table 1). No patient was taking diuretics. The mean preoperative calcium level was 2.91 mmol/L, mean phosphate was 0.77 mmol/L, and mean PTH was 155 pg/mL. No renal leak hypercalciuria was observed. The operative finding was an isolated adenoma in 85 cases, cancer in 1 case, and multiple gland disease in 11 cases, including double adenoma in 1 case, diffuse hyperplasia in 4 cases, and mixed disease involving various degrees of associated adenoma and hyperplasia in 6 cases. All procedures were carried out with the patients under general anesthesia. All four glands were looked for in all cases but 8; these 8 were reoperations for which limited exploration was performed on the basis of the previous surgical findings or topographic examination. The indication for reoperation was persistent pHPT following incomplete surgical exploration in 4 cases and previous thyroid surgery in 4 cases. Measurements Calcium, phosphate, and PTH levels were measured the first day after surgery (D1), the second day after surgery (D2), 1 week after surgery (D8), and then monthly for 6
months (M1 to M6). Intact PTH was measured using a sandwich type or immunoradiometric assay, (ELSA, PTHCIS Bio International), which quantifies only intact PTH 1-84. The intra-assay and interassay coefficients of variation were 4% and 5%, respectively. The lower limit of sensitivity was 5 pg/mL. Normal values ranged between 10 and 65 pg/mL. Serum creatinine, alkaline phosphatase, osteocalcin, and vitamin D (25-OH cholecalciferol) levels were determined before surgery and 6 months after, as were urinary calcium levels. In 46 patients, bone density was evaluated by biphotonic absorptiometry using radiography before surgery and 7 months after with a Hologic QDR-1000 densitometer (Hologic Inc., Waltham, Massachusetts). These measurements were made from the anteroposterior angle on the L1 to L4 segment of the lumbar spine. Crushed vertebrae were not studied. To detect a normocalcemic persistent pHPT, oral calcium loading, which has been proved a valuable adjunct in confirming the diagnosis of pHPT in patients with normocalcemia [8,9], was tested at the sixth month in 71 patients. Calcium, phosphate, and PTH levels were measured every 30 minutes for 3 hours after oral ingestion of 20 mg/kg of calcium gluconolactate. Regulation was considered to be normal if PTH reached either 20 pg/mL before the end of testing or 50% of baseline. Statistical analysis Values are given as mean ⫾ SEM. Statistical analysis was performed using the paired t test for temporal changes within individuals, Student’s t test for two groups comparison, Newman-Keuls test for multiple groups comparison, and the chi-square test for categorical data. Correlation was calculated as Spearman’s coefficient. A P value of ⬍0.05 was considered statistically significant.
A. Denizot et al. / The American Journal of Surgery 182 (2001) 15–19
17
Fig. 1. Evolution of parathyroid parathormone (PTH) levels after surgery. Circles represent mean PTH in all 97 patients; squares represent mean PTH in 67 patients whose highest PTH value during follow-up was normal; and triangles represent mean PTH in 30 patients whose highest PTH value during follow-up exceeded normal range (normal PTH range 10 to 65 pg/mL). D ⫽ day; M ⫽ month.
Results
Comparison of patients with high and normal PTH maximum
Outcome of surgery The outcome of surgery is summarized in Table 1. Hypercalcemia was normalized within a few days after surgery in all 97 patients. Hypocalcemia requiring treatment was observed in 3 patients, including 1 in whom it persisted until the sixth month. Improvement in bone status at 6 months after surgery was documented by normalization of osteocalcin and alkaline phosphatase, increase in bone density, and decrease in z score. Urinary calcium levels significantly decreased. Renal function and 25-OH-vitamin D levels did not differ significantly before surgery and 6 months after. PTH kinetics were the same in all cases (Fig. 1). After surgery, PTH levels decreased; they were almost undetectable on D1. PTH levels began to increase on D2 and continued to rise until M2. PTH levels remained stable between M1 and M4 and then decreased slightly in M5 and M6. Elevated serum PTH levels were observed at least once in 30 of the 97 patients. The highest postoperative serum PTH level was 250 pg/mL, which is fourfold normal. In some patients elevated PTH level was continuous while in others it was intermittent. In all patients with intermittent elevation, however, serum PTH levels were always in the high normal range. In 2 patients elevated serum PTH levels occurred on the second day after surgery. Regulation of PTH was normal in 48 of the 71 patients (68%) who underwent the calcium loading test 6 months after surgery. No patient developed recurrent pHPT until June 1998, the end of the follow-up period.
We defined “PTH maximum” as the individual highest postoperative value. We compared data before surgery and PTH reactivity at M6 in two groups according to whether PTH maximum exceeded the normal range (high PTH maximum group; n ⫽ 30) or remained within the normal range (normal PTH maximum group; n ⫽ 67; Table 2). In these Table 2 Preoperative data and postoperative reactivity of parathyroid hormone (PTH) in patients whose highest value (PTH max) during follow-up was normal range and in patients whose highest value exceeded normal
Age (years) Male/female Weight of resection specimen (g) Calcium (mmol/L) Urinary calcium (mmol/day) Parathyroid hormone (pg/mL) Alkaline phosphatase (UI/L) Osteocalcin (ng/mL) Bone Density (g/cm3) Z Score Creatinine (mol/L) 25-OH-Vitamin D (ng/mL) Normal PTH regulation (% of group) * P ⬍0.05 (Student’s t test). † P ⫽ 0.05 (Student’s t test). ‡ P ⬍0.05 (chi-square test).
Normal PTH max (n ⫽ 67) mean ⫾ SEM
High PTH max (n ⫽ 30) mean ⫾ SEM
59 ⫾ 11 12/55 1771 ⫾ 1957 2.90 ⫾ 0.2 6.55 ⫾ 3.11 120 ⫾ 65 188 ⫾ 68 15 ⫾ 6 0.872 ⫾ 0.154 ⫺0.49 ⫾ 1.23 92 ⫾ 18 15 ⫾ 7 78%
59 ⫾ 9 6/24 2400 ⫾ 2541* 2.93 ⫾ 0.2 6.35 ⫾ 3.02 230 ⫾ 158* 223 ⫾ 74* 22 ⫾ 13* 0.783 ⫾ 0.135* ⫺1.35 ⫾ 1.07* 102 ⫾ 32* 12 ⫾ 6† 48%‡
18
A. Denizot et al. / The American Journal of Surgery 182 (2001) 15–19
two groups, age, gender, and serum and urinary calcium levels at the time of surgery did not significantly differ. Preoperative serum PTH levels and weight of the resection specimen were greater in the high PTH maximum group than in the normal PTH maximum group. The high PTH maximum group also demonstrated more extensive bone involvement (elevated alkaline phosphatases and osteocalcin and lower bone density), higher creatinine levels, and lower vitamin D levels in the preoperative period. Osteocalcin at M6 did not differ in both groups. Regulation of PTH ascertained by oral calcium loading test at M6 was normal in 12 of 25 patients (48%) in the high PTH maximum group and in 36 of 46 patients (78%) in the normal PTH maximum group. Creatinine and vitamin D levels Statistical analysis showed that preoperative PTH level correlated with creatinine levels and with bone status, reflected by osteocalcin and alkaline phosphatase levels (Spearman test, r ⫽ 0.47, 0.73, and 0.65 respectively, P ⬍0.05). For this reason another method was necessary to determine whether any of these preoperative factors plays a determinant role in the occurrence of postoperative elevated PTH levels. We studied the role of creatinine and vitamin D alone, because these two parameters did not change after surgery. Patients were divided into three groups. The first group included 14 patients who presented renal failure with creatinine values greater than 110 mol/L (high C group). The second group included 38 patients in whom renal function was normal but 25-OH-vitamin D levels were lower than 12 ng/mL (low D group). The third group included 33 patients in whom both renal function and vitamin D levels were normal (normal C and D group). Mean PTH maximum was 47 pg/mL in the normal C and D group. In comparison mean PTH maximum was 83 pg/mL in the high C group and 67 pg/mL in the low D group (P ⫽ 0.0009 and 0.004, respectively, Newman-Keuls test). The proportion of patients presenting serum PTH levels higher than 65 pg/mL was 50% (7 of 14) in the high C group, 42% (16 of 38) in the low D group, and 15% (5 of 32) in the normal C and D group (chi-square test, P ⬍0.01). Regulation of PTH during the calcium loading test was abnormal in 72% (8 of 11) of patients in the high C group, 37% (8 of 30) in the low D group, and 23% (5 of 22) in the normal C and D group (chi-square test, P ⬍0.01). Correlation of PTH maximum with preoperative data in the normal C and D group showed that postoperative PTH maximum depended on preoperative PTH values because mean PTH maximum was 39 pg/mL in 18 patients having preoperative PTH lower than 100 pg/mL as compared with 57 pg/mL in 15 patients having preoperative PTH higher than 100 pg/mL (Student’s t test, P ⫽ 0.03). In this group, no correlation (Spearman coefficient) was found between
PTH maximum and bone status as assessed by osteocalcin, alkaline phosphatases, or bone density.
Comments As in previous reports [1– 4], our series documents elevated serum PTH levels after surgery for pHPT in 30% of patients. We demonstrate that PTH maximum (ie, highest PTH value observed after surgery for each patient) correlates with three preoperative variables: highest PTH maximum values were observed in patients with high preoperative creatinine levels, low preoperative vitamin D levels, and high preoperative PTH levels. Several conditions may account for an increased PTH secretion: primary hyperparathyroidism, renal insufficiency, osteomalacia, and hypocalcemia. For the last three situations, increased PTH levels may be considered a normal adaptative response to supranormal stimuli, which should be identified prior to establishing the diagnosis of pHPT. Renal insufficiency even at early stages with glomerular filtration rate of 80 mL/mn/1.73 m2 has been associated with increased PTH levels [10]. The role of this factor in postoperative elevated PTH levels was confirmed by Lundgren [2] in a study with a mean follow-up of 15 years in which elevated PTH levels were associated with high creatinine levels several years after surgery for pHPT. Osteomalacia due to vitamin D deficiency is another documented cause of elevated PTH levels [11]. Bergenzfeld et al [1] emphasized the importance of low 25-OH-vitamin D levels in inducing elevated serum PTH levels after surgery. Indeed, we also found that elevated postoperative serum PTH levels correlated with low 25-OH-vitamin D levels in both the preoperative and postoperative period. Unlike them, however, we did not find an improvement in 25-OH-vitamin D levels after surgery for pHPT. We have no explanation for this discrepancy. Once believed to act simply as a precursor for more active products, 25-OHvitamin D is now thought to play an important role in absorption of calcium by the intestine [12]. Note also that pHPT is a possible cause of 25-OHvitamin D deficiency and renal dysfunction. In such cases pHPT could be concomitant with secondary hyperparathyroidism, which is revealed after surgical cure of pHPT, and which may be evidence of serious and neglected pHPT that should have been treated sooner. A third possible cause of adaptive elevated PTH level is hypocalcemia. Unlike Tisell [4] who studied ionized calcium levels, we found no evidence that hypocalcemia plays a role in postoperative elevated PTH levels. Nevertheless two mechanisms could lead to hypocalcemia after surgery for pHPT. The first is injury or functionally diminished activity of the residual parathyroid tissue, which had been suppressed by chronic hypercalcemia [13], but in this case PTH should also be low. The second possible cause for hypocalcemia is bone repair. Indeed, there is after surgery
A. Denizot et al. / The American Journal of Surgery 182 (2001) 15–19
for pHPT improvement in bone density with intense remineralization and a relative depletion of serum calcium [14]. These repair processes, called the “hungry bone syndrome,” can lead to hyperstimulation of residual parathyroid tissue. Although it would provide a convenient explanation for transient elevated PTH level after surgery, this hypothesis is not supported by our findings, as we observed no correlation between PTH maximum and indicators of preoperative bone status in 33 patients with normal renal function and normal vitamin D levels. Larger studies should be conducted to evaluate this hypothesis. In the absence of any adaptive cause for postoperative elevated PTH levels, a pathological process such as spontaneous secretion of PTH should be suspected. Failure of surgery to cure pHPT is more frequent in patients with multiple gland disease (15% failure versus up to 30% in cases involving familial polyendocrine disease) than in patients with adenoma (less than 5% failure) [15,16]. Persistent pHPT due to residual disease tissue can be suspected in patients presenting hypercalcemia immediately after surgery. Recurrent pHPT due to development of tumor in a gland that was normal at the end of the procedure is characterized by reappearance of hypercalcemia more than 6 months after surgery. In some cases procedures may have been mistakenly considered curative [17] either because some diseased tissue was overlooked or because hyperplasia was not adequately treated. In such cases, it has been suggested that the first manifestation of inadequate resection is elevation of serum PTH levels with normal calcemia [17]. Current follow-up, however, is insufficient for this hypothesis to be established, as recurrence may take many years [18]. Results of the oral calcium loading test 6 months after surgery demonstrated poor regulation of PTH in 52% of patients with high postoperative serum PTH levels. Most of these patients, however, also had high creatinine levels or low 25-OH-vitamin D levels, which both make this test inapplicable [19]. Intravenous calcium loading test would probably have been a more suitable method for evaluating patients with vitamin D deficiency. However, because it is difficult to interpret results, studying postoperative PTH reactivity in patients with elevated serum PTH levels seems of little value. Conclusion In the postoperative follow-up of patients undergoing parathyroidectomy for pHPT, serum PTH levels and calcemia must both be measured. Normal PTH and calcium levels in the first month after surgery is a reliable indicator that surgery was successful and that further surveillance is unnecessary at least for patients who had an isolated adenoma. In patients with elevated serum PTH but normal calcium levels, one should determine creatinine and 25-OHvitamin D levels. In most cases elevated PTH levels are an adaptive reaction to renal dysfunction or vitamin D deficiency. If, however, no adaptive cause is found, persistent or
19
recurrent hyperparathyroidism should be suspected and long-term surveillance is warranted. References [1] Bergenzeld A, Valdemarsson S, Tibblin S. Persistent serum elevated levels of intact parathyroid hormone after operation for sporadic adenoma: evidence of detrimental effect of severe parathyroid disease. Surgery 1996;119:624 –33. [2] Lundgren E, Rastad J, Ridefelt P, et al. Long-term effects of parathyroid operation on serum calcium and parathyroid hormone values in sporadic primary hyperparathyroidism. Surgery 1992;112:1123– 8. [3] Mulder H, Hackeng WHJ, Koster J, Van der Shaar H. Normocalcemia with persistent increase of parathyroid hormone: a prospective study. Calcif Tissue Int 1992;51:27–9. [4] Tisell LE, Jansson S, Nilsson B, et al. Transient rise in intact parathyroid hormone concentration after surgery for primary hyperparathyroidism. Br J Surg 1996;83:665–9. [5] Denizot A, Brue T, Henry JF. Hyperparathormone´mie et normocalce´mie apre`s traitement chirurgical d’une hyperparathyroı¨die primaire. Presse Med 1993;22:1930. [6] Clark OH, editor. Endocrine surgery of the thyroı¨d and parathyroid glands. Toronto: Mosby, 1985, pp 202–3. [7] Denizot A, de Boissezon C, Henry JF. Dosage per ope´ratoire de parathormone. Intereˆt dans la chirurgie de l’hyperparathyroı¨disme primaire. J Chir (Paris) 1995;132:346 –52. [8] Monchik JM, Lamberton RP, Roth U. Role of the oral calciumloading test with measurement of intact parathyroid hormone in the diagnoses of symptomatic subtle primary hyperparathyroidism. Surgery 1992;112:1103–10. [9] Thome JF, Bilezikian JP, Clemens TL, et al. Suppression of parathyroid hormone secretion with oral calcium in normal subjects and in patients with primary hyperparathyroidism. J Clin Endocrinol Metab 1990;70:951– 6. [10] Ritz E, Merke J, Mehls O. Secondary (renal) hyperparathyroidism— diagnosis and medical management. In: Rothmund M, Wells SA, editors. Progress in surgery, parathyroid surgery. Basel: Karger, 1986, pp 165– 85. [11] Aurbach GD, Marx SJ, Spiegel AM. Parathyroid hormone, calcitonin and the cholecalciferols. In: Wilson JD, Foster DW, editors. Williams textbook of endocrinology. 8th ed. Philadelphia: WB Saunders, 1992, p 1445. [12] Barger-Lux JM, Heaney RP, Lanspa SJ, et al. An investigation of sources of variation in calcium absorption efficiency. J Clin Endocrinol Metab 1995;80:406 –11. [13] Mandal AK, Udelsman R. Secondary hyperparathyroidism is an expected consequence of parathyroidectomy for primary hyperparathyroidism: a prospective study. Surgery 1998;124:1021–7. [14] Silverberg SJ, Gartenberg F, Jacobs TP et al. Increased bone mineral density after parathyroidectomy in primary hyperparathyroidism. J Clin Endocrinol Metab 1995;80:729 –34. [15] Granberg P, Cedermark B, Farnbo L, Hamberger B. Surgery for primary hyperparathyroidism: indications, intra operative decisionmaking and results. In: Rothmund M, Wells SA, editors. Progress in surgery, parathyroid surgery. Basel: Karger, 1986, pp 93–105. [16] Henry JF, Denizot A. Cervicotomie premie`re pour hyperparathyroı¨disme primaire. In: Barbier J, Henry JF, editors. L’hyperparathyroı¨disme primaire, monographies de l’association franc¸aise de chirurgie. Paris: Spinger-Verlag, 1991, pp 75–98. [17] Bruining HA, Ong GL. Causes des echecs en chirurgie parathyroı¨dienne. Chirurgie 1987;113:549 –55. [18] Bruining HA, Bierckenha¨ger JC, Ong GL, Lamberts SW. Causes of failure in operations for hyperparathyroidism. Surgery 1987;101:562– 65. [19] Valette S, Feraud C, Conte-Devolx B. Calcium charge test in secondary hyperparathyroidism [letter]. Presse Med 1997;26:164.