Intraoperative decay profile of intact (1-84) parathyroid hormone in surgery for renal hyperparathyroidism—a consecutive series of 80 patients

Intraoperative decay profile of intact (1-84) parathyroid hormone in surgery for renal hyperparathyroidism—a consecutive series of 80 patients Jonathan Lokey, MD, François Pattou, MD, Alejandro Mondragon-Sanchez, MD, Michele Minuto, MD, Barbara Mullineris, MD, François Wambergue, MD, Philippe Foissac-Geroux, MD, Christian Noel, MD, Henri Le Monies de Sagazan, MD, Pierre VanHille, MD, and Charles A. G. Proye, MD, Lille, France

Background. The utility of intraoperative parathyroid hormone (PTH) monitoring is unclear in the surgical management of renal hyperparathyroidism. Our goal was to define the normal pattern of decay during operation for renal hyperparathyroidism by using the rapid intact (1-84) parathyroid hormone (PTH) assay. Methods. Eighty consecutive patients underwent neck exploration for renal hyperparathyroidism. Intact PTH levels were monitored with a rapid immunochemiluminometric assay. Samples were assayed at the induction of anesthesia, after dissection before resection, and 20 and 40 minutes after resection. Followup ranged from 3 to 24 months. Results. Twenty minutes after resection, PTH levels remained many-fold supranormal. Seventy-seven patients (96%) were cured. Of these, 75 patients (94%) had PTH decay of more than 50% from the preoperative level; 74 (99%) were cured. Only 1 of 3 patients (33%) in whom the PTH level decreased less than 40% from the preoperative level was cured. Two patients had intermediate values and both were cured. Conclusions. The intraoperative decay of PTH during operation for renal hyperparathyroidism is slower than for patients with normal renal function. However, 20 minutes after resection, a decline to less than 50% of the preoperative level predicts cure, while a level greater than 60% predicts failure. (Surgery 2000;128:1029-34.) From the Surgical Professorial Unit, General and Endocrine Surgery, University of Lille, France

SECONDARY HYPERPARATHYROIDISM (HPT) is a pervasive complication of chronic renal failure. Medical management is the mainstay, but in 5% to 10% of cases, recalcitrant symptoms such as pruritis and bone pain require surgical therapy.1 In experienced hands, both subtotal parathyroidectomy and total parathyroidectomy with autotransplantation have good results.2 Recurrence indicates an inadequate resection of hyperfunctioning tissue (often a remnant that is too large or that has furPresented at the 21st Annual Meeting of the American Association of Endocrine Surgeons (jointly hosted with the British Association of Endocrine Surgeons), London, United Kingdom, and Lille, France, May 22-25, 2000. Reprint requests: Charles A. G. Proye, Service de Chirurgie Generale et Endocrinienne, Hôpital Claude Huriez, F59037 Lille, France. Copyright © 2000 by Mosby, Inc. 0039-6060/2000/$12.00 + 0 doi:10.1067/msy.2000.110431

11/6/110431

ther hypertrophied), a missed gland, or a supernumerary gland.3,4 The rapid intraoperative determination of intact parathyroid hormone (PTH) with a 2-site immunochemiluminometric assay (ICMA) has become widely established in the operation of primary HPT.5 Removal of all hyperfunctioning parathyroid tissue is confirmed by a prompt decline in circulating intact PTH within 5 to 10 minutes.6-9 However, the role of the quick intact PTH assay is not clearly defined in the treatment of renal HPT. Early research with quick PTH in secondary HPT demonstrated a stepwise decrease in circulating intact PTH corresponding to sequential gland resection.10 Subsequent studies have shown a reproducible decrease in PTH by using the rapid ICMA, but these have been small series with limited follow-up.11,12 A decline in serum PTH 20 minutes after excision to less than 50% of the original value has been observed, but the ability to detect SURGERY 1029

1030 Lokey et al

Table I. Patient characteristics Age (mean) 49.6 (range, 14-78) Sex 41 male; 39 female Preoperative intact PTH 947 pg/mL (range, 65-2960) Preoperative calcium 10.3 mg/dL (range, 5.9-12.5) Secondary HPT 73 Tertiary HPT 7 Years of dialysis 7.2 (range, 1-18) Years of renal graft (n = 15) 2 (range, 1-15)

an inadequate parathyroid resection or supernumerary glands has not been proved. The goal of our study was to define the decay profile of intact PTH in patients undergoing subtotal parathyroidectomy for renal HPT and to evaluate the utility of the rapid intraoperative PTH assay in such interventions. PATIENTS Eighty consecutive patients undergoing parathyroidectomy for renal HPT from January 1998 to February 2000 were studied. Patient demographics are shown in Table I. Typical for this patient population, the mean PTH value was markedly elevated, nearly 16-fold above the normal value, and few patients demonstrated hypercalcemia. Seven of 15 patients who had undergone renal transplantation had functioning grafts. Seventy-four patients (93%) underwent primary (first-time) neck exploration in which the therapeutic goal was bilateral exploration, subtotal resection of 4 or more parathyroid glands, and bilateral transcervical thymectomy. Four patients (5%) had 1 previous parathyroid operation (including 3 referrals), and 2 (2%) were explored for the third time (both referred.) METHODS The PTH levels obtained in the preoperative evaluation (PTHpreop) were determined by a standard immunoradiometric assay. Intraoperative intact PTH levels were determined by using a rapid 2-site ICMA (Allegro PTH and intraoperative PTH; Nichols Institute Diagnostics, San Juan Capistrano, Calif). The normal range in our hospital is 10 to 60 pg/mL. This technique and its accuracy have been previously described.6 Baseline peripheral venous blood samples were obtained just after induction of anesthesia (PTHind), and after all parathyroid tissue had been identified but before excision. Subtotal parathyroidectomy, leaving a viable remnant approximately twice the size of a normal gland (~60-100 mg) and bilateral transcervical thymectomy were performed. Half of 1 parathyroid

Surgery December 2000

specimen was cryopreserved. No frozen section analysis was used. PTH values were determined 20 (PTH20) and 40 minutes after final resection, and later samples were also obtained when necessary. PTH levels were expressed as absolute values and as the percentage decline from pre-excision values. All patients had postoperative laryngoscopy. PTH and serum calcium values were determined on the first postoperative day and again later in the first postoperative month. Intermediate follow-up ranged from 3 to 24 months. RESULTS Overall surgical results. Seventy-seven patients (96%) were considered cured after operation. Cure was defined as a PTH level that remained less than 4 times the normal level and the absence of hypercalcemia. A slightly supranormal level of PTH secretion was the specific goal of surgical therapy to avoid the postoperative development of adynamic bone disease.13 Three patients (4%) had persistent HPT (ie, PTH greater than 4 times the normal level within 6 months of operation). There were no recurrences during the follow-up period (HPT occurring more than 6 months after operation). A summary of the surgical failures is presented in Table II. One patient had a permanent unilateral recurrent laryngeal nerve palsy. One patient (a referral for a third-time operation) required eventual autotransplantation of cryopreserved tissue for hypoparathyroidism. There was 1 hematoma, in a patient who underwent a concomitant thyroidectomy. There were no deaths. Number of parathyroid glands identified. Overall, at least 4 glands were identified in 77 patients (96%). Supernumerary glands were found in 28 patients (35%), 3 grossly evident and 25 small thymic rests. In patients undergoing exploration for the first time, 4 or more glands were seen in 71 of 74 patients (96%), and in the remaining 3 patients, 3 glands were identified. In patients undergoing repeat operation, between 1 and 3 glands were identified at the time of reoperation, accounting for a total of 4 glands in 2 cases and 5 glands in each of the remaining 4 patients. Correlation of PTHpreop and PTHind. There was some degree of correlation (r = 0.66) between the preoperative PTH level and the level determined after induction of anesthesia (Fig 1). However the values did vary by as much as 3-fold. The preoperative value (PTHpreop) was greater than the postinduction level (PTHind) in 50 of 80 cases (63%). Intraoperative PTH. PTH levels 20 minutes after resection remained quite elevated (mean, 182 ± 160 pg/mL), but significantly reduced from pre-

Lokey et al 1031

Surgery Volume 128, Number 6

Fig 1. Correlation of preoperative and postinduction PTH levels. PTH levels determined preoperatively by immunoradiometric assay are plotted against levels obtained immediately after induction of anesthesia by immunochemiluminometric assay.

Table II. Surgical failures Patient AC JW HC

Preoperative PTH (pg/mL)

PTH20 (pg/mL)

700 711 405 (903)*

147 491 387

Decay in PTH Parathyroids (% of PTHpreop) (n) 79 31 4 (57)*

3 4 3 (4)*

Persistence or recurrence?

Reason for failure

Predictive value of I/O PTH

Persistence Persistence Persistence

Missed gland Supernumerary Missed gland

Incorrect Correct Correct

*See results section

excision values. Seventy-five patients (94%) had PTH levels decay to more than 50% of the preoperative value (PTH20 < 50% PTHpreop), and all but 1 (99%) of these patients were cured (patient A.C. in Table II). Four or more glands were identified in all but 2 patients in this group. In the 1 surgical failure in this group, only 3 parathyroid glands were identified at operation. Three patients (4%) had PTH20 that remained greater than 60% of the preoperative value (PTH20 > 60% PTHpreop). Two of these 3 patients (67%) were not cured. One of the 2 surgical failures in this group (patient J.W. in Table II) had a preoperatively identified mediastinal fifth gland, and a combined cervical and mediastinal procedure was planned. After successful cervical exploration and the resection of 4 enlarged glands, the planned resection of the fifth gland was postponed because

of the patient’s tenuous medical condition. A large mediastinal parathyroid was removed the following week. Fig 2 depicts the PTH level through the patient’s course. Twenty minutes after resection of 4 cervical glands, the PTH dropped noticeably, but not below 50% of the preoperative baseline. Twenty minutes after the mediastinal gland was resected; the PTH had dropped more than 50% of the new baseline value. In the second surgical failure of this group (patient H.C. in Table II), there was discordance between the 2 baseline PTH values. Fig 3 depicts the PTH levels in this case. The postinduction PTH level was more than 2 times greater than the preoperative value. The PTH 20 minutes after resection had fallen to less than 50% of the postinduction level, but not of the preoperative level. Four parathyroid glands were thought to have been resected, the preoperative level was

1032 Lokey et al

Surgery December 2000

Fig 2. Intraoperative PTH of patient JW. This patient with secondary HPT had 5 glands identified by preoperative sestamibi scintigraphy. A cervicomediastinal approach was planned. The graph depicts the PTH level during the cervical exploration and resection of 4 eutopic glands (after which the initial operation was halted because of the patient’s medical status) and the subsequent resection of the mediastinal fifth gland. The finely dashed line indicates 50% of the preoperative PTH value. The heavy dashed line indicates 50% of the new baseline established between operations.

Fig 3. Intraoperative PTH of patient HC. In this patient, the PTH20 did not decay to less than 50% of the preoperative level (heavy dashed line), but did drop to less than 50% of the postinduction baseline (fine dashed line). In fact, this patient had only 3 parathyroid glands resected and was not cured.

interpreted as spurious, and the procedure was concluded. However, histopathologic examination revealed 3 parathyroid glands; the fourth specimen was thyroid tissue. The PTH level remained greater than 4 times normal in follow-up (3 months). The third patient in this group was cured after resection of 5 glands, but demonstrated a very slow decay in circulating PTH, reaching a normal value 36 hours after resection. There were also 2 patients (2%) in an intermediate region with PTH decay at 20 minutes of between 40% and 50% of the preoperative level. Both of these patients had 4 glands identified at operation and both were cured. A similar analysis of PTH decay can be performed by using the postinduction PTH as the baseline: 73 patients (91%) had PTH values fall more than 50%. Seventy-one were cured. However,

2 of the 3 failures were also in this group (patients A.C., H.C.) and, conversely, none of the 3 patients who had a PTH decrease less than 40% of the postinduction PTH was a failed operation. Furthermore, 4 patients (5%) remained in the intermediate zone (decay between 40% and 50% of PTHind), including 1 patient not cured (J.W.). Patients with tertiary HPT (n = 7) demonstrated a decline in PTH levels similar to that of primary HPT. Ten minutes after resection, PTH values had not only decreased more than 60% from the preoperative level, but were also within the normal range in all but 1 patient. This 1 exception, in chronic rejection of her transplant and with limited renal function, had a PTH decay profile similar to the patients receiving hemodialysis. Each of these patients had at least 4 glands identified and all were cured.

Lokey et al 1033

Surgery Volume 128, Number 6 DISCUSSION The intraoperative decay of circulating intact PTH after subtotal parathyroidectomy for secondary HPT in the patient receiving hemodialysis is reproducible but markedly slower than in the patient with normal renal function. This is in accordance with the increased half-life of intact PTH in renal failure (6.6 vs 2.2 minutes).14 Although there was correlation between preoperative intact PTH values (immunoradiometric assay) and postinduction intact PTH levels (ICMA), the former were generally higher. The degree of correlation is similar to that seen in studies of primary HPT.9,15 The potential benefits of using the rapid intact intraoperative PTH assay in the operation of secondary HPT are 2-fold: to confirm the adequate resection of hyperfunctioning parathyroid tissue and to alert the surgeon to the possibility of supernumerary glands. In this study, a decline in circulating intact PTH of more than 50% from the preoperative level at 20 minutes after resection was highly predictive of cure, with a positive predictive value of 93% and a sensitivity of 96%. The same analysis using the postinduction PTH value as baseline was slightly less sensitive, but more specific (positive predictive value 92%, sensitivity 97%). While careful pathologic examination of all specimens revealed the presence of more than 4 glands in 28 cases (35%), the vast majority were small rests (< 3 mm) found in the thymus. These small deposits may have clinical significance in long-term follow-up.16 Of greater immediate relevance are missed glands and supernumerary glands that are hyperfunctioning. In this study, the persistent elevation of PTH after subtotal resection of 4 glands was indicative of the presence of a large (1320 mg) mediastinal fifth gland in 1 patient (supernumerary gland) and represented the misidentification of a thyroid nodule for parathyroid tissue in another (missed gland). Specifically, an intact PTH level that decreases less than 40% from the preoperative value has a positive predictive value of 67% for failure or recurrence. Oddly, in our study the same cannot be said using the postinduction baseline; a decline of less than 40% from the postinduction level did not accurately indicate the failures. It is more difficult to make assertions regarding those patients who have a decline in PTH by 20 minutes in the intermediate range (40% and 50% of preoperative value). This group may reflect the wide patient-to-patient variability of PTH half-life14 or its molecular heterogeneity and biphasic metabolism.17 Later samples, at 30 and 40 minutes, had

decayed more than 50% from the baseline value, confirming adequate therapy in this small intermediate group. CONCLUSIONS The intraoperative decay of intact PTH after operation for renal HPT is slower than that seen in the patient with normal renal function. However, 20 minutes after resection, a decay of more than 50% of the preoperative baseline is predictive of cure, while a decrease of less than 40% suggests a missed or hyperfunctioning supernumerary gland and is predictive of failure. REFERENCES 1. Packman KS, Demeure MJ. Indications of parathyroidectomy and extent of treatment for patients with secondary hyperparathyroidism. Surg Clin North Am 1995;75:465-82. 2. Rothmund M, Wagner PK, Schark C. Subtotal parathyroidectomy versus total parathyroidectomy and autotransplantation in secondary hyperparathyroidism: a randomized trial. World J Surg 1991;15:745-50. 3. Henry JFR, Denizot A, Audiffret J, France G. Results of reoperations for persistent or recurrent secondary hyperparathyroidism in hemodialysis patients. World J Surg 1990;14:303-7. 4. Donckier V, Decoster-Gervy C, Kinnaert P. Long-term results after surgical treatment of renal hyperparathyroidism when fewer than four glands are identified at operation. J Am Coll Surg 1997;184:70-4. 5. Irvin GL III. American Association of Endocrine Surgeons. Presidential address: chasin’ hormones. Surgery 1999; 126:993-7. 6. Irvin GL III, Deriso GT. A new, practical, intraoperative parathyroid hormone assay. Am J Surg 1994;169:466-8. 7. Kao PC, van Heerden JA, Taylor RL. Intraoperative monitoring of parathyroid procedures by a 15-minute parathyroid immunochemiluminometric assay. Mayo Clin Proc 1994;69:532-7. 8. Boggs JE, Irvin GL III, Molinari AS, Deriso GT. Intraoperative parathyroid hormone monitoring as an adjunct to parathyroidectomy. Surgery 1996;129:954-8. 9. Garner SC, Leight GS Jr. Initial experience with intraoperative PTH determinations in the surgical management of 130 consecutive cases of primary hyperparathyroidism. Surgery 1999;126:1132-8. 10. Proye CAG, Goropoulos A, Franz C, Carnaille B, Vix M, Quievreux JL, et al. Usefulness and limits of quick intraoperative measurements of intact (1-84) parathyroid hormone in the surgical management of hyperparathyroidism: sequential measurements in patients with multiglandular disease. Surgery 1991;110:1035-42. 11. Koeberle-Wuehrer R, Haid A, Spenger-Mahr H, Neyer U, Zimmerman G. Intraoperative blood sampling for parathyroid hormone measurement during total parathyroidectomy and autotransplantation in patients with renal hyperparathyroidism. Wein Klin Wochenser 1999;111:246-50. 12. Clary BM, Garner SC, Leight GS. Intraoperative parathyroid hormone monitoring during parathyroidectomy for secondary hyperparathyroidism. Surgery 1997;122:1034-9. 13. Hruska K. New concepts in renal osteodystrophy. Nephron Dial Transplant 1998;13:2755-60.

1034 Lokey et al

14. Brossard JH, Cloutier M, Roy L, Lepage R, Gascon-Barre M, D’Armour P. Accumulation of a non-(1-84) molecular form of parathyroid hormone (PTH) detected by intact PTH assay in renal failure: importance in the interpretation of PTH values. J Clin Endocrinol Metab 1996;3923-9. 15. Libutti SK, Alexander R, Bartlett DL, Sampson ML, Ruddel ME, Skarulis M, et al. Kinetic analysis of the rapid intraoperative parathyroid hormone assay in patients during operation for hyperparathyroidism. Surgery 1999;126:1145-51. 16. Reddick RL, Costa JC, Marx SJ. Parathyroid hyperplasia and parathyromatosis [letter]. Lancet 1997;1:549. 17. Davies C, Demeure MJ, St John A, Edis AJ. Study of intact (1-84) parathyroid hormone secretion in patients undergoing parathyroidectomy. World J Surg 1990;14:355-60.

DISCUSSION Dr Michael Demeure (Milwaukee, Wis). Would you clarify the times that you actually drew the labs in relation to the 20-minute interval after resection of the last parathyroid gland? What you are seeming to say is that the half-life intact PTH may be different in a patient with renal failure, which would be something novel. Several reports, including a project we did with the not-so-quick dual antibody PTH assay in 1988, showed after resection of each abnormal gland, you would have a decrease. So results could really depend on the timing of your blood draws. Would you clarify that? Dr Lokey. It is a little more difficult dealing with renal HPT because obviously all glands are affected. The dissection may devascularize some of the glands, but the 20minute PTH is after resection of all of the tissue. So that is when the clock starts, or the zero time value, if you will. Dr Demeure. So then, do you really think the half-life of PTH is longer? Dr Lokey. It is difficult to say that we see a slower decay, but this has not been analyzed with a mathematical model to determine half-life. I think it is more appropriate to say that the decay is slower, but I would be reluctant to say that the actual half-life is slower. Dr Collin Weber (Atlanta, Ga). I would like to second the conclusions drawn from the data presented. This slide outlines a case referred to me with a parathyroid scan that showed a lesion in the mediastinum confirmed by a CAT scan. At surgery, we began with a small neck incision and split the sternum to find the lesion in the mediastinum. The PTH fell from 135 to 55. This raised the question, should I explore the neck in a patient in

Surgery December 2000 whom I had already found a mediastinal tumor? I decided to explore the neck and found a second parathyroid tumor, and thereafter the patient was cured. I raise this as an example that I think is somewhat similar to Dr Proye’s cases. There are patients in whom the PTH will fall dramatically, yet an additional tumor will still be in the neck or chest. Dr Barry Inabnet (New York, NY). Thank you for a very important presentation. I beg to differ with your conclusions and with those of Dr Weber’s. I think it is very important to follow the assay out to 60 minutes, at which time the PTH assay will predict a multigland disease. In the data that I presented at the meeting last year, this phenomenon occurred in 4 patients in a series of 230 patients, which was less than 1% of all patients undergoing operation. I think that if you see a value at the upper end of normal, you must follow it out to 30 or 60 minutes or convert to a conventional parathyroidectomy (bilateral neck exploration). Dr Geoffrey Thompson (Rochester, Minn). I would agree with the authors and Dr Weber. I have now had the opportunity in 5 patients to test the sequential drop in PTH. When you take out the largest of the 4 glands, the PTH will often drop to below 50% of baseline, even though you know that other enlarged glands remain. It is only after you have performed a subtotal parathyroidectomy that you get a PTH level that is back in the normal range. I really think that for multigland disease a 50% drop is insufficient. You really need to see the level back in the normal range within 15 to 20 minutes after you have removed what you think is all the abnormal tissue. Dr Mondragon-Sanchez. From looking at our data, we would agree with that entirely. And 50% is a man-made figure. It is a made-up figure, but we are taking it as our baseline. You may want to choose 20%. Dr Barbara Kinder (New Haven, Conn). I agree completely with Dr Thompson, on the basis of our experience with a primary disease where you have suppressed glands left, the 50% or 55% may be a reasonable drop. But in secondary disease or primary hyperplasia, you are left with abnormal glands and you have to get down well within the normal range before you have cured the patient. I completely agree, Dr Thompson, that has been our experience.