- Email: [email protected]
Impact of complementary parathyroid scintigraphy and ultrasonography on the surgical management of hyperparathyroidism
Impact of complementary parathyroid scintigraphy and ultrasonography on the surgical management of hyperparathyroidism L. Barry Arkles, FRACP, Trevor Jones, FRACS, Rodney J. Hicks, FRACP, Mario A. De Luise, FRACP, a~zdSheung T. Chou, FRCPA, Mdbourne, Australia
Background. An ongoing debate exists regarding the relative merits of fuU versus limited neck exploration in the surgical management of parathyroid adenomata. The aim of this study was to assess the imP act o'f localization studies on the subseq uent surgical management o'f hyp eyparathyroidism " 2 Methods. The accuracy of complementary ultrasonography and O1TI/ 99m Tc parathyroid subtraction scintigraphy in hyperparathyroidism was evaluated retrospectively during a l O~yearperiod in patients referred for localization studies. Surgical and pathologic confirmation of the diagnosis was possible in 121 patients, and these data Jbrmed the basis of this study. Operative procedure, times, outcome, and complications were recorded. Results. The sensitivity, specificity, and accuracy for combined scintigraphy and ultrasonography were 86%, 98%, and 96%, respectively. Limited ~ c k exploration was performed in 61 of 121 patients, and 60 patients underwent fidl neck exploration. In primary hyperparathyroidism 59 of 105 patients underwent limited and 46 underwent full neck exploration with average operative times of 70 and 109 minutes, respectively. (p < 0.0001). Complications developed in five patients who underwent full neck exploration. Conclusions. Confident localization of parathyroid adenomata fadlitated successful limited neck exploration in most of the patients, questioning the need for full neck exploration in all patients with primary hyperparathyroidism. (Surgery 1996;120:845-51.) From the Departments of Nuclear Medicine and Ultrasound, Surgery, Endocrinology and Anatomical Pathology, Austin and Repatriation Medical Centre, Repatriation Campus, Heidelberg West, Victoria, Australia
THE VALUEOF PREOPERATWElocalization studies in the surgical management of primary hyperparathyroidism remains controversial as does the practice of limited neck exploration (LNE). This is because of the high success rate of full neck exploration (FNE) performed by experienced surgeons and the wide variation in accuracy of studies. 1 In contrast, preoperative localization is acknowledged to be of particular value in recurrent disease. ~ In most series the merits of parathyroid subtraction scintigraphy (PSS) versus ultrasonography, computed tomography (CT), or magnetic resonance imaging (MRI) are presented as competing studies, but the value of complementary studies, achieving detection rates of 90%, also has been highlighted, s Scintigraphy is based on the functional activity of the adenoma, whereas ultrasonography Accepted tor publication April 2, 1996. Reprint requests: L. BarryArkles, FRACP,Bundoora Radiodiagnostic, 115 Plenty Rd., Bundoora VIC 3083, MeIboume,Australia. Copyright 9 1996 by Mosby-YearBook, Inc. 0039-6060/96/$5.00 + 0 11/56/73857
provides a high resolution image of small adenomata, with an accurate depiction of their anatomic location. 4 On the basis of experience with preoperative parathyroid localization using PSS and ultrasonography, the primary parathyroid surgeon at the Repatriation Campus elected to perform LNE in those patients in whom confident localization could be made by imaging studies. The preference o f other experienced parathyroid surgeons for FNE provided us an opportunity, as a tertiary referral center for parathyroid imaging, to evaluate the impact of preoperative characterization and localization of parathyroid adenomata on subsequent surgery. Accordingly, we reviewed our experience with combined parathyroid scintigraphy and ultrasonography (CPSU) and the outcome results in those patients who had surgical exploration. METHODS
We retrospectively reviewed all patients referred for investigation of probable parathyroid adenoma by the Department of Nuclear Medicine and Ultrasound durSURGERY 845
846
Arkles et aL
Surgery November 1996
Fig. 1. Concordant study depicts 420 mg adenoma. A, Ultrasonogram, longitudinal left cervical cut. LLPA, Left lower parathyroid adenoma subtending the lower pole of the left thyroid lobe. B, Parathyroid subtraction scan; arrow, left lower parathyroid adenoma.
ing the period January 1985 to December 1994. We were able to obtain detailed surgical and pathologic follow-up data in 121 patients. These patients, the subjects of this study, were hypercalcemic and had elevated levels of parathyroid hormone. Ultrasonography immediately preceded PSS and involved the use o f 7.5 m H z linear array and 5 mHz convex probes and a flexible approach. 5 The PSS involved three sequential 150 k scans of the neck and mediastin u m after intravenous administration of 74 MBq ')~ with subsequent 99mTcO4 acquisition and thyroid subtraction. Optimal regional normalization was ensured by excluding thyroid and parathyroid adenomas on the basis of the ultrasonographic findings. The CPSU report documented the size, echocharacter, site, and functional status of the adenoma. Operative times were evaluated from the anesthetic
chart. Comparison of operative times was performed with an unpaired t test and a Wilcoxon test. Values of p < 0.05 were considered significant. RESULTS Patient population. Surgical and pathologic follow-up data were obtained for 121 patients, 69 women and 52 men. One hundred and five patients had primary and 16 had secondary hyperparathyroidism. Histologic findings indicated parathyroid adenomata in 100 patients and parathyroid hyperplasia in 5 patients. The age of the patients ranged from 22 to 90 years with 51 patients younger than 60 years of age and 54 patients older than 60 years of age. The weights of the adenomata were recorded in 75 of the 105 patients. The range was 50 mg to 6.8 gm; 36 (48%) adenomata weighed more than 500 mg, 39 (52%) weighed less than 500 mg,
Surgery Volume 120, Number 5
and 24 (32%) weighed less than 300 mg. Three adenomata showed cystic change. No carcinomas were observed. Diagnostic performance. In the 100 patients with surgically proven parathyroid adenomata, the sensitivity, specificity, and accuracy for ultrasonography were 76%, 98%, and 94%, respectively, versus 81%, 98%, and 95% for PSS and 86%, 98%, and 96% for CPSU. The positive predictive value for CPSU findings overall was 94%. Five patients with concordant CPSU results were referred from other hospitals after indeterminate localization. Multiple adenomata were present in two patients with primary disease. Ultrasonography and PSS localized two separate adenomata in the first patient, which were confirmed with FNE. In the other patient only a single adenoma was localized. The missed adenoma was retropharyngeal and was localized at reoperation 3 months later. In patients with primary adenomata, the CPSU results were concordantly positive in 63 of 92 patients (69%) who had both studies (Figs. 1 and 2). Localization proved correct in 61 of 63 patients (positive predictive value, 97%). Sixteen patients had discordant CPSU results. Nine patients had a positive PSS result but a negative ultrasonographic result with 6 of 9 adenomata identified on scanning inaccessible to ultrasonography. Seven patients had positive ultrasonographic results but negative PSS results. In the latter group the average weight of the adenomata in six patients was only 253 mg (Figs. 3 and 4). The remaining patient had a large cystic adenoma weighing more than 1 gm. Five patients had concordantly negative results. Sensitivity in five patients with parathyroid hyperplasia was only 50% (8 of 16), but CPSU located multiple hyperfunctioning glands in three patients and a single adenoma in the remaining two patients. All five patients underwent FNE, three on the basis of CPSU results and two on the basis of operative findings. In secondary hyperparathyroidism, sensitivity was 65% (34 of 52 glands). At least one adenoma was detected in 15 of 16 patients and multiple adenomata in 10 of 16 patients (63%). Six patients were thought to have a solitary autonomous adenoma on the basis of concordant CPSU results. Surgical Outcome. In the group of patients with primary hyperparathyroidism, 46 of 105 patients (44%) underwent FNE on the basis of the operating surgeon's philosophy (33 patients), CPSU results, or operative findings (13 patients). Of these, 17 (42%) had concordant positive CPSU results. Fifty-nine patients (56%) underwent CPSU-guided LNE. Operative times were obtainable in 76 of 105 patients with primary hyperparathyroidism. In 41 patients, the duration of the LNEs ranged from 40 to 120 minutes (mean, 70 minutes)
Arkles et al.
847
Fig. 2. Concordant study depicts 900 mg adenoma. A, Ultrasonogram, transverse left cervical cut. LPA, Left upper parathyroid adenoma; arrow heads, left lobe of thyroid; CC, left common carotid artery; LJV,,left jugular vein. B, Parathyroid subtraction scan; arrow, left upper parathyroid adenoma. compared with the duration of 35 FNEs, which ranged in duration from 60 to 120 minutes (mean, 109 minutes) (p < 0.0001). No significant difference was noted in the average weight of adenomata removed by FNE (1.25 gm) and LNE (1.34 gm) or between upper (2 gin) and lower adenomata (1.84 gm). We observed 68 (67%) inferior adenomata, including 2 intrathymic and 1 intrathyroidal, and 34 (33%) upper adenomata. In 72 of 105 patients (69%) with primary hyperparathyroidism, surgery was performed by the primary parathyroid surgeon. Surgical cure was achieved in 69 of 72
848
Azkles et al.
Surgery November 1996
Fig. 3. Ultrasonogram depicts 65 mg adenoma subjacent to left lobe of thyroid gland, longitudinal cervical cut. LUPA, Left upper parathyroid. Adenoma was not detected on scanning. patients (96%) overall and 66 of 67 patients (98%) with a solitary parathyroid adenoma. FNE was performed in 13 patients, and the remaining 59 of 72 (82%) patients underwent CPSU-guided LNE. Surgical cure was achieved in 56 of 59 patients (95%) undergoing LNE. In one patient a missed ectopic (retropharyngeal) second adenomawas uneventfully removed at reoperation, and two adenomata recurred in a different gland at intervals of 2 and 4 years. In the FNE group, 45 of 46 (98%) patients underwent a successful FNE. No complications were recorded in the 59 patients who underwent LNE compared with 5 of 46 patients (11%) who underwent FNE (persistent hypocalcemia in 4 patients and recurrent laryngeal nerve palsy and severe bleeding in 1, requiring hemithyroidectomy). Multinodular goitre was present in 16 of 105 patients (15%), of whom 9 proceeded to FNE and 2 required hemithyroidectomy. Localization with CPSU was correct in 10 of 16 patients (63%). One of these patients had a coexisting parathyroid adenoma and a solitary toxic thyroid nodule. Preoperative CPSU successfully localized and characterized both adenomata (Fig. 5). In patients with multinodular goitre the mean operating times for FNE and LNE were 105 a n d 63 minutes, respectively. Five of six patients with concordant CPSU results underwent successful LNE, including 2 patients who previously had undergone hemithyroidectomy. Five patients had recurrent adenomata (2 to 12 years) after surgery. Two of these patients experienced recurrences after LNE, whereas 3 patients had recurrences after FNE. CPSU correctly localized the five patients with recurrent adenomata, of whom two underwent FNE and three underwent LNE. No operative complications were recorded in these patients.
In the patient group with secondary hyperparathyroidism 6 of 16 patients had concordant CPSU findings of solitary adenoma. In two of these patients one half of the "CPSU autonomous" adenoma was left after subtotal parathyroidectomy. Both these patients had subsequent recurrence of hypercalcemia. DISCUSSION
T h e debate regarding the value of preoperative parathyroid imaging is ongoing, with only a minority of surgeons acknowledging its value. This reflects the feeling of the consensus statement on the m a n a g e m e n t of parathyroid adenomata, which did not advocate preoperative imaging. 2 Furthermore, many surgeons who do use localization still routinely proceed to FNE despite excellent preoperative parathyroid localization results with parathyroid 2~ subtraction scinfigraphy and further improvement in sensitivity using 99mTc-sestamibi.3' 6 This raises doubts as to the value and cost effectiveness of a study that does not alter surgical management. It should be recognised, however, that poor results obtained with parathyroid ultrasonography and scintigraphy in some published series do not necessarily mean that these modalities are intrinsically unreliable, but rather that they may be inaccurate as practised in a particular institution. By the same token, surgical cure and complication rates may be subject to audit in a given institution to determine whether results are comparable with those achieved by experienced parathyroid surgeons. A rational approach to assessing the relative merits of surgical management with and without preoperative localization would be to compare best practice results of FNE versus LNE guided by well validated parathyroid imaging.
Surgery Volume 120, Number 5
Fundamental to this approach is the accuracy of the localizing studies. Ultrasonography is the least expensive of the anatomic imaging modalities. In our series, patients with positive ultrasonography but negative PSS results had small and cystic adenomata, both recognized causes of false-negative scans. This exemplifies the advantage of complementing scintigraphy with ultrasonography. Furthermore, without correlative ultrasonography, false-positive scans caused by underlying thyroid pathologic conditions cannot be excluded. In our patients multinodular goitre did not preclude accurate localization of parathyroid adenomata. Scintigraphic detection of ectopic adenomata before operation enables avoidance of unsuccessful or prolonged FNE. Thus CPSU allowed early definitive surgical assessment of the intrathyroidal and intrathymic adenomata and enabled successful LNE. Thyroid nodules and cystic and small adenomata remain the Achilles' heel of both thallium and sestamibi scintigraphy. The 95% surgical cure rate of CPSU-guided LNE compared favorably with the best reported cure rate of FNE. 7 The shortened operative times and decreased morbidity after LNE observed in our series is in accord with the experience of other researchers. 8-11 Complications may have been avoided in three patients who underwent FNE despite prior concordant CPSU results. Operating r o o m times and costs have been reported to be significantly lower for patients with an accurate preoperative ultrasonography scan than for patients with inaccurate localization. 12 Documentation of unhelpful preoperative localization was associated with suboptimal imaging accuracy. 13, 14It is acknowledged, however, that the differences in operating times were in two groups of patients who were not randomly allocated to FNE or LNE, but in whom the type of surgery was determined partly on the basis of imaging results. A missed diagnosis of primary parathyroid hyperplasia or multiple adenomata is cited as a key reason for avoiding LNE. This did not prove to be the case in our experience, where all five patients proceeded to FNE, three with multiple adenomata identified on CPSU and two as a consequence of operative findings. The 5% incidence of parathyroid hyperplasia in our series was lower than the commonly quoted 15% but corresponds with the 4% incidence reported by Attie et al. 15 This discrepancy may lie in the difficulty in pathologic differentiation of hyperplasia from adenoma. In primary hyperparathyroidism, surgical removal of only the enlarged glands has been f o u n d to be of greater prognostic significance than pathologic findings. The advent of the "quick" parathyroid h o r m o n e assay17 complements accurate preoperative localization and allows exclusion of multiple adenomata and hyperplastic glands without proceeding to FNE.
Arkles et al.
849
Fig. 4. Ultrasonogram depicts 90 mg adenoma subjacent to right lobe of thyroid gland and lateral to trachea, oblique and transverse cervical cuts. RLPA, Right lower parathyroid gland; CCA, right common carotid artery. Adenoma was not.detected on scanning-
Parathyroid adenomata in the setting of multinodular goitre also may be successfully treated with LNE after concordant localization. CPSU enabled differentiation between parathyroid and thyroid nodules in most cases. Many of these patients were elderly and benefited from the mean shortened operative time for LNE (63 minutes) versus FNE (105 minutes). Both toxic nodules and hyperparathyroidism may cause hypercalcemia, and CPSU enabled both these causative factors to be recognised before operation. Thus doubly curative surgery could be performed in a single operation (Fig. 5). In contrast to previous experience, 18 CPSU avoided the need for reoperation. The frequency of multinodal gob tre in association with hyperparathyroidism is likely to increase as the n u m b e r of elderly patients increases. 99mTc-sestamibi parathyroid scintigraphy alone remains unreliable in the presence of multinodal goitre. Confident concordant CPSU localization resulted in successful LNE in several patients who because of their age, severity ofintercurrent illness, paucity of symptoms, and reluctance to undergo surgery would otherwise have been treated medically. No significant complications were reported in the very elderly or sick patients undergoing LNE. The five patients in our series who were referred from other hospitals after indeterminate localization studies had concordantly positive CPSU results and subsequently proceeded to successful neck exploration. Thus CPSU clearly influenced the timing of operation intervention. The conservative management of asymptomatic patients with hyperparathyroidism in whom the a d e n o m a has been precisely located may warrant review. 19 Confident concordant CPSU localization may influence tile
850
Arkles et al.
Surgery November 1996
Fig. 5. Left upper parathyroid adenoma and toxic nodule. A, Uhrasonogram, left oblique cervical cut. Large arrow head, Toxic thyroid nodule; small arrow head, left upper parathyroid adenoma. B, 2~ parathyroid subtraction scan. Arrow, Left upper parathyroid adenoma; arrowhead, toxic thyroid nodule; LPA, left upper parathyroid adenoma; TOXIC NODULE, left lobe toxic nodule. physician to seek operative intervention rather than pursue medical treatment. Improved medical management has decreased the need for surgery in secondary hyperparathyroidism in recent years. Thus only 5 of 16 patients with kidney failure in our series had CPSU in the last 5 years, contrasting with 11 in the first 5 years. In 2 of these 5 patients, half the culprit gland was left behind after subtotal parathyroidectomy and in both patients hypercalcemia subsequently recurred. Our experience thus suggests that functional imaging may be useful in cases of suspected tertiary hyperparathyroidism, with concordant CPSU suggesting a solitary autonomous adenoma. Thus CPSU may affect the timing of surgical intervention. Furthermore, it may influence surgical management by ensuring that half the culprit autonomous adenoma is not left behind after subtotal parathyroidectomy. In the final analysis the value of any investigative study lies in its impact on subsequent management. The pos-
itive predictive value of preoperative imaging is probably the most important factor likely to influence surgical management of hyperparathyroidism. Concordant CPSU results showing a single parathyroid adenoma would thus suggest suitability for LNE in the first instance. Operative and, where available, intraoperativ e values of parathyroid hormone may then dictate the necessity for FNE. Concordant CPSU should be particularly helpful to less experienced parathyroid surgeons. The 2% to 3% incidence of parathyroid carcinoma previously quoted may no longer be accurate. We have yet to find a carcinoma in more than 350 patients referred to our Department during the past 23 years, 2~ in accord with the experience of other researchers. 21 CONCLUSION This study suggests that complementary functional parathyroid scintigraphy and high resolution ultrasonography are superior to either modality alone and
Surgery Volume 120, Number 5
h a v e b e e n s h o w n to b e o f p a r t i c u l a r v a l u e in t h e d e t e c t i o n o f small, cystic, a n d e c t o p i c a d e n o m a t a a n d i n p a t i e n t s w i t h c o e x i s t i n g t h y r o i d disease. C o n c o r d a n t localization is m o s t likely to positively affect surgical m a n a g e m e n t , facilitating L N E w i t h a r e d u c e d c o m p l i c a t i o n rate. T h e legitimate c o n c e r n o f missing cases o f parathyroid hypel-plasia o r m u l t i p l e a d e n o m a t a has n o t b e e n j u s lifted in o u r e x p e r i e n c e a n d s h o u l d f u r t h e r d e c r e a s e with t h e availability o f r a p i d assays for p a r a t h y r o i d h o r m o n e . T h e p e r f o r m a n c e o f FNE for all cases o f p r i m a r y parathyr o i d a d e n o m a r e q u i r e s review in t h o s e hospitals w h e r e h i g h accuracy localization studies are available. O u r e x p e r i e n c e raises t h e possibility t h a t L N E also m a y b e a realistic a l t e r n a t i v e to c o n s e r v a t i v e m a n a g e m e n t in t h o s e p a t i e n t s w i t h c o n c o r d a n t l o c a l i z a t i o n a n d i n w h o m t h e clinical i n d i c a t i o n s f o r s u r g e r y a r e b o r d e r line. It also m a y b e h e l p f u l i n t h e surgical m a n a g e m e n t o f tertiary h y p e r p a r a t h y r o i d i s m , b u t f u r t h e r e x p e r i e n c e is r e q u i r e d to a d e q u a t e l y assess t h e p o t e n t i a l r o l e o f localization in b o t h t h e s e clinical settings. We thank Mrs. L. J o h n s o n , Chief Nuclear Medicine Technologist and Ultrasonographer, and Ms. M. Garra, Senior Nuclear Medicine Research Technologist, for their valuable assistance in the performance of these studies and collation of the data. REFERENCES
1. Coakley AJ. Parathyroid localisation---how and when? EurJ Nucl Med 1991;18:151-2. 2. Consensus Development Confi:rence statement. J Bone Miner Res 1991;6(suppl 2):9-13. 3. Geatti O, Shapiro B, Orsolon PG, et aI. Localisation of parathyroid enlargement: experience with technetium-99m methoxyisobutylisonitrile, and thallium-201 scintigraphy, uhrasonography and computed tomography. EurJ NucI Med 1994;21:17-22. 4. Taillandier J, Kauffinan P, Raynaud F, et al. Preoperative localisation of parathyroid adenomas---role of cervical ultrasonography. Presse Med 1994;23:116-20. 5. Arkles LB. An approach to parathyroid ultrasound. Australian Society for Ultrasound in Medicine Newsletter. Feb 1994;16:2-3.
Arkles et al.
851
6. Hindie E, Melliere D, Simon D, et al. Primary hyperparathyroidism: is technetium-99m-sestamibi-iodine-123 subtraction scanning the best procedure to locate enlarged glands before surgery? J Clin Endocrinol Metab 1995;80:302-7. 7. Kaplan EL, Yashiro T, Salti G. Primary hyperparathD-oidism in the 1990s: choice Of sm'gical procedures for this disease. Ann Surg 1992;215:300-16. 8. Worsey MJ, Cart}, SE, Watson CG. Success of unilateral neck exploration for sporadic primary hyperparathyroidism. Surgery 1993;114:1024-30. 9. Davis RK, Hoffmann J, Dart D, et al. Unilateral parathyroidectomy: the role of thallium-technetium subtraction scans. Otolaryngol Head Neck Surg 1990;102:635-8. 10. Russell CF, LairdJD, Ferguson WR. Scan-directed unilateral cei~ vical exploration tbr parathyroid adenoma: a legitimate approach? WorldJ Surg 1990;14:406-9. 11. Morioka WT. Trends in primary hyperparathyroid surgery. Laryngoscope 1992;102:422-5. 12. Kairaluoma MV, Makarainen H, Kellosalo J, et al. Preoperative ultrasound in patients undergoing initial neck exploration for primary hyperparathyroidism. Ann Chir Gynaeco11993;82:17l-6. 13. Hasselgren PO, Fidler JP. Further evidence against the routine use of parathyroid ultrasonography prior to initial neck exploration for hyperparathyroidism. AmJ Surg 1992;164:337-40. 14. Roe SM, Burns RP, Graham ID, Brock WB, Russell WL. Cost-effectiveness of preoperative localization studies in primary hyperparathyroid disease. Ann Surg 1994;219:582-6. 15. AttieJN, Bock G, Auguste L. Multiple parathyroid adenomas: report of 33 cases. Surgery 1990;108:1014-20. 16. Hosking SW, .Jones H, Du Boulay CE, McGinn FP. Surgery for parathyroid adenoma and hyperplasia: relationship of histology to outcome. Head Neck 1993;15:24-8. t7. Irvin GL, Dembrow VD, Prudhomme DL. Clinical usefulness of an introperative "quick parathyroid hormone" assay. Sm'gery 1993;114:1019-22. 18. Toursarkissian B, Sloan DA, Schwartz RW. Coexisting hyperthyroidism and primary hyperparathyroidism. Surgery 1993;113: 716-8. 19. Davies M. Primary hyperparathyroidism: aggressive or conservative treatment? Clin Endocrinol (Oxf) 1992;6:325-32. 20. Arkles LB. Experience in parathyroid scanning. AmJ Roentgenol Rad Ther Nuc Med 1975;125:634-9. 2l. Salti GI, Fedorak I, Yashiro T, et al. Continuing evolution in the operative management of primary hyperparathyroidism. Arch Surg 1992;127(7):831-6; 836-7.
Background. An ongoing debate exists regarding the relative merits of fuU versus limited neck exploration in the surgical management of parathyroid adenomata. The aim of this study was to assess the imP act o'f localization studies on the subseq uent surgical management o'f hyp eyparathyroidism " 2 Methods. The accuracy of complementary ultrasonography and O1TI/ 99m Tc parathyroid subtraction scintigraphy in hyperparathyroidism was evaluated retrospectively during a l O~yearperiod in patients referred for localization studies. Surgical and pathologic confirmation of the diagnosis was possible in 121 patients, and these data Jbrmed the basis of this study. Operative procedure, times, outcome, and complications were recorded. Results. The sensitivity, specificity, and accuracy for combined scintigraphy and ultrasonography were 86%, 98%, and 96%, respectively. Limited ~ c k exploration was performed in 61 of 121 patients, and 60 patients underwent fidl neck exploration. In primary hyperparathyroidism 59 of 105 patients underwent limited and 46 underwent full neck exploration with average operative times of 70 and 109 minutes, respectively. (p < 0.0001). Complications developed in five patients who underwent full neck exploration. Conclusions. Confident localization of parathyroid adenomata fadlitated successful limited neck exploration in most of the patients, questioning the need for full neck exploration in all patients with primary hyperparathyroidism. (Surgery 1996;120:845-51.) From the Departments of Nuclear Medicine and Ultrasound, Surgery, Endocrinology and Anatomical Pathology, Austin and Repatriation Medical Centre, Repatriation Campus, Heidelberg West, Victoria, Australia
THE VALUEOF PREOPERATWElocalization studies in the surgical management of primary hyperparathyroidism remains controversial as does the practice of limited neck exploration (LNE). This is because of the high success rate of full neck exploration (FNE) performed by experienced surgeons and the wide variation in accuracy of studies. 1 In contrast, preoperative localization is acknowledged to be of particular value in recurrent disease. ~ In most series the merits of parathyroid subtraction scintigraphy (PSS) versus ultrasonography, computed tomography (CT), or magnetic resonance imaging (MRI) are presented as competing studies, but the value of complementary studies, achieving detection rates of 90%, also has been highlighted, s Scintigraphy is based on the functional activity of the adenoma, whereas ultrasonography Accepted tor publication April 2, 1996. Reprint requests: L. BarryArkles, FRACP,Bundoora Radiodiagnostic, 115 Plenty Rd., Bundoora VIC 3083, MeIboume,Australia. Copyright 9 1996 by Mosby-YearBook, Inc. 0039-6060/96/$5.00 + 0 11/56/73857
provides a high resolution image of small adenomata, with an accurate depiction of their anatomic location. 4 On the basis of experience with preoperative parathyroid localization using PSS and ultrasonography, the primary parathyroid surgeon at the Repatriation Campus elected to perform LNE in those patients in whom confident localization could be made by imaging studies. The preference o f other experienced parathyroid surgeons for FNE provided us an opportunity, as a tertiary referral center for parathyroid imaging, to evaluate the impact of preoperative characterization and localization of parathyroid adenomata on subsequent surgery. Accordingly, we reviewed our experience with combined parathyroid scintigraphy and ultrasonography (CPSU) and the outcome results in those patients who had surgical exploration. METHODS
We retrospectively reviewed all patients referred for investigation of probable parathyroid adenoma by the Department of Nuclear Medicine and Ultrasound durSURGERY 845
846
Arkles et aL
Surgery November 1996
Fig. 1. Concordant study depicts 420 mg adenoma. A, Ultrasonogram, longitudinal left cervical cut. LLPA, Left lower parathyroid adenoma subtending the lower pole of the left thyroid lobe. B, Parathyroid subtraction scan; arrow, left lower parathyroid adenoma.
ing the period January 1985 to December 1994. We were able to obtain detailed surgical and pathologic follow-up data in 121 patients. These patients, the subjects of this study, were hypercalcemic and had elevated levels of parathyroid hormone. Ultrasonography immediately preceded PSS and involved the use o f 7.5 m H z linear array and 5 mHz convex probes and a flexible approach. 5 The PSS involved three sequential 150 k scans of the neck and mediastin u m after intravenous administration of 74 MBq ')~ with subsequent 99mTcO4 acquisition and thyroid subtraction. Optimal regional normalization was ensured by excluding thyroid and parathyroid adenomas on the basis of the ultrasonographic findings. The CPSU report documented the size, echocharacter, site, and functional status of the adenoma. Operative times were evaluated from the anesthetic
chart. Comparison of operative times was performed with an unpaired t test and a Wilcoxon test. Values of p < 0.05 were considered significant. RESULTS Patient population. Surgical and pathologic follow-up data were obtained for 121 patients, 69 women and 52 men. One hundred and five patients had primary and 16 had secondary hyperparathyroidism. Histologic findings indicated parathyroid adenomata in 100 patients and parathyroid hyperplasia in 5 patients. The age of the patients ranged from 22 to 90 years with 51 patients younger than 60 years of age and 54 patients older than 60 years of age. The weights of the adenomata were recorded in 75 of the 105 patients. The range was 50 mg to 6.8 gm; 36 (48%) adenomata weighed more than 500 mg, 39 (52%) weighed less than 500 mg,
Surgery Volume 120, Number 5
and 24 (32%) weighed less than 300 mg. Three adenomata showed cystic change. No carcinomas were observed. Diagnostic performance. In the 100 patients with surgically proven parathyroid adenomata, the sensitivity, specificity, and accuracy for ultrasonography were 76%, 98%, and 94%, respectively, versus 81%, 98%, and 95% for PSS and 86%, 98%, and 96% for CPSU. The positive predictive value for CPSU findings overall was 94%. Five patients with concordant CPSU results were referred from other hospitals after indeterminate localization. Multiple adenomata were present in two patients with primary disease. Ultrasonography and PSS localized two separate adenomata in the first patient, which were confirmed with FNE. In the other patient only a single adenoma was localized. The missed adenoma was retropharyngeal and was localized at reoperation 3 months later. In patients with primary adenomata, the CPSU results were concordantly positive in 63 of 92 patients (69%) who had both studies (Figs. 1 and 2). Localization proved correct in 61 of 63 patients (positive predictive value, 97%). Sixteen patients had discordant CPSU results. Nine patients had a positive PSS result but a negative ultrasonographic result with 6 of 9 adenomata identified on scanning inaccessible to ultrasonography. Seven patients had positive ultrasonographic results but negative PSS results. In the latter group the average weight of the adenomata in six patients was only 253 mg (Figs. 3 and 4). The remaining patient had a large cystic adenoma weighing more than 1 gm. Five patients had concordantly negative results. Sensitivity in five patients with parathyroid hyperplasia was only 50% (8 of 16), but CPSU located multiple hyperfunctioning glands in three patients and a single adenoma in the remaining two patients. All five patients underwent FNE, three on the basis of CPSU results and two on the basis of operative findings. In secondary hyperparathyroidism, sensitivity was 65% (34 of 52 glands). At least one adenoma was detected in 15 of 16 patients and multiple adenomata in 10 of 16 patients (63%). Six patients were thought to have a solitary autonomous adenoma on the basis of concordant CPSU results. Surgical Outcome. In the group of patients with primary hyperparathyroidism, 46 of 105 patients (44%) underwent FNE on the basis of the operating surgeon's philosophy (33 patients), CPSU results, or operative findings (13 patients). Of these, 17 (42%) had concordant positive CPSU results. Fifty-nine patients (56%) underwent CPSU-guided LNE. Operative times were obtainable in 76 of 105 patients with primary hyperparathyroidism. In 41 patients, the duration of the LNEs ranged from 40 to 120 minutes (mean, 70 minutes)
Arkles et al.
847
Fig. 2. Concordant study depicts 900 mg adenoma. A, Ultrasonogram, transverse left cervical cut. LPA, Left upper parathyroid adenoma; arrow heads, left lobe of thyroid; CC, left common carotid artery; LJV,,left jugular vein. B, Parathyroid subtraction scan; arrow, left upper parathyroid adenoma. compared with the duration of 35 FNEs, which ranged in duration from 60 to 120 minutes (mean, 109 minutes) (p < 0.0001). No significant difference was noted in the average weight of adenomata removed by FNE (1.25 gm) and LNE (1.34 gm) or between upper (2 gin) and lower adenomata (1.84 gm). We observed 68 (67%) inferior adenomata, including 2 intrathymic and 1 intrathyroidal, and 34 (33%) upper adenomata. In 72 of 105 patients (69%) with primary hyperparathyroidism, surgery was performed by the primary parathyroid surgeon. Surgical cure was achieved in 69 of 72
848
Azkles et al.
Surgery November 1996
Fig. 3. Ultrasonogram depicts 65 mg adenoma subjacent to left lobe of thyroid gland, longitudinal cervical cut. LUPA, Left upper parathyroid. Adenoma was not detected on scanning. patients (96%) overall and 66 of 67 patients (98%) with a solitary parathyroid adenoma. FNE was performed in 13 patients, and the remaining 59 of 72 (82%) patients underwent CPSU-guided LNE. Surgical cure was achieved in 56 of 59 patients (95%) undergoing LNE. In one patient a missed ectopic (retropharyngeal) second adenomawas uneventfully removed at reoperation, and two adenomata recurred in a different gland at intervals of 2 and 4 years. In the FNE group, 45 of 46 (98%) patients underwent a successful FNE. No complications were recorded in the 59 patients who underwent LNE compared with 5 of 46 patients (11%) who underwent FNE (persistent hypocalcemia in 4 patients and recurrent laryngeal nerve palsy and severe bleeding in 1, requiring hemithyroidectomy). Multinodular goitre was present in 16 of 105 patients (15%), of whom 9 proceeded to FNE and 2 required hemithyroidectomy. Localization with CPSU was correct in 10 of 16 patients (63%). One of these patients had a coexisting parathyroid adenoma and a solitary toxic thyroid nodule. Preoperative CPSU successfully localized and characterized both adenomata (Fig. 5). In patients with multinodular goitre the mean operating times for FNE and LNE were 105 a n d 63 minutes, respectively. Five of six patients with concordant CPSU results underwent successful LNE, including 2 patients who previously had undergone hemithyroidectomy. Five patients had recurrent adenomata (2 to 12 years) after surgery. Two of these patients experienced recurrences after LNE, whereas 3 patients had recurrences after FNE. CPSU correctly localized the five patients with recurrent adenomata, of whom two underwent FNE and three underwent LNE. No operative complications were recorded in these patients.
In the patient group with secondary hyperparathyroidism 6 of 16 patients had concordant CPSU findings of solitary adenoma. In two of these patients one half of the "CPSU autonomous" adenoma was left after subtotal parathyroidectomy. Both these patients had subsequent recurrence of hypercalcemia. DISCUSSION
T h e debate regarding the value of preoperative parathyroid imaging is ongoing, with only a minority of surgeons acknowledging its value. This reflects the feeling of the consensus statement on the m a n a g e m e n t of parathyroid adenomata, which did not advocate preoperative imaging. 2 Furthermore, many surgeons who do use localization still routinely proceed to FNE despite excellent preoperative parathyroid localization results with parathyroid 2~ subtraction scinfigraphy and further improvement in sensitivity using 99mTc-sestamibi.3' 6 This raises doubts as to the value and cost effectiveness of a study that does not alter surgical management. It should be recognised, however, that poor results obtained with parathyroid ultrasonography and scintigraphy in some published series do not necessarily mean that these modalities are intrinsically unreliable, but rather that they may be inaccurate as practised in a particular institution. By the same token, surgical cure and complication rates may be subject to audit in a given institution to determine whether results are comparable with those achieved by experienced parathyroid surgeons. A rational approach to assessing the relative merits of surgical management with and without preoperative localization would be to compare best practice results of FNE versus LNE guided by well validated parathyroid imaging.
Surgery Volume 120, Number 5
Fundamental to this approach is the accuracy of the localizing studies. Ultrasonography is the least expensive of the anatomic imaging modalities. In our series, patients with positive ultrasonography but negative PSS results had small and cystic adenomata, both recognized causes of false-negative scans. This exemplifies the advantage of complementing scintigraphy with ultrasonography. Furthermore, without correlative ultrasonography, false-positive scans caused by underlying thyroid pathologic conditions cannot be excluded. In our patients multinodular goitre did not preclude accurate localization of parathyroid adenomata. Scintigraphic detection of ectopic adenomata before operation enables avoidance of unsuccessful or prolonged FNE. Thus CPSU allowed early definitive surgical assessment of the intrathyroidal and intrathymic adenomata and enabled successful LNE. Thyroid nodules and cystic and small adenomata remain the Achilles' heel of both thallium and sestamibi scintigraphy. The 95% surgical cure rate of CPSU-guided LNE compared favorably with the best reported cure rate of FNE. 7 The shortened operative times and decreased morbidity after LNE observed in our series is in accord with the experience of other researchers. 8-11 Complications may have been avoided in three patients who underwent FNE despite prior concordant CPSU results. Operating r o o m times and costs have been reported to be significantly lower for patients with an accurate preoperative ultrasonography scan than for patients with inaccurate localization. 12 Documentation of unhelpful preoperative localization was associated with suboptimal imaging accuracy. 13, 14It is acknowledged, however, that the differences in operating times were in two groups of patients who were not randomly allocated to FNE or LNE, but in whom the type of surgery was determined partly on the basis of imaging results. A missed diagnosis of primary parathyroid hyperplasia or multiple adenomata is cited as a key reason for avoiding LNE. This did not prove to be the case in our experience, where all five patients proceeded to FNE, three with multiple adenomata identified on CPSU and two as a consequence of operative findings. The 5% incidence of parathyroid hyperplasia in our series was lower than the commonly quoted 15% but corresponds with the 4% incidence reported by Attie et al. 15 This discrepancy may lie in the difficulty in pathologic differentiation of hyperplasia from adenoma. In primary hyperparathyroidism, surgical removal of only the enlarged glands has been f o u n d to be of greater prognostic significance than pathologic findings. The advent of the "quick" parathyroid h o r m o n e assay17 complements accurate preoperative localization and allows exclusion of multiple adenomata and hyperplastic glands without proceeding to FNE.
Arkles et al.
849
Fig. 4. Ultrasonogram depicts 90 mg adenoma subjacent to right lobe of thyroid gland and lateral to trachea, oblique and transverse cervical cuts. RLPA, Right lower parathyroid gland; CCA, right common carotid artery. Adenoma was not.detected on scanning-
Parathyroid adenomata in the setting of multinodular goitre also may be successfully treated with LNE after concordant localization. CPSU enabled differentiation between parathyroid and thyroid nodules in most cases. Many of these patients were elderly and benefited from the mean shortened operative time for LNE (63 minutes) versus FNE (105 minutes). Both toxic nodules and hyperparathyroidism may cause hypercalcemia, and CPSU enabled both these causative factors to be recognised before operation. Thus doubly curative surgery could be performed in a single operation (Fig. 5). In contrast to previous experience, 18 CPSU avoided the need for reoperation. The frequency of multinodal gob tre in association with hyperparathyroidism is likely to increase as the n u m b e r of elderly patients increases. 99mTc-sestamibi parathyroid scintigraphy alone remains unreliable in the presence of multinodal goitre. Confident concordant CPSU localization resulted in successful LNE in several patients who because of their age, severity ofintercurrent illness, paucity of symptoms, and reluctance to undergo surgery would otherwise have been treated medically. No significant complications were reported in the very elderly or sick patients undergoing LNE. The five patients in our series who were referred from other hospitals after indeterminate localization studies had concordantly positive CPSU results and subsequently proceeded to successful neck exploration. Thus CPSU clearly influenced the timing of operation intervention. The conservative management of asymptomatic patients with hyperparathyroidism in whom the a d e n o m a has been precisely located may warrant review. 19 Confident concordant CPSU localization may influence tile
850
Arkles et al.
Surgery November 1996
Fig. 5. Left upper parathyroid adenoma and toxic nodule. A, Uhrasonogram, left oblique cervical cut. Large arrow head, Toxic thyroid nodule; small arrow head, left upper parathyroid adenoma. B, 2~ parathyroid subtraction scan. Arrow, Left upper parathyroid adenoma; arrowhead, toxic thyroid nodule; LPA, left upper parathyroid adenoma; TOXIC NODULE, left lobe toxic nodule. physician to seek operative intervention rather than pursue medical treatment. Improved medical management has decreased the need for surgery in secondary hyperparathyroidism in recent years. Thus only 5 of 16 patients with kidney failure in our series had CPSU in the last 5 years, contrasting with 11 in the first 5 years. In 2 of these 5 patients, half the culprit gland was left behind after subtotal parathyroidectomy and in both patients hypercalcemia subsequently recurred. Our experience thus suggests that functional imaging may be useful in cases of suspected tertiary hyperparathyroidism, with concordant CPSU suggesting a solitary autonomous adenoma. Thus CPSU may affect the timing of surgical intervention. Furthermore, it may influence surgical management by ensuring that half the culprit autonomous adenoma is not left behind after subtotal parathyroidectomy. In the final analysis the value of any investigative study lies in its impact on subsequent management. The pos-
itive predictive value of preoperative imaging is probably the most important factor likely to influence surgical management of hyperparathyroidism. Concordant CPSU results showing a single parathyroid adenoma would thus suggest suitability for LNE in the first instance. Operative and, where available, intraoperativ e values of parathyroid hormone may then dictate the necessity for FNE. Concordant CPSU should be particularly helpful to less experienced parathyroid surgeons. The 2% to 3% incidence of parathyroid carcinoma previously quoted may no longer be accurate. We have yet to find a carcinoma in more than 350 patients referred to our Department during the past 23 years, 2~ in accord with the experience of other researchers. 21 CONCLUSION This study suggests that complementary functional parathyroid scintigraphy and high resolution ultrasonography are superior to either modality alone and
Surgery Volume 120, Number 5
h a v e b e e n s h o w n to b e o f p a r t i c u l a r v a l u e in t h e d e t e c t i o n o f small, cystic, a n d e c t o p i c a d e n o m a t a a n d i n p a t i e n t s w i t h c o e x i s t i n g t h y r o i d disease. C o n c o r d a n t localization is m o s t likely to positively affect surgical m a n a g e m e n t , facilitating L N E w i t h a r e d u c e d c o m p l i c a t i o n rate. T h e legitimate c o n c e r n o f missing cases o f parathyroid hypel-plasia o r m u l t i p l e a d e n o m a t a has n o t b e e n j u s lifted in o u r e x p e r i e n c e a n d s h o u l d f u r t h e r d e c r e a s e with t h e availability o f r a p i d assays for p a r a t h y r o i d h o r m o n e . T h e p e r f o r m a n c e o f FNE for all cases o f p r i m a r y parathyr o i d a d e n o m a r e q u i r e s review in t h o s e hospitals w h e r e h i g h accuracy localization studies are available. O u r e x p e r i e n c e raises t h e possibility t h a t L N E also m a y b e a realistic a l t e r n a t i v e to c o n s e r v a t i v e m a n a g e m e n t in t h o s e p a t i e n t s w i t h c o n c o r d a n t l o c a l i z a t i o n a n d i n w h o m t h e clinical i n d i c a t i o n s f o r s u r g e r y a r e b o r d e r line. It also m a y b e h e l p f u l i n t h e surgical m a n a g e m e n t o f tertiary h y p e r p a r a t h y r o i d i s m , b u t f u r t h e r e x p e r i e n c e is r e q u i r e d to a d e q u a t e l y assess t h e p o t e n t i a l r o l e o f localization in b o t h t h e s e clinical settings. We thank Mrs. L. J o h n s o n , Chief Nuclear Medicine Technologist and Ultrasonographer, and Ms. M. Garra, Senior Nuclear Medicine Research Technologist, for their valuable assistance in the performance of these studies and collation of the data. REFERENCES
1. Coakley AJ. Parathyroid localisation---how and when? EurJ Nucl Med 1991;18:151-2. 2. Consensus Development Confi:rence statement. J Bone Miner Res 1991;6(suppl 2):9-13. 3. Geatti O, Shapiro B, Orsolon PG, et aI. Localisation of parathyroid enlargement: experience with technetium-99m methoxyisobutylisonitrile, and thallium-201 scintigraphy, uhrasonography and computed tomography. EurJ NucI Med 1994;21:17-22. 4. Taillandier J, Kauffinan P, Raynaud F, et al. Preoperative localisation of parathyroid adenomas---role of cervical ultrasonography. Presse Med 1994;23:116-20. 5. Arkles LB. An approach to parathyroid ultrasound. Australian Society for Ultrasound in Medicine Newsletter. Feb 1994;16:2-3.
Arkles et al.
851
6. Hindie E, Melliere D, Simon D, et al. Primary hyperparathyroidism: is technetium-99m-sestamibi-iodine-123 subtraction scanning the best procedure to locate enlarged glands before surgery? J Clin Endocrinol Metab 1995;80:302-7. 7. Kaplan EL, Yashiro T, Salti G. Primary hyperparathD-oidism in the 1990s: choice Of sm'gical procedures for this disease. Ann Surg 1992;215:300-16. 8. Worsey MJ, Cart}, SE, Watson CG. Success of unilateral neck exploration for sporadic primary hyperparathyroidism. Surgery 1993;114:1024-30. 9. Davis RK, Hoffmann J, Dart D, et al. Unilateral parathyroidectomy: the role of thallium-technetium subtraction scans. Otolaryngol Head Neck Surg 1990;102:635-8. 10. Russell CF, LairdJD, Ferguson WR. Scan-directed unilateral cei~ vical exploration tbr parathyroid adenoma: a legitimate approach? WorldJ Surg 1990;14:406-9. 11. Morioka WT. Trends in primary hyperparathyroid surgery. Laryngoscope 1992;102:422-5. 12. Kairaluoma MV, Makarainen H, Kellosalo J, et al. Preoperative ultrasound in patients undergoing initial neck exploration for primary hyperparathyroidism. Ann Chir Gynaeco11993;82:17l-6. 13. Hasselgren PO, Fidler JP. Further evidence against the routine use of parathyroid ultrasonography prior to initial neck exploration for hyperparathyroidism. AmJ Surg 1992;164:337-40. 14. Roe SM, Burns RP, Graham ID, Brock WB, Russell WL. Cost-effectiveness of preoperative localization studies in primary hyperparathyroid disease. Ann Surg 1994;219:582-6. 15. AttieJN, Bock G, Auguste L. Multiple parathyroid adenomas: report of 33 cases. Surgery 1990;108:1014-20. 16. Hosking SW, .Jones H, Du Boulay CE, McGinn FP. Surgery for parathyroid adenoma and hyperplasia: relationship of histology to outcome. Head Neck 1993;15:24-8. t7. Irvin GL, Dembrow VD, Prudhomme DL. Clinical usefulness of an introperative "quick parathyroid hormone" assay. Sm'gery 1993;114:1019-22. 18. Toursarkissian B, Sloan DA, Schwartz RW. Coexisting hyperthyroidism and primary hyperparathyroidism. Surgery 1993;113: 716-8. 19. Davies M. Primary hyperparathyroidism: aggressive or conservative treatment? Clin Endocrinol (Oxf) 1992;6:325-32. 20. Arkles LB. Experience in parathyroid scanning. AmJ Roentgenol Rad Ther Nuc Med 1975;125:634-9. 2l. Salti GI, Fedorak I, Yashiro T, et al. Continuing evolution in the operative management of primary hyperparathyroidism. Arch Surg 1992;127(7):831-6; 836-7.