- Email: [email protected]
Value of high resolution real-time ultrasonography in secondary hyperparathyroidism
SCIENTIFIC PAPERS
Value of High Resolution Real-Time Ultrasonography in Secondary Hyperparathyroidism
Orlo H. Clark, MD, San Francisco, California David A. Stark, MD, Boston, Massachusetts Wan-Y. Duh, MD, San Francisco, California Claude D. Arnaud, MD, San Francisco, California Gretchen A. W. Gooding, MD, San Francisco, California
Good medical treatment is effective in preventing secondary hyperparathyroidism in most patients with chronic renal failure, but in some patients, progressive secondary hyperparathyroidism develops with accompanying fatigue, arthralgias, pruritus, and bone pain, as well as a myriad of other symptoms [I-3]. Parathyroidectomy is an effective means of treating these problems, but absolute criteria for selecting patients for surgical rather than medical treatment are not universally accepted. Certainly symptomatic patients, patients with a calciumphosphate product greater than 70 mg/dl, patients with osteitis fibrosa cystica, those with persistent hypercalcemia (calcium level greater than 11 mg/dl), and those with extensive soft tissue calcification benefit from parathyroidectomy [4]. During the past several years, we have been interested in the value of ultrasonography for the localization of parathyroid tumors and for the assessment of the severity of the secondary hyperparathyroidism. Our studies demonstrate that patients with larger hyperplastic parathyroid glands as From the Surgical, Radiology, and Endocrine Services, Veterans Administration Medical Center, and the Departments of Surgery, Radiology, and Endocrinology, University of California. San Francisco, California, and tha Department of Radiology, Massachusetts General Hospital, So&on. Massachusetts. Supported In part by the Me&al Research Service of the Veterans Admlnistration htedlcal Center. Requests for reprints shcutd be addressed to Crlo H. Clark, MD, Veterans Administratbn h4sdkaf Center, Swglcal Service (112). 4150 Clement Street, San Francisco, California 94121. Presented at the 56th Annual Meeting of the Pacific Coast Surgical Association, Monterey, California, February 17-20. 1985.
Volume 150, July 1995
demonstrated by preoperative ultrasonography are unlikely to benefit from continued medical therapy. Most of these patients are symptomatic and have markedly elevated parathyroid hormone levels. Thus, the demonstration of large parathyroid glands by ultrasonography not only identifies the position of many of these abnormal glands, but also suggests that surgical treatment will be necessary. Material and Methods From January 1981 until August 1984,32 consecutive patients aged 17 to 69 years (mean age 48 years) with secondary or tertiary hyperparathyroidism were treated surgically, and 27 of these patients were evaluated by preoperative ultrasonography. There were I3 women and 19 men. Two patients had persistent and four patients had recurrent hyperparathyroidism after one (four patients) or two (two patients) previous parathyroid explorations with removal of from 2% to 3% hyperplastic parathyroid glands. Four of these patients were referred from other institutions. All patients were symptomatic with fatigue, bone pain, pruritus, and arthralgias or had bone disease. Preoperative calcium levels ranged from 9.9 to 13.9 mg/dl (normal 8.5 to 10.1 mg/dl), phosphate 1.5 to 9.9 mg/dl (normal 2.5 to 4.5 mg/dl), parathyroid hormone 1,006 to 19,006 ~1 Eq/ml (normal less than 40 ~1 Eq/ml), and alkaline phosphatase levels were elevated in all but one patient (Table I). Ultrasonography was performed using high resolution (10 mHz) real-time ultrasonography (Picker Microview, Northfield, CT). Ultrasonography with this equipment in experienced hands can detect abnormal parathyroid glands as small as 0.5 cm. Hyperplastic and adenomatous parathyroid glands are generally oval and of low echogenicity. They differ from cysts, which cause en-
9
Clark et al
TABLE I Patient 1 2 3 4 5 6 7 6 9 10 11 12 13 14 15 16 17 16 19 20 21 22 23 24 25 26 27 26 29 30 31 32
Preoperative Laboratory Data In 32 Patients With Secondary or Tertiary Hyperparathyroidism* Age (yr) & Sex 60, F 45, M 47, M 59, M 69, M 56, M 35, M 45, M 36, M 30, F+ 50, M 39. F 52, M 47, M 60, M 26, M+ 31, F 61, F 47, F 42, M 42, M 17, M 49, F 49, M 46, F+ 61, F 5 1, F Persistent 56, F Persistent 34, F Recurrent 36, F Recurrent 61, M Recurrent 50, M Recurrent
Calcium
Phosphorus
PTH
11 10.6 9.9 11.3 11 10.7 11 10.6 11 12 10.7 11 10.6 10.6 10.6 10.5 12.6 12.2 11.9 11.9 9.9 11.9 12.9 10.5 10.6 11.6 10.4 11 10.6 11 11.3 13.9
3.7
2,600 2,960 6,700 7,200 2,000 6,000 7,200 7,360 19,000 166 2,600
z 6:9 1.5 5.1 3.7 4.5 4 1.9 9 6.2 2.9 2.9 6.7 2.6 5.9 3.7 5.7 4.2 6.4 9.3 7 3.6 3.1 5.6 2.5 5.3 4.5 5 5.4 9.9
Normal range for serum calcium 6.5 to 10.1 mg/dl, phosphorus 2.5 to 4.5 mg/dl, and alkaline phosphatase (AP) less than 73 units/liter. + Tertiary status after successful renal transplantation. t = increased level of PTH or AP performed by different assay or method. l
hancement of echoes posterior to a fluid collection (Figure 1, top left and right), and there is no enhancement of echoes posterior to a solid parathyroid gland (Figure 1, bottom left and right). Parathyroid glands are best identified by ultrasonography when they are in contact with the thyroid gland, and about 80 to 85 percent are situated in that position. Parathyroid tumors that are not in contact with the thyroid gland are more difficult to detect by ultrasonography. Parathyroid tumors behind the trachea or esophagus are usually obscured since sound does not penetrate or is deflected by these structures. Enlarged parathyroid glands beneath the sternum also cannot be identified. Serum immunoreactive parathyroid hormone values were determined by radioimmunoassay in one of our laboratories (CDA) at the Veterans Administration Medical Center, San Francisco using a highly sensitive antibody for midregional fragment. Results Twenty-seven of the 32 patients having operations for secondary or tertiary hyperparathyroidism were examined preoperatively by ultrasonography. Fifty-nine of a possible 84 hyperplastic parathyroid
10
parathyroid
t 5,060 5,760 t 156 5,000 t 5,400 4,570 2,500 1,000 6,000 4,400 1,190 t 2,600 1,100 7,000 7,000 6,200 2,360 hormone
AP 347 347 512 1: 2,350 196 135 640 94 122 202 1 202 3:: 366 236 >lZ 120 160 100 97 305
i 249
t
(PTH) less than 40 ~1 Eq/ml,
glands (70 percent) were identified (assuming there were 4 glands per patient) in the 21 patients without
previous parathyroid operations. In seven patients, four glands were observed by ultrasonography, in six patients three glands, in six patients two glands, in one patient one gland, and in one patient no enlarged glands were observed (Table II). The one patient (Patient 11) who had no abnormal parathyroid tissue identified by ultrasonography had a 1.2 by 0.9 by 0.8 cm hyperplastic parathyroid gland situated within the mediastinum and two minimally enlarged (0.7 by 0.5 by 0.3 cm) upper hyperplastic parathyroid glands. At operation, 22 of these 26 patients (85 percent) had four hyperplastic glands identified, 3 patients (12 percent) had three glands identified, and 1 patient (4 percent) had five glands identified (Table II). The ultrasonographic and operative findings in the patient with five parathyroid glands (Patient 9) are noted in Figures 2 and 3. The size of the abnormal glands identified by ultrasonography ranged from 0.5 by 0.4 by 0.4 cm to 2.5 by 2 by 1.5 cm. Most glands identified by ultrasono-
The Amerlcsn Journal of Surgery
Parathyroid Localization
Flgure 1. Top left, uftrasonogram obtained with a 10 mHz real-time scanner of a thyroid cyst in a IongMudlnal plane from a 60 year old man w/th hyperparathyroidism. Note enhancement of echoes posterior to small hypoechoelc thyroid cyst (arrows). Top right, in the transverse plane, tfte thyroid cyst can again be seen with enhancement posterior to the cyst (arrows). 7?tecarotid artery is atso observed (slngh? anow) with enhancement of ecfwes posteriorly. Bottom, a small hyperplastlc parathyroid gland in a 50 year old man with chronic renal la/lure in the hmgltudinalplane (left) and the transverse plane (right). The hypoechoeic parathyroid gland in the transverse plane Is seen In the lefl panel medial and posterior to the left carol/d artery (small arrow) and lateral to the left lobe of the thyroid gland.
Volume 150, July 1995
11
Clark et
al
TABLE II
No. of Hyperplastic Glands Identified
Results of Uitrasonographyand Surgery in 47 Patlents With Secondary or Tertiary HyperparathyroldismWho Did Not Have Previous Parathyrold Surgery
Comments Ultrasonography’ n %
Surgery n
%
5
0
4 3 2 I 0
7 8 8 1 1
33 33 29 29 4 4
1 22 3 ... ...
4 85 12 ... ...
...
...
21
...
28
...
Total
l Assuming 4 parathyroid glands per patient, ultrasonography identified 59 of 84 or 70 percent.
graphy were about 1.5 by 1 by 1 cm in size. The 25 parathyroid glands that were missed by ultrasonography were either substernal (9 glands, 36 percent), 1 cm or less in diameter (7 glands, 28 percent), or situated behind the trachea or esophagus or posteriorly in the neck (8 glands, 32 percent). Serum parathyroid hormone values were higher in patients with the largest hyperplastic parathyroid glands (Figure 4). Six abnormal parathyroid glands were identified by preoperative ultrasonography in five of the six patients who had previous parathyroid operations. They ranged in size from 1.1 by 0.8 by 0.4 cm to 3.2 by 2 by 1.2 cm (Figure 5). The only abnormal parathyroid gland not identified by ultrasonography in patients requiring reoperation was a 0.9 by 0.8 by 0.8 cm hyperplastic parathyroid gland in the left thymus within the mediastinum (Patient 27). Serum calcium levels decreased to within normal limits after parathyroidectomy in all patients although in four, increased parathyroid hormone levels have redeveloped and three have recurrent hypercalcemia. One of the latter patients (Patient 30) has subsequently had hyperplastic autotransplanted parathyroid tissue removed from her arm. The second patient (Patient 13) had only three parathyroid glands identified at operation and a nonrecurrent laryngeal nerve. His fourth hyperplastic gland is probably in an ectopic position. The third patient (Patient 19) redeveloped mild hypercalcemia despite removal of 34/5 hyperplastic parathyroid glands, probably secondary to a fifth hyperplastic parathyroid gland. Complications included postoperative hemorrhage requiring reoperation in one patient (Patient 19) and temporary vocal cord paralysis in one patient (Patient 17). In the latter patient, hoarseness started 1 week after parathyroidectomy and resolved spontaneously after 8 months. In most patients, marked clinical improvement developed after successful
12
parathyroidectomy with loss of bone pain, loss of pruritus, increased energy, decreased depression with fewer headaches, and improved strength.
High resolution real-time ultrasonography is valuable for identifying hyperplastic parathyroid glands in patients with secondary hyperparathyroidism. This is especially true for patients with recurrent or persistent secondary hyperparathyroidism in whom surgery is more difficult because the normal tissue planes may be obliterated and the hyperplastic parathyroid tissue is more likely to be in an ectopic position. Most abnormal parathyroid glands situated in the neck can be identified by ultrasonography. This makes surgery easier. As demonstrated by this study, patients with very large parathyroid glands usually have higher serum parathyroid hormone values and more severe secondary hyperparathyroidism. Such patients are unlikely to respond to medical therapy. Surgery rather than continued medical treatment is therefore recommended in these patients, even if they are asymptomatic. The ultrasonographic findings along with the serum parathyroid hormone levels thus help select patients for appropriate treatment. Takagi and co-workers [!I] stated that “the demonstration of hypertrophied parathyroid glands in itself provides undeniable evidence of secondary hyperparathyroidism.” These investigators identified 28 of 32 excised hyperplastic parathyroid glands (87.5 percent) that weighed more than 500 mg. They were less successful, however, in demonstrating smaller hyperplastic parathyroid glands since only 7 of 19 glands (37 percent) were identified. Despite the excellent ability of 10 mHz real-time ultrasonography to localize hyperplastic parathyroid glands, the diagnosis of secondary hyperparathyroidism should not be doubted if other clinical data support this diagnosis yet no parathyroid glands are identified. When ultrasonography fails to reveal any abnormal parathyroid glands in patients with clinical evidence suggesting severe secondary hyperparathyroidism, then the enlarged hyperplastic parathyroid glands are usually situated beneath the sternum, behind the trachea or esophagus, or deep within the neck. When enlarged glands are identified by this noninvasive technique, the clinician and the patient are often more likely to accept the fact that medical treatment will no longer be effective. Parathyroid ultrasonography, although very helpful, is not necessary for patients before operation, as demonstrated by the five patients who had no preoperative localizing studies. All were successfully treated at operation, and 19 hyperplastic parathyroid glands were histologically confirmed in the patients who had not had previous parathyroid operations. Ultrasonography and other localizing techniques
The American Journal of Surgery
Parathyroid Localization
Figure 2. tJttrasonogram of hyperplastfc parathyroid glands from a 38 year oki man (Patient 9) wtth severe secondary hyperparathyrvkttsm, seen in the right neck in the longitudinal plane (top left) and in the transverse plane (top right). A large hyperplastic gland is seen at the tower end of the right thyroid lobe in the longitudinal plane and a second parathyroid gland is located behind the thyroid superlory on the transverse plane. The carotid artery (small arrow) seen in the transverse plane is lateral to the right lobe of the thyrow and anterior and lateral to the large hypoechoeic parathyroid gland (two arrows). Large hyperplastic parathyroid glands were also identifted In the left side of the neck in longitudinal (bottom reft) and transverse (bottom right) planes.
such as thallium-201 and technetium-99m pertechnetate scanning, and computerized tomographic scanning are recommended, however, for virtually all patients who have recurrent or persistent hyperparathyroidism. If the lesion identified by ultraso-
Volume 150, July 1995
nography is suggestive but not diagnostic of a parathyroid gland, its origin can be confirmed by cytologic examination of the aspiration biopsy specimen or by parathyroid hormone bioassay or immunoassay of the aspirated cells [6,7].
13
Clark et al
Figure 3. Operative view of two hyperplastkparathyrvktgtandsobsenmdlnths rlght slde of the neck (arrows) (top left) and three glands in the /e/t stde of the neck (arrows) (top right). The thyrold g/and is retracted medla/ly wlth a stitch (top left) or wlth the fhgers (top rlght ). Bottom, the five removed hyperplastic glaab. A smallportkm (apptvxlmate& 50 mg) of the left lower gland remalns.
.
18Y=ox+b
16-z *
k0.71 p< 0.001
14-
g 5 12c 0 “> loE’ 8_ .-E; --t E 6-
. I
1
8 Tumor 14
Size
1
I
I
10
12
Figure 4. The cot&at/on of parathyrokt hormune(Pnf)andtlmwrslzelnpatlents with secondary hyperparathyroldlsm. Serum lmmunoreactlve parathyfold hormone level (F/ Eg/ml, normal less than 40 pl Eqfml on vertical axls ls plotted agalnst the total dlameter In centkneters of the parathyrokf glands (assuming four glandsperpatlent) onthehorkontalaxls Pat&ntswMbh&fterP7Hvakh9stendedto have larger hyperplastlc parathyrold glands. The American Journal of Surgery
Parathyroid
Localization
Figure 5. Uftrasonogram of a 7.1 by 0.8 by 0.4 cm right lower hyperpiastic parathyroid g/and in a 58 year old woman with persistent secondary hyprparathyroidism (left) and a 3.22 by 2 by 1.2 cm left lower hyperplastic parathyroid g/and (right) in a 50 year old man with recurrent secondary hyperparathyroidism. This large hyperpiastic parathyroid gland displaced the carotid artery posteriorly.
Summary Thirty-two patients were treated surgically for symptomatic secondary or tertiary hyperparathyroidism, and 27 of these patients had high resolution (10 mHz) real-time ultrasonography before parathyroidectomy. This preoperative localization study identified one or more enlarged hyperplastic parathyroid glands in all but one patient who had not had a previous parathyroid operation, and in five of six patients who did have previous parathyroid operations. In both of the patients in whom no parathyroid glands were identified by ultrasonography the only abnormal enlarged parathyroid glands were those situated within the superior mediastinum. When large glands are not observed by ultrasonography in patients with severe secondary hyperparathyroidism, the glands are usually situated in the superior mediastinum, behind the trachea or esophagus, or deeply within the neck. The size of the parathyroid glands correlated positively with the serum parathyroid hormone level and with the severity of the secondary hyperparathyroidism. Thus, the preoperative identification of parathyroid glands by ultrasonography not only localizes the site of most hyperplastic parathyroid glands (70 percent of patients), but also detects those patients who have enlarged parathyroid glands, elevated serum parathyVolume
150,July
1995
roid hormone levels, and severe secondary hyperparathyroidism. These are the patients who are thus unlikely to respond to further medical therapy. Acknowledgmenti We thank our colleagues in the departments of nephrology and endocrinology who helped us care for the patients. We also thank Brenda Brown, Freda Trevett, Wayne Leonhardt, and Dr. Gregory Linkowski for their excellent assistance.
References 1. Clark OH. Secondary and tertiary hyperparathyroidism. 2.
3.
4. 5.
6.
In: Najarian JS, Delaney JP, eds. Endocrine surgery. Miami: Symposia Specialists, 1981:239-47. Johnson WJ, Goldsmith RS, Beabolt JW. et al. Prevention and reversal of progressive secondary hyperparathyroidism in patients maintained by hemodialysis. Am J Med 197456: 827-32. Slatopolsky E, Coburn JW. Secondary hyperparathyroidism: pathogenesis, diagnosis and treatment. In: Kaplan EL, ed. Surgery of the thyroid and parathyroid glands. Edinburgh: Churchill Livingstone, 1983:224-42. David DS. Calcium metabolism in renal failure. Am J Med 1975;58:48-56. Takagi H, Tominaga Y, Uchida K, et al. Image diagnosis of parathyroid glands in chronic renal failure. Ann Surg 1983; 198:74-9. Clark OH, Stark DD, Gocding GAW. et al. Localization procedures in patients requiring reoperation for hyperparathyroidism. World J Surg 1984;8:509-21.
15
Clark et al
7. Gocding GAW, Clark OH, Stark DID, et al. Parathyroid aspiration biopsy via ultrasonic guidance in hyperparathyroid patients with prior thyroid or parathyroid surgery. Radiology 1965; 155:193-6.
Discussion Leonard Rosoff (Los Angeles, CA): My critique will be confined essentially to the conclusions of the authors regarding the value of ultrasonography in the management of secondary and tertiary hyperparathyroidism-conclusions that I do not believe are warranted by the results of this study. The statistics presented are confusing. Dr. Clark, you have stated that 27 patients had preoperative ultrasonography and then claimed an accuracy rate of 70 percent because 59 glands are identified in 21 patients out of a possible 84 glands, based on an estimate of four glands per patient. About 15 years ago, I was doing total parathyroidectomies for this condition, as was Ogg in England and others. To achieve totality, a meticulous search for all parathyroid glands was necessary. In approximately 50 percent of the relatively small series of patients (approximately 20), five or more glands were found. Dr. Clark, if you would have estimated that half of the patients had five instead of four glands instead of identifying 59 of a possible 84 glands, you would have identified 59 of a possible 94 glands-an accuracy rate of 63 percent instead of 70 percent. Obviously, such projections are not truly scientific and are meaningless, as well as misleading. More importantly, the real question is how valuable is ultrasonography to the individual patient? There is a current tendency, especially by radiologists, to assess the value of parathyroid localization on the basis of the number of glands identified rather than on an evaluation on a case-by-case basis. This manner of reportage is confusing and deceptive. In the series reported herein, the ultrasound studies were correct in only 6 of the 21 patients not operated on previously. In other words, in 70 percent of these patients, the findings were imprecise and inaccurate. Such dismal results hardly justify the statement proclaiming ‘the excellent ability of 10 mHz real-time ultrasonography for localizing hyperplastic parathyroid glands.’ And although, Dr. Clark, you state that the technology can demonstrate glands as small as 0.5 cm diameter, you excuse some of the failures on the basis that the glands were small, measuring 0.7 cm in diameter. But most disturbing is the statement that ultrasonography only identifies a large parathyroid gland in contact with the thyroid and has difficulty identifying glands deep in the neck or in the mediastinum. This is hardly an excellent ability to identify parathyroid glands when, as any one who does parathyroid surgery knows, an important problem at operation is whether a missing fourth gland or possibly a fifth gland may be present deep in the neck or in the mediastinum. I find it difficult to believe that you, Dr. Clark, or any experienced parathyroid surgeon with a 95 percent or better success rate for the initial operation for primary hyperparathyroidism, would have any difficulty in identifying the characteristic large, yellow, hypertrophied glands in the neck in contact with the thyroid. With or without ultrasonography, one would have to explore deep in the neck or in the superior mediastinum to find a fourth gland or rule out a supernumerary parathyroid gland in these locations. So how does ultrasonography aid the surgeon?
16
Admittedly, there is a role for high resolution ultrasonography in patients with recurrent or persistent hypercalcemia after a previous neck operation for either primary or secondary hyperparathyroidism, and it is especially helpful in the differential diagnosis of thyroid nodules. However, if I could use only one test for the identification of parathyroid glands-it would be high resolution computerized tomography and not ultrasonography. And, finally, there was nothing indicating that the glands of patients being treated medically were measured ultrasonographically and compared with the glands in the surgically-treated patients. Dr. Clark, what proof is there in your study or any other study for the statement that enlarged parathyroid glands are unlikely to regress with medical therapy? And, without such a comparison, how do you conclude that determining the size of the glands by ultrasonography is of help in predicting which patients need surgical treatment?
Orlo H. Clark (closing): The results of ultrasound using a 10 mHz real-time scanner are comparable to those of computerized tomographic scanning and thallium-201 chloride and technetium-99m pertechnetate scanning. Thus, in the best of circumstances, approximately 70 to 80 percent of hyperplastic or adenomatous parathyroid glands will be demonstrated by any of these techniques as we have reported in previous prospective studies. All of these preoperative noninvasive localization techniques have the most difficulty, however, in identifying the parathyroid glands of patients who have multiple abnormal parathyroid glands; frequently, the largest glands may be identified but the smaller abnormal glands may be missed. This point should not be underemphasized and, Dr. Rosoff, you have appropriately noted that although 70 percent of our patients had their abnormal parathyroid glands identified by ultrasonography, 30 percent did not. Surgeons would obviously be very unhappy with a curative rate of only 70 percent since most studies demonstrate that more than 90 percent of patients with hyperparathyroidism are successfully treated at the initial operation. Dr. Rosoff, you have also appropriately noted that patients may have more than four glands. This fact is especially important in patients with primary and secondary hyperplasia. In these patients, in contrast to those with solitary adenoma, I remove both lobes of the thymus to prevent persistent hyperparathyroidism since about 15 percent of these patients will have a fifth parathyroid gland and this parathyroid gland is frequently situated within the thymus. Although ultrasonography, computerized tomographic scanning, and thallium and pertechnetate scanning have comparable overall results, I prefer to use ultrasonography for localization in patients who have not had previous operations. There are four reasons for this. First, the results of ultrasonography help me even when they are negative. When no parathyroid gland is identified in the neck, I know that the abnormal parathyroid gland is either situated beneath the sternum, usually in the thymus, or deep in the neck, often in a retrotracheal or retroesophageal position. When the computerized tomographic scans or thallium and technetium scans are negative, the abnormal parathyroid gland may be anywhere. Second, ultrasonography can identify small intrathyroid lesions, and although an intrathyroidal position is unusual, it is difficult at operation to identify a gland there. Third, ultrasonography is con-
The American Journal of Surgery
Parathyroid Localization
siderably less expensive than computerized tomography or thallium and pertechnetate scans. Fourth, there is virtually no biologic hazard from ultrasonography. In patients with persistent or recurrent primary or secondary hyperparathyroidism all these studies are helpful, and the results of the various localizing studies are complementary. Ultrasonography identifies glands adjacent to or within the thyroid. Computerized tomographic scans identify larger glands and are excellent for identifying substernal parathyroid glands and glands in ectopic positions. Thallium and pertechnetate scans identify abnormal parathyroid glands with a rich blood flow. It is good for both cervical and large substernal parathyroid tumors. Dr. Rosoff, you asked whether we can make the conclusion that identifying large parathyroid glands by ultrasonography means that these patients probably will require surgical rather than medical therapy. This conclusion is supported by the strong correlation between the degree of elevation in serum parathyroid hormone values and the size of the parathyroid glands. Since there is usually a strong
Volume 150, July 1995
correlation between the serum parathyroid hormone level and the severity of the secondary hyperparathyroidism, this seems acceptable. We have also reviewed 10 other patients receiving long-term hemodialysis or peritoneal dialysis who also were examined by ultrasonography. In these patients, either no enlarged parathyroid glands were identified or the abnormal gland or glands were less than 1 cm. The serum parathyroid hormone levels were also less than 1,000 ~1 Eq/ml in these patients in contrast with considerably higher levels in the patients we treated surgically. Although these 10 patients had mild secondary hyperparathyroidism, their only slightly elevated serum parathyroid hormone levels, minimal to absent symptoms, and normocalcemia strongly suggest that they would be better managed medically than surgically. The identification of one or more large parathyroid glands by preoperative localization techniques is another objective finding along with symptoms, bone disease, and laboratory data that strongly suggests that surgery rather than continued medical therapy is indicated.
17
Value of High Resolution Real-Time Ultrasonography in Secondary Hyperparathyroidism
Orlo H. Clark, MD, San Francisco, California David A. Stark, MD, Boston, Massachusetts Wan-Y. Duh, MD, San Francisco, California Claude D. Arnaud, MD, San Francisco, California Gretchen A. W. Gooding, MD, San Francisco, California
Good medical treatment is effective in preventing secondary hyperparathyroidism in most patients with chronic renal failure, but in some patients, progressive secondary hyperparathyroidism develops with accompanying fatigue, arthralgias, pruritus, and bone pain, as well as a myriad of other symptoms [I-3]. Parathyroidectomy is an effective means of treating these problems, but absolute criteria for selecting patients for surgical rather than medical treatment are not universally accepted. Certainly symptomatic patients, patients with a calciumphosphate product greater than 70 mg/dl, patients with osteitis fibrosa cystica, those with persistent hypercalcemia (calcium level greater than 11 mg/dl), and those with extensive soft tissue calcification benefit from parathyroidectomy [4]. During the past several years, we have been interested in the value of ultrasonography for the localization of parathyroid tumors and for the assessment of the severity of the secondary hyperparathyroidism. Our studies demonstrate that patients with larger hyperplastic parathyroid glands as From the Surgical, Radiology, and Endocrine Services, Veterans Administration Medical Center, and the Departments of Surgery, Radiology, and Endocrinology, University of California. San Francisco, California, and tha Department of Radiology, Massachusetts General Hospital, So&on. Massachusetts. Supported In part by the Me&al Research Service of the Veterans Admlnistration htedlcal Center. Requests for reprints shcutd be addressed to Crlo H. Clark, MD, Veterans Administratbn h4sdkaf Center, Swglcal Service (112). 4150 Clement Street, San Francisco, California 94121. Presented at the 56th Annual Meeting of the Pacific Coast Surgical Association, Monterey, California, February 17-20. 1985.
Volume 150, July 1995
demonstrated by preoperative ultrasonography are unlikely to benefit from continued medical therapy. Most of these patients are symptomatic and have markedly elevated parathyroid hormone levels. Thus, the demonstration of large parathyroid glands by ultrasonography not only identifies the position of many of these abnormal glands, but also suggests that surgical treatment will be necessary. Material and Methods From January 1981 until August 1984,32 consecutive patients aged 17 to 69 years (mean age 48 years) with secondary or tertiary hyperparathyroidism were treated surgically, and 27 of these patients were evaluated by preoperative ultrasonography. There were I3 women and 19 men. Two patients had persistent and four patients had recurrent hyperparathyroidism after one (four patients) or two (two patients) previous parathyroid explorations with removal of from 2% to 3% hyperplastic parathyroid glands. Four of these patients were referred from other institutions. All patients were symptomatic with fatigue, bone pain, pruritus, and arthralgias or had bone disease. Preoperative calcium levels ranged from 9.9 to 13.9 mg/dl (normal 8.5 to 10.1 mg/dl), phosphate 1.5 to 9.9 mg/dl (normal 2.5 to 4.5 mg/dl), parathyroid hormone 1,006 to 19,006 ~1 Eq/ml (normal less than 40 ~1 Eq/ml), and alkaline phosphatase levels were elevated in all but one patient (Table I). Ultrasonography was performed using high resolution (10 mHz) real-time ultrasonography (Picker Microview, Northfield, CT). Ultrasonography with this equipment in experienced hands can detect abnormal parathyroid glands as small as 0.5 cm. Hyperplastic and adenomatous parathyroid glands are generally oval and of low echogenicity. They differ from cysts, which cause en-
9
Clark et al
TABLE I Patient 1 2 3 4 5 6 7 6 9 10 11 12 13 14 15 16 17 16 19 20 21 22 23 24 25 26 27 26 29 30 31 32
Preoperative Laboratory Data In 32 Patients With Secondary or Tertiary Hyperparathyroidism* Age (yr) & Sex 60, F 45, M 47, M 59, M 69, M 56, M 35, M 45, M 36, M 30, F+ 50, M 39. F 52, M 47, M 60, M 26, M+ 31, F 61, F 47, F 42, M 42, M 17, M 49, F 49, M 46, F+ 61, F 5 1, F Persistent 56, F Persistent 34, F Recurrent 36, F Recurrent 61, M Recurrent 50, M Recurrent
Calcium
Phosphorus
PTH
11 10.6 9.9 11.3 11 10.7 11 10.6 11 12 10.7 11 10.6 10.6 10.6 10.5 12.6 12.2 11.9 11.9 9.9 11.9 12.9 10.5 10.6 11.6 10.4 11 10.6 11 11.3 13.9
3.7
2,600 2,960 6,700 7,200 2,000 6,000 7,200 7,360 19,000 166 2,600
z 6:9 1.5 5.1 3.7 4.5 4 1.9 9 6.2 2.9 2.9 6.7 2.6 5.9 3.7 5.7 4.2 6.4 9.3 7 3.6 3.1 5.6 2.5 5.3 4.5 5 5.4 9.9
Normal range for serum calcium 6.5 to 10.1 mg/dl, phosphorus 2.5 to 4.5 mg/dl, and alkaline phosphatase (AP) less than 73 units/liter. + Tertiary status after successful renal transplantation. t = increased level of PTH or AP performed by different assay or method. l
hancement of echoes posterior to a fluid collection (Figure 1, top left and right), and there is no enhancement of echoes posterior to a solid parathyroid gland (Figure 1, bottom left and right). Parathyroid glands are best identified by ultrasonography when they are in contact with the thyroid gland, and about 80 to 85 percent are situated in that position. Parathyroid tumors that are not in contact with the thyroid gland are more difficult to detect by ultrasonography. Parathyroid tumors behind the trachea or esophagus are usually obscured since sound does not penetrate or is deflected by these structures. Enlarged parathyroid glands beneath the sternum also cannot be identified. Serum immunoreactive parathyroid hormone values were determined by radioimmunoassay in one of our laboratories (CDA) at the Veterans Administration Medical Center, San Francisco using a highly sensitive antibody for midregional fragment. Results Twenty-seven of the 32 patients having operations for secondary or tertiary hyperparathyroidism were examined preoperatively by ultrasonography. Fifty-nine of a possible 84 hyperplastic parathyroid
10
parathyroid
t 5,060 5,760 t 156 5,000 t 5,400 4,570 2,500 1,000 6,000 4,400 1,190 t 2,600 1,100 7,000 7,000 6,200 2,360 hormone
AP 347 347 512 1: 2,350 196 135 640 94 122 202 1 202 3:: 366 236 >lZ 120 160 100 97 305
i 249
t
(PTH) less than 40 ~1 Eq/ml,
glands (70 percent) were identified (assuming there were 4 glands per patient) in the 21 patients without
previous parathyroid operations. In seven patients, four glands were observed by ultrasonography, in six patients three glands, in six patients two glands, in one patient one gland, and in one patient no enlarged glands were observed (Table II). The one patient (Patient 11) who had no abnormal parathyroid tissue identified by ultrasonography had a 1.2 by 0.9 by 0.8 cm hyperplastic parathyroid gland situated within the mediastinum and two minimally enlarged (0.7 by 0.5 by 0.3 cm) upper hyperplastic parathyroid glands. At operation, 22 of these 26 patients (85 percent) had four hyperplastic glands identified, 3 patients (12 percent) had three glands identified, and 1 patient (4 percent) had five glands identified (Table II). The ultrasonographic and operative findings in the patient with five parathyroid glands (Patient 9) are noted in Figures 2 and 3. The size of the abnormal glands identified by ultrasonography ranged from 0.5 by 0.4 by 0.4 cm to 2.5 by 2 by 1.5 cm. Most glands identified by ultrasono-
The Amerlcsn Journal of Surgery
Parathyroid Localization
Flgure 1. Top left, uftrasonogram obtained with a 10 mHz real-time scanner of a thyroid cyst in a IongMudlnal plane from a 60 year old man w/th hyperparathyroidism. Note enhancement of echoes posterior to small hypoechoelc thyroid cyst (arrows). Top right, in the transverse plane, tfte thyroid cyst can again be seen with enhancement posterior to the cyst (arrows). 7?tecarotid artery is atso observed (slngh? anow) with enhancement of ecfwes posteriorly. Bottom, a small hyperplastlc parathyroid gland in a 50 year old man with chronic renal la/lure in the hmgltudinalplane (left) and the transverse plane (right). The hypoechoeic parathyroid gland in the transverse plane Is seen In the lefl panel medial and posterior to the left carol/d artery (small arrow) and lateral to the left lobe of the thyroid gland.
Volume 150, July 1995
11
Clark et
al
TABLE II
No. of Hyperplastic Glands Identified
Results of Uitrasonographyand Surgery in 47 Patlents With Secondary or Tertiary HyperparathyroldismWho Did Not Have Previous Parathyrold Surgery
Comments Ultrasonography’ n %
Surgery n
%
5
0
4 3 2 I 0
7 8 8 1 1
33 33 29 29 4 4
1 22 3 ... ...
4 85 12 ... ...
...
...
21
...
28
...
Total
l Assuming 4 parathyroid glands per patient, ultrasonography identified 59 of 84 or 70 percent.
graphy were about 1.5 by 1 by 1 cm in size. The 25 parathyroid glands that were missed by ultrasonography were either substernal (9 glands, 36 percent), 1 cm or less in diameter (7 glands, 28 percent), or situated behind the trachea or esophagus or posteriorly in the neck (8 glands, 32 percent). Serum parathyroid hormone values were higher in patients with the largest hyperplastic parathyroid glands (Figure 4). Six abnormal parathyroid glands were identified by preoperative ultrasonography in five of the six patients who had previous parathyroid operations. They ranged in size from 1.1 by 0.8 by 0.4 cm to 3.2 by 2 by 1.2 cm (Figure 5). The only abnormal parathyroid gland not identified by ultrasonography in patients requiring reoperation was a 0.9 by 0.8 by 0.8 cm hyperplastic parathyroid gland in the left thymus within the mediastinum (Patient 27). Serum calcium levels decreased to within normal limits after parathyroidectomy in all patients although in four, increased parathyroid hormone levels have redeveloped and three have recurrent hypercalcemia. One of the latter patients (Patient 30) has subsequently had hyperplastic autotransplanted parathyroid tissue removed from her arm. The second patient (Patient 13) had only three parathyroid glands identified at operation and a nonrecurrent laryngeal nerve. His fourth hyperplastic gland is probably in an ectopic position. The third patient (Patient 19) redeveloped mild hypercalcemia despite removal of 34/5 hyperplastic parathyroid glands, probably secondary to a fifth hyperplastic parathyroid gland. Complications included postoperative hemorrhage requiring reoperation in one patient (Patient 19) and temporary vocal cord paralysis in one patient (Patient 17). In the latter patient, hoarseness started 1 week after parathyroidectomy and resolved spontaneously after 8 months. In most patients, marked clinical improvement developed after successful
12
parathyroidectomy with loss of bone pain, loss of pruritus, increased energy, decreased depression with fewer headaches, and improved strength.
High resolution real-time ultrasonography is valuable for identifying hyperplastic parathyroid glands in patients with secondary hyperparathyroidism. This is especially true for patients with recurrent or persistent secondary hyperparathyroidism in whom surgery is more difficult because the normal tissue planes may be obliterated and the hyperplastic parathyroid tissue is more likely to be in an ectopic position. Most abnormal parathyroid glands situated in the neck can be identified by ultrasonography. This makes surgery easier. As demonstrated by this study, patients with very large parathyroid glands usually have higher serum parathyroid hormone values and more severe secondary hyperparathyroidism. Such patients are unlikely to respond to medical therapy. Surgery rather than continued medical treatment is therefore recommended in these patients, even if they are asymptomatic. The ultrasonographic findings along with the serum parathyroid hormone levels thus help select patients for appropriate treatment. Takagi and co-workers [!I] stated that “the demonstration of hypertrophied parathyroid glands in itself provides undeniable evidence of secondary hyperparathyroidism.” These investigators identified 28 of 32 excised hyperplastic parathyroid glands (87.5 percent) that weighed more than 500 mg. They were less successful, however, in demonstrating smaller hyperplastic parathyroid glands since only 7 of 19 glands (37 percent) were identified. Despite the excellent ability of 10 mHz real-time ultrasonography to localize hyperplastic parathyroid glands, the diagnosis of secondary hyperparathyroidism should not be doubted if other clinical data support this diagnosis yet no parathyroid glands are identified. When ultrasonography fails to reveal any abnormal parathyroid glands in patients with clinical evidence suggesting severe secondary hyperparathyroidism, then the enlarged hyperplastic parathyroid glands are usually situated beneath the sternum, behind the trachea or esophagus, or deep within the neck. When enlarged glands are identified by this noninvasive technique, the clinician and the patient are often more likely to accept the fact that medical treatment will no longer be effective. Parathyroid ultrasonography, although very helpful, is not necessary for patients before operation, as demonstrated by the five patients who had no preoperative localizing studies. All were successfully treated at operation, and 19 hyperplastic parathyroid glands were histologically confirmed in the patients who had not had previous parathyroid operations. Ultrasonography and other localizing techniques
The American Journal of Surgery
Parathyroid Localization
Figure 2. tJttrasonogram of hyperplastfc parathyroid glands from a 38 year oki man (Patient 9) wtth severe secondary hyperparathyrvkttsm, seen in the right neck in the longitudinal plane (top left) and in the transverse plane (top right). A large hyperplastic gland is seen at the tower end of the right thyroid lobe in the longitudinal plane and a second parathyroid gland is located behind the thyroid superlory on the transverse plane. The carotid artery (small arrow) seen in the transverse plane is lateral to the right lobe of the thyrow and anterior and lateral to the large hypoechoeic parathyroid gland (two arrows). Large hyperplastic parathyroid glands were also identifted In the left side of the neck in longitudinal (bottom reft) and transverse (bottom right) planes.
such as thallium-201 and technetium-99m pertechnetate scanning, and computerized tomographic scanning are recommended, however, for virtually all patients who have recurrent or persistent hyperparathyroidism. If the lesion identified by ultraso-
Volume 150, July 1995
nography is suggestive but not diagnostic of a parathyroid gland, its origin can be confirmed by cytologic examination of the aspiration biopsy specimen or by parathyroid hormone bioassay or immunoassay of the aspirated cells [6,7].
13
Clark et al
Figure 3. Operative view of two hyperplastkparathyrvktgtandsobsenmdlnths rlght slde of the neck (arrows) (top left) and three glands in the /e/t stde of the neck (arrows) (top right). The thyrold g/and is retracted medla/ly wlth a stitch (top left) or wlth the fhgers (top rlght ). Bottom, the five removed hyperplastic glaab. A smallportkm (apptvxlmate& 50 mg) of the left lower gland remalns.
.
18Y=ox+b
16-z *
k0.71 p< 0.001
14-
g 5 12c 0 “> loE’ 8_ .-E; --t E 6-
. I
1
8 Tumor 14
Size
1
I
I
10
12
Figure 4. The cot&at/on of parathyrokt hormune(Pnf)andtlmwrslzelnpatlents with secondary hyperparathyroldlsm. Serum lmmunoreactlve parathyfold hormone level (F/ Eg/ml, normal less than 40 pl Eqfml on vertical axls ls plotted agalnst the total dlameter In centkneters of the parathyrokf glands (assuming four glandsperpatlent) onthehorkontalaxls Pat&ntswMbh&fterP7Hvakh9stendedto have larger hyperplastlc parathyrold glands. The American Journal of Surgery
Parathyroid
Localization
Figure 5. Uftrasonogram of a 7.1 by 0.8 by 0.4 cm right lower hyperpiastic parathyroid g/and in a 58 year old woman with persistent secondary hyprparathyroidism (left) and a 3.22 by 2 by 1.2 cm left lower hyperplastic parathyroid g/and (right) in a 50 year old man with recurrent secondary hyperparathyroidism. This large hyperpiastic parathyroid gland displaced the carotid artery posteriorly.
Summary Thirty-two patients were treated surgically for symptomatic secondary or tertiary hyperparathyroidism, and 27 of these patients had high resolution (10 mHz) real-time ultrasonography before parathyroidectomy. This preoperative localization study identified one or more enlarged hyperplastic parathyroid glands in all but one patient who had not had a previous parathyroid operation, and in five of six patients who did have previous parathyroid operations. In both of the patients in whom no parathyroid glands were identified by ultrasonography the only abnormal enlarged parathyroid glands were those situated within the superior mediastinum. When large glands are not observed by ultrasonography in patients with severe secondary hyperparathyroidism, the glands are usually situated in the superior mediastinum, behind the trachea or esophagus, or deeply within the neck. The size of the parathyroid glands correlated positively with the serum parathyroid hormone level and with the severity of the secondary hyperparathyroidism. Thus, the preoperative identification of parathyroid glands by ultrasonography not only localizes the site of most hyperplastic parathyroid glands (70 percent of patients), but also detects those patients who have enlarged parathyroid glands, elevated serum parathyVolume
150,July
1995
roid hormone levels, and severe secondary hyperparathyroidism. These are the patients who are thus unlikely to respond to further medical therapy. Acknowledgmenti We thank our colleagues in the departments of nephrology and endocrinology who helped us care for the patients. We also thank Brenda Brown, Freda Trevett, Wayne Leonhardt, and Dr. Gregory Linkowski for their excellent assistance.
References 1. Clark OH. Secondary and tertiary hyperparathyroidism. 2.
3.
4. 5.
6.
In: Najarian JS, Delaney JP, eds. Endocrine surgery. Miami: Symposia Specialists, 1981:239-47. Johnson WJ, Goldsmith RS, Beabolt JW. et al. Prevention and reversal of progressive secondary hyperparathyroidism in patients maintained by hemodialysis. Am J Med 197456: 827-32. Slatopolsky E, Coburn JW. Secondary hyperparathyroidism: pathogenesis, diagnosis and treatment. In: Kaplan EL, ed. Surgery of the thyroid and parathyroid glands. Edinburgh: Churchill Livingstone, 1983:224-42. David DS. Calcium metabolism in renal failure. Am J Med 1975;58:48-56. Takagi H, Tominaga Y, Uchida K, et al. Image diagnosis of parathyroid glands in chronic renal failure. Ann Surg 1983; 198:74-9. Clark OH, Stark DD, Gocding GAW. et al. Localization procedures in patients requiring reoperation for hyperparathyroidism. World J Surg 1984;8:509-21.
15
Clark et al
7. Gocding GAW, Clark OH, Stark DID, et al. Parathyroid aspiration biopsy via ultrasonic guidance in hyperparathyroid patients with prior thyroid or parathyroid surgery. Radiology 1965; 155:193-6.
Discussion Leonard Rosoff (Los Angeles, CA): My critique will be confined essentially to the conclusions of the authors regarding the value of ultrasonography in the management of secondary and tertiary hyperparathyroidism-conclusions that I do not believe are warranted by the results of this study. The statistics presented are confusing. Dr. Clark, you have stated that 27 patients had preoperative ultrasonography and then claimed an accuracy rate of 70 percent because 59 glands are identified in 21 patients out of a possible 84 glands, based on an estimate of four glands per patient. About 15 years ago, I was doing total parathyroidectomies for this condition, as was Ogg in England and others. To achieve totality, a meticulous search for all parathyroid glands was necessary. In approximately 50 percent of the relatively small series of patients (approximately 20), five or more glands were found. Dr. Clark, if you would have estimated that half of the patients had five instead of four glands instead of identifying 59 of a possible 84 glands, you would have identified 59 of a possible 94 glands-an accuracy rate of 63 percent instead of 70 percent. Obviously, such projections are not truly scientific and are meaningless, as well as misleading. More importantly, the real question is how valuable is ultrasonography to the individual patient? There is a current tendency, especially by radiologists, to assess the value of parathyroid localization on the basis of the number of glands identified rather than on an evaluation on a case-by-case basis. This manner of reportage is confusing and deceptive. In the series reported herein, the ultrasound studies were correct in only 6 of the 21 patients not operated on previously. In other words, in 70 percent of these patients, the findings were imprecise and inaccurate. Such dismal results hardly justify the statement proclaiming ‘the excellent ability of 10 mHz real-time ultrasonography for localizing hyperplastic parathyroid glands.’ And although, Dr. Clark, you state that the technology can demonstrate glands as small as 0.5 cm diameter, you excuse some of the failures on the basis that the glands were small, measuring 0.7 cm in diameter. But most disturbing is the statement that ultrasonography only identifies a large parathyroid gland in contact with the thyroid and has difficulty identifying glands deep in the neck or in the mediastinum. This is hardly an excellent ability to identify parathyroid glands when, as any one who does parathyroid surgery knows, an important problem at operation is whether a missing fourth gland or possibly a fifth gland may be present deep in the neck or in the mediastinum. I find it difficult to believe that you, Dr. Clark, or any experienced parathyroid surgeon with a 95 percent or better success rate for the initial operation for primary hyperparathyroidism, would have any difficulty in identifying the characteristic large, yellow, hypertrophied glands in the neck in contact with the thyroid. With or without ultrasonography, one would have to explore deep in the neck or in the superior mediastinum to find a fourth gland or rule out a supernumerary parathyroid gland in these locations. So how does ultrasonography aid the surgeon?
16
Admittedly, there is a role for high resolution ultrasonography in patients with recurrent or persistent hypercalcemia after a previous neck operation for either primary or secondary hyperparathyroidism, and it is especially helpful in the differential diagnosis of thyroid nodules. However, if I could use only one test for the identification of parathyroid glands-it would be high resolution computerized tomography and not ultrasonography. And, finally, there was nothing indicating that the glands of patients being treated medically were measured ultrasonographically and compared with the glands in the surgically-treated patients. Dr. Clark, what proof is there in your study or any other study for the statement that enlarged parathyroid glands are unlikely to regress with medical therapy? And, without such a comparison, how do you conclude that determining the size of the glands by ultrasonography is of help in predicting which patients need surgical treatment?
Orlo H. Clark (closing): The results of ultrasound using a 10 mHz real-time scanner are comparable to those of computerized tomographic scanning and thallium-201 chloride and technetium-99m pertechnetate scanning. Thus, in the best of circumstances, approximately 70 to 80 percent of hyperplastic or adenomatous parathyroid glands will be demonstrated by any of these techniques as we have reported in previous prospective studies. All of these preoperative noninvasive localization techniques have the most difficulty, however, in identifying the parathyroid glands of patients who have multiple abnormal parathyroid glands; frequently, the largest glands may be identified but the smaller abnormal glands may be missed. This point should not be underemphasized and, Dr. Rosoff, you have appropriately noted that although 70 percent of our patients had their abnormal parathyroid glands identified by ultrasonography, 30 percent did not. Surgeons would obviously be very unhappy with a curative rate of only 70 percent since most studies demonstrate that more than 90 percent of patients with hyperparathyroidism are successfully treated at the initial operation. Dr. Rosoff, you have also appropriately noted that patients may have more than four glands. This fact is especially important in patients with primary and secondary hyperplasia. In these patients, in contrast to those with solitary adenoma, I remove both lobes of the thymus to prevent persistent hyperparathyroidism since about 15 percent of these patients will have a fifth parathyroid gland and this parathyroid gland is frequently situated within the thymus. Although ultrasonography, computerized tomographic scanning, and thallium and pertechnetate scanning have comparable overall results, I prefer to use ultrasonography for localization in patients who have not had previous operations. There are four reasons for this. First, the results of ultrasonography help me even when they are negative. When no parathyroid gland is identified in the neck, I know that the abnormal parathyroid gland is either situated beneath the sternum, usually in the thymus, or deep in the neck, often in a retrotracheal or retroesophageal position. When the computerized tomographic scans or thallium and technetium scans are negative, the abnormal parathyroid gland may be anywhere. Second, ultrasonography can identify small intrathyroid lesions, and although an intrathyroidal position is unusual, it is difficult at operation to identify a gland there. Third, ultrasonography is con-
The American Journal of Surgery
Parathyroid Localization
siderably less expensive than computerized tomography or thallium and pertechnetate scans. Fourth, there is virtually no biologic hazard from ultrasonography. In patients with persistent or recurrent primary or secondary hyperparathyroidism all these studies are helpful, and the results of the various localizing studies are complementary. Ultrasonography identifies glands adjacent to or within the thyroid. Computerized tomographic scans identify larger glands and are excellent for identifying substernal parathyroid glands and glands in ectopic positions. Thallium and pertechnetate scans identify abnormal parathyroid glands with a rich blood flow. It is good for both cervical and large substernal parathyroid tumors. Dr. Rosoff, you asked whether we can make the conclusion that identifying large parathyroid glands by ultrasonography means that these patients probably will require surgical rather than medical therapy. This conclusion is supported by the strong correlation between the degree of elevation in serum parathyroid hormone values and the size of the parathyroid glands. Since there is usually a strong
Volume 150, July 1995
correlation between the serum parathyroid hormone level and the severity of the secondary hyperparathyroidism, this seems acceptable. We have also reviewed 10 other patients receiving long-term hemodialysis or peritoneal dialysis who also were examined by ultrasonography. In these patients, either no enlarged parathyroid glands were identified or the abnormal gland or glands were less than 1 cm. The serum parathyroid hormone levels were also less than 1,000 ~1 Eq/ml in these patients in contrast with considerably higher levels in the patients we treated surgically. Although these 10 patients had mild secondary hyperparathyroidism, their only slightly elevated serum parathyroid hormone levels, minimal to absent symptoms, and normocalcemia strongly suggest that they would be better managed medically than surgically. The identification of one or more large parathyroid glands by preoperative localization techniques is another objective finding along with symptoms, bone disease, and laboratory data that strongly suggests that surgery rather than continued medical therapy is indicated.
17