Parathyroid scintigraphy findings in chronic kidney disease patients with recurrent hyperparathyroidism

Dear Author

Here are the proofs of your article.

•

You can submit your corrections online or by fax.

•

For online submission please insert your corrections in the online correction form.

Always indicate the line number to which the correction refers.

•

Please return your proof together with the permission to publish confirmation.

•

For fax submission, please ensure that your corrections are clearly legible. Use a fine black pen and write the correction in the margin, not too close to the edge of the page.

•

Remember to note the journal title, article number, and your name when sending your response via e-mail, fax or regular mail.

•

Check the metadata sheet to make sure that the header information, especially author names and the corresponding affiliations are correctly shown.

•

Check the questions that may have arisen during copy editing and insert your answers/corrections.

•

Check that the text is complete and that all figures, tables and their legends are included. Also check the accuracy of special characters, equations, and electronic supplementary material if applicable. If necessary refer to the Edited manuscript.

•

The publication of inaccurate data such as dosages and units can have serious consequences. Please take particular care that all such details are correct.

•

Please do not make changes that involve only matters of style. We have generally introduced forms that follow the journal’s style.

Substantial changes in content, e.g., new results, corrected values, title and

authorship are not allowed without the approval of the responsible editor. In such a case, please contact the Editorial Office and return his/her consent together with the proof.

•

If we do not receive your corrections within 48 hours, we will send you a reminder.

Please note

Your article will be published Online First approximately one week after receipt of your corrected proofs. This is the official first publication citable with the DOI.

Further changes are, therefore, not possible.

After online publication, subscribers (personal/institutional) to this journal will have access to the complete article via the DOI using the URL:

If you would like to know when your article has been published online, take advantage of our free alert service. For registration and further information, go to:

http://www.springerlink.com.

Due to the electronic nature of the procedure, the manuscript and the original figures will only be returned to you on special request. When you return your corrections, please inform us, if you would like to have these documents returned.

The printed version will follow in a forthcoming issue.

http://dx.doi.org/10.1007/s00259-009-1313-8

To:

Re:

Permission to publish

I have checked the proofs of my article and

‰ I have no corrections. The article is ready to be published without changes.

‰ I have a few corrections. I am enclosing the following pages:

‰ I have made many corrections. Enclosed is the complete article.

Date / signature: _________________________________________________

European Journal of Nuclear Medicine and Molecular Imaging DOI 10.1007/s00259-009-1313-8 Parathyroid scintigraphy findings in chronic kidney disease patients with recurrent hyperparathyroidism

Hindie · Zanotti-Fregonara · Just · Sarfati · Melliere · Toubert · Moretti · Jeanguillaume · Keller · Urena-Torres

Springer Customer Support 2 E-mail: [email protected] Fax: +1-703-5621873

SPi

SPi Building, Sacsac Bacong Oriental Negros 6216

Philippines

AUTHOR'S PROOF!

Metadata of the article that will be visualized in Online

1 Article Title Parathyroid scintigraphy findings in chronic kidney disease patients with recurrent hyperparathyroidism

2 Article Sub- Title 3 Article Copyright -

Year

Springer-Verlag 2009

(This will be the copyright line in the final PDF) 4 Journal Name European Journal of Nuclear Medicine and Molecular Imaging 5

Corresponding Author

Family Name Hindié

6 Particle

7 Given Name Elif

8 Suffix

9 Organization Hôpital Saint-Louis

10 Division Service de Médecine Nucléaire

11 Address 1, av Claude Vellefaux, Paris 75475 Paris cedex-10, France

12 Organization Université Paris 7

13 Division Nuclear Medicine, Hôpital Saint Louis

14 Address Paris , France

15 e-mail [email protected]

16

Corresponding Author

Family Name Ureña-Torres

17 Particle

18 Given Name Pablo

19 Suffix

20 Organization Clinique du Landy

21 Division Service de Néphrologie-Dialyse

22 Address 23, Rue Landy, Saint Ouen 93400, France

23 e-mail [email protected]

24

Author

Family Name Zanotti-Fregonara

25 Particle

26 Given Name Paolo

27 Suffix

28 Organization Université Paris VI

29 Division Nuclear Medicine, Hôpital Saint Antoine

30 Address Paris , France

31 e-mail

32 Family Name Just

33 Particle

Page 1 of 3 Testsite

11/8/2009

file://C:\Programs\ITSPMetadata2PDF\temp\HEJ91313.htm

AUTHOR'S PROOF!

34

Author

Given Name Pierre-Alexandre

35 Suffix

36 Organization Université Paris 7

37 Division Nuclear Medicine, Hôpital Saint Louis

38 Address Paris , France

39 e-mail

40

Author

Family Name Sarfati

41 Particle

42 Given Name Emile

43 Suffix

44 Organization Université Paris 7

45 Division Endocrine Surgery, Hôpital Saint Louis

46 Address Paris , France

47 e-mail

48

Author

Family Name Mellière

49 Particle

50 Given Name Didier

51 Suffix

52 Organization Université Paris XII

53 Division Endocrine Surgery, Hôpital Henri Mondor

54 Address Créteil , France

55 e-mail

56

Author

Family Name Toubert

57 Particle

58 Given Name Marie-Elisabeth

59 Suffix

60 Organization Université Paris 7

61 Division Nuclear Medicine, Hôpital Saint Louis

62 Address Paris , France

63 e-mail

64

Author

Family Name Moretti

65 Particle

66 Given Name Jean-Luc

67 Suffix

68 Organization Université Paris 7

69 Division Nuclear Medicine, Hôpital Saint Louis

70 Address Paris , France

71 e-mail

72 Family Name Jeanguillaume

73 Particle

Page 2 of 3 Testsite

11/8/2009

file://C:\Programs\ITSPMetadata2PDF\temp\HEJ91313.htm

AUTHOR'S PROOF!

Author

74 Given Name Christian

75 Suffix

76 Organization Université d’Angers

77 Division Centre Paul Papin

78 Address Angers , France

79 e-mail

80

Author

Family Name Keller

81 Particle

82 Given Name Isabelle

83 Suffix

84 Organization Université Paris VI

85 Division Nuclear Medicine, Hôpital Saint Antoine

86 Address Paris , France

87 e-mail

88

Schedule

Received 29 August 2009

89 Revised

90 Accepted 21 October 2009

91 Abstract Background: Parathyroidectomy (PTX), either subtotal or total with forearm autografting, is a well-established treatment for refractory renal

hyperparathyroidism (RHPT). However, 20–30% of patients develop persistent or recurrent disease. Obtaining accurate localization before reoperation is difficult.

Patients and methods: The study group comprised 21 consecutive adult patients (18 undergoing haemodialysis and 3 with a renal graft) imaged using 99mTc-sestamibi/¹²³I subtraction scintigraphy. Of the 21 patients, 12 had undergone one previous PTX and the other 9 between two and four parathyroid operations. All patients had symptoms and signs of severe RHPT. The mean serum PTH level was 1,142 pg/ml. ^99mTc-Sestamibi and ¹²³I images were recorded simultaneously. Imaging views comprised a planar view of the neck and mediastinum, followed by a magnified pinhole view over the thyroid bed area. If parathyroid ectopy was detected, SPECT or SPECT-CT was performed.

The forearm was imaged in case of autograft.

Results: Parathyroid scintigraphy was negative in one patient and positive in the other 20 (sensitivity 95.2%). One patient had uptake corresponding to two unresected parathyroid glands. Recurrence at the site of the partially resected gland or autograft was seen in 11 patients. However, six of them had a second 99mTc-sestamibi focus corresponding to a supernumerary parathyroid gland.

Seven other patients had a supernumerary parathyroid gland as the sole cause of relapse. Three of the supernumerary glands showed major ectopy

(intrathyroidal, low mediastinal, undescended within the vagus nerve). One patient had parathyromatosis with multiple parathyroid nodules scattered over the left side of the neck. Reoperation was possible in 13 patients, with no false- positive findings.

Conclusion: Many patients referred with the hypothesis of hyperplasia of a subtotally resected parathyroid gland or autograft were found to harbour a supernumerary parathyroid gland missed at the initial surgery.

92 Keywords separated by ' - '

Secondary hyperparathyroidism - CKD - Sestamibi scanning - Subtraction scanning - Surgery - Parathyroidectomy - Recurrent - Reoperative surgery 93 Foot note

information

Page 3 of 3 Testsite

11/8/2009

file://C:\Programs\ITSPMetadata2PDF\temp\HEJ91313.htm

AUTHOR'S PROOF!

UNCORRECTED

PROOF

1

23 ORIGINAL ARTICLE

4

Parathyroid scintigraphy findings in chronic kidney disease

5

patients with recurrent hyperparathyroidism

6 Elif Hindié&Paolo Zanotti-Fregonara&

7 Pierre-Alexandre Just_&Emile Sarfati_&Didier Mellière_&

8 Marie-Elisabeth Toubert_&Jean-Luc Moretti_&

9 Christian Jeanguillaume&Isabelle Keller&

10 Pablo Ureña-Torres

11 Received: 29 August 2009 / Accepted: 21 October 2009 12 #Springer-Verlag 2009

13

14 Abstract

15 Background Parathyroidectomy (PTX), either subtotal or 16 total with forearm autografting, is a well-established treatment 17 for refractory renal hyperparathyroidism (RHPT). However, 18 20–30% of patients develop persistent or recurrent disease.

19 Obtaining accurate localization before reoperation is difficult.

20 Patients and methods The study group comprised 21 21 consecutive adult patients (18 undergoing haemodialysis and

22 3 with a renal graft) imaged using ^99mTc-sestamibi/¹²³I

23 subtraction scintigraphy. Of the 21 patients, 12 had under-

24 gone one previous PTX and the other 9 between two and

25 four parathyroid operations. All patients had symptoms and

26 signs of severe RHPT. The mean serum PTH level was

27 1,142 pg/ml.^99mTc-Sestamibi and¹²³I images were recorded

28 simultaneously. Imaging views comprised a planar view of

29 the neck and mediastinum, followed by a magnified pinhole

30 view over the thyroid bed area. If parathyroid ectopy was

31 detected, SPECT or SPECT-CT was performed. The forearm

32 was imaged in case of autograft.

33 Results Parathyroid scintigraphy was negative in one

34 patient and positive in the other 20 (sensitivity 95.2%).

35 One patient had uptake corresponding to two unresected

36 parathyroid glands. Recurrence at the site of the partially

37 resected gland or autograft was seen in 11 patients.

38 However, six of them had a second ^99mTc-sestamibi focus

39 corresponding to a supernumerary parathyroid gland. Seven

40 other patients had a supernumerary parathyroid gland as the

41 sole cause of relapse. Three of the supernumerary glands

42 showed major ectopy (intrathyroidal, low mediastinal,

43 undescended within the vagus nerve). One patient had

44 parathyromatosis with multiple parathyroid nodules scat-

45 tered over the left side of the neck. Reoperation was

46 possible in 13 patients, with no false-positive findings.

47 Conclusion Many patients referred with the hypothesis of

48 hyperplasia of a subtotally resected parathyroid gland or

49 autograft were found to harbour a supernumerary parathy-

50 roid gland missed at the initial surgery.

51 Keywords Secondary hyperparathyroidism . CKD .

52 Sestamibi scanning . Subtraction scanning . Surgery .

53 Parathyroidectomy . Recurrent . Reoperative surgery

E. Hindié

:

:

:

Université Paris 7, Paris, France P. Zanotti-Fregonara

:

Nuclear Medicine, Hôpital Saint Antoine, Université Paris VI, Paris, France E. Sarfati

Endocrine Surgery, Hôpital Saint Louis, Université Paris 7, Paris, France D. Mellière

Endocrine Surgery, Hôpital Henri Mondor, Université Paris XII, Créteil, France C. Jeanguillaume

Centre Paul Papin, Université d’Angers, Angers, France P. Ureña-Torres (*)

Service de Néphrologie-Dialyse, Clinique du Landy, 23, Rue Landy, 93400 Saint Ouen, France e-mail: [email protected] E. Hindié (*)

Service de Médecine Nucléaire,

Hôpital Saint-Louis, 1, av Claude Vellefaux, 75475 Paris cedex-10 Paris, France e-mail: [email protected] Eur J Nucl Med Mol Imaging DOI 10.1007/s00259-009-1313-8

JrnlID 259_ArtID 1313_Proof# 1 - 06/11/2009

AUTHOR'S PROOF!

UNCORRECTED

PROOF

54 Introduction

55 Renal hyperparathyroidism (RHPT) is a frequent and 56 serious complication of long-term dialysis. RHPT starts 57 with a decline in kidney function and is driven by impaired 58 excretion of phosphorus, calcitriol deficiency and hypo- 59 calcaemia [1]. Increased FGF-23, a phosphaturic hormone 60 secreted by osteoblasts, further reduces calcitriol production 61 [2, 3]. Diffuse hyperplasia of parathyroid glands initially 62 develops. However, with continuing stimulation, there is a 63 progressive shift to nodular autonomous parathyroid 64 growth, with reduced expression of vitamin D and 65 calcium-sensing receptors [4–6]. In advanced disease, the 66 biochemical profile can shift to hypercalcaemia.

67 There is a strong association between elevated levels of 68 serum phosphorus, calcium, calcium×phosphorus product, 69 parathyroid hormone (PTH) and mortality in dialysis patients 70 [7–10]. Disturbances of bone and mineral metabolism lead to 71 cardiovascular calcifications [11–13]. Other complications of 72 severe RHPT include skeletal fractures, soft-tissue calcifica- 73 tions, and calciphylaxis (cutaneous necrosis induced by 74 calcific uraemic arteriolopathy) [14–17]. The arsenal of 75 medical treatment of RHPT includes diet, calcium- 76 containing and calcium-free intestinal phosphate binders, 77 native and active vitamin D analogues and calcimimetics 78 [18–21]. However, maintaining mineral parameters and PTH 79 levels within the targets recommended by the National 80 Kidney Foundation/Kidney-Dialysis Outcomes Quality 81 Initiative (NKF/K-DOQI) in dialysis patients (phosphorus 82 3.5–5.5 mg/dl, calcium 8.4–9.5 mg/dl, calcium×phosphorus 83 product <55 mg²/dl², PTH 150–300 pg/ml) [22] remains a 84 difficult task [23].

85 Parathyroidectomy (PTX) is an important tool for 86 treating refractory disease [6, 17, 18, 20, 24]. Standard 87 surgical procedures for RHPT include subtotal PTX 88 (SubPTX) leaving a small remnant of one gland, and total 89 PTX with autografting (TPTX+AG) of small parathyroid 90 fragments, usually to the forearm. No definite data show 91 superiority of either technique over the other [25, 26].

92 Successful PTX, with a fall in serum PTH levels and a shift 93 of calcium and phosphorus to bone, is associated with rapid 94 improvement in the disabling symptoms of RHPT such as 95 bone and joint pain, itching, muscle weakness, psycholog- 96 ical irritability and easy fatigue [27,28], an increase in bone 97 mineral density and a reduction in bone fracture risk [29], a 98 decrease or stabilization of cardiovascular calcification [30]

99 and improvement in other cardiovascular risk factors such 100 as high blood pressure, anaemia and dyslipidaemia [20,31].

101 There is a decrease in serum FGF-23 levels, an increase in 102 which has recently been associated with high mortality risk 103 [32,33]. PTX can lead to a dramatic improvement in some 104 severe conditions such as dilated cardiomyopathy [34] and 105 calciphylaxis [15, 28]. Finally, observational studies sug-

106 gest that PTX is associated with a reduction in long-term

107 mortality risk [24,35–37]. Considering patients with renal

108 graft, PTX might be required if hyperparathyroidism

109 persists for more than 6 months after transplantation [38].

110 Unfortunately, a substantial proportion of patients who

111 have received surgery develop persistent or recurrent

112 disease and reoperation may become necessary [39, 40].

113 Rothmund and Wagner defined “persistent disease”as the

114 reappearance of typical symptoms, laboratory abnormalities

115 and radiological signs within 6 months of PTX, and

“recurrence”as a later occurrence of surgical failure [39]. 116 117 Persistent disease usually results from an ectopic or supernu-

118 merary parathyroid gland. Recurrent disease is mostly

119 attributed to overgrowth of a remnant or a parathyroid

120 autograft. However, it can also be due to a supernumerary

121 gland or to the growth of parathyroid tissue inadvertently spilled in the operative field (“parathyromatosis”). The 122

123 multiplicity of potential causes leads to considerable difficulty

124 in diagnosing the site of recurrence [39–41]. Kessler et al.

125 reported on 12 patients requiring repeat surgery after

126 SubPTX; 19 reoperations on the neck and/or mediastinum

127 were performed, with cure in nine patients, persistent RHPT

128 in two, and one surgical death [40].

129 The rates of persistent and recurrent disease in recent

130 series remain high [42–45]. Reoperation is a difficult

131 decision and is often delayed. Therefore, the reoperation

132 rate is a serious underestimate of the real rate of persistent/

133 recurrent disease [16, 25]. For example, in the series of

134 Jofré et al., 33 out of 148 patients (22.3%) developed

135 persistent or recurrent RHPT. However, only 17 (11.5% of

136 the total series) received reoperative surgery [16]. Because

137 of extensive scarring of the neck tissue planes, complication

138 rates of reoperative surgery for RHPT are markedly greater

139 than those of initial surgery [41]. Seehofer et al. reported

140 seven surgical complications in 21 patients who received

141 reoperative surgery [41]. Reoperation also entails high

142 failure rates [44, 46]. In a series reported by Dotzenrath

143 et al. [44], the first reoperation failed to cure the disease in

144 7 of 26 patients (27%).

Obtaining an accurate localization is the Achilles’ heel 145 146 of reoperative surgery. The aim of this study was to

147 examine the yield of parathyroid scintigraphy in patients

148 with persistent/recurrent RHPT.

149 Patients and methods

150 From January 1995 to June 2008, 21 consecutive patients

151 with severe RHPT, in whom reoperation seemed necessary,

152 were referred to us for parathyroid scanning from various

153 dialysis or renal transplantation units in the Paris area.

154 Parathyroid subtraction scintigraphy was performed, based

155 on simultaneous recording of^99mTc-sestamibi and¹²³I [47].

Eur J Nucl Med Mol Imaging JrnlID 259_ArtID 1313_Proof# 1 - 06/11/2009

AUTHOR'S PROOF!

UNCORRECTED

PROOF

156 There were 15 women and 6 men. The mean age of the 157 patients at the time of imaging was 54.4 years (range 30–

158 77 years). Of the 21 patients, 18 were undergoing dialysis (16 159 on haemodialysis, 2 on peritoneal dialysis) with an average of 160 12 years duration (range 3–24 years), and the other 3 had a 161 renal graft. Concerning their history of parathyroid surgery, 12 162 patients had received one operation, 6 had received two 163 operations, 1 had received three operations, and 2 had each 164 received four operations (Table1).

165 All patients had symptoms and signs of severe RHPT.

166 The most frequent symptoms were bone or joint pain and 167 pruritus. All patients had radiological signs of renal 168 osteodystrophy and some had a history of multiple bone 169 fractures. Six patients were reported to have vascular and 170 valvular calcifications or coronary artery disease. One 171 patient had a history of myocardial infarction and another 172 of cerebral stroke. Hyperparathyroidism was deemed 173 refractory to medical treatment. The mean plasma PTH at 174 the time of referral was 1,142 pg/ml (range: 127 to 175 1,980 pg/ml). Two patients with renal grafts had only a 176 moderate increase in PTH levels but associated with 177 hypercalcaemia.

178 Among the 12 patients with only one previous operation, 179 the initial operation consisted of SubPTX in 11 and 180 resection of only two parathyroid glands in 1 (the other 181 two glands were considered of normal size on surgical 182 inspection). The time elapsed since the first operation in 183 these 12 patients was 6.7 years (range 0.5–15 years). The 184 pattern of evolution of serum PTH levels following this 185 initial operation was often not available. Early failure in 186 two patients (≤1 year) was suggestive of persistent disease.

187 Among the nine patients with more than one previous 188 operation, the initial operation consisted of SubPTX in 189 three, TPTX+AG in four, and resection of three glands (one 190 not found) in two. The details of subsequent operations are 191 given in Table1. The time elapsed since the first operation 192 was 10.3 years (2–16 years), and from the most recent 193 operation 2.8 years (0.5–8 years). Early failure in four 194 patients after the most recent operation was suggestive of 195 persistent disease.

196 Parathyroid^99mTc-sestamibi/¹²³I scanning was performed 197 as follows. ¹²³I (12 MBq) was administered first. ^99mTc- 198 Sestamibi (600 MBq) was injected 2 h later. The¹²³I image 199 and the^99mTc-sestamibi image were recorded simultaneously 200 [47] using dual-isotope acquisition in two separate narrow- 201 energy windows without overlap (140 keV ±7% for^99mTc- 202 sestamibi, and 159 keV with a 4% lower limit and a 10%

203 upper limit for ¹²³I). An anterior view of the neck and 204 mediastinum (from the angle of the mandible to the heart) 205 was started 5 min after^99mTc-sestamibi injection and lasted 206 5 min. A magnified view of the thyroid region was then 207 obtained for 10 min using a pinhole collimator. The ¹²³I 208 image was subtracted from the^99mTc-sestamibi image. The

209 degree of subtraction was chosen interactively, with a real-

210 time display. In the event of ectopic uptake,^99mTc-sestamibi

211 SPECT or SPECT/CT (after 2006) was performed. ^99mTc-

212 Sestamibi planar image over the forearm was obtained in

213 case of parathyroid autograft.

214 Sensitivity was expressed per patient in the whole series

215 and per lesion in patients who could receive surgery. The

216 positive predictive value was determined in patients who

217 received surgery.

218 Results

219 Scintigraphy was negative in one patient and positive in 20

220 patients (sensitivity per patient 95.2%).

221 Findings in the 12 patients with only one prior operation

222 Two patients with early failure (≤1 year after surgery)

223 showed a focus of^99mTc-sestamibi uptake corresponding to

224 a supernumerary gland. This fifth parathyroid gland was

225 located lateral to the left thyroid lobe in patient 1 and in the

226 upper anterior mediastinum in patient 2 (Table1). Patient 3,

227 who relapsed after removal of the only two enlarged

228 parathyroid glands, showed two foci corresponding to the

229 remaining parathyroid glands. The other nine patients had

230 received SubPTX three or more years before and were

231 referred with suspicion of remnant hyperplasia. All nine

232 showed a sestamibi-avid focus at the site of the remnant. In

233 four of them, however, a second unexpected focus was seen

234 corresponding to a supernumerary fifth parathyroid gland.

235 The supernumerary gland was located in the upper anterior

236 mediastinum in one patient and in the neck in the other

237 three patients (Table1, Fig.1).

238 Findings in the nine patients with more than one prior

239 parathyroid operation

240 One patient had negative ^99mTc-sestamibi/¹²³I subtraction

241 imaging. ^99mTc-sestamibi SPECT/CT was also performed

242 and was negative, as was US and MRI. In seven patients,

243 scintigraphy showed a supernumerary gland (Table 1).

244 Based on findings from previous operations, the supernu-

245 merary parathyroid gland was a fifth gland in four patients

246 and a sixth parathyroid gland in three patients. The culprit

247 gland was seen in the neck in five patients, being ectopic in

248 three of them (embedded in the thyroid parenchyma in

249 patient 14, Fig.2; lateral to the left thyroid lobe in patient

250 16, Fig.3; located very high in the right neck“undescended

251 gland” in patient 21, Fig. 7). In two patients, the

252 supernumerary parathyroid gland was located in the

253 mediastinum. Both patients had a second site of failure

254 (graft hyperplasia in patient 17, Fig.4; remnant hyperplasia Eur J Nucl Med Mol Imaging

JrnlID 259_ArtID 1313_Proof# 1 - 06/11/2009

AUTHOR'S PROOF!

UNCORRECTED

PROOF

t1.1Table1Patientsdata t1.2PatientSexAge (years)Dialysis/transplantPreviousparathyroidoperationsPTH (pg/ml)No.of fociLocationof sestamibifociSurgeryPTHat 6months (pg/ml)

Follow-up (months)Patientstatusat lastfollow-up t1.3NumberType t1.41M41Peritoneal dialysis (1year)

1SubPTX(upperrightremnant) 6monthspreviously1,0001Lowerleftfocus (fifthgland)cFifthparathyroid behindleft carotid(4,064mg)

14630Jointpainand softtissue calcifications resolved t1.52F73Haemodialysis (16years)1SubPTX(lowerleftremnant) 1yearpreviously1,9801Upperanterior mediastinum (fifthgland)

No–60Deceased t1.63M59Haemodialysis (6years)1Resectionoftwoglands (lowerright,lowerleft) 2yearspreviously

1,5922Upperrightfocus, upperleftfocusUpperrightgland (1,845mg).Upper leftgland(380mg) 1020Pruritusresolved. Received renalgraft t1.74F30Haemodialysis (3years)1SubPTX(upperrightremnant) 3yearspreviously9591Upperrightfocus (remnanthyperplasia)Resectionofupper rightremnant (468mg)

7024Bonedisease improved t1.85F77Haemodialysis (16years)1SubPTX(lowerrightremnant) 15yearspreviously7561Lowerright(remnant hyperplasia)Resectionofremnant (750mg)<464Deceased t1.96M37Renalgraft (1year)1SubPTX(lowerleftremnant) 5yearspreviously1,482a1Lowerleft(remnant hyperplasia)No–18Deceased t1.107F39Renalgraft (11years)1SubPTX(lowerrightremnant) 10yearspreviously285a1Lowerright(remnant hyperplasia)Partialremnant resection(2cm)656Hypercalcaemia resolved t1.118F46Secondrenal graft(1year)1SubPTX(upperleftremnant) 11yearspreviously127a1Upperleft(remnant hyperplasia)Partialremnant resection(1cm)4013Hypercalcaemia resolved t1.129M60Haemodialysis (17years)1SubPTX(lowerrightremnant) 15yearspreviously6502Upperright (fifthgland)+ lowerright (remnant).Fig.1

No–11Deceased t1.1310F71Haemodialysis (15years)1SubPTX(upperleftremnant) 10yearspreviously1,2502Middleleft (fifthgland)+ upperleft(remnant hyperplasia) Fifthgland retrotracheal (1,598mg).Upper left(350mg) 3536Deceased t1.1411F68Haemodialysis (2years)1SubPTX(lowerleftremnant) 6yearspreviously1,1612Uppermediastinum (fifthgland)+ lowerleft(remnant)

No–44Deceased t1.1512F64Haemodialysis (3years)1SubPTX(lowerleftremnant) 2yearspreviously6892Beneaththyroid (fifthgland)+ lowerleft(remnant) No–6Deceased (hepatocarcinoma) t1.1613F42Peritoneal dialysis (7years)

2PTX(lowerleftnotfound) 3yearspreviously. Failedcervicotomy 1yearpreviously 1,0170NegativestudyNo–6Urgentrenal transplantation t1.1714M66Haemodialysis (16years)2TPTX+AG10yearspreviously. Graftresection6months previously

1,0501Middleright (fifthgland; Fig.2) Fifthglandinthe thyroidparenchyma (500mg)

428Improvementin bonepainand asthenia t1.1815F34Haemodialysis (12years)2PTX(lowerrightnotfound) 2yearspreviously.Resection oflowerright1year previously

8001Upperleftfocus (fifthgland)No–6Receivedrenal transplantation Eur J Nucl Med Mol Imaging JrnlID 259_ArtID 1313_Proof# 1 - 06/11/2009

AUTHOR'S PROOF!

UNCORRECTED

PROOF

255 in patient 18, Fig. 5). Finally, one patient (patient 19)

256 showed multiple ^99mTc-sestamibi uptake foci disseminated

257 over the left neck, suggestive of parathyromatosis (Fig.6).

258 Unguided ultrasonography was less sensitive than

259 scintigraphy and was misleading in many patients (see

260 Figs. 1, 3 and 4). However, because ultrasonography was

261 performed by various operators, no formal comparison with

262 scintigraphy was made.

263 Follow-up data

264 Scintigraphy findings, surgical findings and follow-up data

265 are summarized in Table 1.

266 Two patients (including the one with negative imaging)

267 received urgent renal transplantation. Parathyroid surgery

268 was not performed in two other patients because of

269 intercurrent disease (hepatocarcinoma and uterine cancer,

270 respectively). Four patients refused parathyroid surgery;

271 their average survival was 33.2 months. In the patient with

272 parathyromatosis and three prior neck operations, the

273 surgeon was faced with severe tissue adherences and

274 fibrosis and was able to remove only a few of the

275 parathyroid implants. PTH levels at 6 months were as high

276 as before surgery. The patient died at 25 months.

t1.1916F62Haemodialysis (17years)2TPTX+AG14yearspreviously. Failednecksurgery 4yearspreviously 1,8601Upperleftfocus inlateralneck (fifthgland;Fig.3) Fifthgland(2cm) closetocarotid artery 30024Marked improvement inbonepain t1.2017F59Haemodialysis (18years)2TPTX+AG16yearspreviously. Graftresection4years previously

1,7002Uppermediastinum (fifthgland)+ graft(Fig.4) No–33Uterinecancer t1.2118F32Haemodialysis (12years)2SubPTX(lowerleft)9years previously.Resectionfifth gland8yearspreviously

1,4442Lowmediastinum (sixthgland)+ lowerleftremnant (Fig.5) Necksurgeryonly. Lowerleftremnant (1cm)

80096Receivedrenal transplantation t1.2219M73Haemodialysis (24years)3(SeeFig.6legend)1,300≥10Multiplefoci suggestiveof parathyromatosis (Fig.6)

Onlysomeparathyroid nodulesresected1,30325Deceased t1.2320F57Haemodialysis (10years)4Surgery7,5,4and 1yearpreviouslyb1,6871Lowerleftneck focus(sixthgland)Resectionofasixth parathyroid(1.8cm)24857Received renalgraft t1.2421F53Haemodialysis (20years)4(SeeFig.7legend)1,2001Atrightmandibular angle(sixth parathyroid;Fig.7) Undescendedsixth parathyroid (1,020mg)

13296Goodoverall condition a Hyperparathyroidismpersistingafterrenaltransplantationwithhypercalcaemia.Allotherpatientswereundergoingdialysis b SubPTXwithupperleftremnant7yearspreviously;remnantablationandforearmgraft5yearspreviously;resectionofanintrathymicfifthparathyroidgland4yearspreviously;unsuccessful neckexploration1yearpreviously. c Theparathyroidforearmautograftwasbarelyvisible

Fig. 1 A 60-year-old man on haemodialysis (patient 9). He had undergone SubPTX 15 years previously, keeping a remnant of the lower right parathyroid gland. He had recurrent RHPT with cardiovascular calcifications, stenosis of the aortic valve, and osteodystrophy. Ultraso- nography confirmed remnant hypertrophy. ^99mTc-sestamibi and ¹²³I images were recorded simultaneously (pinhole collimator over the neck). Subtraction images show two clear-cut sestamibi foci, one corresponding to the lower right parathyroid remnant, and the other to a fifth right parathyroid gland (dominant focus). The patient refused surgery; he died 11 months later with myocardial infarction

Eur J Nucl Med Mol Imaging

JrnlID 259_ArtID 1313_Proof# 1 - 06/11/2009

AUTHOR'S PROOF!

UNCORRECTED

PROOF

277 In the other 12 patients, 14 parathyroid glands were 278 resected (Table1). The positive predictive value of^99mTc- 279 sestamibi/¹²³I imaging was 100%. Also, no additional 280 parathyroid lesions were discovered at surgery. In patient

281 18, the surgeon resected only the neck focus which led to a

282 partial decrease in PTH level (PTH at 6 months 800 pg/ml

283 vs. 1,444 pg/ml preoperatively). She later received renal

284 transplantation. PTH levels at 6 months in the other patients

285 ranged from <4 to 300 pg/ml (mean 96 pg/ml). Although PTH levels did not exceed the target (150–300 pg/ml) 286

287 recommended by the National Kidney Foundation in dialysis

288 patients [22], they were very low in three patients (patients

289 3, 5 and 14) consistent with postsurgical hypoparathyroid-

290 ism. Surgery was followed by an improvement in clinical

291 condition in all patients. Patient 5 died 64 months after

292 surgery, at the age of 82 years. Patient 10 died 36 months

293 after surgery, at the age of 74 years. The other ten patients

294 were alive at the time of this report with follow-up periods

295 ranging from 6 to 96 months (mean 41.2 months).

296 Discussion

297 PTX is a well-established treatment for RHPT when medical

298 therapy has failed. Successful surgery results in a rapid decline

299 in PTH, improvement in mineral parameters (phosphorus and

300 calcium), relief of symptoms, and a beneficial impact on

301 clinical hard outcomes such as bone health, cardiovascular

302 morbidity, and mortality [15–17, 24, 27, 29, 34–37].

Fig. 4 A 59-year-old woman on haemodialysis (patient 17). She had undergone two prior parathyroid operations, consisting of TPTX+AG 16 years previously and ablation of the autograft 4 years previously.

PTH at referral was 1,700 pg/ml. Ultrasonography showed hyperplasia at the graft site. ^99mTc-sestamibi/¹²³I subtraction scanning shows an intense focus in the upper mediastinum corresponding to an intra- thymic fifth parathyroid gland. A second focus is seen over the left forearm (lower right image), suggesting that previous graft resection was incomplete. Uterine cancer was later diagnosed and parathyroid surgery was postponed

Fig. 3 A 62-year-old woman on haemodialysis (patient 16). TPTX+AG was performed 14 years previously. A second operation on the neck 10 years later for recurrent disease was negative. She had severe RHPT (PTH 1,860 pg/ml) resistant to large doses of calcimimetics. Ultrasonog- raphy was negative. ^99mTc-sestamibi/¹²³I subtraction scanning shows uptake lateral to the upper left thyroid pole, suggesting an ectopic parathyroid gland. Uptake of the forearm parathyroid autograft was moderate. Guided ultrasonography showed a small mass between the carotid artery and jugular vein. This fifth parathyroid gland was resected.

The PTH level at 6 months was 300 pg/ml reflecting the presence of the functioning forearm autograft. Bone pain markedly improved

Fig. 2 A 66-year-old man on haemodialysis (patient 14). Initial parathyroid surgery consisted of TPTX+AG 10 years previously. The forearm autograft was resected 6 months before the present study following a positive Casanova test [72]. He had persistent hyperpara- thyroidism (PTH 1,050 pg/ml). ^99mTc-sestamibi/¹²³I subtraction scanning shows an intense focus at the mid-level of the right thyroid lobe. This lesion was confirmed by ultrasonography. At surgery, a fifth parathyroid gland, located beneath the thyroid capsule, was resected.

Bone pain and asthenia markedly improved

Eur J Nucl Med Mol Imaging JrnlID 259_ArtID 1313_Proof# 1 - 06/11/2009

AUTHOR'S PROOF!

UNCORRECTED

PROOF

303 Unfortunately, even when operated on by experienced 304 surgeons, 20–30% of patients develop persistent or recurrent 305 disease [16, 25]. These patients are often in poor clinical 306 condition and repeat surgery should not be postponed.

307 However, reoperation for RHPT is difficult, has a lower 308 success rate than initial operation [44, 46] and a higher 309 complication rate [40, 41]. About half of the patients with 310 severe recurrence do not receive repeat surgery [16,25].

311 The importance of accurate localization before reopera- 312 tion needs to be emphasized. Many patients in the present 313 series who were referred with the hypothesis of remnant 314 hyperplasia (or graft hyperplasia) were found to have a 315 parathyroid gland at an unsuspected location. Moreover, six 316 patients (29%) had mixed failure, with two clear-cut^99mTc- 317 sestamibi uptake foci, one corresponding to hyperplasia of 318 known residual tissue and the other to a missed supernu- 319 merary parathyroid gland. The finding of fewer than four 320 glands at initial operation is a classic cause of persistent 321 disease [48]. However, identification of four glands by no 322 means guarantees adequate resection [45, 49]. Akerström 323 et al. established the frequency of supernumerary parathy- 324 roid glands at 13% (64/503 random autopsy cases) and

325 described their wide anatomical distribution [50]. In the

326 present study, ^99mTc-sestamibi uptake corresponding to a

327 supernumerary parathyroid gland was seen in 13 out of 21

328 patients who received repeat surgery. Based on data from

329 prior surgery, the culprit gland was a fifth parathyroid gland

330 in ten patients and a sixth gland in three patients.

331 Many teams now use extensive initial surgery in an

332 attempt to remove hidden supernumerary parathyroid

333 glands, including: the excision of fat surrounding the

334 parathyroid glands, routine removal of all thymic tissue

335 accessible from the neck incision, and opening of the

336 carotid sheaths bilaterally. Despite these manoeuvres, the

337 percentage of patients with missed supernumerary glands

338 remains substantial [16,51]. Matsuoka et al. reported on 44

339 patients who had TPTX+AG. Three patients had serum

340 PTH levels of >60 pg/ml on day 1 after surgery consistent

341 with a missed parathyroid gland, since a parathyroid graft is

342 not immediately functional [49], and another patient had a

343 PTH level of >300 pg/ml at 6 months. This high rate of Fig. 6 A 73-year-old man on haemodialysis who had undergone three prior parathyroid operations (patient 19). The first operation, 16 years previously, consisted of SubPTX keeping a lower left parathyroid remnant. The second operation 5 years later found parathyromatosis on the left side; multiple parathyroid nodules and the left thyroid lobe were resected. The third operation, 3 years before the present study, consisted of resection of pretracheal parathyroid nodules, thyroid isthmectomy and partial right thyroid lobectomy. The patient had hyperparathyroidism resistant to medical therapy including calcimi- metics (PTH 1,300 pg/ml) with osteodystrophy and bone fractures. On

99mTc-sestamibi/¹²³I scanning there is a small right thyroid remnant that has disappeared in the subtraction image (arrow). There are multiple foci of sestamibi uptake disseminated over the left side of the neck consistent with parathyromatosis. Only some of the parathyroid nodules could be resected. PTH levels at 6 months were as high as preoperatively. The patient died 25 months later (cardiac arrest) Fig. 5 A 32-year-old woman on haemodialysis with recurrent RHPT

(PTH 1,444 pg/ml) after two previous parathyroid operations (patient 18). SubPTX with lower left remnant had been performed nine years previously. A fifth left cervical parathyroid gland was resected in a second operation 1 year later. ^99mTc-sestamibi/¹²³I subtraction scanning shows two intense foci, one in the neck corresponding to remnant hyperplasia and the other in the lower anterior mediastinum seen on planar and SPECT images. The neck focus was confirmed by ultrasonography. A CT scan of the thorax confirmed the presence of a 2.5 cm lesion anterior to the arch of the aorta. Cervical surgery only was performed, and a 1-cm parathyroid mass corresponding to remnant hyperplasia was resected. The PTH level decreased to 800 pg/ml. The patient received renal transplantation. The mediastinal parathyroid gland was not resected

Eur J Nucl Med Mol Imaging

JrnlID 259_ArtID 1313_Proof# 1 - 06/11/2009

AUTHOR'S PROOF!

UNCORRECTED

PROOF

344 short-term surgical failure (4/44, 9%), even by a most 345 experienced team, indicates that adjunct to surgery might be 346 required [51].

347 Based on the examination of routinely resected thymic 348 tissue, some authors have reported very high percentages of 349 supernumerary parathyroid glands in patients with renal 350 failure [52, 53]. However, in the great majority of cases, 351 intrathymic parathyroid tissue was microscopic. In our 352 opinion, mixing up microscopic parathyroid tissue islets in 353 the thymus with macroscopic supernumerary glands is not 354 desirable as it blurs the real significance of supernumerary 355 glands and the potential role of imaging before initial 356 surgery. In the study by Pattou et al. [52], the average 357 weight of ten supernumerary glands identified at repeat 358 surgery was 3,126 mg (450–6,635 mg), suggesting that all 359 these supernumerary glands were potentially detectable by 360 imaging before first surgery, while the significance of the 361 resected microscopic islets of intrathymic parathyroid tissue 362 in terms of recurrence is less clear.

363 Major ectopy of a parathyroid adenoma (low mediasti- 364 nal, intrathyroidal, inside the carotid sheath, undescended 365 gland) is encountered in about 1–2% of patients with 366 primary hyperparathyroidism [54, 55]. This incidence is 367 expected to be higher in patients with RHPT taking into 368 consideration that any of the four or additional parathyroid 369 glands can be ectopic. Many of the parathyroid lesions in 370 the present reoperation series showed ectopy. Patient 14 had 371 a fifth gland partially embedded in the thyroid parenchyma 372 (Fig. 2). Intrathyroidal glands that are fully included in 373 the thyroid parenchyma are rare [54]. In two other patients, 374 the ectopic gland was located lateral to the upper pole of the 375 thyroid lobe near the carotid sheath (Fig. 3). This is a 376 frequent ectopy. Undescended parathyroid glands are rare 377 and usually located within the carotid sheath, high in the 378 neck, at or superior to the carotid bifurcation [56]. In a 379 Japanese series of RHPT, an undescended gland was 380 identified in about 1% of patients (13/1,666), either during 381 initial surgery after routine opening of the carotid sheaths or 382 during repeat surgery [57]. The undescended parathyroid 383 gland in patient 21 was clearly visible on ^99mTc-sestamibi 384 images (Fig. 7). It was confirmed by angiography and 385 approached selectively. After dissection of the carotid 386 sheath, the undescended parathyroid gland was found to 387 be embedded in the vagus nerve, quite a rare occurrence 388 [58]. According to Pawlik et al., intravagal parathyroid 389 glands are distinct from“classical”undescended glands in 390 being always supernumerary“accessory”glands [58]. It is 391 particularly interesting that in our patient 21 a “fifth”

392 parathyroid gland had been removed from the left vagus 393 nerve 1 year before the removal of the present sixth gland 394 from the right side. To our knowledge, this is the first report 395 of symmetrical supernumerary intravagal parathyroid 396 glands. Undescended glands might be difficult to detect

397 on ^99mTc-sestamibi scans due to the proximity of the

398 submandibular salivary glands. The two“intravagal”glands

399 described by Pawlik et al. were missed by sestamibi [58].

400 Again, in the three patients with undescended glands

401 reported by Axelrod et al. [59], the culprit gland was

402 identified by selective venous sampling assessing PTH

403 concentration, and only retrospectively was sestamibi found

404 positive upon reviewing the scans. Lateral views or

405 SPECT/CT should be helpful in resolving suspicious cases.

406 Low mediastinal parathyroid glands in primary hyper-

407 parathyroidism are now often removed by the thoraco-

408 scopic route to avoid median sternotomy or thoracotomy

409 [60]. Experience with thoracoscopy is more limited in

410 RHPT; the risk of parathyroid fracture and spilling should

411 be considered.

412 Rupture of the parathyroid capsule in the operative field

413 may result in regrowth of multiple nodules of adenomatous

414 parathyroid tissue “parathyromatosis”[39–41,44, 61,62].

415 It is more frequent in case of SubPTX [62]. Parathyroma-

416 tosis lesions are densely adhesive to surrounding tissues Fig. 7 This 53-year-old woman on haemodialysis had undergone four prior parathyroid operations (patient 21). The first, 16 years previous- ly, consisted of TPTX+AG. The second neck operation 10 years later failed. In the third operation the forearm parathyroid graft was ablated.

In the fourth operation, 3 years before the present examination, an undescended fifth parathyroid gland on the left side was ablated.

However, the PTH level increased again (1,200 pg/ml). ^99mTc- sestamibi/¹²³I subtraction scanning shows an ectopic sestamibi focus located high in the right side of the neck (arrow). The lateral view (lower right image) shows that this ectopic parathyroid gland (arrow) lies posterior to the right submandibular salivary gland. Angiography confirmed the presence of a nodule behind the jugular and carotid vessels. At surgery, a 1,020-mg parathyroid tumour (sixth gland) was resected. This parathyroid gland was located within the carotid sheath, embedded in the vagus nerve. The PTH level 1 year after surgery was 132 pg/ml

Eur J Nucl Med Mol Imaging JrnlID 259_ArtID 1313_Proof# 1 - 06/11/2009

AUTHOR'S PROOF!

UNCORRECTED

PROOF

417 and scars, making re-excision difficult [61,62]. Prognosis 418 is poor [61,62]. In the patients reported by Stehman-Breen 419 et al., clinical manifestations were dramatic and the disease 420 progressed relentlessly despite numerous operations [61]. In 421 patient 19 with parathyromatosis, subtraction scintigraphy 422 showed multiple^99mTc-sestamibi uptake foci scattered over 423 the left neck (Fig. 6). The surgeon could not perform a 424 complete resection; the patient died at 25 months. It might 425 be difficult to image parathyromatosis lesions. In the series 426 of Matsuoka et al., parathyromatosis was detected preoper- 427 atively in 40% of patients [62].

428 Remnant or parathyroid graft recurrence is frequent in 429 countries with low rates of kidney transplantation, such as 430 Japan, reaching 30% at 7 years [17]. In the present series, 11 431 patients had recurrence at the site of known residual 432 parathyroid tissue. During imaging, however, it is important 433 to search for additional foci. Indeed, 6 of the 11 patients with 434 remnant or parathyroid graft hyperplasia harboured a second 435 cervical or mediastinal sestamibi focus corresponding to a 436 missed supernumerary gland (Figs.1,4 and5).

437 Reducing the need for reoperation should be a major 438 objective given the high surgical morbidity [40, 41, 44].

439 Richards et al. performed a meta-analysis of 53 publications 440 and 501 reoperations for RHPT [26]. The reoperations 441 showed that the initial cervical operation had been 442 inadequate in 42% of patients who had undergone SubPTX 443 and in 34% of patients who had undergone TPTX+AG. The 444 authors concluded that many failures occur because of 445 limitations in preoperative localization and inadequate 446 exploration [26]. We and others have shown that parathy- 447 roid scintigraphy prior to initial surgery can provide 448 anatomical and functional information that could help 449 avoid failure [63–66]. In addition to major parathyroid 450 ectopy, minor aberrancy (retro-oesophageal, in the vicinity 451 of the carotid sheath, in the upper mediastinum), or simply 452 asymmetry in gland position, may be the reason for 453 extensive dissection during initial surgery and sometimes 454 failure [63–66]. Preoperative imaging may also reveal a 455 supernumerary gland [63]. Parathyroid glands in a single 456 patient can show differences in size, nodularity, prolifera- 457 tive activity, and autonomy [4–6, 17]. Those parathyroid 458 glands with the highest tracer uptake should probably not 459 be used as remnant tissue or for graft constitution [63,64].

460 In a study by Fuster et al. [66], in which the surgeon was 461 blinded to the preoperative sestamibi results, 9 of 48 patients 462 (19%) had surgical failure at 1 year (PTH 884±217 pg/ml).

463 The authors considered that, if the sestamibi information 464 had been taken into account, the 1-year surgical failure rate 465 would have been reduced to 2% [66]. Sestamibi imaging 466 protocols have variable sensitivities before first surgery in 467 RHPT. Torregrosa showed that, although sensitivity with the 468 single-isotope“wash-out”technique may approach 89% on a 469 patient basis, in no patient (n=27) were all four glands

470 visualized [64]. Sensitivity of dual-isotope scanning is higher

471 [63,67]. Using simultaneous acquisition of the two isotopes

472 [47], that avoids artefacts due to shift or tilt of the patient’s

473 neck during imaging, four glands or more were visualized in

474 82% of RHPT patients [63]. This technique has also proven

475 to be sensitive for detecting multiple gland disease in

476 primary hyperparathyroidism [68].

477 In the case of reoperation for RHPT, it is generally agreed

478 that ^99mTc-sestamibi provides the highest accuracy among

479 noninvasive localization techniques [41, 43, 44, 69]. Few

480 studies have shown false-positives [41]. Sestamibi imaging

481 was positive in 71% of patients (5/7) in the series of

482 Dotzenrath et al. [44], 75% (6/8) in the series of Seehofer

483 et al. [41], 85% (11/13) in the series of Chou et al. [43], and

484 93% (13/14) in the study of Neumann et al. [69]. These

485 sensitivity rates are higher than those reported before initial

486 surgery, probably because residual glands causing recurrence

487 are uniformly autonomous and highly hyperactive. The

488 sensitivity per patient in the present study was 95% (20 of

489 21 patients). In contrast with previous series, our study

490 showed that many patients (29%) harbour two sites of failure.

491 The sensitivity of sonographic exploration in repeat

492 parathyroid surgery is low, close to 50% [41], with an

493 inability to detect parathyroid glands in the mediastinum,

494 and in the retrotracheal or retro-oesophageal areas. The false-

495 positive rate is about 18% [41] with possible confusion with

496 thyroid nodules, lymphatic nodes etc. However, ultrasonog-

497 raphy can be very useful to provide anatomical confirmation

498 for a sestamibi focus in the neck (Fig. 3). It can also guide

499 fine-needle biopsy, and fluid aspiration for PTH measure-

500 ments, to confirm the parathyroid origin before reoperation.

501 On the other hand, CT or MRI can provide landmarks and

502 confirmation for mediastinal or para-/retro-oesopheageal

503 lesions. ^99mTc-sestamibi SPECT and CT can now be

504 performed on the same machine, with fusion of functional

505 and anatomical images. SPECT-CT cannot replace pinhole

506 acquisition, however, which is still necessary to obtain a

507 magnified view of the thyroid bed area, enhancing both

508 resolution and sensitivity [70].¹¹C-Methionine PET has had

509 more limited use because of its scarce availability. Hessman

510 et al. found lower sensitivity than with sestamibi [71].¹¹C-

511 Methionine might, however, give useful information in some

512 patients with negative sestamibi findings as shown by

513 Rubello et al. [72].

514 Other studies that may help identify graft recurrence are:

515 measurements of PTH gradient from the graft-bearing arm

516 versus the contralateral arm [43] and the Casanova test that

517 requires ischaemic blockade of the arm with the autograft

518 [73]. Invasive studies should be restricted to patients with

519 negative or unclear noninvasive studies. Selective venous

520 sampling seems to offer high sensitivity when performed by

521 specialized teams [41]. Finally, intraoperative PTH mea-

522 surement and the gamma probe may also be helpful, as Eur J Nucl Med Mol Imaging

JrnlID 259_ArtID 1313_Proof# 1 - 06/11/2009