Optimizing the Minimally Invasive Approach to Mediastinal Parathyroid Adenomas

Transcription

1 Optimizing the Minimally Invasive Approach to Mediastinal Parathyroid Adenomas Benjamin Wei, MD, William Inabnet, MD, James A. Lee, MD, and Joshua R. Sonett, MD Division of Cardiothoracic Surgery, Duke University Hospital, Durham, North Carolina; Division of Metabolic, Endocrine, and Minimally Invasive Surgery, Mount Sinai School of Medicine, New York, New York; and the Divisions of Endocrine/Thyroid Surgery and Thoracic Surgery, New York Presbyterian Hospital Columbia, New York, New York Background. Patients with refractory hyperparathyroidism after neck exploration may have a mediastinal parathyroid gland that has not been identified reliably with a single radiologic study. We report 17 patients who underwent minimally invasive resection for mediastinal parathyroid adenomas after confirmatory multipoint radiologic imaging. Methods. Fifteen patients underwent thoracoscopic procedures and 2 patients underwent mediastinoscopic procedures for resection of suspected mediastinal parathyroid adenoma. Preoperative localizing studies included sestamibi scan, computed tomography scan of the neck and chest, and selective venous sampling of parathyroid hormone levels. Once a mediastinal location was determined, thoracoscopic or mediastinoscopic resection was performed. Successful removal of parathyroid tissue was confirmed with a 50% or greater reduction in intraoperative parathyroid hormone levels. Results. Parathyroid adenoma was resected in 88% of patients after the operation. The cure rate was 100% in patients with two or more concordant studies locating parathyroid tissue in the mediastinum and 60% in those with one positive study. The thoracostomy tube was removed on median postoperative day 1 (range, 0 to 2 days). Median hospital stay was 3 days (range, 2 to 7 days). The most common complication was temporary hypocalcemia, which occurred in 18% of patients. Conclusions. Minimally invasive parathyroidectomy is an effective treatment of hyperparathyroidism caused by mediastinal parathyroid tissue. Targeted exploration depends on the guidance of preoperative localization studies and measurement of intraoperative parathyroid hormone levels to verify successful resection. Selective venous sampling and high-resolution computed tomography scanning can be helpful in patients with negative sestamibi scans. (Ann Thorac Surg 2011;92:1012 7) 2011 by The Society of Thoracic Surgeons Approximately 95% of patients with primary hyperparathyroidism are cured after neck exploration. In many of those in whom a cure is not achieved after parathyroidectomy, an ectopic or supernumerary parathyroid is in a different location. An ectopic mediastinal parathyroid gland may be present in as many as 25% of patients with primary hyperparathyroidism, although only about 2% of them are inaccessible through a standard cervical incision and require a thoracic approach [1, 2]. The use of video assisted thoracic surgery (VATS) for the removal of ectopic mediastinal parathyroid gland has been described in several case reports and case series [3]. We report now our experience with the minimally invasive treatment of patients with documented mediastinal parathyroid tissue. In most of these patients, prior neck exploration for primary hyperparathyroidism has failed. We performed VATS parathyroidectomy in 15 patients, and mediastinoscopic removal of parathyroid tissue in 2 patients. This is the largest series of mediastinal explorations for parathyroid adenoma to date. Patients and Methods Between July 1, 2005, and March 31, 2010, 784 patients underwent a surgical exploration for hyperparathyroidism in our department. Of these, 15 patients underwent VATS exploration, and 2 patients had mediastinoscopic exploration for parathyroid adenomas that could not be resected through a standard cervical incision. In these patients, preoperative studies included one or more of the following: (1) Tc-99m sestamibi scan, (2) neck ultrasound imaging, (3) computed tomography (CT) scan of the neck and chest, and (4) selective venous sampling of parathyroid hormone levels. Three, if not all four of these studies, were performed in most of the patients. We retrospectively reviewed the preoperative characteristics, intraoperative data, and postoperative outcomes for these patients. The Columbia University Medical Center Institutional Review Board approved the study. Diagnostic Studies For the sestamibi scan, iodine-123 was injected intravenously 24 hours before the scan, and technetium-99 was Accepted for publication April 22, Address correspondence to Dr Sonett, Division of Thoracic Surgery, NY Presbyterian Hospital Columbia, PH Rm 104, 14th Flr, 622 W 168th St, New York, NY 10032; js2106@columbia.edu. Dr Sonnett discloses that he has a financial relationship with Covidien by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg WEI ET AL 2011;92: MEDIASTINAL PARATHYROID ADENOMA RESECTION 1013 injected at the time of the scan. When the scan was taken, the iodine counts were subtracted from the technetium counts. Single photon-emission CT images of the neck and chest were obtained in certain instances for additional anatomic information. CT scan of the neck and chest with 2.5-mm-thin slices was performed with and without intravenous contrast. Sagittal and coronal images were reformatted from the initial data acquired. Selective venous sampling (SVS) was performed through access at the femoral vein. A sheath was placed in the femoral vein, a guidewire was passed into the vein, and a catheter was passed over the guidewire for sampling of venous blood. Parathyroid hormone (PTH) levels were measured at the bilateral internal jugular veins at upper, middle, and lower locations; the right brachiocephalic vein, the left brachiocephalic vein at proximal and distal locations; the superior vena cava at upper and lower locations, and the inferior vena cava. SVS was considered confirmatory for a mediastinal parathyroid gland if there was a twofold gradient at an intrathoracic location. Operative Technique Once a mediastinal location was determined from the concordant localization studies, patients were taken to the operating room for VATS or mediastinoscopic exploration. Exploration and resection with conventional mediastinoscopy was used if the parathyroid adenoma appeared to be located in the middle mediastinum on preoperative imaging, whereas VATS was used when the gland was localized in the anterior mediastinum or part of the middle mediastinum that would be difficult to access with the mediastinoscope. The decision to perform right-sided vs left-sided VATS was also determined by the location of the parathyroid adenoma on the preoperative CT scan. The patient was intubated with a double-lumen endotracheal tube, placed in a standard lateral decubitus position, and 3 or 4 thoracoscopic ports were created. If the adenoma was located in the anterior mediastinum, thymectomy and excision of mediastinal fat tissue was performed with the assistance of the Ligasure V (Valleylab, Boulder, CO) laparoscopic instrument. If the parathyroid adenoma was located in a part of the middle mediastinum not easily accessible with the mediastinoscope, we removed it in a more targeted fashion with VATS and omitted the thymectomy and resection of mediastinal fat. We resected 1 adenoma in the aortopulmonary window through the left chest and another from near the innominate artery through the right chest in this manner. The specimen was removed from the thorax inside an Endo CATCH bag (Covidien, Mansfield, MA) through the 10-mm port, and in most cases, the specimen was sent for frozen section. Intraoperative venous PTH values were obtained at baseline, when the specimen was removed (ie, T-0), and at 5, 10, and at the discretion of the surgeon, 20 minutes after removal of the specimen. After hemostasis was achieved, the lung was reexpanded and a chest tube left in place. One patient also received a pulmonary wedge resection for a 0.6-cm carcinoid tumor on the left side during the operation. Another patient had no appreciable drop in PTH levels and no parathyroid adenoma was seen on the frozen section, so a cervical incision was made and cervical thymus extracted, with successful removal of the ectopic parathyroid gland. Patients were routinely prescribed supplemental calcium postoperatively for hypocalcemia prophylaxis. We considered mediastinal exploration for primary hyperparathyroidism successful if we achieved (1) an intraoperative decrease in PTH levels of 50% or greater from the higher of the baseline and T-0 values, (2) pathologic confirmation of removal of parathyroid gland or adenoma from the mediastinum, and (3) postoperative normalization of calcium levels. Results Patient characteristics are reported in Table 1. Patients were a median age of 55 years (range, 29 to 79 years), and 14 of 17 (82%) were women. The indications for operation were primary hyperparathyroidism in all cases. Ten patients (59%) had undergone at least one prior neck exploration with parathyroidectomy, which had yielded parathyroid hyperplasia or normal parathyroid tissue. The median preoperative calcium level was 11 mg/dl (range, 9.8 to 12.8 mg/dl; normal reference range, 8.4 to 9.8 mg/dl). The median preoperative PTH level was 113 pg/ml (range, 56 to 5583 pg/dl; normal reference range, 8 to 51 pg/dl). No intraoperative complications occurred. Intraoperative blood loss was minimal. No patients were converted to thoracotomy. Excluding the patient for whom intraoperative PTH level data were unavailable, PTH levels decreased by 50% compared with baseline in 14 of 16 patients (88%). In the 2 patients whose PTH levels did not decline intraoperatively, pathology revealed no evidence of parathyroid tissue in the specimens, and unsurprisingly, hyperparathyroidism persisted in these patients after the operation. All patients were extubated in the operating room. Chest tubes in 14 of 15 (93%) patients who underwent VATS were removed on postoperative day 0 or 1, and the median hospital stay was 3 days (range, 2 to 7 days). One patient who underwent mediastinoscopic exploration went home the same day of the operation, and the other underwent simultaneous thyroidectomy for papillary thyroid cancer and was discharged on postoperative day 1. Temporary hypocalcemia developed in 3 of 17 patients (18%) after the operation, and 1 patient required a continuous intravenous infusion of calcium. This patient s first operation for hyperparathyroidism was a total parathyroidectomy with reimplantation of parathyroid tissue in the sternocleidomastoid muscle, which was subsequently removed when recurrent hyperparathyroidism developed. This patient required intubation after VATS thymectomy because glottic obstruction developed related to temporary left vocal cord palsy and edema,

3 1014 WEI ET AL Ann Thorac Surg MEDIASTINAL PARATHYROID ADENOMA RESECTION 2011;92: Table 1. Characteristics of Patients Receiving Mediastinoscopic or Videothoracoscopic Resection for Mediastinal Parathyroid Adenoma Patients, No. 17 Age (median SD, years) Gender, % Female 82 Male 18 Previous neck exploration, % Yes 59 No 41 Preoperative calcium (median SD, mg/dl) Location of gland Ultrasound, % Negative 88 Neck 6 Not available 6 Sestamibi, % Mediastinal 71 Negative 23 Neck 6 Computed tomography scan, % Mediastinal 71 Negative 23 Neck 6 Selective venous sampling, % Mediastinal 59 Not available 41 Method, % Video-assisted thoracoscopic surgery 88 Right 60 Left 40 Mediastinoscopy, % 12 2 concordant studies Yes 71 No 29 50% reduction in intra-op PTH levels, % Yes 82 No 12 Not available 6 Pathology-confirmed parathyroid adenoma, % Yes 88 No 12 Resolution of hypercalcemia post-op, % Yes 82 No 12 Not available 6 Postoperative calcium (median SD, mg/dl) Complications, % Hypocalcemia 18 Recurrent laryngeal nerve injury 6 Myocardial infarction 6 Pneumonia 6 Chest tube removal (median SD, postop day) Hospital stay (median SD, days) 3 2 Postop postoperative; PTH parathyroid hormone; SD standard deviation. which was visualized by laryngoscopy. A preoperative laryngoscopic examination had noted a preexisting right vocal cord paresis, which was related to prior neck exploration. Her left vocal cord paresis resolved spontaneously; she was extubated on postoperative day 2. Permanent hypoparathyroidism did not develop in this patient. No incisional or wound complications occurred, and no patient required readmission for issues related to the operation. There were no perioperative deaths. Excluding the patient for whom the follow-up calcium level was unavailable, hypercalcemia resolved in 14 of 16 patients (88%). Pathology analysis identified parathyroid adenoma or hyperplasia in all of the successful explorations, as well as in the patient lost to follow-up. No parathyroid tissue was seen in 2 of 17 patients (12%). Both patients had undergone only a single preoperative localization study before VATS exploration that demonstrated a mediastinal parathyroid gland. Results of selective venous sampling appeared to be consistent with a mediastinal location for the gland, but neither the CT nor sestamibi scans were able to localize the adenoma. Previous neck exploration for primary hyperparathyroidism had been unsuccessful in both patients, so the decision was made to proceed with mediastinal exploration with VATS based on a single localizing study. These patients both had persistent hyperparathyroidism after VATS exploration. One patient s hyperparathyroidism subsequently resolved after she received a left thyroid lobectomy for a suspected intrathyroidal parathyroid adenoma, which was confirmed histologically. The other patient has not yet undergone a curative parathyroid resection. Comment The superior parathyroid glands are normally found posterior to the upper and middle lobes of the thyroid gland, whereas the inferior parathyroid glands are typically found within 1 cm from where the inferior thyroid artery and recurrent laryngeal nerve cross. Ectopic parathyroid glands, however, can be found anywhere along the line of descent: in the retroesophageal space and the tracheoesophageal groove in the case of superior parathyroid glands and in the thymus and intrathyroidal locations for inferior parathyroid glands. Although it is estimated that mediastinal glands in primary hyperparathyroidism may approach 20%, only 2% require mediastinal exploration and extraction [1, 4]. Most parathyroid glands in the thymus can be resected through a transcervical approach. In many cases, a mediastinal parathyroid gland is suspected when a neck exploration is unsuccessful in finding an abnormal gland or when persistent or recurrent hyperparathyroidism occurs after a parathyroidectomy. Recently, as imaging techniques have become more advanced at determining the location of these ectopic mediastinal glands, some have challenged the notion that a negative bilateral neck exploration should be undertaken before considering a mediastinal or tho-

4 Ann Thorac Surg WEI ET AL 2011;92: MEDIASTINAL PARATHYROID ADENOMA RESECTION 1015 racic approach to seek out the ectopic parathyroid tissue [5, 6]. Preoperative localization studies are necessary before mediastinal exploration for parathyroid adenoma: older studies have reported a 33% to 40% failure rate of exploration through sternotomy without preoperative imaging [7 9]. The optimal approach to preoperative localization has not been determined. Patients with primary hyperparathyroidism generally receive neck ultrasound imaging and a sestamibi scan to evaluate for possible locations of a parathyroid adenoma. The cervical ultrasound result will be negative in almost all patients who have mediastinal parathyroid adenomas because the adenoma lies below bony structures of the chest, such as the sternum and clavicles, which prevent it from being seen. Only one of our patients had an ultrasound study that demonstrated a parathyroid adenoma in the neck; the CT scan, however, showed the gland at the aortopulmonary window, a location that was confirmed at mediastinoscopy. The sensitivity of sestamibi scan for detecting a mediastinal parathyroid gland was 80% in our series, which is comparable to reported sensitivities of 70% to 81% for detection of ectopic parathyroid glands [10, 11]. CT scan can be used to corroborate a sestamibi scan that demonstrates a mediastinal parathyroid gland. Parathyroid adenomas are usually well circumscribed on CT scan, and enhance intensely with the administration of intravenous contrast [12]. The accuracy of CT scan appears to be between 73% and 84%; however, in general it is less specific of a modality than a sestamibi scan [13, 14].Inour experience, the usefulness of CT scan has been optimized by working with a radiologist with a special interest in parathyroid localization with this modality. The sensitivity of CT scan in our series was 80%. In the future, combination sestamibi and CT scan (single photonemission CT/CT) may prove to be more commonly used, as a single diagnostic examination that achieves both anatomic and functional localization and which may not require the expertise that finding an ectopic parathyroid by CT alone seems to require [15]. SVS may not be necessary in patients who undergo concordant sestamibi and CT scans; however, SVS has been demonstrated to have a very high sensitivity (95%) and fairly good specificity (86%) [9]. For patients with only one other positive localization study, SVS may be useful as a confirmatory test. In the patient in whom all other localization studies are unrevealing, a positive SVS may be justification for mediastinal exploration, with the caveat that these patients appear to have a lower rate of successful cure of hyperparathyroidism. In our series, 4 patients with negative results on sestamibi and CT scans went for VATS exploration on the basis of SVS results alone, and hyperparathyroidism resolved in 2 (50%). In one patient in whom VATS failed, the sestamibi scan demonstrated uptake in the left lobe of the thyroid gland that was suggestive of a parathyroid adenoma. The CT scan showed nodules in the thymic bed and in the left thyroid. We decided to perform a VATS exploration after SVS showed a high PTH level in her superior vena cava compared with the brachiocephalic and internal jugular veins. The intraoperative frozen section was negative for parathyroid tissue, and intraoperative PTH levels failed to decrease appropriately. The patient was subsequently cured of her hyperparathyroidism after a left thyroidectomy for an intrathyroidal parathyroid adenoma. In this case, the sestamibi scan was more accurate than the CT scan or SVS. The second patient for whom VATS exploration failed also had only one study localizing suspected parathyroid tissue in the mediastinum. SVS demonstrated a high PTH gradient at the left proximal brachiocephalic vein. Sestamibi and CT scans did not reveal any locations suggestive for parathyroid adenoma. Pathologic examination of the specimen obtained through VATS revealed that no parathyroid gland was resected. This patient has not yet undergone successful a operation for her hyperparathyroidism, so it is not clear whether she had a parathyroid gland located in another position or if she truly had a mediastinal parathyroid gland that was missed at VATS. Previous authors have commented on the risk inherent in thoracoscopic exploration for mediastinal parathyroid tissue in patients with unclear or discordant preoperative localization studies [16]. On the basis of our results, we too urge caution before proceeding with VATS exploration in patients who have discordant results of preoperative localization studies. In our two unsuccessful cases, SVS yielded a false-positive result that resulted in an unnecessary exploration and delay in treatment. A patient with a sestamibi scan that demonstrates increased uptake in the region of the neck, despite other discordant studies, should likely undergo repeat neck exploration with possible intraoperative ultrasound imaging or radioguidance before being considered for VATS exploration. Video assisted thoracoscopic surgery was 100% successful in patients with two or more concordant studies and was 60% in patients who had only one study suggesting a mediastinal location for ectopic parathyroid tissue. Even with a lower success rate, VATS exploration may be justified in these latter patients who remain symptomatic or otherwise affected by their primary hyperparathyroidism, usually after a prior neck exploration has failed. We suggest, however, that any patient with only one positive localization study, SVS or otherwise, should be counseled about the decreased likelihood of a successful VATS procedure. Our diagnostic and treatment algorithm is shown in Figure 1. The superiority of VATS over sternotomy for mediastinal exploration for parathyroid adenoma is fairly well established. Although no randomized studies of VATS vs sternotomy have been performed, historical data show that VATS has a decreased complication rate and quicker recovery [2, 17, 18]. Mediastinoscopy offers an even shorter hospital stay and quicker recovery time compared with VATS. Extractions of mediastinal parathyroid tissue through an anterior mediastinotomy and subxiphoidal laparoscopic approaches have also been reported [19, 20].

5 1016 WEI ET AL Ann Thorac Surg MEDIASTINAL PARATHYROID ADENOMA RESECTION 2011;92: Fig 1. Diagnostic and treatment algorithm for patients presenting with suspected mediastinal parathyroid adenoma. (CT computed tomography; PTH parathyroid hormone; VATS video-assisted thoracoscopic surgery.) In deciding between mediastinoscopy and VATS, we prefer mediastinoscopy for parathyroid adenomas that appear to be located in the middle mediastinum (most commonly, in a paratracheal location), and VATS for those that are located in the anterior mediastinum. We typically perform VATS through the right chest because there is generally more working space for exploration of the anterior mediastinum. We approach the adenoma through the left chest, however, if the preoperative CT scan demonstrates that it would be more accessible from this side. Generally, we perform a complete thymectomy in these patients. In a few instances, the gland was clearly visible and easily isolated, and we selectively resected the ectopic mediastinal parathyroid gland without thymectomy. The average hospital length of stay (3.3 days) and success rate of exploration (88%) for our VATS patients compares favorably with the largest series of thoracoscopic mediastinal parathyroid resections in the literature to date, which reported an average length of stay of 4.7 days and a 77% success rate among 13 patients [16]. Various modalities for demonstrating successful resection of parathyroid adenomas exist, including intraoperative PTH monitoring and frozen section, and are readily used for parathyroidectomies performed through a cervical incision. We believe that intraoperative PTH monitoring is useful for VATS or mediastinoscopic parathyroidectomy as well. Resection of the abnormal gland was successful in all patients in our series who had a 50% or greater drop in the intraoperative PTH level. We believe that the use of frozen section can therefore be reserved for patients in whom the PTH levels fail to drop. A negative examination for parathyroid tissue on frozen section can lead to one or more of three outcomes: (1) further resection of any residual thymic tissue in the chest through VATS, (2) exploration and removal of thymic tissue through a cervical incision at the time of VATS, or (3) termination of the operation and early discussion with the patient in the postoperative period regarding likely failure of VATS to cure hyperparathyroidism and the need to pursue further diagnostic and treatment strategies. If an experienced surgeon is available at the time of VATS, patients could conceivably be prepared for repeat neck exploration in the case of intraoperative failure of the VATS resection to result in an appropriate decrease in PTH levels. No pulmonary or incisional complications occurred postoperatively, and day of chest tube removal (median, postoperative day 1) and length of stay (median, 3 days) in our series was comparable to other studies. The major complication that could be attributed to technical factors was temporary left vocal cord paresis and reintubation in a patient who had known preexisting right vocal cord paresis. Recurrent laryngeal nerve palsy is an uncommon problem after VATS parathyroidectomy, having been reported in only 2 patients previously [1, 4]. Our series shows that temporary postoperative hypocalcemia may be common after successful VATS removal of a parathyroid adenoma, occurring in 18% of our patients despite routine calcium supplementation. This rate falls within the range of quoted rates of hypocalcemia in the literature of between 8% and 42% after parathyroidectomy for primary hyperparathyroidism [12, 21]. Calcium levels should be monitored carefully

6 Ann Thorac Surg WEI ET AL 2011;92: MEDIASTINAL PARATHYROID ADENOMA RESECTION 1017 in the postoperative period in VATS parathyroidectomy, and vigilant attention to patient complaints suggestive of hypocalcemia after discharge from the hospital is warranted. In conclusion, mediastinal exploration for primary hyperparathyroidism requiring the expertise of a thoracic surgeon is somewhat uncommon. Fewer than 2% of all patients who underwent parathyroidectomy required a mediastinoscopic or VATS approach. In these patients, the use of multiple preoperative localization studies is crucial for the confirmation of a truly mediastinal location to give the patient the best chance at successful VATS or mediastinoscopic exploration. Current modalities for identifying a mediastinal position of ectopic parathyroid tissue include sestamibi, high-resolution CT or single photon-emission CT scanning, MRI, and SVS. Mediastinal exploration was successful in all patients in our series who had two or more concordant studies but was only 60% successful in patients who had only one confirmatory study. We believe that practitioners should be cautious when they counsel patients who present with only one positive preoperative location study about their expectations of success. Patients who undergo mediastinoscopic exploration are normally discharged on the same day of the operation, whereas patients who undergo VATS exploration can usually be discharged on day 1 or 2. Our group has recently begun to perform this operation on an outpatient basis, with removal of chest tube and discharge home on the same day as the operation. In our experience, the most common postoperative complication of mediastinoscopic or VATS parathyroidectomy is temporary hypocalcemia, which usually responds to outpatient oral calcium treatment and is generally self-limiting. With careful preoperative patient selection, VATS and mediastinoscopy can act as safe and effective modalities for the treatment of hyperparathyroidism caused by suspected mediastinal parathyroid tissue. References 1. Prinz RA, Lonchyna V, Carnaille B, et al. Thoracoscopic excision of enlarged mediastinal parathyroid glands. Surgery 1994;116: Conn JM Goncalves MA, Mansour KA, et al. The mediastinal parathyroid. Am Surg 1991;57: Alesina PF, Moka D, Mahlstedt J, Walz MK. Thoracoscopic removal of mediastinal hyperfunctioning parathyroid glands: personal experience and review of the literature. World J Surg 2008;32: Soler R, Bargiela A, Cordido F, Aguilera C, Argueso R, Cao I. MRI of mediastinal parathyroid cystic adenoma causing hyperparathyroidism. J Comput Asssist Tomogr 1996; 20: Liu RC, Hill ME, Ryan JA. One-gland exploration for mediastinal parathyroid adenomas: cervical and thoracoscopic approaches. Am J Surg 2005;189: Amar L, Guignat L, Tissier F, et al. Video-assisted thoracoscopic surgery as a first-line treatment of mediastinal parathyroid adenomas: strategic value of imaging. Eur J Endocrinol 2004;150: Wang C, Gaz RD, Moncure AC. Mediastinal parathyroid exploration: a clinical and pathologic study of 47 cases. World J Surg 1986;10: Edis AJ, Sheedy PF, Beahrs OH, Van Heerdan JA. Results of reoperation for hyperparathyroidism with evaluation of preoperative localization studies. Surgery 1978;84: Clark O. Mediastinal parathyroid tumors. Arch Surg 1988; 123: Phitayakorn R, McHenry CR. Incidence and location of ectopic abnormal parathyroid glands. Am J Surg 2006;191: Amar L, Guignat L, Tissier F, et al. Video-assisted thoracoscopic surgery as a first-line treatment for mediastinal parathyroid adenomas: strategic value of imaging. Eur J Endocrinol 2004;150: Mittendorf EA, Merlino JI, McHenry CR. Post-parathyroidectomy hypocalcemia: incidence, risk factors, and management. Am Surg 2004;70: Numerow LM, Morita ET, Clark OH, Higgins CB. Persistent/ recurrent hyperparathyroidism: a comparison of sestamibi scintigraphy, MRI, and ultrasonography. J Magn Reson Imag 1995;5: Mazzeo S, Cappelli C, Caramella D, et al. Multidetector CT in diagnostic work-up of patients with primary hyperparathyroidism. Radiol Med 2007;112: Eslamy HK, Ziessman HA. Parathyroid scintigraphy in patients with primary hyperparathyroidism: 99mTc sestamibi SPECT and SPECT/CT. Radiographics 2008;28: Randone B, Costi R, Scatton O, et al. Thoracoscopic removal of mediastinal parathyroid glands: a critical appraisal of an emerging technique. Ann Surg 2010;251: Medrano C, Hazelrigg SR, Landreneau RJ, Boley TM, Shawgo T, Grasch A. Thoracoscopic resection of ectopic parathyroid glands. Ann Thorac Surg 200;69: Russell CF, Edis AJ, Scholz DA, Sheedy PF, Van Heerden JA. Mediastinal parathyroid tumors: experience with 38 tumors requiring mediastinotomy for removal. Ann Surg 1981;193: Schlinkert RT, Whitaker MD, Argueta R. Resection of selected mediastinal parathyroid adenomas through an anterior mediastinotomy. Mayo Clin Proc 1991;66: Wei JP, Gadacz TR, Weisner LF, et al. The subxiphoidal laparoscopic approach for resection of mediastinal parathyroid adenoma after successful localization with Tc-99msestamibi radionuclide scan. Surg Laparosc Endosc 1995;5: Kald BA, Mollerup CL. Risk factors for severe postoperative hypocalcemia after operations for primary hyperparathyroidism. Eur J Surgery 2002;168:552 6.