Early (A) and delayed (B) planar sestamibi scan images read as nonlocalizing. Focused parathyroidectomy based on a surgeon's reading revealed a single left inferior parathyroid adenoma.
Planar sestamibi scan (A, early; B, delayed) and select single-photon emission computed tomographic images (C and D) read as nonlocalizing. Focused parathyroidectomy based on a surgeon's reading revealed a single right inferior parathyroid adenoma.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Anderson SR, Vaughn A, Karakla D, Wadsworth JT. Effectiveness of Surgeon Interpretation of Technetium Tc 99m Sestamibi Scans in Localizing Parathyroid Adenomas. Arch Otolaryngol Head Neck Surg. 2008;134(9):953–957. doi:10.1001/archotol.134.9.953
To evaluate the ability of surgeons to predict the laterality of parathyroid adenomas from technetium Tc 99m sestamibi scans and compare their results with the radiologic interpretations.
Retrospective medical record review with single- blinded review of sestamibi scans.
Tertiary care academic medical center
The study population included 110 consecutive parathyroidectomy cases from January 2001 to June 2004. Inclusion criteria were biochemically proven primary hyperparathyroidism, with documented serum hypercalcemia and elevated parathyroid hormone levels. Only cases due to a single adenoma that were cured with a single surgical procedure were included.
All patients underwent resection of a parathyroid adenoma following a preoperative sestamibi localization study and serum calcium and parathyroid hormone level analysis.
Main Outcome Measure
Adenoma location was determined from a review of operative and pathological reports. Two head and neck surgeons performed a blinded review of all scans, and their findings were compared with the radiology reports.
Of 82 adenomas, 51 (62%) were correctly lateralized in the radiology report, while the other 31 were interpreted as normal scans. The sensitivity and specificity of the radiologic interpretations for parathyroid adenomas in all patients with primary hyperparathyroidism were 62% and 83%, respectively. The scan interpretation of the 2 surgeons produced accurate lateralization of 91% and 91% of these single adenomas. Of the 31 single adenoma scans read as normal by the radiologist, the surgeons correctly lateralized 22 of 29 (76%) and 21 of 28 (75%) of the adenomas.
The review of sestamibi scans by surgeons allows accurate localization of parathyroid adenomas that may not be identified by standard radiologic interpretations.
Bilateral neck exploration has been the standard treatment for primary hyperparathyroidism, with an established cure rate of 95% or greater for experienced surgeons. However, it requires more intervention than is necessary for more than 85% of patients, whose hyperfunctioning condition is due to a single adenoma.1,2 With the development of new localization and diagnostic technologies, the treatment standard has shifted to more directed surgery, so that unilateral neck explorations and directed, or focused, parathyroidectomies are now becoming the standard of care. These minimally invasive procedures depend on accurate preoperative localization of the offending parathyroid tissue.
The modern era of parathyroid imaging began in 1989, when Coakley et al3 reported the advantages of technetium Tc 99m–labeled sestamibi over thallium in parathyroid imaging. Once reserved for recurrent cases, sestamibi scintigraphy has proven to be the best initial preoperative modality for locating hyperfunctioning parathyroid tissue and is now commonly performed on all patients with primary hyperparathyroidism.4-7
The sensitivity of the sestamibi scan is widely reported to be 43% to 100%, with more recent results consistently measuring near 90%.1,7-11 The diagnostic accuracy is improving as physicians become more familiar with the scan protocol, interpretation, and correlation with operative findings. An important aspect of reported sestamibi scan accuracy is the fact that the studies typically compare clinical findings with the radiologists' interpretation of the parathyroid scan.
This study focuses on independent surgeon interpretation of sestamibi scans, with the goal of predicting adenoma laterality. By augmenting the radiologist's reading of the scan, we hypothesize that the localization and/or laterality accuracy for single adenomas will improve.
Study approval was obtained from the institutional review board of the Eastern Virginia Medical School, Norfolk. We retrospectively reviewed the 110 consecutive parathyroidectomy cases of 2 fellowship-trained head and neck surgeons (D.K. and J.T.W.) from January 2001 to June 2004. Inclusion criteria were biochemically proven primary hyperparathyroidism, with documented serum hypercalcemia and elevated intact parathyroid hormone levels. Diagnosis was typically made by the referring physicians, using serum calcium and/or ionized calcium measurements. We routinely performed follow-up measurements of our patients' serum calcium levels, adjusted for serum albumin level. Only cases due to a single adenoma that were cured with a single surgery were included, with a cure being defined as normalization of postoperative serum calcium and parathyroid hormone (PTH) levels.
All patients underwent preoperative sestamibi scanning, which was performed at 10 different local hospitals, using various protocols. Scintigraphy was performed using 5.9 to 38.0 mCi of intravenous sestamibi, with early images captured at 15 minutes and late images at either 2, 3, or 5 hours. Planar images of the head, neck, and upper thorax were obtained. Additional views and single-photon emission computed tomographic (SPECT) scans were frequently performed at the discretion of the radiologist, and these are included in this study. All scans were interpreted by a radiologist as per hospital protocol and were also reviewed by the operating surgeon before the procedure.
Surgery began as a focused parathyroidectomy directed to the suspect region, according to the scan interpretation. Adenoma removal was confirmed with frozen section pathologic examination and intraoperative PTH measurements at baseline and 15 minutes after removal. If a single adenoma was positively identified and removed, the procedure was terminated. Unilateral neck explorations were performed when a single adenoma was not found at the initial suspected site. Bilateral explorations were performed when single adenomas were not identified on the initial operative side. No additional intraoperative localization techniques, such as the gamma probe, were used. The diagnosis and location of each single adenoma were obtained from a review of the operative findings, intraoperative PTH values, histopathologic diagnosis, and findings from the clinical follow-up examination.
Each surgeon retrospectively reviewed all scans independently, blinded to the patients' identifying data. The films were presented to the surgeons in a random fashion by the primary author (S.R.A.), with patient names obscured from view. Patient diagnosis was not available during the film reviews. After the surgeons' radiologic interpretations were rendered and recorded, the pathologic diagnoses were then determined. Then the scan interpretations were compared with the true adenoma location. Radiologists' interpretations of all scans were also examined, which included some scans that were unavailable for blinded study surgeon review.
Sestamibi scan interpretations were classified as true positive when the report correctly identified the side of the neck containing the adenoma and as true negative when normally functioning glands were not visualized. False-positive reports incorrectly designated a focus of scintigraphic activity as a hyperfunctioning parathyroid gland, and false-negative reports did not identify a hyperfunctioning gland. Sensitivity was determined as the standard ratio of true positives to the sum of true positives and false negatives. Specificity was calculated as the ratio of true negatives to the sum of false positives and true negatives.
Of the 110 parathyroidectomy patients, 2 were excluded for tertiary hyperparathyroidism, 12 were excluded for secondary hyperparathyroidism, and 2 cases were unresolved with initial surgery. Of the 94 patients with primary hyperparathyroidism who were cured with a single procedure, 93 had the full study data available for review. Of these 93 patients, 82 (88%) were found to have a single adenoma, which was the focus of this study. The patient ages ranged from 33 to 91 years, with a mean of 58.4 years (Table).
Of these 82 adenomas, 51 (62%) were correctly lateralized in the radiology report, while the other 31 were interpreted as normal scans. There were 10 true-negative and 2 false-positive readings. The sensitivity and specificity of the radiology interpretations for parathyroid adenomas in all primary hyperparathyroidism patient scans were 62% and 83%, respectively. One radiology interpretation scored both a false negative and a false positive because it reported an adenoma on the left, but the adenoma was actually located on the right side.
The scans for 76 of the 82 patients with single adenomas were available for retrospective blinded surgeon review. Despite being unavailable for blinded review, some scans had their preoperative surgeon interpretation recorded at the history review and physical examination and were therefore included in this study. The scan interpretations of the 2 surgeons produced accurate lateralization of 71 of 78 (91%) and 70 of 77 (91%) of these single adenomas. Of the 31 single adenoma scans read as normal by the radiologist (false negatives), surgeon review and/or interpretations were performed by 2 head and neck surgeons for 29 and 28. From these films, the 2 surgeons correctly lateralized 22 of 29 (76%) and 21 of 28 (75%) of the adenomas (Figure 1 and Figure 2).
The advantages of the focused parathyroidectomy over the traditional bilateral neck exploration have been well described and include decreased operative time and length of hospital stay, smaller incision, decreased incidence of postoperative hypocalcemia, lower cost, and the option of avoiding general anesthesia.12-17 Even when a focused parathyroidectomy case develops into a unilateral neck exploration, most of these advantages over bilateral exploration still apply.9
Preoperative sestamibi scintigraphy has emerged as the optimal evaluation of hyperfunctioning parathyroid tissue. The study's ability to direct a surgeon to a single adenoma currently averages 85% to 90%. Our study measured the radiology lateralization accuracy at 61%, which is lower than that quoted in most recent reports. This finding could be because our review includes scans performed several years ago and refinements in scan performance and interpretation have improved localization accuracy. In addition, our scans were performed and interpreted at 10 different facilities, including some that perform this examination infrequently and lack the interaction with endocrine surgeons from which larger facilities benefit. The radiology departments used differing scan protocols, as described previously, and occasionally included SPECT scanning. Some of the scan protocols included in this study, such as allowing 3 to 5 hours for tracer washout, have been found to be suboptimal. These scans were included in this study because this is a review of consecutive surgery patients. We have since altered our protocol to have all patients undergo sestamibi scanning at our institution to establish uniformity in scintigraphy results. All reviewed scans were hard copies. This resulted in the surgeons reviewing a variety of film sizes, image number, and orientation, which allows for a broad applicability of our results. With the advent and widespread installment of digital radiographic systems, which allow infinite adjustment of the digital images, the current scan localization accuracy may be even higher.
Sestamibi parathyroid scanning is a relatively new technology and continues to improve with time. The single isotope, double-phase (washout) technique, initially reported in 1992, is the standard technique.18 Single-photon emission computed tomography improves sensitivity and localization accuracy, and SPECT–computed tomography fusion promises to further improve localization. Dual radionuclide subtraction imaging has been found to benefit localization in patients with thyroid abnormalities.6 Although its sensitivity for multiglandular disease is notably lower, misdiagnosis can be avoided with supplemental preoperative imaging and/or intraoperative rapid PTH measurements.
Although the accuracy of the sestamibi scan regularly approaches 90% in the literature, we propose that surgeons can improve the localization accuracy. They have a more thorough understanding of developmental and clinical anatomy and are not bound to strict radiologic interpretation criteria. This study validates the concept that the surgeons' lateralization accuracy of 91% exceeds published radiology values. Our surgeons based their scan interpretations on scintigraphic asymmetry. For scans that did not have an obvious adenoma radiotracer collection, the surgeons used minute differences in radiotracer retention to correctly lateralize the adenoma. Some authors advocate supplemental imaging, such as ultrasound, for nonlocalizing sestamibi scans. Our study shows that surgeon review of the sestamibi scan will considerably improve study accuracy: our surgeons were able to correctly lateralize 75% of the adenomas in “nonlocalizing scans.” This occurs without any additional cost or patient involvement.
The quoted accuracy of sestamibi scans in published studies is based on radiologists' scan interpretations, and some authors even suggest that the surgical approach and operative success depend on the skill of an institution's radiology staff.1,12,19 Some studies have specifically examined the role that the nuclear medicine radiologist plays in parathyroid adenoma localization, finding better results with experienced scintigraphers12 and when the scans are read by the surgeon and nuclear medicine physician together.4 Our study appears to be the first to measure the surgeon's ability to localize parathyroid adenomas through independent sestamibi scan interpretation.
One obvious potential bias in this study is the fact that the radiologists reviewed all sestamibi scans, while the surgeons retrospectively reviewed the scans for this study, knowing that it includes only patients with single adenomas. However, this “bias” exists in reality, as radiologists interpret the scans according to radiologic criteria, including the absence or presence of an adenoma. Surgeons, meanwhile, review the scans in preparation for opening the neck. They are responsible for deciding which side to explore first and where to focus the search. Therefore, they should approach scan interpretation as the surgeons did in the present study—by using any scintigraphic asymmetry to predict adenoma laterality, whether or not there is an obvious adenoma. We propose that surgeons scrutinize all scans preoperatively, especially nonlocalizing scans, to increase the chance of a successful adenoma localization and unilateral exploration.
We conclude that the sestamibi scan is a proven, reliable method for preoperative localization of parathyroid adenomas and that the review of sestamibi scans by surgeons allows accurate localization of parathyroid adenomas that may not be identified by standard radiologic interpretations.
Correspondence: Scott R. Anderson, MD, Department of Otolaryngology–Head and Neck Surgery, Eastern Virginia Medical School, 825 Fairfax Ave, Ste 510, Norfolk, VA 23507 (firstname.lastname@example.org).
Submitted for Publication: May 3, 2007; final revision received October 24, 2007; accepted October 30, 2007.
Author Contributions: Dr Anderson had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Anderson and Karakla. Acquisition of data: Anderson and Vaughn. Analysis and interpretation of data: Anderson, Karakla, and Wadsworth. Drafting of the manuscript: Anderson and Vaughn. Critical revision of the manuscript for important intellectual content: Anderson, Karakla, and Wadsworth. Statistical analysis: Anderson. Administrative, technical, and material support: Anderson and Vaughn. Study supervision: Karakla and Wadsworth.
Financial Disclosure: None reported.
Previous Presentation: This study was presented at the 2007 Annual Meeting of the American Head and Neck Society; April 28-29, 2007; San Diego, California.