Lesion based diagnostic performance of dual phase ^99mTc-MIBI SPECT/CT imaging and ultrasonography in patients with secondary hyperparathyroidism

Secondary hyperparathyroidism (SHPT) occurs due to a progressive increase in the level of parathyroid hormone (PTH) in diseases that affect the metabolism of calcium or phosphorus and is a common complication in patients with chronic kidney disease [1, 2]. Patients with severe SHPT often develop high turnover bone disease, which can cause bone pain and skeletal fractures risk [3]. SHPT is also associated with an increased risk of cardiovascular calcification and mortality [4]. Traditional therapies such as administration of calcium salts and vitamin D are limited by hypercalcemia, hyperphosphatemia and lack of long-term efficacy [5].

Surgical parathyroidectomy remains the only option in 5% to 10% of patients with chronic renal failure who are treated with long-term dialysis and fail to respond to medical therapy [6, 7]. Successful surgical treatment can reduce the PTH levels and improve the clinical symptoms such bone pain. Different from primary hyperparathyroidism (PHPT) in which patients generally have only one lesion [8], patients with SHPT have multiple lesions [9]. The unguided bilateral neck exploration is effective in 90–95% patients [10]: surgical failure is due to ectopic glands and undetected multi-glandular disease. Surgical outcome is often unfavorable, resulting in persistent or recurrent hyperparathyroidism in 10–30% cases [11]; therefore, pre-surgical localization is necessary in patients with SHPT.

Parathyroid glands can be detected with multiple modalities, such as imaging, high-resolution (7.0–14.0 MHz) ultrasonography (USG), thin-section CT and MRI [12]. USG is an advantageous modality as a non- radiation emitting, and widely available technique in clinic. ^99mTc-methoxyisobutyl isonitrile (^99mTc-MIBI) is a common radiopharmaceutical used in parathyroid imaging. Dual-phase parathyroid imaging with ^99mTc-MIBI currently represents the current diagnostic modality for parathyroid glands. The diagnostic performance of ^99mTc-MIBI SPECT/CT in patients with SHPT showed a higher sensitivity and specificity compared with ultrasonography in detecting hyperplastic glands on a per-patient-based analysis [9, 13].

There were already some studies have reported the diagnostic performance of ultrasonography and ^99mTc-MIBI dual phases imaging. But the results of sensitives had a vary difference. The sensitives of ultrasonography and ^99mTc-MIBI dual phases imaging had a wide range of 35.9–91.5% and 36.6–66%, respectively [14‐16]. Now we study the diagnostic performance of ^99mTc-MIBI SPET/CT and ultrasonography in patients and lesions with SHPT were all studied, to explore the factors that affect the diagnostic performance of SPECT/CT, such as PTH level and diameter. The patients with secondary hyperparathyroidism that all had renal failure. Patients were undergoing both ^99mTc-MIBI dual phases SPECT/CT imaging and ultrasonography before surgery.

The clinic pathological characters and ^99mTc-MIBI data of the 50 patients with SHPT who underwent surgery are summarized in Table 1. For the patients included in the study, the primary diseases of all patients were chronic renal failure. And among them, there were 42 patients with peritoneal dialysis or hemodialysis. Among the 50 patients, 49 (98.00%) patients had a positive ^99mTc-MIBI imaging, and only 1 patient had a negative ^99mTc-MIBI imaging. ^99mTc-MIBI imaging found an ectopic parathyroid in one patient (Fig. 2). For ultrasonography, there were 39 (78.00%) patients were found had at least one increased parathyroid. In patients with known serum calcium and phosphorus levels, 34.00% (17/50) and 76.00% (38/50) patients showed higher serum calcium and phosphorus levels, respectively.

In 50 patients who underwent surgery, two patients had parathyroid subtotal resection, 48 patients had complete parathyroid resection and autologous transplantation. A total of 199 glands were resected in the surgery of the 50 patients. One hundred and eighty-three glands were proved to be parathyroid lesions. Four glands were confirmed to be normal parathyroid glands, twelve were lymph nodes. In 100 and 83 confirmed lesions, 179 were parathyroid hyperplasia, and 4 lesions in 3 patients were confirmed to be parathyroid adenoma (Fig. 3). Among the fifty patients, the pathological results revealed one lesion from one patient (2%), two lesions from 4 patients (4%), three lesions from 6 patients (12%), and four lesions from the rest 39 patients (78%). ^99mTc-MIBI imaging showed 112 positive lesions in 50 patients, and 108 lesions confirmed to be true positive. For the 4 false positive lesions found in three patients, two lesions were confirmed to be thyroid tissue, two lesions were lymph nodes. Seventy-four lesions found in 39 patients showed negative ^99mTc-MIBI uptake but were the pathological confirmed to be parathyroid hyperplasia or adenoma. On per-lesion basis analysis, the sensitivity and specificity of ^99mTc-MIBI were 59.34% and 75.00%, respectively. Eighty-nine enlarged lesions in ultrasonography, 86 lesions were confirmed to be true positive. Ninety-seven normal size lesions were pathological confirmed to be parathyroid hyperplasia. Therefore, on per-lesion basis analysis, the sensitivity and specificity of ultrasonography were 46.24% and 80.00%, respectively. A representative ultrasonography imaging showed in the Fig. 4.

Abnormal ^99mTc-MIBI uptake was considered to be positive on visual analysis, and the different uptake strength of each lesion was graded to three levels. Each lesion was scored according to ^99mTc-MIBI uptake level. And then the lesions number of each patient and lesion uptake value scores of one patient were grouped for the 50 patients included; the numbers of patients in each group were listed in Table 2. Furthermore, for these patients, the number of lesions for each patient was counted on the basis of the pathological results. The correlation between PTH level and laboratory test items such as calcium, phosphorus and alkaline phosphatase were analyzed. Pearson correlation results showed a significant linear association between the serum AKP level and PTH level (r = 0.699, P < 0.001). The lesions diameter and the number of lesions were confirmed by pathological results, and the lesion uptake value scores of each patient showed no significant linear association with PTH levels.

Following ROC analysis, the optimal cutoff values of the maximum tumor diameter for predicting positive ^99mTc-MIBI was 8.05 mm (AUC 0.731, 95% CI 0.618–0.843; sensitivity = 82.20% and specificity = 52.90%) (Fig. 5). The cutoff values of 8.05 mm was used for lesion diameter in the 118 lesions as measured by SPECT/CT. Eighteen lesions (66.67%) of 27 lesions were negative in ^99mTc-MIBI SPECT/CT in patients with a lesion diameter lower than the cutoff value (P < 0.0001) (Table 3).

We studied rare cases in patients with secondary hyperparathyroidism that all had renal failure. All the patients were undergoing both ^99mTc-MIBI dual phases SPECT/CT imaging and ultrasonography before surgery. ^99mTc-MIBI parathyroid imaging is a valuable method based on the differential washout rate between the thyroid tissue and parathyroid hyperplasia [12]. Parathyroid hyperplasia has very high metabolic rate despite its small size, and it shows intense ^99mTc-MIBI uptake. More recently, SPECT/CT systems combining state-of-the-art multi-detector CT and state-of-the-art gamma cameras are being produced, and guidelines for image acquisition, interpretation, and reporting for patients with SHPT have been published [13, 18]. SPECT/CT imaging also influences the surgical approach, enabling to choose different surgical approaches for lesions in different areas [19]. In our study, imaging detected an ectopic parathyroid in one patient. The major significance of ^99mTc-MIBI imaging pre-surgery was the detection of ectopic parathyroid, which guided the surgery. In our study, the sensitivities of ^99mTc-MIBI imaging on a per-patient basis and on a per-lesion basis were 98.00% and 59.34%, respectively. For USG the results were 78.00% and 46.24%, respectively.

In our study the sensitivity of ^99mTc-MIBI imaging on a per-lesion basis was 59.34% that exactly matches to a meta-analysis publication on planar ^99mTc-MIBI imaging the pooled sensitivity was 58% in SHPT [20]. There were many researches results showed that the sensitivity of SPECT/CT should be higher than planar ^99mTc-MIBI imaging [21]. There was also study showed that single-phase SPECT/CT (early or delayed) had not significant in superior to dual-phase planar imaging for sensitivity. Dual-phase SPECT/CT had statistically significant in superior to single-phase SPECT/CT and superior to dual-phase planar imaging and dual-phase SPECT [22, 23]. Therefore, ^99mTc-MIBI imaging show a limitation in the diagnosis of lesions in patients with SHPT. Usually we used the ^99mTc-MIBI uptake value as the guide information to diagnose whether the parathyroid lesions were; and CT information only was used to confirm the diagnosis. However, it was very difficult to differentiate the lymph node and the tumors only by CT for the lesions without ^99mTc-MIBI uptake in SPECT imaging. ^99mTc-MIBI has a lower sensitivity in multiple gland disease [24]. But the mechanism is not clear at present. The pre-surgical localization determines the success rate of surgery and the risk of recurrence. Compared to patients with PHPT who mostly have only one lesion [8], most of the patients with SHPT have multiple lesions [9]. The surgical results in our research also proved that 78% of the patients had four lesions. If the surgeon determines the surgical resection according to the ^99mTc-MIBI imaging results, many lesions would be missed. Intraoperative PTH assay is routinely used in clinical practice, and Ohe’s study showed that the decrease in the level of intraoperative PTH could predict the rate of successful parathyroidectomy [25].

In the per-lesion-based analysis, the specificity of ^99mTc-MIBI imaging was 75.00%. The most common cause false-positive results were the presence of thyroid nodules. Because MIBI can be taken up by both hyper-functioning parathyroid and thyroid tissue, the differential diagnosis is based on its elimination from the hyper-functioning thyroid and parathyroid glands. Normal parathyroid tissue without uptake in either of the two phases. The possibility of clearances in similar times for both tissues, which could lead to false negatives, has also been described [11, 26]. Specificity of parathyroid imaging is highly dependent on the prevalence of thyroid nodular disorders in the evaluated patient population [23]. The factors may affect the diagnostic accuracy rate were the followings: the location of parathyroid glands, the size of the lesion, and the functional activity of the lesions, and the prevalence of mitochondria-rich oxyphil cells [27, 28]. Lesion size is important as it relates to the system resolution and to the amount of tracer taken up by the parathyroid tissue [29]. Our results showed that lesion diameter was a statistically significant predictive factor in predicting positive ^99mTc-MIBI SPECT/CT. The larger the lesion diameter, the higher was the MIBI uptake value.

Serum PTH is a major regulator of mineral metabolism that plays a critical role in the maintenance of calcium and phosphate levels [30]. The major physiologic function of PTH is the circulation of ionized calcium. PTH has effects on the gut, kidney, and bone, serving to maintain serum calcium level within a tight range [31]. The pathogenesis of patients with PHPT is independent secretion of PTH by the parathyroid tissue. Unlike patients with PHPT, patients with SHPT have a compensatory PTH secretion for hypocalcaemia. Therefore, in patients with SHPT, the serum calcium level is uncertain; it may increase or remain normal. Pearson correlation analysis between PTH level and various factors showed that the serum AKP level and PTH level exhibit a significant linear association. PTH increased the bone metabolism conversion to elevate the serum AKP level. High level of AKP is the marker of increased bone conversion [32].

Our study demonstrated that the dual phase ^99mTc-MIBI SPECT/CT imaging had a higher sensitivity in the diagnosis of patients with SHPT than ultrasonography on a per-patient-based analysis (98.00% vs. 78.00%). On a per-lesion-based analysis, the sensitivity and specificity of ^99mTc-MIBI were 59.34% and 75.00%, respectively. The results of ultrasonography were 46.24% and 80.00%. Dual phase ^99mTc-MIBI scan for positioning the lesion is an effective pre-surgical tool for patients with SHPT. The lesion diameter maybe a limitation for ^99mTc-MIBI SPECT/CT imaging sensitivity, when the lesion diameter was above 8.05 mm, the higher positive result can be obtained. Serum AKP and PTH level exhibited a significant linear association. That means the higher PTH level, the higher AKP level.