Algorithm for preoperative localization workup in recurrent or persistent hyperparathyroidism 
When re-operative parathyroid surgery is necessary, it has been historically associated with an increased risk of vocal fold paresis and permanent hypoparathyroidism , as well as lower cure rates compared with primary surgery . Thus an “ounce” of prevention in primary parathyroid surgery is worth a “pound” of cure with revision surgery. While parathyroid imaging for preoperative localization has improved greatly over the past 2 decades, John Doppman’s observation that “the greatest challenge in parathyroid localization is to localize a good parathyroid surgeon”  holds true, especially within the context of revision surgery. Prior to embarking on re-operative parathyroid surgery, the surgeon should (1) review the prior and current biochemical workup to confirm the diagnosis, and (2) review the prior operative, pathology, imaging, and intraoperative PTH results in attempt to understand the extent of original surgery and potential cause for failure. Parathyroid embryology and anatomy, intraoperative PTH monitoring, preoperative imaging, radiopharmaceutical guidance, jugular venous sampling, and RLN monitoring are all important considerations and adjuncts which may be helpful in re-operative parathyroid surgery. Lastly, a modified lateral approach avoiding scar tissue from prior surgery is helpful when possible .
The indications for re-operative parathyroidectomy are technically the same as for initial exploration; however given the higher operative risks and lower likelihood of cure, a thoughtful evaluation of the risks and benefits of surgery as well as alternative treatment options is imperative. It is important to confirm that the patient has symptoms that justify further surgery or may suffer end-organ damage such as bone disease or renal stones if left untreated.
It is important to confirm the diagnosis of primary hyperparathyroidism and consideration should be given to familial hypocalciuric hypercalcemia, secondary hyperparathyroidism from vitamin D deficiency, or other causes of hypercalcemia, which would not benefit from revision surgery. In patients with confirmed persistent or recurrent primary hyperparathyroidism, consideration for hereditary etiologies such as MEN1, MEN2a, or hyperparathyroid jaw tumor syndrome should be considered, as well as the possibility of an incompletely excised adenoma or parathyroid carcinoma.
Biochemical confirmation of persistent hyperparathyroidism should be obtained. Review of prior records (operative, pathology, laboratory, and imaging) to ascertain the findings and extent of the procedure carried out at the initial operation including which parathyroid glands were identified and removed should be clarified. Additionally, areas which were and were not explored should be defined. Consider re-review of prior pathology slides. Assess preoperative vocal fold mobility, as the need for re-operation in a patient with unilateral vocal fold immobility should be carefully considered when weighing the risks and benefits of exploring the contralateral side.
It may be necessary to repeat localization scans.
High-resolution ultrasound is an excellent method for preoperative localization of parathyroid adenomas in re-operative cases. Its utility has been established, despite the potential effect of scar tissue obscuring views of the tissue planes and vascularity important for identification of an adenoma. Ghaheri et al. reviewed their ultrasound localization results in re-operative explorations and found similar positive predictive values between patients with and patients without prior thyroid or parathyroid surgery, 84 % vs. 90 % . In conjunction with ultrasound, fine-needle aspiration biopsy may be analyzed for intact parathyroid hormone (iPTH) for localization of the suspected abnormal parathyroid gland, especially when intrathyroidal .
Tc 99m-sestamibi scan (MIBI) using single-photon emission computed tomography (SPECT) with or without fusion CT is very helpful when available, but may have some institutional variability in sensitivity and specificity. Addition of the CT fusion adds better anatomical localization for surgical planning. In the re-operative setting the accuracy of the test is reduced. In patients with persistent hyperparathyroidism, MIBI SPECT accurately localized a pathological parathyroid gland in 33 % of cases before re-operative parathyroidectomy, compared to 61 % in a primary surgical setting . Recently, reports of up to 79 % sensitivity for MIBI preoperative imaging have been given . Multiglandular disease is challenging to detect; Siperstein et al. found that even when the preoperative location suggested a single adenoma, the final diagnosis for those US-identified single-gland patients was only correct 75 % of the time and 70 % of the time for MIBI patients. When MIBI and US were concordant, 16 % of patients were found to have additional parathyroid pathology on further exploration .
“Four-dimensional” contrast-enhanced dynamic CT (4D CT) provides another modality of parathyroid localization. Generally the patient has a non-contrast CT, followed by iodinated contrast CT scans in both the early arterial and subsequent venous phases. Some concerns about increased radiation dose have been raised, although lower dose protocols have been used with success . It may help identify multiglandular disease, which could be an advantage over MIBI imaging and can identify abnormal parathyroid glands missed by US and MIBI . Additionally, it was not found to have an accuracy difference between primary and re-operative cases . It does have the advantage over US of being able to image the mediastinum.
Enlarged parathyroid glands typically demonstrate increased intensity on T2-weighted images on MRI, but is generally less sensitive than US, MIBI, and/or CT in part due to long scan times and motion artifact related to swallowing and respiration. In one review of patients with persistent/recurrent hyperparathyroidism, MIBI was compared with MRI and found to have similar sensitivity and positive predictive value .
Venous Sampling and Angiography
When noninvasive techniques are unsuccessful, catheterization and blood sampling for PTH can provide information about the region of the pathologic gland. Angiography may reveal a blush localizing an adenoma. Experience with this technique is limited to specialized centers however.
Intraoperatively, bilateral inferior jugular vein sampling with (intraoperative PTH) intraoperative parathyroid hormone (IoPTH) can be helpful in indicating the laterality of hyperfunctioning parathyroid tissue. In cases where IoPTH samples were >5 % different from each other, hyperfunctioning parathyroid tissue was successfully localized, though the senior author finds at least a 50 % difference more convincing. This has been found to be especially useful for localization in the setting where initial removal of a suspected adenoma does not yield an adequate IoPTH response, or where preoperative imaging is non-localizing. This is a simple task if exposure for bilateral neck exploration is used . It can also be performed in the office using US guidance. In cases of jugular vein PTH symmetry, the patient may have a mediastinal adenoma or bilateral multigland disease.
Re-operation may be undertaken within a week of the initial surgery or postponed to greater than 3 months from the initial procedure. The strategy is to either avoid fibrosis and inflammation or allow it to subside and obtain further localization studies. Generally there is no rush to get back in unless the patient had undergone bilateral exploration with identification of all four glands and was undertreated with removal of only a single gland.
After the diagnosis and indication for surgery are confirmed, localization studies assist in forming a strategy to approach the remaining diseased parathyroid tissue. MIBI/SPECT CT and US provide complementary information for localization and are good for initial investigation before re-operative cases. Discordant or non-localizing studies may prompt the need for alternative imaging, generally CT, or an invasive localization study if experience is available (Table 24.1).
Apply a systematic approach to the distribution of remaining disease
Distribution of remaining disease
Multiple abnormal glands
Incomplete resection or parathormatosis
Most re-operative parathyroid surgery is performed for a missed, solitary, eutopic adenoma [1, 9, 16, 17]. In a review by Richards et al. of 2065 patients who underwent parathyroidectomy for primary hyperparathyroidism, 228 patients (11 %) were identified that underwent re-operative procedures, the majority of which (57 %) were found to have solitary gland disease. The adenomas were typically located in eutopic position lateral to the thyroid gland . It is notable that in this review 8.5 % of patients had MEN 1, which perhaps accounts for why this is skewed to represent multiglandular disease, and other reports describe up to 78 % of patients presenting with solitary disease in re-operative parathyroid surgery . In another re-operative review of 237 patients adenomas were found in ectopic locations in 32 % of cases and were most frequently in the thymus .
Typical locations for ectopic parathyroid tissue have been described. Ectopic inferior parathyroid glands can vary in position due to abnormal descent from the third pharyngeal pouch during embryological development. Most commonly, ectopic inferior parathyroid glands are intrathymic, but they can also be mediastinal, and intrathyroidal, within the thyrothymic ligament, and in the submandibular triangle. Superior parathyroid glands are less variable in position. During embryologic migration the superior parathyroids migrate with the thyroid gland along a relatively short path arising from the fourth branchial pouch. Ectopic superior glands can be located in the tracheoesophageal groove, retroesophageal space, posterosuperior mediastinum, thyroid gland, carotid sheath, and paraesophageal . The location of missed parathyroid glands was defined in a recent review. In a re-operative parathyroidectomy cases 5 (10 %) were adherent to the posterior thyroid capsule, 11 (22 %) were behind the thyroid in the tracheoesophageal groove, 7 (14 %) were close to the clavicle in the prevertebral space, 3 (6 %) were directly over the recurrent laryngeal nerve, 9 (18 %) were easy to identify near the inferior thyroid pole, 13 (26 %) had fallen into the thymus, and 2 (4 %) were within the thyroid gland .
In a review of the American College of Surgeons national surgical quality improvement program database (2008–2011), the annual rate of re-operation was 3.6–4.8 %. Patients undergoing re-operative parathyroid surgery were identified to more likely be obese, have a longer operative time, have a longer postoperative stay, and were more likely to be readmitted in 30 days .
Operating in a neck with scar tissue and altered tissue planes can be challenging and carries a higher risk of complications. Due to the increased likelihood of a prolonged procedure and increased difficulty of the dissection, the operation is generally performed under general anesthesia. The surgical approach for re-operative parathyroid surgery should be designed to avoid dissecting through scar tissue if possible and tailored to the preoperative localization studies. Since most initial parathyroidectomies are performed through a midline approach mobilizing the strap muscles laterally and retracting the thyroid medially, re-explorations may need to be performed through a different approach. Fighting through scar tissue can result in bleeding from the surface of the thyroid and obscure the surgical field, as well as surgeon fatigue. Unilateral approach or bilateral neck explorations are considered for re-operation depending on the surgeon’s confidence in the preoperative localization studies. If a prior midline approach was used, re-operation through the prior incision can be performed with a modified lateral approach by retracting the strap muscles and thyroid medially to avoid bleeding from peeling the straps off the thyroid. A midline incision facilitates a bilateral neck exploration if needed. In the lateral approach the patient’s head is turned contralateral to the adenoma side and dissection is carried out lateral to the strap muscles and medial to the carotid sheath. The recurrent laryngeal nerve can be identified more easily in virgin tissue using this approach. The vagus nerve can be easily identified and stimulated in the carotid sheath through this approach. Stimulation of the vagus can confirm that the nerve monitoring system is functioning and can be relied upon to help locate the recurrent laryngeal nerve. The anatomic and electrophysiologic algorithm for identification of the nonrecurrent laryngeal nerve entailing distal and proximal vagal nerve stimulation has been defined by Kamani et al. .