Pediatric Endocrine Surgery: A 20-Year Experience at the Mayo Clinic
A. Kundel, G. B. Thompson, M. L. Richards, L. X. Qiu, Y. Cai, F. W. Schwenk, A. N. Lteif, S. T. Pittock, S. Kumar, P. J. Tebben, I. D. Hay, and C. S. Grant Mayo Clinic, Rochester, Minnesota 55905
Context: Surgically managed endocrinopathies are rare in children. Most surgeons have limited experience in this field. Herein we report our operative experience with pediatric patients, per- formed over two decades by high-volume endocrine surgeons.
Setting: The study was conducted at the Mayo Clinic (a tertiary referral center).
Patients: Patients were <19 years old and underwent an endocrine operation (1993-2012).
Main Outcome Measures: Demographics, surgical procedure, diagnoses, morbidity, and mortality were retrospectively reviewed.
Results: A total of 241 primary cases included 177 thyroid procedures, 13 neck dissections, 24 para- thyroidectomies, 14 adrenalectomies, 7 paragangliomas, and 6 pancreatic procedures. Average age of patients was 14.2 years. There were 133 total thyroidectomies and 40 hemithyroidectomies. Fifty-three cases underwent a central or lateral neck dissection. Six-month follow-up was available for 98 total thyroidectomy patients. There were four cases of permanent hypoparathyroidism (4%) and no per- manent recurrent laryngeal nerve (RLN) paralyses. Sequelae of neck dissections included temporary RLN neurapraxia and Horner’s syndrome. Parathyroidectomy was performed on 24 patients: 20 with primary hyperparathyroidism (HPT), three with tertiary HPT, and one with familial hypocalciuric hy- pocalcemia. Three patients (16%) had recurrent HPT, all with multiglandular disease. One patient had temporary RLN neurapraxia. We performed seven bilateral and seven unilateral adrenalectomies; eight were laparoscopic. Indications included pheochromocytoma, Cushing’s syndrome, adrenocorti- cal carcinoma, congenital adrenal hyperplasia, and ganglioneuroma. One death was due to adreno- cortical carcinoma. Five paraganglioma patients had succinate dehydrogenase subunit B mutations, and one recurred. Six patients with insulinoma underwent enucleation (n = 5) or distal pancreatectomy (n = 1). A single postoperative abscess was managed nonoperatively.
Conclusion: Pediatric endocrine procedures are uncommon but can be safely performed with complication rates comparable to those of the adult population. It is imperative that these oper- ations be performed by high-volume surgeons. (J Clin Endocrinol Metab 99: 399-406, 2014)
E ndocrine neoplasms are a rarity in childhood, with an incidence of 1000 to 2000 per 100 000 for thyroid nodules, 2 to 5 per 100 000 for hyperparathyroidism (HPT), 2 per 10 000 000 for adrenocortical tumors, and even less frequent for islet cell neoplasms (1-4). Because children and adolescents represent a healthy population, disease is rarely discovered as an incidental finding. The
usual presentation is with disease-specific symptomatol- ogy and a more advanced stage as compared to adults (3, 5, 6). Thus, initial surgical management may be signifi- cantly more extensive and may carry with it an increasing risk of complications when performed by a less experi- enced surgeon. Pediatric patients are not the same as small adult patients, and standards must be set in order to avoid
Abbreviations: FHH, familial hypocalciuric hypercalcemia; FNA, fine-needle aspiration; HPT, hyperparathyroidism; LN, lymph node; MEN, multiple endocrine neoplasia; MTC, medul- lary thyroid cancer; PPNAD, primary pigmented nodular adrenal hyperplasia; PTC, papillary thyroid carcinoma; RLN, recurrent laryngeal nerve; US, ultrasound; VHL, Von Hippel-Lindau.
devastating consequences. Several recent studies reported superior outcomes when endocrine surgery is performed at a specialty center (7, 8). Herein we report the Mayo Clinic experience with pediatric and adolescent endocrine surgical disorders over two decades. Our hypothesis is that the rate of postoperative complications in pediatric pa- tients can be comparable to those of adults when per- formed by high-volume surgeons.
Patients and Methods
With the approval of the institutional review board, the records of all patients 18 years old and younger who underwent a pro- cedure of the thyroid, parathyroids, pancreas, adrenal, or para- ganglia identified in the Mayo Clinic Rochester surgical database between January 1993 and June 2012 were retrospectively re- viewed. All operations were performed by one of three endocrine surgeons on staff.
HPT patients treated at the Mayo Clinic between 1993 and 2003 had PTH levels reported in picomoles per liter. A double- antibody capture chemiluminometric assay was used. From 2003 until the present, a change in the method to a double- antibody capture using the Roche Elecsys assay reported results as picograms per milliliter. Therefore, PTH results initially re- ported as picomoles per liter have been converted to picograms per milliliter (pmol/L/0.1061). All other laboratory units were consistent over the two decades. Hypocalcemia was defined by calcium levels below the normal values in our laboratory with or without the presence of symptoms. Hypocalcemia was consid- ered transient if it occurred for less than 6 months.
Demographics, initial symptomatology, surgical procedure, pathological diagnoses, family history and genetic composition, recurrence, morbidity, and mortality were assessed.
Results
Weidentified 241 patients that met the inclusion criteria. The overwhelming majority (n = 177) of patients underwent some type of thyroid resection. Fourteen patients had adre- nalectomies, 7 underwent resection of a paraganglioma, 24 underwent parathyroid resection, 6 had a neuroendocrine pancreatic procedure, and 13 underwent some form of neck dissection. A patient was placed into the “neck dissection” group if it was the primary surgery performed at Mayo Clinic, whereas patients who underwent a thyroidectomy and subsequently had a neck dissection were placed into the “thyroid” group. Eighty percent of cases were performed by one surgeon (G.B.T.). Follow-up was on average 62 months and was obtained via medical records.
Thyroid disease
Patient demographics/indication
We performed 177 thyroid procedures on patients un- der the age of 19 years over two decades. The average age
was 14 years (range, 2-18). We observed a 3.2:1 female to male ratio. Most common indications for surgical inter- vention were a thyroid nodule (39%) and a diagnosis of hyperthyroidism (40%). Two patients had initial hemi- thyroidectomy at an outside institution and underwent a completion thyroidectomy at Mayo Clinic for follicular and Hürthle cell thyroid carcinoma, respectively. All sus- picious thyroid nodules detected on initial assessment were preoperatively evaluated with fine-needle aspiration (FNA). Indications for surgery are listed in Table 1. One patient with a multinodular goiter had an FNA of the dominant nodule that showed a Hürthle cell neoplasm; hence, this patient is seen in both groups. Twenty-six of 177 patients who underwent a thyroid procedure were under age 10, and four patients were less than 5 years of age. The two most common indications were Graves’ dis- ease and multiple endocrine neoplasia (MEN) 2A.
Surgery
Our surgical numbers reflect only the primary opera- tion performed at Mayo Clinic. If a second operation was necessary, it was considered a reoperation and not counted in the initial set. As noted in Table 2, the vast majority of procedures were near-total thyroidectomies. All subtotal thyroidectomies were performed for either Graves’ disease or multinodular goiter. In all cases, the specimens were intraoperatively evaluated using the Mayo frozen section protocol. Completion thyroidectomy was never required because of this approach. One patient
| Indication | Females | Males | Total |
|---|---|---|---|
| Nodule | 57 | 14 | 70 (39%) |
| PTC | 27 | 5 | 32 |
| Suspicious for PTC | 2 | 1 | 3 |
| Suspicious for follicular | 4 | 3 | 7 |
| neoplasm | |||
| Follicular adenoma | 3 | 1 | 4 |
| Hürthle cell adenoma | 3 | 0 | 3 |
| Indeterminant FNA | 3 | 1 | 4 |
| Large benign nodule | 13 | 3 | 16 |
| MTC | 1 | 0 | 1 |
| Hyperthyroidism | 58 | 13 | 71 (40%) |
| Toxic MNG | 10 | 1 | 11 |
| Graves' disease | 48 | 12 | 60 |
| Goiter-MNG | 8 | 3 | 11 (6%) |
| Familial-prophylactic for | 8 | 13 | 21 (12%) |
| MEN2A | |||
| Other | 5 | 0 | 5 (3%) |
| Completion for FTC | 1 | 0 | 1 |
| Completion for HTC | 1 | 0 | 1 |
| Hashimoto's thyroiditis | 3 | 0 | 3 |
| Total | 135 | 42 | 177 |
Abbreviations: FTC, follicular thyroid carcinoma; HTC, Hürthle cell thyroid carcinoma; MNG, multinodular goiter. Boldface data represent the totals for each subgroup.
| Surgery Performed | n (%) |
|---|---|
| Thyroid | |
| Bilateral | 133 (75) |
| Total/near-total thyroidectomy | 128 (72) |
| Subtotal thyroidectomy | 5 (2.8) |
| Unilateral | 44 (25) |
| Lobectomy | 40 (23) |
| Nodulectomy | 3 (1.7) |
| Isthmusectomy | 1 (0.5) |
| Lymphadenectomy | |
| Central compartment | 40 |
| Unilateral lateral neck dissection | 10 |
| Bilateral lateral neck dissection | 3 |
| Other | |
| Nerve graft (ansa cervicalis) | 2 |
| Tracheal repair | 1 |
with extensive papillary thyroid carcinoma (PTC) with metastasis to bilateral lateral neck compartments also had invasion of the larynx and recurrent laryngeal nerve (RLN) requiring intentional nerve sacrifice and ansa cer- vicalis nerve grafting. Another patient with similarly ex- tensive disease also required nerve grafting. Ninety-three percent of patients with a diagnosis of PTC underwent either a prophylactic or therapeutic central compartment lymphadenectomy. All lateral neck dissections were per- formed on patients with PTC, and all had either clinically palpable or enlarged lymph nodes (LNs) on ultrasound (US), with involvement confirmed by FNA. Six patients with hereditary medullary thyroid cancer (MTC) under- went prophylactic thyroidectomy at a young age and thus pathologically had a benign thyroid gland. The final his- tological diagnoses are listed in Table 3.
Genetics
Twenty-two patients had hereditary MTC, 18 with MEN2A and four with familial MTC. Two patients had a preoperative diagnosis of MTC; the rest underwent pro- phylactic thyroidectomy. No case was discovered de novo. The codons associated with MEN2A diagnosis were avail-
| Histology | n |
|---|---|
| Thyroid cancer | 47 |
| PTC | 40 |
| MTC | 7 |
| Benign disease | 130 |
| Follicular adenomas | 39 |
| Hürthle cell adenomas | 6 |
| C-cell hyperplasia | 9 |
| Hereditary MTC, benign | 6 |
| Colloid nodule | 1 |
| Hyperplasia/thyroiditis | 69 |
able for 18 patients: seven cases with 618 codon identified; three cases of codons 631, 634, and 666; and one case each of codons 533 and 804.
Complications/follow-up
The average follow-up for this group was 41.7 months (range, <1 to 181). Average follow-up for patients with a diagnosis of cancer was 50.5 months (range, 1-187), with one lost to follow-up-a patient who underwent a prophy- lactic thyroidectomy for MEN2A who was found to have a micro-MTC on pathology. Out of 133 total thyroidecto- mies, 98 patients had ≥ 6 months of follow-up. Of the 58 patients with transient hypoparathyroidism, four patients underwent hemithyroidectomy, and 50 underwent total thy- roidectomy (14 patients > 6-mo follow-up); however, all resolved by the last follow-up date. During this time period, there were five recurrences requiring reoperation, all with a diagnosis of PTC. All recurrences were in the lateral neck, and one had an additional LN in the central compartment. Four occurred within 4 years and one at 10 years from pri- mary surgery. Three of these patients underwent resection of their metastatic disease at Mayo Clinic. Subsequent fol- low-up indicated no persistent disease. Two patients under- went surgery elsewhere. We reviewed our rates of temporary and permanent hypocalcemia and RLN injury (see Table 4). There was one unrelated accidental death in a patient with Graves’ disease. Of 26 patients younger than 10 years old, nine had transient hypocalcemia (34%).
Neck dissection
Patient demographics/indication
All but two of these patients underwent primary thy- roid resection elsewhere and presented to Mayo Clinic with recurrent or persistent disease that required re-exci- sion. There were 13 patients, with a 1:2.3 female to male ratio. The average age of these patients was 15.1 years (range, 5-18), and only one child was younger than 10
| Complication | n | % of total (n = 177) |
|---|---|---|
| Hypocalcemia | ||
| Transient | 58 | 32.7 |
| Permanent | 4 | 2.3 |
| RLN paralysis | ||
| Transient | 4 | 2.3 |
| Permanent | 0 | 0 |
| Other | ||
| Horner's syndrome, transient | 1 | 0.6 |
| Hematoma | 1 | 0.6 |
| Drain stitch abscess | 1 | 0.6 |
| Total permanent complications | 4 | 2.3 |
| Symptoms | n (%) |
|---|---|
| Major symptoms | 6 (30) |
| Kidney stones | 5 (25) |
| Pancreatitis | 1 (5) |
| Nonspecific symptoms | 13 (65) |
| Neurological symptoms (fatigue, depression, | 9 (30) |
| panic attacks, skipped heartbeat, headache) Gastrointestinal symptoms (nausea, vomiting, constipation, abdominal pain) | |
| 4 (20) | |
| Asymptomatic | 3 (15) |
| Total | 20 (100) |
years. Indications for surgery included 12 patients with recurrent or persistent PTC in the lateral neck nodes and one patient with recurrent MTC in the lateral neck.
Surgery
The surgical procedures included left, right, or bilateral neck LN dissection with or without central compartment dissection. All neck compartments were preoperatively evaluated with US, and worrisome nodes were biopsied to confirm the presence of disease. Resection of lateral lym- phatics was achieved by performing a level III, IV, and anterior V lymphadenectomy. Level II was included if US showed nodal involvement.
Genetics
One patient with an extensive family history had a ge- netic mutation in the Ret proto-oncogene (codon 634), consistent with the diagnosis of MEN2A.
Complications/follow-up
Mean follow-up was 73.8 months (range, 6-160). One patient with PTC had metastasis to the lung at presenta- tion. One patient required a completion left modified rad- ical neck dissection after previously undergoing a selective level IV neck dissection at Mayo Clinic. There was one RLN injury and one case of Horner’s syndrome; both sub-
sequently resolved. There was one death unrelated to sur- gical management.
Hyperparathyroidism
Patient demographics/indication
We identified 24 patients 18 years of age and younger with HPT. Average age was 15.4 years (range, 10-18) with a 1:1.1 female to male ratio. Primary HPT was the cause of hypercalcemia in 20 patients; 16 children had single gland disease, and four had multigland disease. Three patients had a diagnosis of tertiary HPT: one after renal transplant, one with hypophosphatemic rickets, and one with severe renal osteodystrophy. Another patient, in retrospect, had symptomatic (pancreatitis) familial hy- pocalciuric hypercalcemia (FHH). The most common clinical finding was nephrolithiasis. The most common overall symptom was fatigue (Table 5).
Asymptomatic patients at presentation were identified by screening for familial disease. One FHH patient ini- tially presented with several severe bouts of pancreatitis and was noted to have high serum plasma calcium levels. Table 6 lists the relevant biochemical data collected from the 24 patients. Eighteen patients had recorded 24-hour urine calcium levels.
All but one patient had imaging studies that included US, sestamibi, or both. Several patients had additional studies (computed tomography, magnetic resonance im- aging) that did not alter management.
Surgery
Out of 20 primary HPT patients, 16 had single gland disease, with one of them having a supernumerary in- trathymic gland requiring median sternotomy at reopera- tion. Those patients with multigland disease all had fa- milial syndromes (one familial HPT, two MEN1, one MEN2A). From the tertiary HPT patients, two had mul- tigland disease as a consequence of end-stage renal failure after renal transplant and severe renal osteodystrophy,
| Biochemical Value | No. of Patients | Mean | Range | Reference Range |
|---|---|---|---|---|
| All hyperparathyroid patients | 24 | |||
| Serum calcium, mg/dL | 24 | 12.3 | 10.4-16.2 | 8.5-10.5 |
| Serum phosphorus, mg/dl | 24 | 3.5 | 1.2-9.5 | 3.1-4.7 |
| PTH, pg/mL | 24 | 226.3 | 43.3-1629 | 15-65 |
| Urinary calcium, mg/24 h | 18 | 453.5 | 45.6-1061 | NA |
| Primary hyperparathyroid patients | 20 | |||
| Serum calcium, mg/dl | 20 | 12.4 | 10.4-16.2 | 8.5-10.5 |
| Serum phosphorus, mg/dL | 20 | 3.1 | 1.8-4.7 | 3.1-4.7 |
| PTH, pg/mL | 20 | 148.1 | 43.3-448 | 15-65 |
| Urinary calcium, mg/24 h | 16 | 490.9 | 45.6-1061 | NA |
Abbreviation: NA, not available.
respectively. One patient with familial hypophosphatemic rickets, surprisingly, had only a single gland disease. The patient with FHH presented with multiple bouts of pan- creatitis that resolved after resection of 31/2 glands; all were normal on pathological evaluation. Subcutaneous auto- transplantation to the anterior chest wall was routinely carried out in subtotal parathyroidectomies. Tissue was removed the following day at bedside if PTH levels were in the normal range or elevated.
Genetics
Nine of 20 patients with primary HPT had a known family history. Five patients had familial primary HPT. The FHH patient had a family history of FHH as well. Two were MEN1 patients, and one belonged to an MEN2A kindred.
Complications/follow-up
Of the entire group undergoing parathyroid surgery (n = 24), four required reintervention. The first patient underwent sternotomy for an intrathymic, supernumerary gland after a negative four-gland exploration. A repeat ses- tamibi scan identified the mediastinal gland that was missed on the initial studies. The second patient had resection of three glands for MEN1 HPT with reimplantation. The fourth, inferior gland, was not identified. Six years later, the patient presented with recurrent graft-dependent primary HPT requiring removal. The third patient with familial HPT required resection of a second gland 8 months subsequent to his initial operation. The final patient requiring reinterven- tion was a male with familial HPT after 31/2 gland resection. He presented with remnant recurrence and hypercalcemia. He underwent multiple ethanol ablative procedures with successful results. The patient with FHH has not had any further bouts of pancreatitis since surgery but remains with a persistently elevated calcium level, consistent with her known syndrome.
One patient had a temporary RLN palsy after single gland resection (4%). No incidence of permanent RLN nerve in- jury or hypocalcemia was noted. Follow-up for this group averaged 54.4 months (range, <1 to 188), with eight patients followed for <6 months; however, all cases of hypoparathy- roidism resolved by the last evaluation date.
Adrenal tumors/paragangliomas
Patient demographics/indication
A total of 14 patients underwent adrenalectomies at an average age of 13.6 years (range, 2-18) and a 1:1.8 female to male ratio. Six patients underwent adrenalectomy for pheochromocytoma. The most common presenting symp- toms of the remaining eight children were weight gain (62.5%), acne (62.5%), hirsutism (50%), and thinning of
the skin (50%). Five children underwent adrenalectomy for Cushing’s syndrome: three diagnoses of primary pig- mented nodular adrenal hyperplasia (PPNAD), one adre- nocortical carcinoma, and one with intractable Cushing’s disease. The other patients included one girl with a T-se- creting adrenocortical carcinoma who presented with pre- cocious adrenarche, one child with 21-hydroxylase defi- ciency presenting with weight loss and lethargy, and one girl with an incidentally discovered adrenal ganglioneu- roma. Three children ages 2, 4, and 8 underwent unilateral adrenalectomy for adrenocortical cancer and bilateral ad- renalectomies for Von Hippel-Lindau (VHL) syndrome and PPNAD, respectively.
Seven patients with an average age of 15 years (range, 12-18) underwent paraganglioma resection. The female to male ratio was 1:1.3. The majority of tumors were in the para-aortic/paracaval location. The most common signs and symptoms were hypertension (69%), excessive sweating (54%), palpitations (54%), and heat intolerance (23%). Pal- pitations were more pronounced in the paraganglioma group, whereas heat intolerance was noted solely in the pheo- chromocytoma group, present in 50% of the patients.
Surgery
All 14 children underwent surgical resection. Bilateral adrenalectomies were performed for seven of the patients. Two were performed for patients with VHL syndrome and bilateral pheochromocytomas via a transabdominal open approach due to large size and multiple nodules. The re- maining five patients underwent bilateral transabdominal laparoscopic adrenalectomy (one Cushing’s disease, one congenital adrenal hyperplasia, and three PPNAD). Of the seven unilateral adrenalectomies, four were performed via a transabdominal open approach for a 6.5-cm pheochro- mocytoma, one for combined adrenal and pancreatic re- section, and two adrenocortical cancers.
All paraganglioma resections were performed via an open approach.
Genetics
None of the children who underwent an adrenalectomy had a known family history or a genetic mutation. How- ever, five were found to have genetic alterations: two chil- dren were diagnosed with VHL; two had a diagnosis of Carney complex; one child was diagnosed with MEN2B; and there was one patient with PPNAD in the absence of the other features of Carney complex.
Five of the patients with paragangliomas were found to have a mutation in the succinate dehydrogenase subunit B protein; of these, three had a positive family history. Two patients had a suspected family history but have not yet been specifically tested.
Complications/follow-up
Two patients required reoperation. One child had an excision of recurrent adrenocortical carcinoma within the adrenal fossa and adjuvant therapy with mitotane, expir- ing shortly thereafter from disease progression. The sec- ond was a patient with MEN2B who had to undergo com- pletion adrenalectomy for a pheochromocytoma in the contralateral gland. Mean follow-up for this group was 90.4 months (range, 6-178).
Two patients with paraganglioma had metastases to LNs. One of these patients had additional liver metastases and subsequently had recurrence of his tumor at the previous site necessitating re-excision. No deaths were noted in this group, with an average follow-up of 60.9 months (range, 1-131).
Pancreas tumors
Patient demographics/indication
Six pancreatic procedures were performed. All patients had an insulinoma, diagnosed by presenting symptom- atology and standard biochemical testing. Average age was 15.8 years (range, 12-18).
Surgery
Enucleation was performed for five of the six insulino- mas (83%), with locations scattered throughout the pan- creas: two in the body, one in the head, one in the uncinate process, one at the junction of the head and uncinate pro- cess. Distal pancreatectomy was performed for the sixth patient with insulinoma because no discreet mass was identified on preoperative or intraoperative imaging; however, a preoperative calcium stimulation test indi- cated elevated insulin levels in the tail of the pancreas. Pathology confirmed the presence of islet cell hyperplasia.
Genetics
No genetic mutations or familial predispositions were identified in this group.
Complications/follow-up
There were no recurrences, reoperations, or death, ob- served in this group. One patient who underwent distal pancreatectomy for insulinoma had a small postoperative intraabdominal abscess treated conservatively with anti- biotics. Average follow-up for these patients was 7.5 months (range, <1 to 25). None have developed features concerning for the MEN1 syndrome.
Discussion
Multiple studies in recent years have looked at the correlation between surgeon experience and patient outcomes (9-14). A
common theme prevails: high-volume surgeons have lower operative mortality and morbidity rates. Birkmeyer et al (9) evaluated patient mortality among widely performed car- diovascular and cancer procedures, noting a 57% improve- ment in mortality attributable to high operating volume. This improvement varied with the type of procedure. If the pro- cedure required more precise technical intraoperative skills, for example, an aortic valve replacement, the difference in morbidity between high- and low-volume surgeons was sub- stantially greater than for lower complexity procedures such as a lung resection (9). A study evaluating the relationship between surgeon experience and outcomes for thyroidec- tomy established a direct link between surgeon volume (>100 thyroidectomies in 5 y) and operative complications with no correlation to hospital volume (14). Similarly, Stavrakis et al (10) found an inverse correlation between surgeon volume and complication rates in patients undergo- ing endocrine surgery (thyroid, parathyroid, adrenal). En- docrine surgery requires significant surgical skills; for exam- ple, precise identification of RLN and parathyroids in thyroidectomy. Most endocrine surgery patients are healthy overall and do not require extensive perioperative in-hospital support, thus supporting the fact that patient outcomes are more directly related to the skill and experience of the oper- ating endocrine surgeon.
With similar results, this relationship has been studied in pediatric populations. Higher surgeon volume has been linked to fewer complications after pyloromyotomy and ap- pendectomy in children (15-17). Surgeons that fell within the “high-volume” bracket for pyloromyotomies, for example, had a 90% lower risk of complications as compared to low- volume surgeons (16). The conclusion of Langer et al (17) echoed the improved surgical outcomes when high-volume surgeons performed the procedure. They noted a 4-fold in- crease in the risk of duodenal mucosal perforations during a pyloromyotomy among general surgeons as compared to pe- diatric surgeons; however, after adjustment for volume, this did not reach statistical significance (17). Alexander at el (15) compared morbidity rates after 175 pediatric appendecto- mies between general surgeons and pediatric surgeons with expertise in more complex pediatric cases. They concluded that children with gangrenous or perforated appendicitis had complication rates of 8% when operated by pediatric sur- geons vs 33% incurred by the group operated by general surgeons. Equally divergent were the outcomes between gen- eral and pediatric surgeons when performing inguinal her- niorrhaphy in a pediatric population; however, high-volume general surgeons had similar complication rates to those of pediatric surgeons (18). These reports underline the close association of outcomes to surgical volume and experience.
Analogous findings emerged from evaluating parathy- roidectomy and thyroidectomy outcomes in this young
population (8, 19), with rates of endocrine-related com- plications at 5.6% (RLN injury, voice disturbance, hypo- parathyroidism, hypocalcemia, and tetany) for high-vol- ume vs 10% for low-volume surgeons (20). Regardless of the surgeon’s subspecialty (general vs pediatric vs head and neck), volume was the unique variable that correlated to superior outcomes (20). Sosa et al (19) likewise iden- tified surgeon volume as an independent predictor of out- come in pediatric thyroid and parathyroid procedures.
Herein we report outcomes of pediatric endocrine sur- gical procedures involving disorders of the thyroid, para- thyroid, adrenal, paraganglia, and pancreas at a tertiary medical center over a period of two decades performed by high-volume (endocrine) surgeons.
Thyroid/neck dissection
Several older studies described a substantially higher risk of postoperative complications in pediatric patients as compared to adults undergoing thyroidectomy (21, 22). Although this has been attributed to the intrinsic higher operative risk in the pediatric population, the more likely explanations include volume and routine parathyroid au- totransplantation. More recent literature suggests com- plication rates of permanent hypoparathyroidism in 1.4-4% (23-25). Our 2.3% rate of permanent hypopara- thyroidism clearly falls within this range. The commonly quoted rate of permanent RLN injury is under 1% (26). We report no unintentional permanent RLN injuries. One patient with extensive bilateral central compartment disease required a temporary (2 wk) tracheostomy due to intentional sacrifice of one RLN and transient dysfunction of the contralateral nerve. Transient RLN paralysis was noted in 2.3% of our thyroidectomy cohort. This is com- parable to the rate of 2% quoted by Rosato et al (27) in the adult population. One patient in the “neck dissection” group who underwent a reoperation for central compart- ment node clearance after thyroidectomy elsewhere had transient hoarseness. Other complications included a transient Horner’s syndrome and one cervical hematoma. All of these complication rates are comparable to those in the adult population published by Grant et al (28) from our institution that demonstrate a 1.2% incidence of per- manent hypoparathyroidism, 0.2% unintentional RLN paralysis, and 0.5% incidence of cervical hematoma.
Parathyroid
HPT in children often presents in a more aggressive manner as compared to adults. Most adults are diagnosed with an incidental finding on routine laboratory evalua- tions, whereas children are more often symptomatic, prompting assessment of calcium levels (29). Similar to the findings by Kollars et al (3), only 15% of children were
asymptomatic and were identified due to known familial syndromes; fatigue was the most prevalent symptom in both series. Sixteen percent (n = 4) of children required reintervention; three were attributed to familial multig- landular disease, and the fourth was due to a supernumer- ary intrathymic gland. One patient (4%) had temporary RLN injury, and none had permanent RLN injury or hy- pocalcemia. These results are comparable to those re- ported by van Heerden et al (30) from our institution in adults with 0.3% incidence of permanent hypocalcemia and 0.8% incidence of unilateral vocal cord paralysis. Both the adult and pediatric populations underwent sur- gery by high-volume surgeons yielding low morbidity rates. Similar to the thyroid data, complications from parathyroid surgery relate to surgeon volume, with supe- rior outcomes observed among surgeons performing > 70-100 endocrine cases per year (10, 20).
Adrenal/paraganglioma/pancreas
None of the patients who underwent resection of ad- renal, paraganglioma, or enucleation for insulinoma had an operative complication. Out of this entire cohort, only three required reoperation, all attributed to the natural history of their disease. The complication rates in patients undergoing intra-abdominal procedures are likely related to the experience of the operative surgeon.
Conclusion
During the course of 20 years, the (endocrine) surgeons at our institution performed approximately 14 000 endo- crine procedures (~700 per year); only 241 of these were carried out in pediatric patients. As several studies have indicated, surgeon volume is inversely related to the rate of complications when operating on adult and pediatric pop- ulations. When comparing our adult outcomes to those of the pediatric population, the complication rates are low and quite comparable. This is especially true for neck pro- cedures. The operative consequences are especially signif- icant in the pediatric population and can bear devastating long-term sequelae. Working alongside pediatric endocri- nologists, pediatric hospitalists, and general surgery resi- dents experienced in the care of pediatric surgery patients, we believe that high-volume (endocrine) surgeons provide the best possible surgical management for pediatric endo- crine surgical patients.
Acknowledgments
Address all correspondence and requests for reprints to: Anna Kundel, Department of Gastroenterologic and General Surgery, Mayo Clinic, 200 First Street SW, Rochester MN, 55905. E- mail: anna.kundel@gmail.com.
No external funding was secured for this study. Funds were allocated by the Department of Surgery at Mayo Clinic. All au- thors have no financial relationships to disclose.
Disclosure Summary: All authors have no conflict of interest.
References
1. Josefson J, Zimmerman D. Thyroid nodules and cancers in children. Pediatr Endocrinol Rev. 2008;6:14-23.
2. Bernstein L, Gurney JG. Cancer Incidence and Survival Among Chil- dren and Adolescents: United States SEER Program 1975-1995. Bethesda, MD: National Cancer Institute; 1999:139-147.
3. Kollars J, Zarroug AE, van Heerden J, et al. Primary hyperparathy- roidism in pediatric patients. Pediatrics. 2005;115:974-980.
4. Yao JC, Eisner MP, Leary C, et al. Population-based study of islet cell carcinoma. Ann Surg Oncol. 2007;14:3492-3500.
5. Harman CR, van Heerden JA, Farley DR, Grant CS, Thompson GB, Curlee K. Sporadic primary hyperparathyroidism in young patients: a separate disease entity? Arch Surg. 1999;134:651-655; discussion 655-656.
6. Michalkiewicz E, Sandrini R, Figueiredo B, et al. Clinical and out- come characteristics of children with adrenocortical tumors: a re- port from the International Pediatric Adrenocortical Tumor Regis- try. J Clin Oncol. 2004;22:838-845.
7. Mittendorf EA, McHenry CR. Complications and sequelae of thy- roidectomy and an analysis of surgeon experience and outcome. Surg Technol Int. 2004;12:152-157.
8. Wang TS, Roman SA, Sosa JA. Predictors of outcomes following pediatric thyroid and parathyroid surgery. Curr Opin Oncol. 2009; 21:23-28.
9. Birkmeyer JD, Stukel TA, Siewers AE, Goodney PP, Wennberg DE, Lucas FL. Surgeon volume and operative mortality in the United States. N Engl J Med. 2003;349:2117-2127.
10. Stavrakis AI, Ituarte PH, Ko CY, Yeh MW. Surgeon volume as a predictor of outcomes in inpatient and outpatient endocrine surgery. Surgery. 2007;142:887-899; discussion 887-899.
11. Schrag D, Panageas KS, Riedel E, et al. Hospital and surgeon pro- cedure volume as predictors of outcome following rectal cancer re- section. Ann Surg. 2002;236:583-592.
12. Begg CB, Riedel ER, Bach PB, et al. Variations in morbidity after radical prostatectomy. N Engl J Med. 2002;346:1138-1144.
13. Hillner BE, Smith TJ, Desch CE. Hospital and physician volume or specialization and outcomes in cancer treatment: importance in quality of cancer care. J Clin Oncol. 2000;18:2327-2340.
14. Sosa JA, Bowman HM, Tielsch JM, Powe NR, Gordon TA, Udels- man R. The importance of surgeon experience for clinical and eco- nomic outcomes from thyroidectomy. Ann Surg. 1998;228:320- 330.
15. Alexander F, Magnuson D, DiFiore J, Jirousek K, Secic M. Specialty versus generalist care of children with appendicitis: an outcome comparison. J Pediatr Surg. 2001;36:1510-1513.
16. Ly DP, Liao JG, Burd RS. Effect of surgeon and hospital character- istics on outcome after pyloromyotomy. Arch Surg. 2005;140: 1191-1197.
17. Langer JC, To T. Does pediatric surgical specialty training affect outcome after Ramstedt pyloromyotomy? A population-based study. Pediatrics. 2004;113:1342-1347.
18. Borenstein SH, To T, Wajja A, Langer JC. Effect of subspecialty training and volume on outcome after pediatric inguinal hernia re- pair. J Pediatr Surg. 2005;40:75-80.
19. Sosa JA, Tuggle CT, Wang TS, et al. Clinical and economic out- comes of thyroid and parathyroid surgery in children. J Clin Endo- crinol Metab. 2008;93:3058-3065.
20. Tuggle CT, Roman SA, Wang TS, et al. Pediatric endocrine surgery: who is operating on our children? Surgery. 2008;144:869-877; discussion 877.
21. Newman KD, Black T, Heller G, et al. Differentiated thyroid cancer: determinants of disease progression in patients <21 years of age at diagnosis: a report from the Surgical Discipline Committee of the Children’s Cancer Group. Ann Surg. 1998;227:533-541.
22. La Quaglia MP, Corbally MT, Heller G, Exelby PR, Brennan MF. Recurrence and morbidity in differentiated thyroid carcinoma in children. Surgery. 1988;104:1149-1156.
23. Palmer BA, Zarroug AE, Poley RN, Kollars JP, Moir CR. Papillary thyroid carcinoma in children: risk factors and complications of disease recurrence. J Pediatr Surg. 2005;40:1284-1288.
24. Popovtzer A, Shpitzer T, Bahar G, Feinmesser R, Segal K. Thyroid cancer in children: management and outcome experience of a refer- ral center. Otolaryngol Head Neck Surg. 2006;135:581-584.
25. Bargren AE, Meyer-Rochow GY, Delbridge LW, Sidhu SB, Chen H. Outcomes of surgically managed pediatric thyroid cancer. J Surg Res. 2009;156:70-73.
26. Reeve T, Thompson NW. Complications of thyroid surgery: how to avoid them, how to manage them, and observations on their possible effect on the whole patient. World J Surg. 2000;24:971-975.
27. Rosato L, Avenia N, Bernante P, et al. Complications of thyroid surgery: analysis of a multicentric study on 14,934 patients operated on in Italy over 5 years. World J Surg. 2004;28:271-276.
28. Grant CS, Stulak JM, Thompson GB, Richards ML, Reading CC, Hay ID. Risks and adequacy of an optimized surgical approach to the primary surgical management of papillary thyroid carcinoma treated during 1999-2006. World J Surg. 2010;34:1239-1246.
29. Roizen J, Levine MA. Primary hyperparathyroidism in children and adolescents. J Chin Med Assoc. 2012;75:425-434.
30. van Heerden JA, Grant CS. Surgical treatment of primary hyper- parathyroidism: an institutional perspective. World J Surg. 1991; 15:688-692.
ENDOCRINE SOCIETY
Register NOW for ICE/ENDO 2014 June 21-24, 2014, Chicago, Illinois www.ice-endo2014.org