World Journal of Surgery

, Volume 31, Issue 3, pp 593–598 | Cite as

Recurrence after Total Thyroidectomy for Benign Multinodular Goiter

  • Kylie L. Snook
  • Peter L. H. Stalberg
  • Stan B. Sidhu
  • Mark S. Sywak
  • Pamela Edhouse
  • Leigh Delbridge



Total thyroidectomy is now the preferred option for the management of benign multinodular goiter (BMNG), and it ought not be associated with recurrent disease. The aim of the present study was to examine the efficacy of total thyroidectomy for BMNG and to review reasons for recurrence.

Material and methods

The study group comprised all patients from January 1980 to December 2005 who underwent a definitive procedure to remove all thyroid tissue for BMNG, and who were subsequently identified as having developed a recurrence. Included were patients who underwent primary total thyroidectomy at our unit, or a two or more stage procedure where a definitive secondary total thyroidectomy was performed at our unit.


There were 3,044 total or secondary total thyroidectomies performed for BMNG during the study period. Ten patients were identified as having developed recurrent BMNG requiring reoperation despite previous complete “total” thyroidectomy. There were 11 sites of recurrence in 10 patients. Only one was a true local recurrence in the thyroid bed. Another 9 recurrences related to the embryology of the thyroid gland, 4 in the pyramidal tract and 5 in the thyrothymic tract. There was one recurrence at another site (submandibular) in a patient with presumed metastatic thyroid cancer despite benign histology. There were no complications in any of the 10 patients.


Total thyroidectomy for BMNG is not only a safe procedure but is efficacious in preventing recurrent disease. Failure to remove embryological remnants such as thyrothymic residue or pyramidal remnants during total thyroidectomy is the major cause of recurrence.

Over two decades ago, total thyroidectomy replaced subtotal thyroidectomy as the preferred option for the management of benign multinodular goiter (BMNG) in the Endocrine Surgical Unit, University of Sydney, Sydney, Australia.1,2 Since then many major endocrine surgery centers worldwide have adopted the same approach,3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13 which should now be considered as part of every endocrine surgeon’s armamentarium.14 Over that time, total thyroidectomy for BMNG has proven to be a very safe procedure with a minimal complication rate.15 What has not been as well studied to date is how efficacious the procedure is.

We define total thyroidectomy as the attempted removal of all visible thyroid tissue in the neck and, as such, it should be associated with negligible rates of recurrence. Likewise near-total thyroidectomy, leaving less than 1 cm of tissue, often posterior to the recurrent laryngeal nerve near the ligament of Berry, should also be associated with a low recurrence rate, although this procedure has not been our practice. Subtotal thyroidectomy, on the other hand, where a remnant of thyroid tissue is intentionally left in the thyroid bed is associated with a significant rate of recurrence.

The aim of the present study was to examine the efficacy of total thyroidectomy for BMNG to and to review reasons for recurrence.


This was a retrospective case series study comprising all patients in the period from January 1980 to December 2005 who underwent a definitive procedure or procedures to remove all thyroid tissue for BMNG, and who were subsequently identified as having developed a recurrence. Included were patients who underwent primary total thyroidectomy at our unit, or who underwent a multiple stage procedure where the definitive secondary (total) thyroidectomy with the intent to remove all remaining visible thyroid tissue was performed at our unit. Excluded were patients in whom the initial pathological diagnosis showed a clinically significant thyroid cancer; however, patients with incidental papillary microcarcinoma found on histological examination were included.

Information was obtained from the University of Sydney Endocrine Surgical Unit Database, as well as from patient records and contacts with general practitioners. All patients provided informed consent in relation to the storage and analysis of data.


From January 1980 to December 2005 a total of 3,044 total thyroidectomies or secondary total thyroidectomies were performed for BMNG in our unit. Within that group we identified 10 patients who developed recurrent BMNG requiring surgery despite a previous “total” thyroid procedure. The details of the patients are listed in Table 1. Nine of the ten patients underwent operation for recurrence in our unit, and the tenth patient had reoperative surgery at another hospital.
Table 1.

Patients who developed recurrent bilateral multinodular goiter (BMNG) despite previous total thyroidectomy



Age at complete thyroidectomy

Age at recurrent surgery

Site/s of recurrence


Mode of presentation

1 JS




Anterior mediastinum

Thyrothmyic rest

Atypical chest pain

2 VB




Anterior mediastinum

Thyrothymic rest

Chest x-ray

3 DK




Left retrosternal

Thyrothymic rest

Hoarse voice

4 IL




Left retrosternal

Thyrothymic rest

Lower neck lump

5 DM




Left retrosternal

Thyrothymic rest

Compression symptoms

6 HA




Midline, pyramidal


Neck lump

7 FB




Midline, pyramidal


Neck lump

8 ME




Midline, pyramidal


Neck lump

9 MR




Right thyroid bed & pyramidal

True recurrence, pyramidal (dyshormonogenesis)

Neck lumps (×2)

10 MN




Right submandibular

Undiagnosed cancer

Neck lumps

There were 2 males and 8 females. Mean age at initial intended total or secondary total thyroidectomy was 45 years (range: 10–64 years). The mean time to reoperation for recurrent disease was 57 months (range: 12–151months). Mean age at operation for recurrent disease was 52 years (range: 15–73 years).

In relation to the initial intended total procedure, 7 patients underwent initial total thyroidectomy and 3 patients underwent more than one procedure on their thyroid gland before complete “total thyroidectomy” was considered achieved.

All patients had BMNG on histopathological examination of the thyroid gland at the original procedure/s. Three patients had additional pathology at the time of the primary or secondary procedure. One patient (IL) had 2 incidental papillary carcinomas (1 mm and 1.5 mm) on histopathology of no clinical significance. One patient (FB) underwent a left hemithyroidectomy for a Hurtle cell adenoma, followed 7 years later by a completion right hemithyroidectomy for BMNG. A third patient (MN, patient 10 in Table 1) underwent secondary total thyroidectomy and neck clearance in our unit for bilateral nodular recurrence after two previous thyroid procedures performed elsewhere for benign disease. The histology demonstrated BMNG with reactive lymphadenopathy. She then presented with a right submandibular swelling, and FNA demonstrated thyroid tissue. After further surgery, histology has demonstrated evidence of bony metastases on radioactive ablation (administered with the intent of eliminating any residual benign disease), although no cancer had ever been identified histologically.

Table 2 documents the sites of recurrence that have been classified as (1) thyroid bed, (2) pyramidal tract, (3) thyrothymic tract (mediastinal/retrosternal), or (4) elsewhere. There were 11 sites of recurrence in 10 patients. There was only one true local recurrence at the ligament of Berry in the thyroid bed (MR). This patient also had a separate pyramidal recurrence and an underlying pathological diagnosis of thyroid dyshormonogenesis. There were three other recurrences in the pyramidal tract, all of which presented as neck masses. Five recurrences were associated with the thyrothymic tract (retrosternal, anterior mediastinal). The last patient (MN, patient 10 in Table 1) presented with the submandibular recurrence and, as noted above, has presumed metastatic thyroid cancer despite benign histology.
Table 2.

Sites of recurrence of BMNG despite total thyroidectomy

Classification of recurrence

True recurrence


  Thyroid bed


Embryological remnant


  Pyramidal lobe


  Thyrothymic tract


Other sites


  Submandibular swelling


There were no complications in any of the 10 patients, specifically no cases of either permanent recurrent laryngeal nerve palsy or permanent hypoparathyroidism.


Total thyroidectomy has been the operation of choice for BMNG in our unit since the early 1980s. It has been well demonstrated that total thyroidectomy is a safe operation with minimal complications if performed by surgeons and trainees well trained in the technique.16, 17, 18 This article examines whether total thyroidectomy for BMNG is also efficacious, i.e. it is associated with minimal rates of recurrence.

Less than total thyroidectomy, either subtotal or near-total thyroidectomy involves removal of most of the thyroid gland, leaving behind a small variable remnant of tissue. It aims to eliminate postoperative hypocalcemia by preserving the blood supply of the superior parathyroid glands by leaving in situ a small amount of thyroid tissue adjacent to these glands and their vascular pedicles. The recurrent laryngeal nerve is also theoretically protected by leaving thyroid tissue around the ligament of Berry in situ, thereby leaving the nerve untouched. Although the incidence of postoperative temporary hypocalcemia is reduced, less than total thyroidectomy has not been shown to be associated with a lower complication rate than total thyroidectomy4,9,19, 20, 21. Moreover, it leads to recurrent goiter in up to 23% of patients.22 This high recurrence rate, along with good evidence that total thyroidectomy is a safe operation with almost the same complication rate, and that reoperative surgery has a higher relative risk of complications than extensive primary surgery,23 has rendered subtotal thyroidectomy for BMNG an operation of the past.21

Total thyroidectomy, which aims to remove all visible evidence of thyroid tissue, ought not in theory at least, to be associated with recurrent BMNG if expertly performed. This study has demonstrated that thyroid bed recurrences are almost completely abolished (1 out of 3,044) with total thyroidectomy. However, recurrences still occur, and in most cases can be explained on the basis of embryological descent of the thyroid gland, either the pyramidal tract or the thyrothymic tract.

The thyroid gland begins to develop as early as 3–4 weeks gestation. It results from the fusion of two parts—the medial and lateral thyroid components. The medial component begins as a diverticulum in the pharyngeal endoderm, which descends and forms a bilobed structure anterior to the trachea. This reaches its final position by 7 weeks and undergoes histological differentiation into the typical follicles during weeks 10 and 11. The lateral thyroid components (also known as ultimo-branchial bodies) develop from proliferation of pharyngeal endoderm and are derived from the ventral portion of the endodermal fourth branchial pouch and the vestigial fifth pouch. It is from these lateral components that the C-cells originate. Ultimately the medial and the lateral components fuse. The pedicles of the lateral components detach themselves from the pharynx and are replaced with mesenchyme; the residual of this posterolateral projection becomes the tubercle of Zuckerkandl.24,25 The right and left lobes of the thyroid grow caudally with growth of the fetus, taking up their final position at either side of the 2nd to 4th tracheal rings. This line of normal embryological descent of the thyroid is known as the thyrothymic tract.

Thyroid rests are present in over 50% of patients and occur within the line of the thyrothymic tract. These rests are more commonly seen as an extension or prolongation of thyroid tissue attached to the lower pole of the thyroid by a narrow pedicle or even just a fibrovascular band.26 They can, however, be located within the anterior mediastinum. They may be mistaken for small lymph nodes or even parathyroid glands, and mostly they cause no problems.

Failure to identify and remove these rests at total thyroidectomy may result in recurrence, as in 5 patients in our series (Fig. 1). The presentation of recurrent thyrothymic tract disease is variable, as seen by the 5 different modes of presentation in this series. Interestingly, one patient (IL) had a recurrence in the anterior mediastinum with its own separate blood supply, a branch of the inferior thyroid artery proximal to ligation at total thyroidectomy. Such an artery may provide a clue to the existence of a thyrothymic rest at primary surgery.
Figure 1.

Recurrence in a thyrothymic rest delivered from the anterior mediastinum via a cervical incision.

There were 4 pyramidal recurrences (Fig. 2). The pyramidal lobe is the embryological remnant of the thyroglossal tract, the pathway the thyroid gland takes as it descends from its place of origin, the foramen cecum at the base of the embryological tongue. The pyramid contains thyroid cells and must be adequately identified and excised along with its fibrous remnant, which may track cranially to the inferior border of the hyoid bone. If it is not excised, the “total thyroidectomy” may fail, with recurrent nodular disease developing as a midline neck swelling.
Figure 2.

Recurrence in the pyramidal tract following total thyroidectomy. The forceps are pointing to the cricoid.

This series has supported our hypothesis that total thyroidectomy almost completely abolishes the incidence of true local recurrence. The only patient in this series who had a true local recurrence (in the thyroid bed) was a 15-year-old male who had a dyshormonogenetic multinodular goiter. He also had a simultaneous pyramidal recurrence. The pathological process behind his goiter differs from that of BMNG. Benign MNG, as is well known, results from heterogeneity of the thyroid suppressing hormone (TSH) receptor, resulting in nodular change between areas of degeneration. In contrast, dyshormonogenetic goiter results from chronic stimulation of the TSH receptor as a result of a disturbance in the feedback system of thyroid hormone synthesis.27 Dyshormonogenetic goiters are characterized by hypercellular nodules that are solid and microfollicular. Nuclei may be enlarged, irregularly shaped, and often bizarre, and they can be difficult to distinguish from thyroid carcinoma.28,29 Whether dyshormonogentic goiter is a greater risk factor for goiter recurrence than BMNG is unknown, as there are few cases reported in the literature, all with inadequate follow-up. Chronic stimulation of all TSH receptors would lead to recurrence even if only minimal numbers of thyroid cells were left behind at thyroidectomy. This explains the two recurrences at different sites in this young man. Interestingly, all patients were under regular review by an endocrinologist, and the TSH levels for the group with recurrence were not noted to be significantly different from those of the remainder of the group as such. Indeed, of note, the TSH levels for the recurrence group as a whole were at the lower end of the normal range before the final procedure (mean 1.27 mIU/l and 0.47 SEM), consistent with regrowth of autonomous thyroid tissue.

Metastatic thyroid cancer with no obvious primary lesion has been previously described.30, 31, 32, 33 The last patient (MN, patient 10 in Table 1) with probable metastatic disease has not had an identified primary lesion or tissue evidence of secondary disease, despite repeated histopathologic review. Although this may represent a metastasis from a small primary cancer not ever sectioned, it may also represent a form of thyroid cancer that is undetectable by current modes of pathological examination, especially given the aggressive nature of the local recurrence.

A limitation of the present study is the possibility that other patients with recurrent BMNG who had a previous total thyroidectomy within our unit have presented elsewhere, as this was not a systematic follow-up of all operative patients. However, given the nature of referral patterns for this unit and the relative stability of the Australian population, the number of patients with recurrence who were lost to follow-up would be very small indeed. Presumably, there is also another unidentified cohort of patients who have asymptomatic recurrences that are not clinically apparent. The only way that this group could ever be identified would be to perform routine follow-up screening ultrasound on every patient after total thyroidectomy. However, given that this group has not presented clinically with symptoms, such a program would not be cost-effective and would only be of academic interest.

Total thyroidectomy for BMNG is not only safe but also efficacious in preventing recurrent disease, although a clear understanding of thyroid embryology is essential to ensure successful surgery. Embryological remnants are the major cause of failure of total thyroidectomy for BMNG. Special attention must be paid to these areas with every routine thyroidectomy. Failure to remove a thyrothymic rest or pyramidal remnant can result in recurrence of BMNG years later.


  1. 1.
    Reeve TS, Delbridge L, Cohen A, et al. Total thyroidectomy. The preferred option for multinodular goiter. Ann Surg 1987;206:782–786PubMedCrossRefGoogle Scholar
  2. 2.
    Khadra M, Delbridge L, Reeve TS, et al. Total thyroidectomy: its role in the management of thyroid disease. Aust N Z J Surg 1992;62:91–95PubMedCrossRefGoogle Scholar
  3. 3.
    Zambudio AR, Rodriguez J, Riquelme J, et al. Prospective study of postoperative complications after total thyroidectomy for multinodular goiters by surgeons with experience in endocrine surgery. Ann Surg 2004;240:18–25PubMedCrossRefGoogle Scholar
  4. 4.
    Pappalardo G, Guadalaxara A, Frattaroli FM, et al. Total compared with subtotal thyroidectomy in benign nodular disease: personal series and review of published reports. Eur J Surg 1998;164:501–506PubMedCrossRefGoogle Scholar
  5. 5.
    Muller PE, Kabus S, Robens E, et al. Indications, risks, and acceptance of total thyroidectomy for multinodular benign goiter. Surg Today 2001;31:958–962PubMedCrossRefGoogle Scholar
  6. 6.
    Mishra A, Agarwal A, Agarwal G, et al. Total thyroidectomy for benign thyroid disorders in an endemic region. World J Surg 2001;25:307–310PubMedCrossRefGoogle Scholar
  7. 7.
    Gough IR, Wilkinson D. Total thyroidectomy for management of thyroid disease. World J Surg 2000;24:62–65CrossRefGoogle Scholar
  8. 8.
    Friguglietti CU, Lin CS, Kulcsar MA. Total thyroidectomy for benign thyroid disease. Laryngoscope 2003;113:1820–1826PubMedCrossRefGoogle Scholar
  9. 9.
    Colak T, Akca T, Kanik A, et al. Total versus subtotal thyroidectomy for the management of benign multinodular goiter in an endemic region. Aust N Z J Surg 2004;74:974–978CrossRefGoogle Scholar
  10. 10.
    Bron LP, O’Brien CJ. Total thyroidectomy for clinically benign disease of the thyroid gland. Br J Surg 2004;91:569–574PubMedCrossRefGoogle Scholar
  11. 11.
    Hisham AN, Azlina AF, Aina EN, et al. Total thyroidectomy: the procedure of choice for multinodular goitre. Eur J Surg 2001;167:403–405PubMedCrossRefGoogle Scholar
  12. 12.
    Gibelin H, Sierra M, Mothes D, et al. Risk factors for recurrent nodular goiter after thyroidectomy for benign disease: case-control study of 244 patients. World J Surg 2004;28:1079–1082PubMedCrossRefGoogle Scholar
  13. 13.
    Ayache S, Tramier B, Chatelain D, et al. [Evolution of the thyroid surgical treatment to the total thyroidectomy. Study of about 735 patients]. Ann Otolaryngol Chir Cervicofac 2005;122:127–133PubMedGoogle Scholar
  14. 14.
    Wheeler MH. Total thyroidectomy for benign thyroid disease. Lancet 1998;351:1526–1527PubMedCrossRefGoogle Scholar
  15. 15.
    Delbridge L. Total thyroidectomy: the evolution of surgical technique. Aust N Z J Surg 2003;73:761–768CrossRefGoogle Scholar
  16. 16.
    Reeve TS, Curtin A, Fingleton L, et al. Can total thyroidectomy be performed as safely by general surgeons in provincial centers as by surgeons in specialized endocrine surgical units? Making the case for surgical training. Arch Surg 1994;129:834–836PubMedGoogle Scholar
  17. 17.
    Martin L, Delbridge L, Martin J, et al. Trainee surgery in teaching hospitals: is there a cost? Aust N Z J Surg 1989;59:257–260PubMedGoogle Scholar
  18. 18.
    Gough IR. Total thyroidectomy: indications, technique and training. Aust N Z J Surg 1992;62:87–89PubMedCrossRefGoogle Scholar
  19. 19.
    Ozbas S, Kocak S, Aydintug S, et al. Comparison of the complications of subtotal, near total and total thyroidectomy in the surgical management of multinodular goitre. Endocr J 2005;52:199–205PubMedCrossRefGoogle Scholar
  20. 20.
    Marchesi M, Biffoni M, Tartaglia F, et al. Total versus subtotal thyroidectomy in the management of multinodular goiter. Int Surg 1998;83:202–204PubMedGoogle Scholar
  21. 21.
    Delbridge L, Guinea AI, Reeve TS. Total thyroidectomy for bilateral benign multinodular goiter: effect of changing practice. Arch Surg 1999;134:1389–1393PubMedCrossRefGoogle Scholar
  22. 22.
    Reeve TS, Delbridge L, Brady P, et al. Secondary thyroidectomy: a twenty-year experience. World J Surg 1988;12:449–453PubMedCrossRefGoogle Scholar
  23. 23.
    Thomusch O, Machens A, Sekulla C, et al. Multivariate analysis of risk factors for postoperative complications in benign goiter surgery: prospective multicenter study in Germany. World J Surg 2000;24:1335–1341PubMedCrossRefGoogle Scholar
  24. 24.
    Pelizzo MR, Toniato A, Gemo G. Zuckerkandl’s tuberculum: an arrow pointing to the recurrent laryngeal nerve (constant anatomical landmark). J Am Coll Surg 1998;187:333–336PubMedCrossRefGoogle Scholar
  25. 25.
    Hisham AN, Aina EN. Zuckerkandl’s tubercle of the thyroid gland in association with pressure symptoms: a coincidence or consequence? Aust N Z J Surg 2000;70:251–253PubMedCrossRefGoogle Scholar
  26. 26.
    Sackett WR, Reeve TS, Barraclough B, et al. Thyrothymic thyroid rests: incidence and relationship to the thyroid gland. J Am Coll Surg 2002;195:635–640PubMedCrossRefGoogle Scholar
  27. 27.
    Cooper DS, Axelrod L, DeGroot LJ, et al. Congenital goiter and the development of metastatic follicular carcinoma with evidence for a leak of nonhormonal iodide: clinical, pathological, kinetic, and biochemical studies and a review of the literature. J Clin Endocrinol Metab 1981;52:294–306PubMedCrossRefGoogle Scholar
  28. 28.
    Rosai J CM, DeLellis RA. Tumours of the thyroid gland. In Rosai J, (eds). Atlas of Tumour Pathology, 3rd series ed, Washington, DC, Armed Forces Institute of Pathology, 1990:302–303Google Scholar
  29. 29.
    Lloyd RV DB, Young WF. Endocrine diseases. Atlas of Nontumor Pathology, 1st edition, Washington, DC, American Registry of Pathology and the Armed Forces Institute of Pathology; 2002:144–146Google Scholar
  30. 30.
    Paksoy N, Ozturk H, Demircan A, et al. Occult papillary carcinoma of the thyroid presenting as an intratracheal tumour. Eur J Surg Oncol 1994;20:694–695PubMedGoogle Scholar
  31. 31.
    Nishikawa M, Toyoda N, Yonemoto T, et al. Occult papillary thyroid carcinoma in Hashimoto’s thyroiditis presenting as a metastatic bone tumor. Endocr J 1998;45:111–116PubMedGoogle Scholar
  32. 32.
    Lopez-Escamez JA, Lopez-Nevot A, Moreno-Garcia MI, et al. Cervical metastasis of occult papillary thyroid carcinoma associated with epidermoid carcinoma of the larynx. ORL J Otorhinolaryngol Relat Spec 1999;61:224–226PubMedGoogle Scholar
  33. 33.
    Homan MR, Gharib H, Goellner JR. Metastatic papillary cancer of the neck: a diagnostic dilemma. Head Neck 1992;14:113–118PubMedCrossRefGoogle Scholar

Copyright information

© Société Internationale de Chirurgie 2007

Authors and Affiliations

  • Kylie L. Snook
    • 1
  • Peter L. H. Stalberg
    • 1
  • Stan B. Sidhu
    • 1
  • Mark S. Sywak
    • 1
  • Pamela Edhouse
    • 1
  • Leigh Delbridge
    • 1
  1. 1.Department of Endocrine and Oncology SurgeryUniversity of Sydney Endocrine Surgical Unit, Royal North Shore HospitalSydneyAustralia

Personalised recommendations