—  ENDOCRINE PATHOLOGY SOCIETY   —

Use and Abuse of Frozen Sections in Diagnosis of Follicular Thyroid Lesions


Virginia A. LiVolsi
University of Pennsylvania
Philadelphia, PA


The intraoperative pathology consultation is an essential tool in surgical management of diseases in many organs. Frozen section can serve as the first diagnostic procedure to differentiate between benign and malignant tumors and helps in further planning of the surgery. [1, 2, 3] The reasons for frozen section request given by surgeons include: diagnosis of the lesion, identification of tissue type (e.g. parathyroid), assessment of margins, determination of operability (i.e., liver metastases present will abort a Whipple procedure), assessment of adequacy of lesional material for definitive workup (i.e. suspected malignant lymphoma, and provision of tissue for special studies (i.e. molecular analysis for sarcoma). In the modern era with advances in radiology and increasing experience with fine needle aspiration cytology, the diagnosis of many lesion is already known preoperatively, so that the most common reason for frozen section request even a decade ago (i.e., diagnosis) is becoming less frequent.

In certain organ systems, the role of intraoperative frozen section has always been and continues to be problematic. One of these areas is in thyroid diseases and especially nodules; herein, the role of intra-operative consultation is controversial; this is especially true for encapsulated follicular patterned lesions which comprise the major reason for thyroid surgery in current practice. [4, 5, 6]

Before the wide use of fine needle aspiration biopsy of thyroid nodules, the frequency of thyroid surgery was much greater than in current surgical practice. [1, 7] The value of frozen section in those cases was not disputed, since the differential diagnosis of papillary carcinoma versus nodular goiter could be relatively easily rendered and appropriate surgery accomplished. [8, 9] Thus in the days when preoperative diagnosis was not easily made, intraoperative frozen section made sense.

In the more modern era, in the thyroid the selection of patients requiring surgery is usually made on the basis of FNA diagnoses of either neoplasm, suspicious for neoplasm or malignant. [7, 10, 11] Patients with a definite FNA diagnosis of malignancy are subjected to total or near total thyroidectomy, whereas, lobectomy is performed in cases diagnosed as follicular neoplasm; upon completion of the histopathologic examination a completion thyroidectomy can be performed if the nodule removed is diagnosed as carcinoma. [11, 12] A similar surgical treatment is also followed in cases diagnosed as suspicious for malignant or indeterminate for neoplasm. [13, 14] However, this two-step surgery consisting of lobectomy followed by completion thyroidectomy does subject the patient to two surgical procedures and the associated risks. [15, 16, 17]

Several authors have shown that intra-operative consultation is necessary for the definite surgical management of thyroid nodules. [5, 18] As mentioned above the primary use of frozen section in surgical pathology is to establish a diagnosis; however, in some instances frozen section may fail the task of establishing a definitive diagnosis; this leads to deferral of diagnosis to final histopathologic examination. [1]

The definite surgical procedure is delayed until a final pathology diagnosis is rendered on permanent sections. (It is obvious that should an abnormal lymph node be identified and on frozen section a diagnosis of metastatic carcinoma consistent with thyroid primary is rendered, no additional diagnostic intraoperative assessment is needed}. [17, 19, 20, 21, 22]

Both the surgical and pathology literature contain institutional experiences regarding the use of intra-operative assessment of thyroid lesions. [4, 5, 6, 21, 23, 24, 25] The experiences and conclusions drawn from these publications can be equally divided into two groups; ones that found that this procedure to be of limited utility and others who concluded that frozen section is useful (of more or of equal diagnostic value compared to fine-needle aspiration).

The arguments against the use of frozen sections in thyroid nodules include: 1) FNA can diagnose papillary carcinoma in more than 90% of cases; therefore, it is not cost effective to perform frozen section in cases diagnosed as definite for malignancy on FNA; 2) frozen section can induce artifactual nuclear clearing, which can be mistaken for papillary thyroid carcinoma leading to false positive intraoperative diagnosis; 3) frozen section is of limited value in the diagnosis of follicular or Hürthle cell carcinoma, because the diagnosis of malignant follicular and Hürthle cell lesion is dependent upon demonstration of true capsular and/or vascular invasion which requires a thorough and detailed examination of the lesion's capsule. [4, 14, 26, 27, 28, 29] Therefore, due to limited sampling at the time of the frozen section a majority of lesions diagnosed on FNA as follicular / Hürthle cell neoplasm or indeterminate for neoplasm will be deferred to final histopathologic examination. [6, 29, 30, 31, 32]

Among the investigators who have suggested that frozen sections can be beneficial in thyroid lesions particularly in follicular/Hürthle cell lesions, the most commonly quoted report is from Mayo Clinic. [5] These authors studied a cohort of 1023 patients undergoing surgery for a diagnosis of follicular and Hürthle cell thyroid neoplasm; 78% of cases were diagnosed as malignant on frozen section, thus leading to a definite surgical management and avoiding the delay of a two step surgical procedure (lobectomy followed by completion thyroidectomy). However, in our view this study describes the experience from a unique pathology laboratory, where the majority of surgical specimens are diagnosed on the basis of multiple frozen sections. Hence if we consider the working environment of usual surgical pathology laboratory in both community practices and academic centers, it will not be cost effective and practically impossible to perform multiple frozen sections to distinguish between benign and malignant follicular and Hürthle cell tumors.

One of the major arguments to prove the limited utility of frozen section in thyroid nodules is the fact that for practical reasons, extensive sectioning of the capsule in follicular patterned thyroid nodules cannot be done. [5, 18, 33, 34, 35] In a recent study from Johns Hopkins University, frozen section was only able to render a diagnosis of malignancy in 1 of 29 cases diagnosed as follicular neoplasm on FNA. [29] In a similar study from our institution 50% of the diagnoses on frozen sections performed on various thyroid lesions were deferred to permanent histologic sections; of these 22% were diagnosed as malignant on final histology requiring additional surgery. [6] Chen et al studied 125 patients with follicular thyroid lesions and from these 87% were deferred to permanent sections after intra-operative assessment. In addition, 5% of cases were misdiagnosed as malignant leading to unnecessary total thyroidectomy. [26] The Hopkins group also assessed the cost of the procedure and found every useful frozen section cost over $12000.

In addition, due to the limited sampling performed at the time of the frozen section, some malignant follicular lesions may be under-diagnosed as benign. [26] This is issue is well addressed in the study by Kingston et al; of 198 follicular neoplasms21% were misinterpreted as benign on frozen section examination. [35] A similar study by Crowe et al showed that 6% of the malignant lesions in their series were not correctly identified on frozen section. [36]

A majority of thyroid nodules that are excised have undergone pre-operative FNA, the most sensitive diagnostic technique available for the management of thyroid lesions. [7, 10] However, it is important to be aware of the fact that FNA induced trauma can lead to a variety of histologic alterations within thyroid nodules. [37] These have been described in various publications and include hemorrhage, infarction, nuclear atypia and capsular and vascular pseudo-invasion. [37, 38] The latter can be mistaken for foci of true tumor invasion on frozen section; similarly the nuclear atypia can also be mistaken for papillary carcinoma. [38, 39] In our experience, the post FNA foci of pseudo-invasion are usually linear and are surrounded by areas of hemorrhage and granulation tissue. [38] However, such distinction between true and false post-FNA foci of capsular and or vascular invasion is only possible in detailed examination of well-fixed permanent histologic sections. [38, 39]

Mchenry et al studied a cohort of 78 patients with thyroid nodules and compared the FNA and frozen section diagnosis along with the cost benefit analysis of frozen section examination. These authors found that diagnosis was deferred in 38 patients (50%), and one to six frozen sections were performed in each patients leading to alteration in surgical management in only 2 patients (3%). [25]A study of 812 patients by Morisini et al comparing the FNA, frozen section and final diagnosis found that frozen section was less sensitive for the diagnosis of papillary cancer and more specific for the diagnosis of follicular carcinoma. [18] In this study there were 21 false negative diagnoses and no false positive diagnosis on frozen section. However, this study fails to provide any details about the number of frozen sections performed and examined and the times required for the analyses.

In view of the above-mentioned studies, we believe that in everyday practice of pathology, frozen section examination is of limited value in the diagnoses of follicular and Hürthle cell tumors and does not lead to any additional information as compared to preoperative FNA. In addition, it is not practical to do multiple frozen sections on a thyroid nodule to possibly reach a definite diagnosis. If one evaluates the risk of a follicular or Hurthle cell neoplasm being malignant, one recognizes that only about 20% will be diagnosed as carcinoma after complete histopathologic examination, [9, 40] so that in the majority of patients, lobectomy will be adequate therapy.

Is there any role for intra-operative assessment in the surgical management of thyroid nodules?

Despite the above-mentioned review outlining the arguments in favor and against the use of frozen section in the surgical management of thyroid nodules, it has been shown by both retrospective and prospective analyses that intra-operative evaluation of thyroid nodules can be useful in cases diagnosed as suspicious for papillary thyroid carcinoma on FNA. [13, 14, 27, 41, 42] Rodriguez et al investigated the value of frozen section in cases diagnosed as benign, suggestive of neoplasm, or malignant on FNA. They found that frozen section examination was particularly helpful in cases with a "suggestive" cytology diagnosis, whereas, there was no significant change in diagnosis in cases with benign or malignant diagnoses. [43] Basolo et al reported the use of intra-operative cytology as an adjunct to frozen section analysis of thyroid. [27] In this publication the intra-operative cytology (IOC) preparation were stained with Ultrafast-Papanicolaou method. The nuclear morphology was comparable to that seen in FNA smears. According to this study there was a 98% correlation between IOC and final histologic diagnosis as compared to 71% correlation between frozen section alone and final histologic diagnosis. Tworek et al studied the impact of IOC and frozen section, and found that IOC was highly sensitive and specific in the diagnosis of papillary thyroid carcinoma. [44] In our experience, Ultrafast-Papanicolaou stained scrape preparations serve as a useful adjunct to the frozen section ; use of this techniquehas led to completion thyroidectomy in some cases , which had been diagnosed as suspicious for papillary thyroid carcinoma on FNA. [13, 14, 41] Interestingly, in our experience and that of others, most of these cases represent follicular variant of papillary thyroid carcinoma on final histopathologic examination. [13, 41, 45] A majority of these are encapsulated follicular patterned lesions without any obvious capsular and/or vascular invasion. Thus these cases can be easily mistaken as follicular adenoma on frozen section alone, since the nuclear features of papillary thyroid carcinoma are frequently not observed in frozen sections. [46]

I have for the most part included studies which agree with my view of the general lack of diagnostic utility for thyroid nodules in the era of FNA. However, it would be remiss and unfair not to include studies in the literature that espouse an opposite point of view. In addition to the study cited above from the Mayo Clinic, several of these report on large numbers of cases and their results deserve to be considered. Within the past five years, a number of reports about FNA and frozen section of thyroid nodules have been published from around the world: these indicate that the yields are equivalent or that frozen section is not cost effective in the era of FNA [47, 48, 49, 50, 51] ; some studies indicate that selective use of frozen section often with intraoperative imprint cytology in lesions considered "suspicious" on FNA may be useful [52, 53] . Only a rare report favors frozen section diagnosis in the thyroid [54, 55] .

In view of the ample evidence presented in the literature regarding the use of intraoperative consultation for the thyroid nodule, one can draw the following major conclusions:
  1. Intraoperative consultation is not indicated in cases diagnosed as definite for malignancy on FNA due to high sensitivity and specificity (indeed, I have often told residents that if surgeons do not believe a definitive cytologic diagnosis of papillary cancer, they should either not have FNA performed or get themselves another cytopathologist; doing both FNA with a definite diagnosis and also a frozen section is wasteful of resources in the medical care environment)

  2. The records of good cytopathologists indicate that the false positive rate for papillary carcinoma approaches 0%

  3. in the majority of cases frozen section is of limited or no value in the diagnosis of follicular or Hürthle cell carcinoma, because the histologic characterization of these lesions requires detail analysis of the tumor capsule for demonstrating capsular and/or vascular invasion-an analysis which is not practical in the daily practice of pathology

  4. intraoperative consultation (frozen section combined with IOC) is most effective in cases, which are diagnosed as suspicious for papillary thyroid carcinoma on FNA
One special case deserves to be mentioned and I have been seeing more of these lesions in the past 2-3 years: this is the situation of either an ultrasound detected subcentimeter nodule [56] with equivocal or no FNA and the incidental (usually subcapsular) white subcentimeter nodule found grossly in the intraoperative period. Frozen section on these small lesions is to be condemned in my opinion, for two reasons:
  1. the freezing artifact often so interferes with the cytomorphology of the permanent sections that a definitive diagnosis may never be made with certainty; and

  2. even if the lesion is a carcinoma, it is a papillary microcarcinoma and of no significance anyway.
Sanctioning a total thyroidectomy for such a lesion on a frozen section seems inappropriate and overtreatment in any event.

References

  1. Nakazawa H, Rosen P, Lane N, Lattes R: Frozen section experience in 3000 cases. Accuracy, limitations, and value in residency training. Am J Clin Pathol 49:41-51., 1968

  2. Barney PL: Histopathologic problems and frozen section diagnosis in diseases of the larynx. Otolaryngol Clin North Am 3:493-515, 1970

  3. Bredahl E, Simonsen J: Routine performance of intra-operative frozen section microscopy, with particular reference to diagnostic accuracy. Acta Pathol Microbiol Scand Suppl 212:104, 1970

  4. Bronner MP Hamilton R, LiVolsi VA.: Utility of frozen section analysis on follicular lesions of the thyroid. Endocr Pathol 5:154-161, 1994

  5. Paphavasit A, Thompson GB, Hay ID, Grant CS, van Heerden JA, Ilstrup DM, Schleck C, Goellner JR: Follicular and Hurthle cell thyroid neoplasms. Is frozen-section evaluation worthwhile? Arch Surg 132:674-8; discussion 678-80, 1997

  6. Montone KT, LiVolsi V: Frozen section analysis of thyroidectomy specimens:experience over a 12-year period. Pathol Case Rev 2:241-245, 1996

  7. Kini SR: Guides to Clinical Aspiration Biopsy Thyroid, 2nd ed edition. New York, NY, Igaku-Shoin, 1996

  8. Kini SR, Miller JM, Hamburger JI, Smith-Purslow MJ: Cytopathology of follicular lesions of the thyroid gland. Diagnostic Cytopathology 1:123-32, 1985

  9. LiVolsi VA: Surgical Pathology of The Thyroid. Philadelphia, PA, WB. Saunders, 1990

  10. Baloch ZW, Sack MJ, Yu GH, Livolsi VA, Gupta PK: Fine-needle aspiration of thyroid: an institutional experience. Thyroid 8:565-9, 1998

  11. Silverman JF WR, Larkin EW, et al.: The role of fine-needle aspiration biopsy in the rapid diagnosis and management of thyroid neoplasms. Cancer 57:1164-1170, 1986

  12. Altavilla G, Pascale M, Nenci I: Fine needle aspiration cytology of thyroid gland diseases. Acta Cytol 34:251-6., 1990

  13. Logani S, Gupta PK, LiVolsi VA, Mandel S, Baloch ZW: Thyroid nodules with FNA cytology suspicious for follicular variant of papillary thyroid carcinoma: follow-up and management. Diagn Cytopathol 23:380-5., 2000

  14. Chen H, Zeiger MA, Clark DP, Westra WH, Udelsman R: Papillary carcinoma of the thyroid: can operative management be based solely on fine-needle aspiration? J Am Coll Surg 184:605-10, 1997

  15. Chonkich GD, Petti GH, Jr., Goral W: Total thyroidectomy in the treatment of thyroid disease. Laryngoscope 97:897-900, 1987

  16. Korun N, Asci C, Yilmazlar T, Duman H, Zorluoglu A, Tuncel E, Erturk E, Yerci O: Total thyroidectomy or lobectomy in benign nodular disease of the thyroid: changing trends in surgery. Int Surg 82:417-9, 1997

  17. Marchetta FC, Sako K: The diagnosis of thyroid carcinoma during the postoperative period after less than total thyroidectomy. Am J Surg 136:455-6, 1978

  18. Morosini PP, Mancini V, Filipponi S, Taccaliti A, et al: Comparison between the diagnostic accuracy in diagnosis of thyroid nodules with fine needle biopsy an intraoperative histological evaluation of frozen tissue. Minerva Endocrinol 22:1-5, 1997

  19. Crile G, Jr.: The relationship of surgeons and pathologists. Am J Clin Pathol 75:458-9, 1981

  20. Sivula A: Diagnostic aspects in 117 patients treated surgically for thyroid carcinoma. Ann Chir Gynaecol Suppl 65:13-21, 1976

  21. Rodigas P, Sufian S, Kaibara N, Matsumoto T: Surgery of the thyroid gland. Int Surg 62:588-91, 1977

  22. Wade H: The treatment and preoperative diagnosis of differentiated thyroid carcinoma presenting as a clinically solitary nodule. Br J Surg 67:728-31, 1980

  23. Hirst E, Coombes BH, Bale PM, Palmer AA, Hambly CK: Diagnosis by frozen section examination. I. Results in breast and thyroid lesions. Aust N Z J Surg 37:325-31, 1968

  24. Hayashi Y, Tsubokura T, Ukita M, Sakamoto Y, Umisa H: [Histological diagnosis of thyroid diseases by rapid frozen section and paraffin section methods]. Rinsho Byori 20:129-34, 1972

  25. McHenry CR, Rosen IB, Walfish PG, Bedard Y: Influence of fine-needle aspiration biopsy and frozen section examination on the management of thyroid cancer. Am J Surg 166:353-6, 1993

  26. Chen H, Nicol TL, Udelsman R: Follicular lesions of the thyroid. Does frozen section evaluation alter operative management? Ann Surg 222:101-6, 1995

  27. Basolo F, Baloch ZW, Baldanzi A, Miccoli P, LiVolsi VA: Usefulness of Ultrafast Papanicolaou-stained scrape preparations in intraoperative management of thyroid lesions. Mod Pathol 12:653-7, 1999

  28. Layfield LJ, Mohrmann RL, Kopald KH, Giuliano AE: Use of aspiration cytology and frozen section examination for management of benign and malignant thyroid nodules. Cancer 68:130-4, 1991

  29. Udelsman R, Westra WH, Donovan PI, Sohn TA, Cameron JL: Randomized prospective evaluation of frozen-section analysis for follicular neoplasms of the thyroid. Annals of Surgery 233:716-22, 2001

  30. Mulcahy MM, Cohen JI, Anderson PE, Ditamasso J, Schmidt W: Relative accuracy of fine-needle aspiration and frozen section in the diagnosis of well-differentiated thyroid cancer. Laryngoscope 108:494-6, 1998

  31. Remsen KA, Lucente FE, Biller HF: Reliability of frozen section diagnosis in head and neck neoplasms. Laryngoscope 94:519-24, 1984

  32. DeMay RM: Frozen section of thyroid? Just say no [editorial; comment]. Am J Clin Pathol 110:423-4, 1998

  33. McHenry CR, Raeburn C, Strickland T, Marty JJ: The utility of routine frozen section examination for intraoperative diagnosis of thyroid cancer. Am J Surg 172:658-61, 1996

  34. Melliere D, Danis RK, Lasry G: Cold nodules of the thyroid. Reevaluation of surgical excision on the basis of a new group of 607 patients. Nouv Presse Med 8:1399-402, 1979.

  35. Kingston GW, Bugis SP, Davis N: Role of frozen section and clinical parameters in distinguishing benign from malignant follicular neoplasms of the thyroid. Am J Surg 164:603-5, 1992

  36. Crowe PJ, Chetty R, Dent DM: Thyroid frozen section: flawed but helpful. Aust N Z J Surg 63:275-8, 1993

  37. LiVolsi VA, Merino MJ: Worrisome histologic alterations following fine-needle aspiration of the thyroid (WHAFFT). Pathol Annu 29:99-120., 1994

  38. Baloch ZW, LiVolsi VA: Post fine-needle aspiration histologic alterations of thyroid revisited. Am J Clin Pathol 112:311-6, 1999

  39. Lopez JI, Pereda E, Rodil MA, Fernandez-Larrinoa A: Histological changes mimicking papillary carcinoma following fine needle aspiration of the thyroid gland. Arch Anat Cytol Pathol 44:98-100, 1996

  40. Rosai J, Carcangui ML, DeLellis RA: Tumors of The Thyroid Gland. In Atlas of Tumor Pathology. Edited by J Rosai and LE Sobin. Vol 3rd Series, Fascicle 5. Washington, DC, Armed Forces Institute of Pathology, 1992

  41. Paessler M, LiVolsi VA, Baloch Z: Role of Ultrafast Papanicolaou stained scrape preparations as an adjunct to frozen section in the surgical management of thyroid lesions. Endocr Practice 7:89-94, 2001

  42. Shen PU, Kuhel WI, Yang GC, Hoda SA: Intraoperative touch-imprint cytological diagnosis of follicular variant of papillary thyroid carcinoma. Diagn Cytopathol 17:80-3., 1997

  43. Rodriguez JM, Parrilla P, Sola J, Bas A, et al.: Comparison between preoperative cytology and intraoperative frozen- section biopsy in the diagnosis of thyroid nodules. Br J Surg 81:1151-4, 1994

  44. Tworek JA, Giordano TJ, Michael CW: Comparison of intraoperative cytology with frozen sections in the diagnosis of thyroid lesions. Am J Clin Pathol 110:456-61, 1998

  45. Lin HS, Komisar A, Opher E, Blaugrund SM: Follicular variant of papillary carcinoma: the diagnostic limitations of preoperative fine-needle aspiration and intraoperative frozen section evaluation. Laryngoscope 110:1431-6, 2000

  46. Baloch ZW, Gupta PK, Yu GH, Sack MJ, LiVolsi VA: Follicular variant of papillary carcinoma. Cytologic and histologic correlation. Am J Clin Pathol 111:216-222., 1999

  47. Duck SD, Goldenbery D. LinnS, Krausz MM, and Hershko, DD. The role of fine needle aspiration and intraoperative frozen section in the surgical management of solitary thyroid nodules. Surg Today 32: 857-861 2002.

  48. Lee TI, Yang, HJ, Lin SY. Lee MT, et al. THe accuracy of fine needle aspiration biopsy and frozen section in patients with thyroid cancer. Thyroid 12: 619-626, 2002.

  49. Cheng MS, Morgan JL, Serpell JW. Does frozen section have a role in the intraoperative management of thyroid nodules? Aust N Zeal J Surg 72: 570-572, 2002

  50. Mandell DL, Genden EM, Mechanick JL, Begman DA, et al. Diagnostic accuracy of fine needle aspiration and frozen section in nodular thyroid disease. Otolaryngol Head Neck Surg 124: 531-536, 2001.

  51. Brooks AD, Shaha AR, DuMornay W, Huvos AG et al. Role of fine needle aspiration biopsy and frozen section analysis in the surgical management of thryoid tumors. Ann Surg Oncol 8: 92-100, 2001.

  52. Lin HS, Komisar A, Opher E, Blaugrund SM. Surgical msnsgement of thyroid masses: assessong the need for frozen section evaluation. Laryngosope 109: 868-873, 1999.

  53. Roach JC, Heller KS, Dubner S, Sznyter LA. THe value of frozen section examinations in determining the extent of thyroid surgery in patients with indeterminate fine needle aspiration cytology. Arch Otolaryngol Head Neck Surg 128: 263-267. 2002.

  54. Taneri F, Poyraz A, Tekin, E, Ersoy, E Dursun A. Accuracy and significance of fine needle aspiration cytology and frozen section in thyroid surgery. Endocr Reg 32: 187-191, 1998.

  55. Caraci P, Aversa S, Mussa A, Pancani G et al: Role of fine needle aspiration and frozen section evaluation in the surgical management of thyroid nodules. Br J Surg 89: 797-801, 2002.

  56. Carpi A, Nicolini A, Casara D, Rubello D, Pelizzo MR. Nonpalpable thyroid carcinoma: Clinical controversies on preoperative selection. Am J Clin Oncol 26: 232-235, 2003.