• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Frozen Section Diagnosis in Pediatric Surgical Pathology: A Decade's Experience in a Children's Hospital

Posted on: Thursday, 15 December 2005, 03:02 CST

By Coffin, Cheryl M; Spilker, Krista; Zhou, Holly; Lowichik, Amy; Pysher, Theodore J

Context.-Intraoperative consultations, including frozen sections (FSs), are essential for patient care and are a key quality component in anatomic pathology. Little data exists about the use, frequency, and type of discrepancies and deferral rates of FS diagnoses in pediatric and adolescent surgical pathology.

Objective.-The purpose of this study was to analyze indications, discrepancies, and deferrals for all FSs performed at a children's hospital during a 10-year period.

Design.-All FSs for 1995-2004 were reviewed for indications, discrepancies, deferred diagnoses, and turnaround time. Discrepancies were categorized into major and minor subtypes according to potential impact on patient care.

Results.-A total of 35 611 surgical pathology cases were accessioned, with 2839 intraoperative consultations, which included 2783 FSs and 56 nonmicroscopic consultations. Most frequent indications included questions related to neoplasms (tumor detection, specimen adequacy, triage, classification, and margins) and suspected Hirschsprung disease. In these consultations, 115 discrepancies (4%) were identified, of which 7 (0.2%) were major, with potentially significant clinical impact, and 108 (3.9%) were minor. The major discrepancies included tumor, ganglion cell, or organism detection. The minor discrepancies involved sampling error, reclassification of benign or malignant neoplasms without clinical consequences, tumor typing or grading, and ganglion cell identification without clinical impact. Deferrals in 718 FSs (25% deferral rate) included tumor classification from generic to specific, identification of organisms, and evaluation of lymph node biopsies for lymphoma. Turnaround time exceeded 20 minutes in 403 cases (14%).

Conclusions.-The FS rate of 7.8% overall and 5% of surgical pathology cases is similar in children's and general hospitals. The major discrepancy (discordance) rate is lower, which may reflect the different indications for FS in children and adolescents. Evaluation of colonic biopsies for ganglion cells is a diagnostic pitfall. The deferral rate of 25% reflects the definition of a deferred diagnosis. Traditional definitions of deferred and discordant FS diagnoses should be refined to reflect the increasing use of adjunct techniques, especially in tumor classification. These findings emphasize that, in children and adolescents, most FSs are performed for tumor classification, triage, detection, and specimen adequacy, and for possible Hirschsprung disease. In children and adolescents, FSs are used infrequently to identify normal or unknown tissue, to analyze a lesion in a radiographically directed specimen, or to detect lymph node metastases. The differences in pediatric and adolescent FS indications and use underscore the importance of focused education in pediatric surgical pathology.

(Arch Pathol Lab Med. 2005;129:1619-1625)

Intraoperative consultations, both microscopic and macroscopic, are critical patient care services provided by surgical pathologists. Interpretation is a complex process that requires specialized histomorphologic knowledge, which must often be complemented by clinical, and sometimes laboratory and radiographie, information. Key quality components include relevance, accuracy, and timeliness.1-4 Although specific reviews and analyses of intraoperative consultations have been published for renal,5 soft tissue,6 and central nervous system7-9 lesions, the emphasis in many of these publications has been on neoplastic diseases and conditions that affect adults more frequently than children. In pediatric surgical pathology, intraoperative microscopic evaluations are often used for solid tumors, including germ cell neoplasms and tumors of the central nervous system, kidneys, bone, soft tissue, liver, and lymph nodes.10,11 Evaluation for Hirschsprung disease is also an important indication for frozen sections (FSs) in children; and because the results can have an immediate effect on surgical management, reliability and error prevention are important concerns for both pediatric surgeons and surgical pathologists.12,13 Despite these considerations, there is little published data available about intraoperative consultations for pediatric and adolescent patients.14

We undertook this study to determine the frequency, indications, timeliness, and accuracy for intraoperative microscopic consultations (FSs and touch imprints) in the high acuity, high complexity, tertiary pdiatrie hospital setting. All intraoperative pathologic consultations for a 10-year period were reviewed and analyzed.

Table 1. Summary of Intraoperative Consultations in a Decade of Pediatric Surgical Pathology Accessions

MATERIALS AND METHODS

All surgical pathology accessions from January 1995 through December 2004 were reviewed from Primary Children's Medical Center, a 232-bed tertiary children's hospital with high acuity and complexity that serves the Intermountain West (Utah, Idaho, Wyoming, and Montana). In 2004, there were more than 11000 inpatient admissions, 152000 outpatient visits, 14000 surgical procedures, and 698 medical staff members. Reports were identified that included FS and intraoperative consultation results. Reports were analyzed to determine whether an FS (intraoperative microscopic consultation including cryostat sections or touch imprints) or intraoperative nonmicroscopic consultation was performed. The number was tabulated for intraoperative consultations that exceeded a turnaround time of 20 minutes from receipt in the surgical pathology laboratory to communicating a verbal report with written documentation. cases in which the reporting time was not documented were considered to exceed 20 minutes.

Indications for FS were analyzed and categorized as FS to evaluate a tumor, FS for possible Hirschsprung disease, or other types of FS. Discrepancies between the FS diagnosis and the permanent diagnosis were categorized as major, defined as having a potentially significant impact on patient care; or minor, defined as lacking a significant impact on patient care. cases in which the final diagnosis was deferred at the time of FS were analyzed to determine the deferral rate and the types of deferrals.

RESULTS

During the 10-year period, 35611 surgical pathology cases were accessioned, and 2839 intraoperative consultations were performed on 1889 accessioned cases. Findings are summarized in Table 1. Of these, 2783 were FSs, and 56 were intraoperative nonmicroscopic consultations. The FS rate was 5% for individual patient procedures (cases) with 1 or more FS and 7.8% for individual FS performed, with an average of 1.5 FSs per case. Turnaround time exceeded 20 minutes in 403 (14%) of the intraoperative consultations. Most frequent indications for FS included questions related to neoplasms (tumor detection, specimen adequacy, triage, classification, or margins in 2165 FSs, 78% of all FSs) or evaluation for Hirschsprung disease in 484 FSs, 17% of all FSs.

A total of 115 discrepancies (4% of all FSs) were identified (Table 2). Seven discrepancies (6% of all discrepancies and 0.2% of all FSs) were classified as major, with a potentially significant impact on patient care. However, in 5 of the 7 cases there was no actual impact on patient care because of timely recognition of the discrepancy between the FS result and the final diagnosis. The types of major discrepancies included tumor classification (3 cases, all low grade), presence or absence of tumor in a margin (1 case), false- positive ganglion cell identification (2 cases), or organism detection (1 case). For the 2 cases of falsepositive ganglion cell identification, one error resulted in the incorrect placement of a colostomy, which was recognized before complications; and the other had the potential for the incorrect placement of a colostomy, without the actual consequence. There were 108 minor discrepancies (94% of all discrepancies and 3.9% of all FSs). Minor discrepancies involved sampling errors in the specimen, block, or slide (14 cases, including 5 with false-negative ganglion cell sampling); reclassification or regrading of benign or malignant neoplasms without clinical consequences (70 cases); recognition of ganglion cells in the transition zone (7 cases, including 4 false-positive cases and 3 false-negative cases) or appendix (1 case) in suspected Hirschsprung disease; or other discrepancies (16 cases). None of the 13 minor discrepancies involving ganglion cell-sampling problems or misidentification resulted in harm, and in most cases, the possibility of transition zone sampling was discussed and managed intraoperatively.

Table 2. Summary of Frozen section Discrepancies

FS diagnosis was deferred in 718 cases, a deferral rate of 25%. A deferred diagnosis with diagnostic reclassification from generic on the FS to specific in the final diagnosis occurred in 685 cases (95% of deferrals), and the generic diagnosis was generally sufficient for clinical managerial purposes. Examples of generic FS deferred diagnoses with a final specific diagnosis based on permanent sections and often requiring ancillary diagnostic tests (such as immunohistochemistry) included: small blue cell tumors (Ewing sarcoma/primitive neuroectodermal tumor, rhabdomyosarcoma, neuroblastoma, an\d medulloblastoma); spindle cell tumors (fibromatoses, infantile fibrosarcoma, synovial sarcoma, and nerve sheath tumors); giant cell-containing lesions of bone; thyroid follicular lesions; germ cell tumors; benign bone cysts; and lymph node biopsies for suspected lymphomas; suspected infections; and evaluation for malformations, such as branchial cleft cysts, or gonadal identification. Uncertainty about the FS diagnosis, with an inability to answer the immediate surgical managerial question, occurred in 33 cases (5% of deferrals). Types of diagnostic uncertainty included the inability to classify a lesion generically for clinical managerial purposes (neoplastic versus nonneoplastic conditions, or subtypes of neoplasms) or problems related to the type or condition of the FS specimen (calcified, bony, necrotic, scanty, crushed, or nondiagnostic tissue samples).

COMMENT

Intraoperative consultations, particularly FSs, are a critical procedure for care of children and adolescents; however, there is a paucity of published literature on this topic. This analysis of a decade of experience with intraoperative consultations in a children's hospital highlights important similarities and differences in intraoperative microscopic consultation use and results in pediatric surgical pathology. The FSs rate of 7% for actual individual FSs (5.7% for patient accessions) and the average of 1.5 FSs per patient case are very similar to rates reported in the literature,3 but are relatively higher when bed size is taken into account. The College of American Pathologists (CAP)1 has set a turnaround time threshold of 20 minutes for a single FS. In the present series, 86% of FSs were completed less than 20 minutes after the receipt of the specimen in the surgical pathology laboratory. This analysis included all FSs, and most cases that exceeded the 20- minute threshold involved multiple simultaneous FSs or FS procedures that occurred at times when no technical assistance was available.

Principal indications for FS this series included tumor evaluation in 78% of the cases and Hirschsprung disease evaluation in 17% of the cases. In their comprehensive study of pediatric FSs in Australia, Preston and Bale15 reported that 40% of FSs were performed for evaluation of central nervous system tumors, 19% involved small round blue cell tumors, and 25% were performed for evaluation of suspected Hirschsprung disease. A more recent series that compared intraoperative cytopathology with FS diagnosis revealed that intraoperative consultations were performed for small round blue cell tumors in 38%, bone tumors in 26%, and lymph nodes in 24% of cases.11 In the present series, other indications for intraoperative microscopic consultations included analysis of developmental cysts, classification of benign bone cysts, identification of maldeveloped gonads, detection of infection, identification of normal tissues or specimen adequacy, evaluation of margins, and assessment of upper or lower respiratory tract mucosal biopsies or brushings for ciliary motility. Mammographically directed breast biopsy specimens, sentinel lymph node analyses, and FS to assess transplant rejection were not encountered. For many of the clinical managerial questions posed at the time of intraoperative consultations in the children's hospital environment, the appropriate triage of tissue is important and a specific intraoperative diagnosis is less of a consideration, because it is understood that ancillary studies are necessary and that a specific diagnosis may not be needed or even requested by the surgeon. These ancillary studies may include cultures for organisms, flow cytometry for cell surface markers or DNA ploidy, cytogenetics, molecular diagnostic tests, or research protocols. Specific information regarding the Children's Oncology Group Protocols is available on their Web site (www.childrensoncologygroup.org).

The comparison of FS diagnosis with the final pathologic diagnosis is an important quality indicator, and both the Association of Directors of Anatomic and Surgical Pathology and the CAP have published suggested accuracy thresholds.1,4 In this series, FS concordance with the final diagnosis was high, with an overall concordance rate of 96% without adjustment for deferred diagnoses. Major discrepancies with a potentially significant effect on patient care accounted for 0.2% of cases, although the majority of these did not have any actual effect on management or outcome. A CAP Q-Probe study of 79647 FSs in 297 institutions yielded a 98.4% concordance rate for all FSs.3 An institutional report of FS diagnoses at the Mayo Clinic for 1 year revealed a 2.2% error rate.2 Specific information about pediatric FS accuracy is limited. In the series by Preston and Bale,15 3.5% of FS diagnoses were inaccurate, and mainly involved classification of central nervous system tumors and evaluation of bile ducts in various pediatric liver diseases. Of note, the intraoperative assessment of bile ductular size is no longer considered necessary for portoenterostomy procedures for extrahepatic biliary atresia. An assessment of 58 cases involving intraoperative cytopathology with or without FS evaluation revealed a 96% accuracy rate in classifying benign versus malignant lesions.11 A summary of FSs performed for pediatric sarcomas in 33 cases revealed a concordance rate of 78%, with a 12% major and 9% minor discrepancy rate.10

Hirschsprung disease is an anxiety-evoking pathologic pitfall for both FS and routine surgical pathologic biopsies. In this series, FS evaluations for Hirschsprung disease were performed both for a leveling colostomy to achieve acute decompression before subsequent definitive excision with anastomosis and for a one-stage surgical treatment or primary resection. Among the 484 Hirschsprung disease FSs in our series, 97% had a complete agreement between the FS diagnosis and the final diagnosis. Only 0.4% had a major discordance caused by ganglion cell misidentification, and although 1 patient experienced an incorrect colostomy placement, early recognition of the error prevented serious medical complications. Minor discordance without a significant clinical effect occurred in 2.6% of cases. Of these 13 minor discrepancies, 8 were caused by biopsy of the transition zone with its irregular ganglion cell distribution; 5 were caused by sampling error, despite multiple levels during the FS procedure; and 1 was caused by performance of an FS on the appendix, which is well known as an unreliable site for evaluation of aganglionosis. The concordance rates for Hirschsprung disease FSs reported by other authors ranges from 89% to 100%.12,13,15 The high level of accuracy reported by Preston and Bale15 occurred in 132 FSs, which were usually performed to select the upper level of resection for previously diagnosed Hirschsprung disease and only occasionally to diagnose or exclude aganglionosis. In another analysis, by Maia,12 of 93 FSs in 80 patients undergoing 132 surgical procedures for Hirschsprung disease, the overall concordance rate was 89%. When the subset of FSs performed on the initial pathologic specimen was separately analyzed, the concordance rate fell to 67%. The discordant cases involved both false-positive and false-negative interpretations for ganglion cells. Thus, there was a 3-fold higher risk of discordance in the interpretation of initial diagnostic biopsies. Challenges in the recognition of ganglion cells included ganglion cell immaturity, differences in the concentration of ganglion cells in the myenteric versus the submucosal plexuses, and artifacts introduced by freezing (Figure 1, A through D). Two principal recommendations were put forth by Maia as a result of her study.12 First, the absence of ganglion cells in a rectal biopsy for possible Hirschsprung disease should be established on well-prepared permanent sections. Second, establishing an initial diagnosis of Hirschsprung disease by FS was not recommended. Another analysis by Shayan of 700 FSs in 304 patients with possible Hirschsprung disease who were evaluated during a 15-year period revealed discordance in 3% of patients.13 Although the error rate in this series was low, the authors emphasized the potentially serious clinical consequences of FS errors, including a need for unexpected surgery or a second operation. Factors that contributed to FS errors included sampling from the transition zone, superficial sectioning of the FS specimen, freezing artifact, technical difficulty in FS preparation, recognition of immature ganglion cells, and inexperience in interpretation of Hirschsprung disease pathologic specimens. In our experience, particular difficulties with Hirschsprung disease FSs are encountered when specimen adequacy is suboptimal; when confounding factors, such as inflammation or a previous biopsy site are present; when unusual subtypes of aganglionosis, such as long segment disease or skip lesions are encountered; or when the transition zone, a markedly dilated bowel segment, or the appendix are sampled (Figure 2, A through D). The distribution of ganglion cells in the transition zone between the aganglionic segment and the normal intestine is variable, and pathologic clues for the FS identification of the transition zone are relatively subtle, including eosinophilic inflammation, irregularly dispersed ganglion cells, and nerve hyperplasia (Figure 3, A and B). The appendix is not a reliable specimen for evaluation for Hirschsprung disease because of the variability in presence and distribution of ganglion cells in both normal individuals and in patients with aganglionosis. If there is a request for FS on an appendix for evaluation of ganglion cells, we discuss the pitfalls with the surgeon and recommend cancellation of the FS. An algorithmic approach to assessment of specimen appropriateness, adequacy, sampling, potential pitfalls and confounding factors, \and requests for second opinions at the time of FS has further contributed to error prevention in our institution.

Figure 1. A, Frozen section of muscularis propria biopsy in suspected Hirschsprung disease reveals small clusters of ganglion cells surrounding small nerve fibers in the myenteric plexus (toluidine blue, original magnification 200). B, The ganglion cells are small and immature, and the toluidine blue stain highlights the cytoplasm (original magnification 400). C, A neural unit in the submucosal plexus is small and contains an aggregate of ganglion cells adjacent to a small nerve fiber (hematoxylin-eosin, original magnification 400). D, A neural unit in the myenteric plexus shows abundant ganglion cells around a small nerve fiber between 2 muscle layers (hematoxylin-eosin, original magnification 400).

The deferral of an FS diagnosis to the permanent or final diagnosis may vary from hospital to hospital depending on specific local indications and how intraoperative microscopic consultations are used.3 In an analysis of the FS deferral rate in a CAP Q-Probe involving 79 647 FSs from 297 institutions, the deferral rate was 4.2% for all FSs, and 92.6% were classified as appropriate and 1.2% as inappropriate deferrals.3 When considering the issue of FS deferred diagnoses, it is important to review the reasons why FSs are requested. These include evaluating specimen adequacy, determining adequacy of excision, making a therapeutic decision, ascertaining whether a lesion is benign or malignant, determining the extent of tumor spread, identifying metastases, establishing a new diagnosis on an unknown disease process, and determining a precise histopathologic diagnosis.16-18 In our series, nearly all deferred FS diagnoses were appropriate. The deferral rate of 25% reflects how the deferred diagnosis was defined. For example, with a generic FS diagnosis of a "small blue cell tumor," the final pathologic diagnosis of Ewing sarcoma would have been classified as an appropriately deferred diagnosis (Figure 4, A through D). If such a tumor involved the central nervous system, particularly the posterior fossa, the differential diagnosis would include medulloblastoma and atypical teratoid/malignant rhabdoid tumor, and the distinction between these neoplasms is not always possible at the time of FS, because of histologic heterogeneity and the importance of ancillary immunohistochemical and molecular genetic tests. Spindle cell neoplasms present a similar challenge, with a morphologic spectrum encompassing myofibroma, other fibromatoses, infantile fibrosarcoma, rhabdomyosarcoma, and other spindle cell sarcomas. It is possible to make an FS diagnosis in cases with classic histologic features and fortuitous sampling at the time of FSs; however, in other cases, deferral is necessary to allow more extensive histopathologic examination and ancillary testing. Of the deferred diagnoses in our series, 95% involved a generic classification at FS that was refined to a specific diagnosis on permanent section, often with the use of ancillary techniques. Only 5% of deferred FSs yielded results that could not address the clinical managerial question for which the FS was requested. This compares with the reported deferral rate of 4.2% in the CAP Q-Probe when the definitional differences are taken into account.3 This finding prompts reconsideration of the definition for FS deferred diagnoses to more accurately reflect the relevance of the FS diagnosis to the type of clinical or surgical managerial question being asked intraoperatively and to the indication for FS. The importance of close communication among surgeons, pediatric oncologists, pediatric radiologists, and pathologists cannot be overemphasized. Frequently, this communication occurs before and during the intraoperative consultation and contributes to a better understanding of how the information from the FS will be used. Clinical decisions about staging with bone marrow biopsies and aspirates, line placement for chemotherapy, other pharmacotherapy, and further surgery may all be influenced by the FS result.

Figure 2. A, Full thickness biopsy of classic Hirschsprung disease shows absence of ganglion cells in both the submucosal and myenteric plexuses associated with hypertrophic nerve fibers (hematoxylin-eosin, original magnification 100). B, Hypertrophic nerve fibers in the region of the myenteric plexus are associated with absent ganglion cells in classic Hirschsprung disease (hematoxylin-eosin, original magnification 400). C, in long-segment Hirschsprung disease, there is an absence of both ganglion cells and hypertrophic nerve fibers in both plexuses (hematoxylineosin, original magnification 40). D, In long-segment Hirschsprung disease, the absence of ganglion cells is associated with flattening of the 2 layers of muscularis propria against each other without intervening tissue or nerve fibers (hematoxylin-eosin, original magnification 400).

Figure 3. A, The transition zone in Hirschsprung disease is characterized by a sparse inflammatory infiltrate including eosinophils and hytpertrophic nerve fibers (hematoxylin-eosin, original magnification 200). B, Ganglion cells are irregularly and sparsely distributed in the transition zone, and the presence of an inflammatory infiltrate with eosinophils is a further clue to the location of the biopsy in the transition zone (hematoxylin-eosin, original magnification 400).

Figure 4. A, An intraoperative touch imprint reveals a small blue cell neoplasm without further distinguishing features (hematoxylin- eosin, original magnification 400). B, A section of a small blue cell tumor that was diagnosed as Ewing sarcoma shows sheets of round to oval, small- to medium-sized cells with scant pale cytoplasm and areas of individual cell necrosis (hematoxylin-eosin, original magnification 200). C, Immunohistochemistry for CD99 (O13) reveals a characteristic diffuse pattern of membranous reactivity in Ewing sarcoma (original magnification 200). D, Immunohistochemistry for Fli-1 shows diffuse nuclear reactivity in Ewing sarcoma (original magnification 400).

In conclusion, these findings underscore important differences in indications for and in pathologic processes encountered in pediatric FS practice. The spectrum of neoplasms and types of questions posed during intraoperative consultations differs significantly from the diseases and situations encountered in the adult or general hospital setting, and evaluation for Hirschsprung disease accounts for a relatively higher proportion of intraoperative consultations. These findings emphasize the importance of specific education in pediatric surgical pathology to focus on neoplasms and other disorders that may be disproportionately encountered in pediatric care.

References

1. Nakhleh RE, Fitzgibbons PL. Quality Improvement Manual in Anatomic Pathology. 2nd ed. Northfield, Ill: College of American Pathologists; 2002.

2. Ferreiro JA, Myers JL, Bostwick DC. Accuracy of frozen section diagnosis in surgical pathology: review of a 1-year experience with 24,880 cases at Mayo Clinic Rochester. Mayo Clin Proc. 1995;70:1137- 1141.

3. Zarbo RJ, Hoffman CG, Howanitz PJ. Interinstitutional comparison of frozen-section consultation: a College of American Pathologists Q-Probe study of 79,647 consultations in 297 North American institutions. Arch Pathol Lab Med. 1991;115:1187-1194.

4. Association of Directors of Anatomic and Surgical Pathology. Recommendations on quality control and quality assurance in anatomic pathology. Am J Surg Pathol. 1991;15:1007-1009.

5. Krishnan B, Lechago J, Ayala C, et al. Intraoperative consultation for renal lesions: implications and diagnostic pitfalls in 324 cases. Am J Clin Pathol. 2003; 120:528-535.

6. Golouh R, Bracko M. Accuracy of frozen section diagnosis in soft tissue tumors. Mod Pathol. 1990;3:729-733.

7. Burger PC, Vogel FS. Frozen section interpretation in surgical neuropathology. Am J Surg Pathol. 1977;1:323-347.

8. Burger PC, Vogel FS. Frozen section interpretation in surgical neuropathology, II: Intraspinal lesions. Am J Surg Pathol. 1978;2:81- 95.

9. Burger PC. Use of cytological preparations in the frozen section diagnosis of central nervous system neoplasia. Am J Surg Pathol. 1985;9:344-354.

10. Fisher JE, Burger PC, Perlman EJ, et al. The frozen section yesterday and today: pediatric solid tumors-crucial issues. Pediatr Dev Pathol. 2001;4:252-266.

11. Wakely PE, Frable WJ, Kornstein MJ. Role of intraoperative cytopathology in pediatric surgical pathology. Hum Pathol. 1993;24:311-315.

12. Maia DM. The reliability of frozen-section diagnosis in the pathologic evaluation of Hirschsprung's disease. Am J Surg Pathol. 2000;24:1675-1677.

13. Shayan K, Smith C, Langer JC. Reliability of intraoperative frozen sections in the management of Hirschsprung's disease. J Pediatr Surg. 2004;39:1345-1348.

14. Hamoudi AC, Rutledge JC, Novak RW, et al. Quality improvement approaches: survey of pathologists serving the pediatric patient. Arch Pathol Lab Med. 1994;118:165-167.

15. Preston HS, Bale PM. Rapid frozen section in pediatric pathology. Am J Surg Pathol. 1985;9:570-576.

16. Horn RC. What can be expected of the surgical pathologist from frozen section examinations? Surg Clin North Am. 1962;42:443- 454.

17. Ackerman LV, Ramirez CA. The indications for and limitations of frozen section diagnosis; a review of 1269 consecutive frozen section diagnoses. Br J Surg. 1959;46:336-350.

18. Sawady J, Berner JJ, Siegler EE. Accuracy of and reasons for frozen sections: a correlative, retrospective study. Hum Pathol. 1988;19:1019-1023.

Cheryl M. Coffin, MD; Krista Spilker, BA; Holly Zhou, MD; Amy Lowichik, MD, PhD; Theodore J. Pysher, MD

Accepted for publication August 11, 2005.

From the Division of Pediatric Pathology, Department of Pathology, Primary Children's Medical Center and the University of Utah, Salt Lake City.

The authors have no rel\evant financial interest in the products or companies described in this article.

Portions of this material were previously reported and presented in abstract form at the United States and Canadian Academy of Pathology Meeting, San Antonio, Tex, March 1, 2005. The abstract was published in Modem Pathology. 2005;18(suppl):322A. Abstract 1498.

Corresponding author: Cheryl M. Coffin, MD, Department of Pathology, Primary Children's Medical Center, 100 N Medical Dr, Salt Lake City, UT 84113 (e-mail: cheryl.coffin@ihc.com).

Reprints not available from the authors.

Copyright College of American Pathologists Dec 2005

Source: Archives of Pathology & Laboratory Medicine

More News in this Category