The article by Ferreiro and associates in this issue of the Mayo Clinic Proceedings (pages 1137 to 1141) succinctly presents several meaningful aspects about the intensive frozen section practice in surgical pathology at the Mayo Clinic. It describes a unique experience with an approach to intraoperative consultation that is not practiced elsewhere. This study is neither a rigorous evaluation of the accuracy of this approach nor, as stated in the report, a careful evaluation of the efficiency of this practice in comparison with other practices elsewhere. Nonetheless, this is a clear exposition of a unique practice pattern.
Routine Use of Frozen Sections
Simply stated, at the Mayo Clinic, an assessment is usually completed and even a final diagnosis is rendered intraoperatively in most cases, except for certain problems such as hematopoietic malignant lesions. In distinct contrast, at most other institutions, frozen sections are requested in a relatively small number of operative cases to provide an answer to a specific question posed by the clinician.
1
Most pathologists believe that frozen sections have more technical and sampling limitations than do permanent, fixed sections.2
Many of these limitations are overcome with extensive technical and professional resources committed to the intraoperative setting, as in the Mayo Clinic Department of Laboratory Medicine and Pathology. The intraoperative capability for consultation based on frozen section technology at the Mayo Clinic seems amazing to most other pathologists, who provide permanent sections of all tissue specimens and rely on these permanent, paraffin-embedded sections in most cases. In contrast, at the Mayo Clinic, only frozen section processing is used in some cases. The surgeons elsewhere are often expected to exercise judgment and to ask unambiguous questions related to an intraoperative decision. Developing technology may eliminate some of the restrictions on the frozen section situation, including the ability to perform some immunoperoxidase evaluations for specific cell markers within the imposed time constraints.Evolution of Role
Intraoperative consultations that involve the preparation of rapid frozen sections for intraoperative pathologic diagnosis are central to the practice of surgical pathology and have been so considered for almost 100 years. As mentioned recently by Wick,
2
however, this role developed gradually over a prolonged period.3
Until approximately 2 decades ago, most frozen section services used techniques similar to those described in this article, with metachromatic or monochromatic water-based stains such as toluidine blue; these techniques remain useful in many practice settings. Most practices, however, have adopted hematoxylin-eosin staining, a procedure that may involve 1 or 2 extra minutes but results in a stable section that lasts indefinitely. These frozen sections are stained in the same way as permanent sections; thus, most pathologists who prepare frozen sections have a ready correlation between the two.Diagnostic Accuracy
The evaluation of accuracy in the current report is not rigorous. Indeed, using so many routine specimens means that many cases have straightforward diagnoses. In most other reports of frozen section experience, only a small percentage of all surgical procedures prompt frozen section analysis because of a specific question about a diagnosis. In these hospitals, the denominator differs so drastically that comparison with Mayo data is difficult. Nevertheless, with experienced use of frozen sections, the number of errors is indeed small. Furthermore, review of the list of the few cases that are thought to involve errors in the report by Ferreiro and colleagues clearly shows that many are unimportant. For example, whether the documentation of a nodule of remaining melanoma in a wide excision is made intraoperatively or later is probably immaterial.
Economic Efficiency
Another issue of general interest is the economic and professional efficiency of these practices. We agree with these authors that this study neither confirms nor establishes the economic efficiency of the Mayo approach. The evaluation of these diagnostic practices for answers to questions about economic efficiency may have to rest on more specific subset analysis and specific issues such as precision in positive and negative diagnoses of thyroid lesions,
4
frequency of missing a low-grade ovarian carcinoma because of sampling errors,5
, 6
sensitivity of lymph nodal evaluation;7
, 8
or assessment of inflammation in hip joints.9
Practice patterns may simply differ from one institution to another, depending on the nature of the organization and the types of patients who receive medical care.Inappropriate Application
Recently, considerable discussion has been generated about the inappropriate use of frozen sections in the setting of a grossly benign breast biopsy specimen without a dominant mass (or without a mass greater than 1 cm in diameter).
10
, 11
, 12
, 13
We support the idea that a frozen section is inappropriate in most such practice settings and will always involve irreparable damage to some percentage of small epithelial lesions of the breast.As all of medicine evolves, continued scrutiny of these matters is warranted. One of the questions approached by Weiss and coworkers
14
relative to practice patterns for use of frozen sections is the extent of the requesting surgeon's insight and understanding of the process involved. We fear that the removal of surgical residents from a rotation in surgical pathology—a trend in recent years—will cause further deterioration of understanding and escalate the need for surgeons and surgical pathologists to maintain two-way communication. Weiss and associates14
stated that the American Board of Surgery and other specialty boards have de-emphasized the postgraduate experience in pathology. Accordingly, many newly trained surgeons lack understanding about appropriate utilization of frozen sections.Wick,
2
in response to the report by Weiss and colleagues,14
reminded us that occasionally surgeons demonstrate an autocratic and unreasonable demand for immediate information, which must be moderated by the surgical pathologist with knowledge of the intraoperative setting and an understanding of what the surgeon actually needs. Wick2
also supports the continued use, in most practice settings, of clear indications for requesting a frozen section—that is, with a “clear-cut purpose or plan to use the results.” Understandably, this “less definitive than permanent section diagnosis” should usually have a specific question to be answered that justifies the inordinate effort mobilized in most settings.1
Weiss and coworkers14
and Wick2
also noted that the use of frozen sections to establish a diagnosis in a setting in which outside material already purportedly provided a diagnosis is an unusual but increasing phenomenon. In addition, a substantial percentage of frozen sections are done to satisfy the curiosity of the family and to avoid a delay of 24 hours in knowing the diagnosis—an indication that many of us would question. The importance of rapid diagnosis of a lymph node dissection interpreted as negative is debatable when it is significantly counterbalanced by the small percentage of patients who will later be told that microscopic metastatic involvement is present. Ferreiro and associates cite this as one of the most common problems in their practice.Conclusion
Vive la difference and the opportunity to discuss divergent practices in various settings. The fact that this experience is unique to the Mayo Clinic is astounding when one considers the number of persons trained in surgery and pathology at the Mayo Clinic who have ultimately had distinguished careers elsewhere. The lack of emigration of this technique bespeaks the special nature of practice at several levels at Mayo. Thus, the conclusion of this editorial evaluation is brief: practices at the Mayo Clinic cannot be compared and contrasted with those elsewhere because they are so different.
REFERENCES
- The indications for and limitations of frozen section diagnosis: a review of 1269 consecutive frozen section diagnoses.Br J Surg. 1959; 46: 336-350
- Intraoperative consultations in pathology: a current perspective [editorial].Am J Clin Pathol. 1995; 104: 239-242
- The life of Joseph Colt Bloodgood, M.D., public surgeon.Surg Gynecol Obstet. 1993; 177: 193-200
- Role of frozen section and clinical parameters in distinguishing benign from malignant follicular neoplasms of the thyroid.Am J Surg. 1992; 164: 603-605
- Accuracy of frozen-section (intraoperative consultation) diagnosis of ovarian tumors.Am J Obstet Gynecol. 1994; 171: 823-826
- The accuracy of frozen section in the diagnosis of ovarian neoplasms.Gynecol Oncol. 1991; 43: 61-63
- Prognostic factors in men with stage Dl prostate cancer: identification of patients less likely to have prolonged survival after radical prostatectomy.J Urol. 1994; 152: 1077-1081
- Frozen section detection of lymph node metastases in prostatic carcinoma: accuracy in grossly uninvolved pelvic lymphade-nectomy specimens.J Urol. 1986; 136: 1234-1237
- Frozen histologic section as a guide to sepsis in revision joint arthroplasty.Clin Orthop. 1994; 304: 229-237
- Accuracy and reliability of frozen section diagnosis in a series of 672 nonpalpable breast lesions.Am J Clin Pathol. 1995; 103: 199-205
- Immediate management of mammographically detected breast lesions.Am J Surg Pathol. 1993; 17: 850-851
- Frozen section examination of breast biopsies: practice parameters.Am J Clin Pathol. 1995; 103: 6-7
- Frozen-section diagnosis of breast biopsy specimens: a necessary procedure?.Arch Surg. 1993; 128: 955-956
- Frozen section consultation: utilization patterns and knowledge base of surgical faculty at a university hospital.Am J Clin Pathol. 1995; 104: 294-298
Article info
Identification
Copyright
© 1995 Mayo Foundation for Medical Education and Research. Published by Elsevier Inc. All rights reserved.
