Trevor Starnes, M.D., Pathology Resident
Dorina Gui, M.D., Ph.D., Director of Surgical Pathology
John Bishop, M.D., Director of Anatomic Pathology



The optimal use of intraoperative consultation, also known as frozen sections (FS), requires thoughtful collaboration between the surgical and pathology teams. A better understanding of the FS can improve its utilization.

The purpose of the frozen section is to answer a specific question that has an immediate impact on the operative and perioperative care of the patient [1]. The FS diagnosis should provide enough information to inform the patient's management, but should not go too far in making a specific diagnosis that may not be correct on final pathology. This occurs partly because FS involves limited sampling and histology that is usually less clear than permanent sections. Special studies, such an immunohistochemistry or special staining, are not available for frozen intraoperative consultation, and this also can make diagnosis less specific.

For the pathology team, an FS is an emergent, highly time-sensitive request. The pathology team must abruptly defer whatever they were doing to make sure the FS is taken care of in a timely and accurate manner. Ideally, they have been able to anticipate the need for FS through their understanding of the needs of specific surgical procedures, or communication from the surgical staff, and so are accordingly well-prepared [2].

The basic FS process is that a tissue specimen ranging from a small biopsy to multiple organs is received from the operating room with a specific question to be answered within about 20 minutes. Particularly for larger specimens, an examination of what can be seen macroscopically is also critical. The samples are often painted with ink (before sampling) that can be seen under the microscope so that pathology can keep track of where things are, e.g. a surgical margin where the other side of the cut tissue remains in the patient. Samples of the tissue are chosen to be reviewed under the microscope, up to approximately an inch squared (2.5 cm) and an eighth of an inch (0.3 cm) thick at a time. These samples of tissue are frozen in a gel and are referred to as “blocks”, and these can then be cut in a cryostat to produce very thin sections to place on a slide, stain, and view under the microscope.

The most common use of intraoperative consultation is the evaluation of tumor margins to determine if additional tissue must be removed. Other uses of FS include the identification of lymph node metastases, the characterization of the nature of a lesion (benign or malignant, sometimes specific diagnoses), diagnostic adequacy of tissue, and evaluation for transplant [1].


Communication and Sampling

Like all complicated collaborations, the FS works best when all parties are proactive about communication. Here are some illustrations of why that is important:

*Particularly for small specimens where the pathologist can glean minimal information from the tissue macroscopically (grossly), it's important to keep in mind that identical microscopic findings in two separate anatomic locations can mean very different things. For example, seeing bland-looking glands in a place where glands do not belong could raise the possibility of a well-differentiated metastatic adenocarcinoma, but would be normal if these glands are normally present in the area sampled. This means that clear and specific descriptions of where a biopsy was taken from are important.

*The pathologist must understand the clinical history. For example, the pathologist should know whether the patient had received prior radiotherapy as this can affect how the cells look. In practice this means the pathologist must take the time to understand this information and the surgeon's notes or other communication should include information relevant to the pathologist such as prior treatments, biopsies, imaging results, prior malignancies, etc.

*An important communication-related issue is margin status. The surgeon can help the pathologist with the timeliness and accuracy of orientation with the communication of the need for margin status evaluation along with unambiguous indicators of orientation. Common methods include different suture lengths, suture colors, and number of clips. What's important is that it is easy to interpret and clearly communicated. For most tumor types, FS on margins is accurate. For margins in gastrointestinal malignancies, the accuracy of FS is approximately 98%, with lower accuracy in diffuse tumors or duodenal specimens [3]. A notable exception to the utility of FS to assess margins is in melanoma, where there is very poor agreement between FS and permanent sections [4].

*For larger specimens, the amount of tissue analyzed under the microscope represents a small fraction of the overall tissue which may itself be distorted by disease. A clearly communicated question to be answered by FS will help target the pathology team to the appropriate lesion in larger samples, and increase the effectiveness of sampling at FS. This will be even more critical for unusual questions applied to specimen types in which pathology is normally looking for something else.

*While FS may sometimes be seen as objective and unambiguous clinical data, in some situations the pathologist must communicate a degree of uncertainty in their diagnosis at FS, and weigh the consequences of a false negative or false positive result. The pathologist should have a basic understanding of the treatment algorithm that they are influencing with their FS diagnosis. When faced with diagnostic uncertainty, it is also the best practice for the pathologist to seek a second opinion if it can be obtained in a timely way, particularly from a pathologist with subspecialty expertise in the specimen type. If an unacceptable level of uncertainty cannot be resolved, it is appropriate for the pathologist to make a deferral so that no frozen section diagnosis is given, but rather the diagnosis will be clarified by permanent sections [1].

*A situation in which pathology will often need to call the surgical staff is during a shift change, e.g. the transition to after normal business hours. The purpose of this call is to make sure that pathology can adequately prepare for a FS as we are starting our on-call shift after a day of doing other duties. If we can receive adequate (e.g. 20 minutes) of warning for an otherwise unexpected frozen section, we can check the basic clinical history and have the frozen room ready to receive the FS specimen to avoid delays in FS diagnosis.

*Miscommunication is a risk in FS result reporting and could have substantial consequences. Therefore, pathologists who communicate diagnoses should receive a read-back to ensure that their diagnosis is understood by the surgical team.


Artifacts and Technical Issues

The FS has limitations and is not equivalent in quality to formalin-fixed, paraffin-embedded (permanent) sections. Artifacts in the context of FS are abnormalities seen microscopically that do not directly reflect an abnormality in the tissue and can make interpretation difficult.

*Freezing the tissue introduces ice crystal artifact. Ice crystals can alter the sizes of the nuclei, and tear holes in membranes that look like vacuoles, fat, or inclusions. The proper temperature and freezing rate of a specimen is a fine balance. This artifact is hard to avoid entirely, but is more likely if too large of a specimen is placed onto a slide given a longer time required for freezing to take place. Particularly if all lesional tissue is frozen, the clarity of the permanent section diagnosis may be compromised by the ice crystals. Thus, it is important for pathology and surgical team to communicate to ensure that lesional tissue is available for permanent sections when possible.

*Adipose tissue is particularly challenging to cut for a frozen section and the slides are of lower quality per time spent than other types of tissue.

*Another artifact is shatter artifact in which the tissue looks like venetian blinds. This can occur due to the sample getting too cold during freezing, or when calcified tissue is cut. As a general rule, tissue that is too hard to cut with a scalpel is suboptimal for FS analysis.

*Air drying artifact causes cells to have indistinct borders and smudgy chromatin. This can occur if there is a delay between taking a section of tissue and fixing it [1].

While pathologists are trained to deal with these artifacts, it is a potential source of challenges for interpretation.


Quality Improvement and Assurance

In one large study, the error of FS was approximately 5%. Approximately half of these errors were due to microscopic misclassification [5].

The special challenges of frozen sections, and the resultant potential for errors, can be at least partially addressed by continuous quality improvement and quality assurance measures that many institutions such as UC Davis have implemented. For example, frozen section slides are re-reviewed in conjunction with the permanent slides to track agreement and determine where systemic or individual improvements can be made. It is important to have a mechanism such as this in place to identify and examine discordant results and minimize future errors.


Key Points

*Frozen sections are done when the result can impact operative or perioperative care.

*Effective communication between the surgical and pathological personnel is critical to optimal patient care in frozen section analysis.

*Frozen sections are not equivalent to permanent sections in certainty or specificity of diagnosis.



  1. Lester et al. Diagnostic Pathology Intraoperative Consultation, Second Edition. Salt Lake City, UT: Elsevier, Inc., 2018.
  2. Taxy JB. Frozen Section and the Surgical Pathologist: A Point of View. Archives of Pathology & Laboratory Medicine: July 2009, 133(7): 1135-1138.
  3. McAuliffe JC, Tang LH, Kamrani K, Olino K, Klimstra DS, Brennan MF, Coit DG. Prevalence of False-Negative Results of Intraoperative Consultation on Surgical Margins During Resection of Gastric and Gastroesophageal Adenocarcinoma. JAMA surgery. October 2018.
  4. Prieto VG, Argenyi ZB, Barnhill RL, Duray PH, Elenitsas R, From L, Guitart J, Horenstein MG, Ming ME, Piepkorn MW, Rabkin MS, Reed JA, Selim MA, Trotter MJ, Jonhson MM, Shea CR. Are en face frozen sections accurate for diagnosing margin status in melanocytic lesions? Am J Clin Pathol. 2003;120(2):203–8.
  5. Sams SB, Wisell JA. Discordance Between Intraoperative Consultation by Frozen Section and Final Diagnosis: A Classification Model to Guide Quality Improvement. International Journal of Surgical Pathology: February 2017; 25(1).