Pathologic diagnosis in head and neck practice: how fast is fast enough?

Kenneth O. Devaney 0 1 Alessandra Rinaldo 0 1 AlWo Ferlito 0 1 0 A. Rinaldo A. Ferlito (&) Department of Surgical Sciences, ENT Clinic, University of Udine , Piazzale S.Maria della Misericordia, 33100 Udine, Italy 1 K. O. Devaney Department of Pathology, Allegiance Health , Jackson, MI, USA - Diagnosis of pathologic specimens has become faster over the past decade or so; nevertheless, diVerences in the speed of diagnosis remain. This article asks a simple question: just what is a reasonable expectation for the turnaround time for a routine pathologic diagnosis? Should it be measured in weeks, days, hours or minutes? There was a time, not awfully long ago, when the rendering of a pathologic diagnosis could be a relatively leisurely aVair: the time from the receipt of a tissue specimen to a Wnal written pathology report could span the range from a few days to a few weeks (in the case of particularly puzzling specimens). Those days are gone forever. Today, there are large commercial laboratories promising same-day service for many types of pathology specimens, while teaching hospital laboratories are somewhat slower. Is one approach better than the other? The gold standard-frozen section Mayo Clinic in Rochester, Minnesota, has raised the process of rendering lightning-fast surgical pathology diagnoses to a real art form [13]. Surgeons removing tumors in their operating rooms can expect a relatively comprehensive pathologic diagnosis within 10 or 15 min in most instances, courtesy of frozen section examination. While permanent sections are examined the following day for conWrmation of the original frozen section diagnosis, as a practical matter a thorough pathologic examination is reported and available in the patients medical record before the patient leaves the operating room to go to the recovery room. As attractive as this process is, it is unlikely to be reproduced widely. One substantial hurdle is posed by the very nature of frozen section examination; and that is its artifactual overlay. As in microscopic images, sections prepared by frozen section are by their very nature less clear than are permanent sections, that is, a trade-oV is accepted between image quality and speed of diagnosis. In return for some diminution of image clarity, Mayo Clinic pathologists return diagnoses within a matter of a very few minutes. As appealing as this is, there are some very real hurdles to extending this practice everywhere. There is a technical dimension: Mayo uses a particular specialized type of freezing microtome, a diVerent piece of equipment from the conventional cryostats employed in most hospital laboratories today. The widely used cryostats freeze tissue a little more slowly than the Mayos freezing microtome and are subject to some artifacts that are not as frequently encountered with the freezing microtome (this latter instrument, for example, is better able to handle adipose tissue than is the widely available cryostat). However, the freezing microtome, with its enclosed refrigerant, is a highly specialized piece of equipment, both expensive to construct and expensive to operate and particularly labor intensive in its daily usage. There is a professional limitation as well: Mayo Clinic pathologists train over the years to interpret their frozen section specimens. By contrast, pathologists elsewhere are grounded in the daily interpretation of permanently prepared images for most of their diagnostic work. This means that frozen section interpretation often becomes something in which they do not spend the bulk of their time practicing, and so are not as experienced as are Mayo Clinic pathologists when it comes to interpreting frozen section specimens. There are, Wnally, some potential limitations to the freezing of the tissue itself. Ancillary techniques for diagnosis and prognosis, such as immunohistochemistry or molecular studies, have in the main been reWned for application to routinely Wxed tissues. It is true that frozen tissue methods have been developed for use with many of the antibodies used in immunohistochemistry, and so this is not an insurmountable problem. Nevertheless, an abrupt shift to frozen section processing of tissues would necessitate a re-evaluation of procedures for the ancillary usage of tissues for a wide range of other uses, which would likely not be a trivial matter. When Mayo Clinic expanded beyond Minnesota to Arizona and Florida in the 1980s, it became apparent that it was actually very diYcult to reproduce the frozen section procedures at work in Rochester in those two sites: the expense of recreating the original procedures and practices proved to be prohibitive. This, in turn, suggests that replacing current conventional pathologic practices with Mayo Clinic Rochester-style frozen section practices would not be a practical means of speeding the pathologic diagnostic process along. The present options The processing of human tissue for pathologic diagnosis ordinarily begins with Wxation, a process that halts cell death and dissolution. This process of Wxation requires some Wnite amount of time, usually measured in hours; larger portions of tissue require longer periods of Wxation, while tiny portions of tissue are Wxed much more rapidly. After Wxation, this same tissue must be made hard enough to cut into exceedingly thin sections that will transmit light. This process of embedding, which usually rests upon inWltration of the tissue with melted paraYn, also requires a period of several hours. Finally, after cutting these thin sections, the sections must be transferred to individual glass slides, stained and coverslipped. Taken together, the time from the initial dissection of the specimen to the production of routinely stained glass slides may range from 9 to 15 h. What can be done to accelerate this process? For one thing, the production process could be started earlier in the day, so as to produce stained slides at midnight, or at 3 a.m. As well, surgical pathology laboratories could be staVed on weekends and holidays. This may not be such a boon, of course, there are few clinicians to be found in their oYces at midnight or 3 a.m., or on a Sunday or a holiday, and so this sort of solution does not seem likely to win too many supporters. Another approach would be to experiment with an alternative to the time-honored tradition of batch processing specimens, in which the entire days work is processed by an automated process at the end of the work day, after the workers have left for the day. An alternative, then, might be to construct a continuous system for processing specimens as they are received, without the need for waiting for a full days batch to accumulate. As it happens, manufacturers are beginning to market instruments designed to do just that: produce slides via a continuous throughput process; this, in turn, is coupled with microwave-assisted tissue processing, which serves to accelerate the overall process yet (...truncated)