Biopsies and frozen sections are particularly dependent on accurate clinical information and correlation (Table 6.2). When evaluating small biopsies, the pathologist must consider the size of the lesion in order to account for lesional heterogeneity. The location of biopsy sampling relative to other structures or organs may provide clues to contaminating tissues and/or adequacy of sampling. Deep-seated lesions, for instance, may have a “rind” of reactive prominent tissue reaction [1]. For core-needle biopsies of the breast, pathologic–radiologic correlation is essential in every case, and every attempt should be made to resolve any discrepancies. For biopsies taken for calcifications, this may include deeper levels, polarization for calcium oxalate, and obtaining radiographs of the paraffin blocks . In cases of a mass lesion, this may include deeper levels [28]. In a 1997 CAP Q-Probes study [29] of breast needle-core biopsies, participants reported that only 45 % of cases had a radiographic report submitted to pathology, and only 62 % of pathology reports documented correlation between histology and mammography. A more recent Q-Probes study [30] of breast biopsies revealed that despite continued emphasis on mammographic correlation, radiologic images were reviewed in only 22 % of cases, and radiology reports were reviewed in only 48 % of cases. Furthermore, in cases with noncorrelation , 46 % of those pathology reports did not indicate there was a lack of correlation , and 46 % also did not indicate what additional steps were taken in evaluating the specimen . The adequacy of clinical information or provided radiologic information was not specifically addressed in the study.

Clinical information and operative findings can also aid in identifying the biopsy specimen mix-ups. Specific information regarding the character of the sampled lesion can be very helpful. For example, a bladder biopsy sampling from an area of mucosal erythema should not have the same histologic appearance as sampling from a gross papillary lesion. Likewise, sampling of polyps in the gastrointestinal tract should look different from flat mucosal sampling. Specific information regarding the location of the sampled lesion can also be helpful. In the breast, a sampling from the subareolar region or nipple complex should contain bundles of muscle, and thus, appear histologically distinct from sampling in the more peripheral breast. Any noncorrelations must be investigated in order to rule out specimen mix-ups.

Intraoperative frozen section interpretation requires a close interaction between the pathologist and the surgeon [31]. Appropriate indications for frozen section fall into three categories: (1) to establish the presence or nature of a lesion, (2) to determine margin adequacy, (3) to determine if adequate lesional material is present and if additional sampling is warranted. [1] Frozen section should not be performed to simply satisfy curiosity, to identify normal structures, or to allow expedient communication with the patient and/or patient’s family members. [1] Knowledge of the reason and the purpose for frozen section is necessary for appropriate handling of the tissue, and potentially to avoid unnecessary attempts at diagnostic overcommitment by the pathologist. Each organ site has unique indications for frozen section and unique limitations to consider, and the reader is referred to standard texts for a thorough discussion. Inappropriate frozen sections may risk loss of small lesions through rough facing of the block (e.g., focal findings such as giant cells in a temporal artery biopsy), or may compromise margins and staging elements. Mishandling may also hinder the permanent section evaluation (e.g., sectioning a lung wedge taken for interstitial disease, rather than inflating the intact specimen with formalin). Intraoperative triage of material for ancillary studies can be vital in appropriate surgical pathology workups, and this process is aided by accurate clinical information. For example, some lymph node biopsies may require only routine permanent sections (e.g., rule out metastatic disease), but other cases may require flow cytometry (e.g., lymphoma workup), or cultures for infectious disease. Cytogenetic studies may be indicated for a number of specimen types, ranging from placenta to bone marrow. Inappropriate triage of tissue may result in loss of vital ancillary study information.

Specific organ types have other requirements for clinical information and clinical correlation, which are detailed in appropriate texts. Only a few organ sites are mentioned herein. Musculoskeletal pathology is dependent on imaging studies and clinical findings, as nicely summarized by Kilpatrick [32]. In some instances, histologically identical soft tissue neoplasms have different names and different prognoses depending on specific anatomic location or depth (e.g., lipomatous tumors). In the liver, clinical and laboratory data is essential in narrowing the differential diagnosis toward a specific cause [33]. For bone marrow evaluation, necessary information may include duration of symptoms, physical findings, occupation, travel and exposure history, and family/personal medical history, and the marrow biopsy should be correlated with the complete blood count, peripheral smear, and aspirate smears. Ideally, an integrated report is generated that includes not only these elements, but also any flow cytometry findings, and any follow-up ancillary studies [34]. Neuropathology biopsy workup also relies on clinical and radiological features such as age/gender, duration and tempo of disease, character of signs and symptoms, prior diagnoses, prior radiation or other treatment, travel or residence in endemic areas, known systemic disease (e.g., immune compromise), and family history. Radiologic characteristics, such as circumscription, infiltration, location, and enhancement features, can all be helpful. The imaging can also aid in determining if the sampling is representative during frozen section evaluation [35]. Investigation of interstitial lung disease requires communication between the clinician, radiologist, and pathologist, and consideration of all factors aids in planning an appropriate biopsy approach (i.e., transbronchial biopsy versus wedge biopsy) [36]. Some histologic patterns of lung disease require that the clinician go back and reassess the patient for occupational exposures, potential sources of hypersensitivity and systemic diseases such as collagen vascular disease. In such cases, a histologic differential diagnosis may be helpful to the clinician.

The list of potential barriers to effective communication of clinical information is lengthy, and includes clinicians, pathologists, technology, logistical issues, and at times, the patients themselves. The easy portability of many surgical pathology specimens, particularly small biopsies, allows easy physical separation from their points of origin. This minimizes face-to-face interaction with the receiving laboratory and may be a barrier to communication. The increasing mobility of patients, both geographically and among local healthcare systems, may make the clinician’s construction of an accurate clinical history very difficult. For example, a breast cancer patient may have her core biopsy performed in a radiology center, receive neoadjuvant therapy at one hospital, and then undergo subsequent excision at a second hospital. Consequences for such a patient might include inaccurate diagnosis, multiple contradictory diagnoses, and duplication of studies. Technology, including laboratory information systems , radiology information systems, and electronic health records (EHRs), can be important sources of clinical information but are limited by interfacing capabilities, ease of use, by the accuracy of the information entered, and the personnel using them. Computerized provider order entry (CPOE) may be complex enough that the physician relies on ancillary personnel to enter vital information, potentially resulting in the entry of inaccurate information. Even when adequate technology is available, the surgical pathologist must be motivated to make use of it, and to construct integrated reports that reconcile all available information.

Perhaps the most important barrier is a combination of overconfidence in the capabilities of the specialty of surgical pathology , and an apparent growing lack of knowledge of pathology among newly trained clinicians. Due to technological changes and budgetary constraints, many medical schools now employ virtual laboratory exercises rather than traditional hands-on microscope work, and clinical resident rotations through pathology have all but disappeared. Consequently, clinical resident pathology exposure is generally limited to review of “pre-packaged” surgical pathology cases at tumor boards. The words of O. N. Rambo [5] still ring true a half-century later:

…because the teaching of pathology used to be relegated primarily to the long-forgotten pre-clinical phase, pathologists traditionally have been regarded to be more scientific than many of their colleagues. A mystic perversion of this assumption prevails among those clinicians who believe that the pathologist, given only a piece of a patient’s tissue, has all of the other ingredients necessary to produce a statement of absolute truth at the end of his report. More dangerous to mankind is a pathologist with the same concept.

The importance of clinical information may be self-evident to the pathologist, but there is little literature regarding the frequency of its absence or the impact of its absence. A CAP Q-Probes study [37] of surgical pathology accessioning processes reported a number of accessioning deficiencies. Among the deficiencies were: 40 % “no clinical history or diagnosis present on requisition slip”; 9.4 % “no tissue source indicated on container or requisition slip”; 6.7 % “no name of submitting physician.” In total, 2.4 % of cases had no clinical information on the requisition slip. In the majority (74 %) of the cases with missing or incorrect information not related to patient identification, laboratories chose to do nothing about the problem. Accessioning error rates were lower in laboratories with a formal plan for error detection. A review [38] of histopathological cases in the United Kingdom revealed that clinical information was inadequate in 6.1 % of overall cases. Statistically significantly higher rates of inadequate clinical information originated from surgeons (9.4 %) compared to non-surgeon clinicians (1.4 %). In addition, surgeons were less likely to provide a differential diagnosis when compared to other clinicians (38.8 versus 74.4 %). Interestingly, there was no significant difference between trainees and consultants for either clinical history or differential diagnosis. Also, surgeons were less likely to provide contact information when submitting a specimen to pathology (27.5 versus 42.0 %). An internal quality audit [39] of histopathology reporting in the United Kingdom revealed that 15 % of the reports contained inadequate or absent clinical details and 19 % contained incomplete or inaccurate clerical details. Similarly, an Australian internal quality audit [40] of surgical pathology reports showed inadequate clinical history in 1.2 % of the audited cases and clerical inadequacies in 3.0 %. Lack of sufficient clinical and clerical details is not only potentially dangerous, it wastes the time of clerical staff, pathologists, and others.