Abstract
Background. To avoid diagnostic errors such as missed diagnosis and errors in staging tumors due to inadequate tissue sampling, pathologists submit additional sections (AS). Objective. This study assessed frequency, diagnostic yield, distribution, and cost of AS. Method. Among 1542 AS cases, we calculated mean AS per case; fraction of AS that altered diagnosis or stage; AS variation by tissue, malignant versus benign lesions, presence or absence of neoadjuvant therapy, mass, margin, lymph nodes, or other source, resident versus pathologist assistant (PA) dissector; and AS cost per case. Results. Overall 9.2 ± 8.8 AS were collected per case. In only 3.8% (58/1542) of cases AS altered diagnosis or stage. Urinary bladder cases provoked the most AS: 19.5 ± 15.1 per case. Significantly more AS came from malignant versus benign lesions (10.8 ± 9.7 vs 7.6 ± 7.5, P = <.0001) and from specimens treated with neoadjuvant therapy versus malignant lesions not so treated (12.3 ± 9.4 vs 10.3 ± 9.8, P = .02). Lymph nodes were sampled more heavily compared with mass, margin, and other sites combined (11.8 ± 11.4 vs 8.9 ± 8.4, P = .003), but in 78.4% (1209/1542) of cases, AS were from mass. Of diagnosis or stage altering AS cases, two thirds (38/58) were from masses, one fifth (11/58) from lymph nodes, a 10th (6/58) from margins, and a 20th (3/58) from other specimen sites. Resident versus pathologist assistant dissection caused no significant AS difference. AS contributed 40% cost per case. Conclusions. AS per case ranged widely; their diagnostic yield was low; they were highest in urinary bladder specimens, in malignant and particularly neoadjuvant-treated lesions. Although lymph nodes were most heavily sampled, most AS were from masses. Resident dissection did not increase AS and cost of AS was high.
Introduction
Adequate sampling of surgical pathology specimens is vital to arriving at the correct pathologic diagnosis. Sampling is adequate when a well-defined protocol is followed during grossing of a specimen and all relevant tissue sections are submitted for processing. Standardized grossing manuals exist1-3; however, there are variations across laboratories and pathologists due to personal preference or grossing style. Inadequate sampling results from failure to submit ample and crucially important sections from the specimens as defined by the grossing manuals. It can lead to diagnostic errors including missed diagnosis; over or under calling benign, borderline, or malignant diagnoses; and erroneous pathologic staging of tumors. To minimize sampling error and to establish an accurate pathologic diagnosis, pathologists use recuts and/or additional tissue sections (ie, more tissue sections in addition to initial representative sections). However, recuts and additional sections (AS) have drawbacks, including increased processing costs and increased turnaround times. Multiple studies have assessed the utility of recuts and levels in improving diagnostic accuracy4-8; however, the diagnostic utility of “additional tissue sections” in a surgical pathology specimen has not been completely elucidated.
In our study, we evaluate the effectiveness of additional tissue sections in arriving at an accurate pathologic diagnosis, as well as proper pathologic staging. In addition, we evaluated whether benign or malignant specimens require more AS and the impact of neoadjuvant therapy on the requirement of AS. The source of AS with respect to whether the AS were from the mass, margin, lymph nodes (second search to find additional lymph nodes), or other parts of the specimens was also examined. We also looked at whether specimens grossed by pathologist assistants or the residents require more AS. Finally, cost of processing AS was also evaluated.
Materials and Method
Following institutional review board approval, we did a retrospective analysis of the archives of surgical pathology specimens received in the pathology department of St John Hospital and Medical Center, Detroit, MI, over a 5-year period from January 1, 2012, to December 31, 2016. Of the 137 984 specimens received, 103 494 were small biopsies that were entirely submitted, 17 512 were big/resection specimens, and 16 978 were small to medium-sized specimens that were not entirely submitted. Additional tissue sections were submitted in 1542 cases. Of these 1314 were big/resection specimens and 228 were small to medium-sized specimens. Pathology reports and glass slides for these specimens were reviewed. Data such as pathologic diagnosis, number of AS, category of grossing individual, that is, resident versus pathologist assistant, year of training for residents, type of specimen (general surgical vs oncologic), and specimen site were collected.
Additional tissue sections are defined as additional tissue submitted in tissue cassettes for processing and microscopic evaluation. They are different from recuts/levels in which more microscopic slides are made from already processed tissue.
In this study, we considered AS “useful” if it was determined that the additional tissue sections contributed to the final diagnosis either by changing the diagnosis altogether, changing a benign entity to borderline or malignant, changing pathologic tumor stage, changing margin status, or identifying lymph-vascular invasion.
Mean number of AS from different locations in the specimens was determined for benign versus malignant specimens, for specimens with and without neoadjuvant treatment, and for specimens grossed by residents versus pathologist assistants.
Processing cost of AS was determined using the reagent cost and consumables (cassette, glass slide, coverslip, and label) cost. Compensation for the histotechnologists and laboratory technicians was also included in calculating the processing cost of AS.
Data were analyzed using Student’s t test, analysis of variance, and χ2 analysis. All data were analyzed using SPSS v. 24.0, and a P value of .05 or less denoted statistical significance.
Results
A total of 1542 specimens were identified in which AS were submitted. These comprised 4.5% of the small to medium-sized and big resection specimens. In these cases, 14 159 additional tissue sections were submitted with a mean of 9.2 ± 8.8 sections submitted per case. AS were found to be “useful” in 58 cases (3.8%) out of the 1542 specimens with AS. Of these cases in which AS were useful, 6 (11%) specimens had a change in the final pathologic diagnosis and 29 (50%) cases had a change in pathologic stage (Figure 1). Breast specimens benefitted the most from AS, as 25 (7.7%) out of 325 total breast specimens requiring AS had a change in pathologic diagnosis. Urinary bladder specimens required the highest number of AS (19.5 ± 15.1); however, AS proved useful in only one of these cases. Table 1 summarizes the numbers of initial and AS for different organs, as well as usefulness of AS in different organs.
Breakdown of “usefulness” of additional sections.
Mean Number of Initial and Additional Sections and Usefulness of Additional Sections (AS) in Different Organs.
Malignant specimens required significantly higher numbers of AS as compared with benign specimens (10.8 ± 9.7 vs 7.6 ± 7.5, P ≤ .0001; Table 2). Specimens in which neoadjuvant therapy was administered prior to resection required a significantly higher number of AS as compared with those without prior treatment (12.3 ± 9.4 vs 10.3 ± 9.8, P = .02; Table 2).
Mean Number of Initial and Additional Sections in Different Types of Specimens.
Analyzing the source of AS, we found 78% of specimens required AS from the mass (Figure 2). Of the 58 specimens in which AS proved useful, 38 had AS from the mass, 11 from additional lymph node searches, 6 from the margin, and 3 from other parts of the specimen. The mean number of AS submitted in second (additional) lymph node searches was significantly higher compared with those submitted from the mass, margin, and other parts of the specimens combined (lymph nodes: 11.8 ± 11.4 vs mass, margin, other parts of specimens combined: 8.9 ± 8.4. P = .003; Figure 3). Of all the specimens requiring AS from lymph nodes, 70.9% were large bowel. Of all the specimens requiring AS from margin, 31.1% were breast, 17.6% were lung, and 16.2% were large bowel. Table 3 summarizes the percentage of specimens with AS from mass, margin, lymph nodes, and other parts of the specimens in benign versus malignant specimens and also in specimens with and without neoadjuvant therapy. Significantly higher percentage of malignant specimens and specimens with neoadjuvant treatment (Tx) had AS from second look lymph node search as compared with benign specimens and specimens without prior treatment.
Source of additional sections.
Mean of additional sections from different sources.
Source of Additional Sections in Different Types of Specimens.
Of the 1542 cases with AS submitted, breast and large bowel accounted for 648 (42%) of the specimens (Table 1). Nearly one quarter (24.6%) of the 14 159 AS submitted were for breast specimens. All 25 breast cases and 12 large bowel cases in which AS proved useful were malignant (Table 4). Malignant large bowel cases required significantly higher number of AS as compared with benign specimens (malignant: 11.5 ± 11.1 vs benign: 6.9 ± 6.6. P ≤.0001); however, for breast specimens there was no significant difference. Table 5 compares the distribution of AS in breast and large bowel cases with and without neoadjuvant therapy.
Data on Benign and Malignant Breast and Large Bowel Cases.
Abbreviation: AS, additional sections.
Test not shown because of nonnormality of 1 mean.
Comparison of Benign and Malignant Breast and Large Bowel Cases With and Without Neoadjuvant Therapy (Tx).
Abbreviation: AS, additional section.
Test not shown because of nonnormality of 1 mean.
Actual number (just 1 specimen with only 1 additional section).
There was no significant statistical difference in mean number of AS among the specimens grossed by residents as compared with those grossed by pathologist assistants (9.4 ± 9.3 vs 9.0 ± 8.2, P = .41). Also, there was no significant difference among the residents with respect to their year in training (Figure 4). However, first year residents were more frequently (392) asked to submit AS as compared with second (233), third (178), and fourth year (62) residents. This is likely influenced by the amount of time spent on the surgical pathology rotation. Even after adjustment for this fact, however, it was found that first year residents were asked for AS more frequently. First year residents submitted AS on 49 specimens for each block they were on surgical pathology rotation, second year residents 42 per block, third year residents 33 per block, and fourth year residents 31 per block.
Mean number of additional sections for different grossers.
The cost of processing one AS in our laboratory is $4.11. This amount does not, however, account for the compensation of residents, pathologist assistants, and pathologists. A mean of 9.2 AS per specimen equates to an additional $37.81 in processing charges. Finally, this translates into $58 303 for all the 1542 specimens requiring AS.
Discussion
Frederick A. Meier has conceptualized surgical pathology as a 12-step production process. 9 The process begins with identifying patients and goes on to obtaining specimen. It then proceeds to labelling, transport, and accessioning. The process continues with sampling the specimen, fixing, embedding, cutting, and mounting processed sections on slides. Further steps include labeling and delivering the slides to the surgical pathologist who then interprets what is seen and may order ancillary studies in some cases before composing a final report that is transmitted to the clinician. This is a complex process with the possibility of errors at each step.
Analytic errors in surgical pathology can occur during gross sampling, fixing, embedding, cutting blocks, mounting, staining and labeling slides, and delivering the slides to the pathologist for microscopy. A recent study by Nakhleh et al 10 evaluated discrepancy rates in surgical pathology diagnoses among 84 studies and found a median discrepancy rate of 18.3%. Most studies evaluating analytical errors elucidated disparity of diagnosis (eg, benign to malignant and vice versa); however, none of the studies have examined grossing or sampling errors specifically. To minimize sampling error, pathologists often request submission of additional tissue sections. We studied the impact of additional tissue sections on final pathologic diagnosis and pathologic staging and tried to evaluate the factors that dictate the need of AS.
Grossing surgical pathology specimens can be a standardized process when sampling recommendations are adopted from published grossing manuals. However, some grey areas and difference of opinion lead to somewhat different practices to the extent that different institutions come up with their own grossing guidelines.11,12 Different protocols are followed for different organs, for benign versus malignant specimens, and for cases with and without prior therapy.13-16 This leads to a wide range of tissue sections being submitted and is depicted by the results of our study (Tables 1 and 2).
There is more standardization for tumor cases than for benign entities. The College of American Pathologists (CAP) has standardized protocols called “cancer protocol templates” for examination of specimens from patients with different kinds of tumors. 17 The American Joint Committee on Cancer (AJCC) has a pathologic staging manual for different tumors. 18 Both CAP cancer protocol templates and the AJCC pathologic staging manual are periodically updated based on latest research data. Both of these pertain more to microscopic findings; however, adequate sampling of the specimen at the time of grossing is needed in order to make proper use of CAP templates and AJCC pathologic staging.
A number of factors dictate the need of AS in various surgical pathology specimens. Additional tissue sections are relatively frequently requested by the pathologists on specimens coming from surgeries performed after neoadjuvant therapy. This usually happens when no tumor is seen at the time of grossing or examination of the initial histologic sections. In these cases, AS are required to look for any residual tumor and assess treatment response. Some pathologists require the entire mass, scar tissue, or area of interest submitted for microscopy; others request submitting even the surrounding tissue if they do not see any viable tumor. Results of our study confirmed that significantly higher mean number of initial as well as AS was submitted on specimens with prior neoadjuvant treatment as compared with those without any treatment. But even when the entire tissue is submitted, there is still a sampling artifact present as tissue blocks contain tissue which is 2 to 4 mm in thickness and only 4 to 5 µm sections are examined on the slide. This argument favors submission of representative sections. In cases with neoadjuvant treatment, more standardization in the way these cases are to be grossed may limit or decrease the need of AS.
Sometimes tumors are not grossly visible. This is true not only for specimen status post neoadjuvant therapy but also for primary pancreaticobiliary, prostatic, and urinary bladder tumors. Most surgical pathology laboratories submit the entire prostatectomy specimen at the time of initial gross examination to overcome this problem.12,19-21 In urinary bladder and pancreatic specimens, however, many pathologists will request AS after examining initial representative sections. 13 There does not appear to be a simple solution to this issue, as it is not feasible monetarily or from a time standpoint to submit these specimens in their entirety.
The need for more lymph nodes is another frequent reason for a pathologist to ask for AS on a given specimen. This happens more so in cases of colorectal specimens as compared with any other organ and more often in a setting of neoadjuvant therapy. In our study, a total of 151 specimens had AS from lymph nodes. Of these, 107 (71%) were large bowel and 44 were from other organs. The mean number of additional lymph node sections from large bowel specimens was significantly greater compared with the mean number of additional lymph node sections from all other organs combined (large bowel: 13.1 ± 12.8 versus all other organs combined: 8.5 ± 5.6, P = .041). However, the lymph nodes found in additional tissue sections are usually negative and rarely upstage a tumor. In our study, 29 cases had a change in pathologic stage based on findings of AS. Of these, 18 were a change in pathologic tumor (pT) stage and 11 were a change in pathologic nodal (pN) stage. The significance of an increased number of negative lymph nodes for the clinician is not clear. One might consider detection of supplementary lymph nodes in additional tissue sections as “usefulness” of AS. In our study, however, with respect to lymph nodes we considered AS useful only when additional lymph nodes upstaged the patient. Finding 20 negative lymph nodes instead of 10 might give both the pathologist and clinician psychological satisfaction, but whether or not it affects the therapy, prognosis, or subsequent management of the patient is questionable. Meticulous lymph node search at the time of initial tissue sampling not only confers adequate patient care but also can eliminate the need of AS in many such specimens. Sometimes incidental tumors are identified in small and large bowel specimens including appendices that are removed for some other reason such as obstruction, gun-shot wounds, or acute appendicitis. In these cases, a lymph node search is usually not performed when the specimen is first grossed. Therefore, these specimens may require AS of the lymph nodes, which would be completely justified. Among 1542 specimens included in our study, incidental tumors were found in 66 (4.3%) specimens and AS were found to be useful in 6 of these.
Sometimes additional tissue sections are ordered due to special circumstances, such as in cases of thyroid follicular adenoma, breast with ductal carcinoma in situ (DCIS), and hysterectomy specimens showing complex hyperplasia of the endometrium. In cases of thyroid follicular adenoma, pathologists want to rule out capsular invasion and thus minimally invasive follicular carcinoma (MI-FTC) and request for submitting the entire lesion or at least entire capsule of the lesion; in breast AS are submitted to rule out any micro-invasive component; and in the uterus AS are used to evaluate the entire endometrium to confirm there is no focus of endometrioid adenocarcinoma. With respect to minimally invasive thyroid carcinoma there is contradictory evidence in the literature. A benign and indolent course is reported by many studies owing to low risk of recurrence and distant metastasis.22,23 On the contrary, some authors do not consider MI-FTC to be purely an indolent disease.24,25 Similarly in DCIS with microinvasion there is controversial evidence in literature with respect to management including sentinel lymph node biopsy. DCIS with microinvasion comprises 0.6% to 3.4% of breast cancer diagnoses26-28 and has a low risk of lymph node metastasis.28-34 In case of endometrial pathology, differential diagnosis of hyperplasia and low-grade endometrioid adenocarcinoma can be problematic even in the eyes of experts. 35 Byun et al 36 found endometrial carcinoma in hysterectomy specimen of 30.8% patients who were diagnosed with atypical endometrial hyperplasia on initial biopsy. From a pathologic and specimen grossing standpoint; however, the argument in favor of representative sections in these scenarios is still the same, even if the entire tissue is submitted; the glass slides are still representative, constituting 4 to 5 µm of an average tissue thickness of 3 mm. Moreover, although the precise management differences are dependent on different factors and are beyond the scope of this discussion, the generalized principles for management of thyroid follicular adenoma versus minimally invasive thyroid follicular carcinoma, DCIS versus micro invasive ductal carcinoma of breast and complex atypical hyperplasia versus FIGO (International Federation of Gynecology and Obstetrics) Grade I endometrioid adenocarcinoma are the same.22,23,27,30,37-42 Another situation in which pathologists require the entire endometrium to be submitted is cases of malignant and borderline primary ovarian neoplasms specifically serous carcinomas. This is to rule out synchronous tumor of the endometrium and/or to confirm that the ovarian tumor is primary and not metastatic from the endometrium. Synchronous endometrial and ovarian carcinomas have been reported in 5% to 10% of cases.43-45 Plausibly synchronous tumors are more frequent when both organs are grossly abnormal. The chances of finding an endometrial primary tumor in the setting of a grossly normal endometrium, however, are not very high. Frequency of grossly normal endometrium in setting of ovarian pathology is an area that may need further investigation. In our study, there was not a single case found in which AS of endometrium changed or modified a final pathologic diagnosis.
In the case of breast specimens, failure to locate the biopsy clip during gross examination and sampling is a common reason why AS are requested. This is true even if a mass is grossly identified as the tumor may be multifocal. Multifocal breast carcinoma is indeed more frequent than thought of, with a prevalence ranging from 9% to 75%.46-48 In our study, AS made a difference in 2 breast cases by changing tumor focality {pT1a to pT1a(m), pT1c to pT1c(m)}. The difference it made to patient management and outcome, however, is uncertain. Identification of multifocal disease is a relative contraindication for breast conservation surgery. Lynch et al, 49 however, found that multifocality was not an independent factor contributing to recurrence-free survival, breast cancer–specific survival, and overall survival.
Management of colonic adenomas, particularly the larger ones and those with high-grade dysplasia, is another controversial area. 50 In colorectal specimens with huge adenomas in which only representative sections are submitted and no invasion is seen, pathologists tend to request submission of the entire lesion to look for an invasive component. This might sound a reasonable approach but the final results still depict representative glass slides of the entire lesion. So there may still be invasion in remainder of the tissue that is in the block, and there is no way to confirm that unless the blocks are exhausted of tissue. In our study, of the 140 benign large bowel cases, AS did to contribute to the diagnosis in a single one.
Additional tissue sections may be needed due to some other reasons. These are elaborated in following examples:
An interesting case requiring additional sections was a right hemicolectomy specimen for invasive adenocarcinoma. Seventeen lymph nodes were identified in this case, which is a number deemed sufficient by most pathologists. Only representative sections were submitted from the largest lymph node, which was grossly and microscopically entirely replaced by the tumor and hence considered a tumor deposit as there was no lymphoid tissue noted in the representative sections. One additional lymph node was positive in the remaining 16 lymph nodes. The presence of one positive lymph node made the pN stage pN1, but was this case pN1a or pN1b was not entirely certain as initially grossed. The pathologist requested to submit the remainder of the largest lymph node, and on additional sections, a thin rim of lymphoid tissue was identified, hence classifying the tissue as positive lymph node rather than a tumor deposit and upstaging the pathologic lymph node stage from pN1a to pN1b. While the additional sections did upstage the patient, it is highly unlikely that this resulted in any change in patient management, as the clinical stage remained unchanged.
Another notable case was a prostatectomy specimen for hyperplasia in an 88-year-old male. Initially 10 representative sections were submitted, but the pathologist requested the entire specimen be submitted which is the protocol for biopsy-proven prostate carcinoma and not for benign prostatic hyperplasia. Therefore, an additional 85 sections were submitted. A single 0.3 cm focus of prostatic adenocarcinoma with a Gleason score of 3 + 3 = 6 was found in AS18. This was one of the few specimens in our study where AS caused a benign diagnosis to be changed to a malignant one. As per recent clinical guidelines, active surveillance is an option for patients with a Gleason score of 3 + 3 = 6 on prostate core biopsies.51-53 So whether this change in diagnosis altered patient management is still questionable.54,55
Additional sections are also requested on benign cases, either to confirm a diagnosis or to rule out malignancy. In our study, one of the benign specimens frequently requiring AS was the thyroid. Identification of minimal capsular or vascular invasion can convert a benign adenoma to carcinoma in case of follicular lesions of the thyroid; however, as mentioned previously there is controversial evidence in the literature with respect to course of disease in MI-FTC.22-25 Pathologists may require additional tissue sections on thyroid specimens for a variety of reasons, evaluation of the entire capsule in cases of follicular lesions being the most common. This can be extremely challenging and may require a lot of sections if the lesion is large. The argument in favor of representative sections is once again the same, “even if the entire capsule is submitted, the resulting glass slides are still representative.” Thin slicing and careful inspection of gross cut surfaces (for possible areas of mushrooming for example) may be more practical in such circumstances.
Finally, another reason for requesting AS may be level of experience and/or confidence of a pathologist and experience of the grossing person. Pathologists just out of training, perhaps require more AS before signing out a case. This, however, is very subjective and further studies need to be performed to evaluate such an association if one exists.
Processing of additional tissue sections comes at cost. A single tissue block costs around $4 to process. This adds $36 additional cost on average for a single case requiring AS and comprises 40% of the processing cost in such cases. This is exclusive of compensation for the time of the individual grossing and the pathologist to review the glass slides. With the stringent regulations and decreasing compensations for hospitals and pathologists, this can be a considerable surplus expense for anatomical pathology laboratories.
Processing of AS adds at least 24 hours to the turnaround time and likely delays the sign-out much longer, when timing of sectioning and staining, especially immunohistochemical staining are taken into account. In our study, we did not study the impact of AS on turnaround time.
Conclusion
AS were requested on 4.5% of large resection and small to medium-sized specimens. Their diagnostic yield was low, contributing to diagnosis of only 3.8% of cases in which they were ordered. Lymph nodes were sampled more heavily than masses, margins, or other sites; nevertheless, more than three fourths (78.4%) of AS came from tumor masses and two thirds (38/58) of AS that altered diagnosis or staging were from masses. Malignant lesions, particularly those that had undergone neoadjuvant therapy, required more AS. They proved most useful when collected from masses in malignant lesions, especially in breast specimens. AS are more frequently requested in breast and gastric lesions treated with neoadjuvant therapy, pancreatobilary, and bladder tumors, which are not discreet lesions. Follicular thyroid neoplasms, endometrial lesions on the borderline between atypical hyperplasia and low-grade adenocarcinoma, and cases where an incidental or unexpected tumor is found are some other situations requiring AS. However, AS usually fail to provide additional diagnostic or staging information in these situations. Resident dissections did not increase the need of AS. The cost of AS was 40% of the overall cost of processing in cases in which they were ordered.
AS have a very low, but definite yield of revised diagnoses and pathologic stages; this low yield, however, comes at relatively high cost.
Footnotes
Author’s Note
Muhammad Siddique Khurram is now affiliated to Vanderbilt University Medical Center, Nashville, TN, USA.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethical Approval
Not applicable, because this article does not contain any studies with human or animal subjects.
Informed Consent
Not applicable, because this article does not contain any studies with human or animal subjects.
Trial Registration
Not applicable, because this article does not contain any clinical trials.