Translate this page into:
An Audit of Diagnostic Disparity between Intraoperative Frozen Section Diagnosis and Final Histopathological Diagnosis of Central Nervous System Lesions at a Tertiary Care Center
Address for correspondence: Vikram Singh, MD (Path), DNB, DM (Histopathology), Department of Pathology, Army Hospital (Research and Referral), Dhaulakuan, Delhi Cantt 110010, New Delhi, India (e-mail: vikram8165n@gmail.com).
This article was originally published by Thieme Medical and Scientific Publishers Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.
Abstract
Introduction
Evaluation of intraoperative squash smear and frozen section (FS) in central nervous system (CNS) neoplasms is consistently practiced for rapid assessment and has several advantages to its credence. It is an invaluable tool to ensure adequacy of tissue obtained to establish the diagnosis. Moreover, it aids in guiding the surgeon for critical decisions regarding the extent of resection. Although molecular markers have been integrated with morphology in the revised 2016 World Health Organization classification of brain tumors, precise morphological assessment still remains the foundation for the diagnosis and rapid intraoperative assessment of morphological details is equally critical and rewarding.
Objective
This study aims to audit the diagnostic disparity between intraoperative diagnoses based on a combination of squash cytology and FS in cases of CNS lesions with gold standard, final diagnosis based on examination of formalin fixed paraffin embedded hematoxylin, and eosin-stained tissue sections.
Materials and Methods
All intraoperative squash cytology and FS reported for CNS lesions from January 2017 to December 2020 were reviewed. The cases were categorized into three groups—group 1: when diagnosis of intraoperative diagnosis based on a combination of squash cytology and FS was same as the final histopathological diagnosis (concordant), group 2: partially concordant, and group 3: discordant cases.
Statistical Analysis
Descriptive statistics was used to classify the data and diagnostic accuracy was calculated.
Results
Complete concordance was present in 69.96% (191/273) cases, 20.1% (55/273) cases showed partial concordance, and 9.89% (27/273) cases were discordant with histopathological diagnosis. Out of the 27 discordant cases, misclassification of tumor type was the most common category (11 cases, 40%), followed by grading mismatch (7 cases, 25.9%), and misdiagnosis of tumor versus nontumor conditions (9 cases, 33.3%).
Conclusion
Our study shows that combination of intraoperative squash cytology and FS shows a high percentage of accuracy in arriving at intraoperative diagnosis in cases of intracranial lesions. Regular audits of discordant cases should be conducted by surgeons and pathologists as part of a quality assurance measure to sensitize themselves with the potential pitfalls, minimizing misinterpretation and helping in providing a more conclusive opinion to the operating surgeons.
Keywords
frozen section
squash smear
central nervous system
intraoperative diagnosis
Introduction
Intracranial tumors are a diverse group of malignancies that are broadly classified as primary or secondary (metastatic). The current World Health Organization (WHO) classification describes more than 130 different central nervous system (CNS) tumors, entities, or variants.[1]
Evaluation of intraoperative squash smear or frozen section (FS) in CNS neoplasms is consistently practiced at many centers for rapid assessment and has several advantages to its credence.[2] It is an invaluable tool to ensure adequacy of tissue obtained to establish the diagnosis. Moreover, it aids in guiding the surgeon for critical decisions regarding the extent of resection.[3,4] Although molecular markers have been integrated with morphology in the revised 2016 WHO classification of brain tumors,[1] precise morphological assessment still remains the foundation for the diagnosis and rapid intraoperative assessment of morphological details is equally critical and rewarding.[2] The usefulness and accuracy of this vital intraoperative technique in the assessment of CNS lesions have been confirmed in many studies.[5,6] Out of the two primary methods for evaluation of intraoperative biopsies, quash smear cytology has recently gained importance because of image-guided stereotactic biopsies. The soft and fragile texture of brain tumors aids in smear preparation providing exquisite nuclear details free of freezing artifact and distortion. Squash cytology is faster to prepare, easy to adopt, requiring minimal tissue, and is an inexpensive and reliable intraoperative technique allowing reliable diagnosis and guidance to neurosurgeon.[7] The FS technique is more reliable and superior in terms of characterization of architectural details and cytomorphology. It has the limitation of difficult section cutting due to soft consistency of tissue, costly equipment, and freezing artifacts due to ice crystal formation.[8]
Advantages and limitation of these techniques must be accepted by both the neurosurgeon and reporting pathologist.[9] The studies have considered combined use of FS and squash as complimentary procedures to assist neurosurgeons in accurate diagnosis as compared with single procedure only.[10] It is also recommended to evaluate intraoperative preparations in the context of clinical impression and neuroimaging for an accurate intraoperative diagnosis.
The aim of the current study is to audit the diagnostic disparity between intraoperative diagnoses based on combination of squash cytology and FS in cases of CNS lesions with final histopathological diagnosis, which is taken as the gold standard.
Materials and Methods
This retrospective analytical study was performed at the department of pathology of a tertiary care hospital, in which all the consecutive intraoperative squash cytology and FSs reported for CNS lesions from January 2017 to December 2020 were reviewed. The squash smears and FS were prepared, examined, and reported intraoperatively for each case. CNS tumors were graded and classified according to WHO 2016 classification. The intraoperative diagnosis was correlated with gold standard, final formalin fixed paraffin embedded (FFPE) hematoxylin, and eosin stained (H&E) tissue sections. Where required, immunohistochemistry was performed on the FFPE tissue specimen for establishing the final diagnosis. Detailed demographic details, clinical history, and radiological findings were compiled for all the cases from the medical records.
Inclusion Criteria
All the intraoperative neurosurgical specimens which were evaluated by preparing both squash smears and FS and confirmed with histopathological examination of postsurgical biopsy tissue.
Exclusion Criteria
Postsurgical biopsy tissue received without intraoperative squash or FS request from the neurosurgical department.
Sample inadequate to perform all the above three procedures.
Intraoperative final diagnoses based on combination of squash cytology and FS were correlated with the histological diagnosis of the paraffin-embedded sections, the latter being regarded as the gold standard, and categorized into three groups. The discordant cases were recognized and the possible causes for the fallacies and shortcomings were analyzed. The diagnostic compatibility and accuracy between the two diagnostic techniques was established on the basis of classification of CNS tumor and grading. The cases were categorized into three groups after correlation:
Complete concordance was present when intraoperative final diagnosis based on a combination of squash cytology and FS was same as the final histopathological diagnosis (concordant).
Intraoperative final diagnosis based on a combination of squash cytology and FS was not wrong but too broad to characterize it as fully concordant (partial concordance). In these cases, partial concordance did not affect the management of case.
Intraoperative final diagnosis based on a combination of squash cytology and FS was not confirmed by final histopathological diagnosis (discordant).
Data collected were statistically analyzed using SPSS software (version 25) and compared between the intraoperative diagnostic procedures (squash smears and FS) and the final histopathological diagnosis.
Results
A total of 1,453 cases were received at our department for intraoperative evaluation during the study period, out of which 283 (19.5%) cases were CNS lesions. Out of the 283 cases, 10 cases were excluded from the study as tissue was exhausted after squash/FS preparation. Hence, a total of 273 cases were included in the study, of which 154 (56.4%) patients were male and 119 (43.6%) were female. The age of the patients ranged from 02 to 81 years, the mean age being 44.7 ± 18.4 years. Majority of the lesions were supra-tentorial, 250 (91.6%), and 23 (8.4%) were infra-tentorial. Most cases presented in the 4th to 7th decades (156/273, 57.1%). There were 24 (8.8%) pediatric cases. In the pediatric age group, medulloblastoma was the most frequent tumor, 11 (45.8%), and among the adults, glioblastoma (GBM; n = 90, 36.1%) was most frequently diagnosed tumor in the present study.
Histomorphological Spectrum of Central Nervous System Lesions
Out of the total 273 cases, 261 (95.6%) cases were found to be neoplastic lesions and 12 (4.4%) were nonneoplastic. Astrocytic and oligodendroglial tumors, ependymomas, meningiomas, tumors of sellar region, medulloblastoma, mesenchymal tumors, tumors of peripheral nerve, neuronal-glial tumors, lymphomas, and metastasis were the neoplastic lesions constituting 56.77% (155/273), 2.56% (7/273), 13.18% (36/273), 3.29% (9/273) 6.95% (19/273), 1.46% (4/273), 4.2% (10/273), 0.42% (1/273), 1.83% (5/273), and 6.22% (17/273), respectively. The nonneoplastic lesions composed of benign cysts, inflammatory lesions, reactive gliosis, and normal brain parenchyma constituting 0.84% (2/273) each, and vascular lesions and inflammatory lesions constituting 1.26% (3.273) each as shown in ►Table 1.
| Type of lesion | Final diagnosis on HPE | N (%) | Concordant (n) | Partial correlation (n) | Discordant (n) |
|---|---|---|---|---|---|
| Diffuse astrocytic oligodendroglial tumors | Pilocytic astrocytoma WHO Gr I | 6 (2.5) | 3 | 3 | 0 |
| Pilomyxoid astrocytoma WHO Gr I | 3 (1.3) | 3 | 0 | 0 | |
| Diffuse astrocytoma WHO Gr II | 30 (1.8) | 22 | 6 | 4 | |
| Pleomorphic xanthoastrocytoma WHO Gr II | 1 (0.42) | 0 | 0 | 1 | |
| Anaplastic astrocytoma WHO Gr III | 9 (3.79) | 7 | 1 | 1 | |
| GBM WHO Gr IV | 90 (37.9) | 83 | 4 | 3 | |
| Small cell GBM WHO Gr IV | 1 (0.42) | 0 | 1 | 0 | |
| Giant cell GBM WHO Gr IV | 1 (0.42) | 0 | 1 | 0 | |
| Gliosarcoma WHO Gr IV | 6 (2.6) | 4 | 2 | 0 | |
| Oligodendroglioma WHO Gr II | 4 (1.7) | 2 | 2 | 0 | |
| Anaplastic oligodendroglioma WHO Gr III | 3 (1.3) | 1 | 2 | 0 | |
| Subependymal giant cell astrocytoma | 1 (0.42) | 0 | 0 | 1 | |
| Ependymomas | Classical ependymoma | 3 (1.3) | 2 | 1 | 0 |
| Myxopapillary ependymoma | 2 (0.84) | 1 | 1 | 0 | |
| Anaplastic ependymoma WHO Gr III | 2 (0.84) | 0 | 2 | 0 | |
| Meningeal tumors | Angiomatous meningioma | 1 (0.42) | 1 | 0 | 0 |
| Atypical meningioma | 8 (3.4) | 2 | 6 | 0 | |
| Anaplastic meningioma | 1 (0.42) | 0 | 1 | 0 | |
| Fibroblastic meningioma WHO Gr I | 3 (1.3) | 3 | 0 | 0 | |
| Psammomatous meningioma WHO Gr I | 1 (0.42) | 0 | 0 | 1 | |
| Transitional meningioma WHO Gr I | 7 (2.9) | 7 | 0 | 0 | |
| Meningothelial meningioma WHO Gr I | 15 (6.3) | 15 | 0 | 0 | |
| Tumors of sellar Region | Pituitary adenoma | 5 (2.1) | 3 | 0 | 2 |
| Pituicytoma | 1 (0.42) | 0 | 1 | 0 | |
| Craniopharyngioma | 2 (0.84) | 2 | 0 | 0 | |
| Granular cell tumor | 1 (0.42) | 0 | 0 | 1 | |
| Embryonal tumor | Medulloblastoma | 17 (6.2) | 4 | 12 | 1 |
| Tumors of peripheral nerves | Schwannoma | 10 (4.2) | 7 | 3 | 0 |
| Mesenchymal nonmeningothelial tumors | Hemangioblastoma | 1 (0.42) | 0 | 0 | 1 |
| Solitary fibrous tumor/hemangiopericytoma | 1 (0.42) | 0 | 1 | 0 | |
| Ewing sarcoma | 2 (0.84) | 0 | 2 | 0 | |
| Neuronal and mixed neuronal–glial tumor | Ganglioglioma | 1 (0.42) | 1 | 0 | 0 |
| Lymphoma | B cell-NHL | 2 (0.84) | 0 | 0 | 2 |
| DLBCL-AGC | 1 (0.42) | 1 | 0 | 0 | |
| High-grade NHL | 1 (0.42) | 1 | 0 | 0 | |
| Primary CNS lymphoma | 1 (0.42) | 0 | 0 | 1 | |
| Metastasis | Metastatic adenocarcinoma | 6 (2.5) | 5 | 1 | 0 |
| Papillary adenocarcinoma | 1 (0.42) | 1 | 0 | 0 | |
| Metastatic carcinoma | 8 (3.4) | 6 | 2 | 0 | |
| Poorly differentiated carcinoma | 1 (0.42) | 0 | 0 | 1 | |
| Metastatic deposits of high-grade sarcoma | 1 (0.42) | 0 | 0 | 1 | |
| Inflammatory lesions | Inflammatory granulation tissue | 1 (0.42) | 1 | 0 | 0 |
| Infarction | 1 (0.42) | 0 | 0 | 1 | |
| Cystic lesions | Epidermoid cyst | 2 (0.84) | 2 | 0 | 0 |
| Vascular lesions | Cavernoma | 2 (0.84) | 0 | 0 | 2 |
| Benign vasoformative lesion | 1 (0.42) | 0 | 0 | 1 | |
| Reactive lesions | Exuberant reactive microglial proliferation | 1 (0.42) | 0 | 0 | 1 |
| Gliosis | 2 (0.84) | 1 | 0 | 1 | |
| Miscellaneous | Benign brain tissue | 2 (0.84) | 2 | 0 | 0 |
| Total | 273 | 191 | 55 | 27 |
Abbreviations: CNS, central nervous system; GBM, glioblastoma; HPE, holoprosencephaly; NHL, non-Hodgkin lymphoma; WHO, World Health Organization.
Degree of Correlation
Intraoperative final diagnosis based on combination of squash cytology and FS was correlated with the final histopathological diagnosis in each case. The overall diagnostic accuracy was 95%.
Complete correlation was present in 69.96% (191/273) cases, 20.1% (55/273) cases showed partial correlation, and 9.89% (27/273) cases were discordant with histopathological diagnosis as shown in ►Table 2.
| Degree of agreement | Frequency (percentage) |
|---|---|
| Completely same diagnosis (concordant cases) | 191 (70%) |
| Partial agreement in diagnosis | 55 (20.1%) |
| No agreement in diagnosis (discordant cases) | 27 (9.9%) |
Abbreviations: FFPE, final formalin fixed paraffin embedded; FS, frozen section; HPE, holoprosencephaly.
Out of the 27 discordant cases, misclassification of tumor type was the most common category (11 cases, 40%), followed by grading mismatch (7 cases, 25.9%) and misdiagnosis of tumor versus nontumor conditions (9 cases, 33.3%) as shown in ►Table 3. In the misclassification category, the most distinct pitfall was in the diagnosis of lymphomas, which were misdiagnosed as high-grade gliomas by intraoperative techniques. Grading of astrocytic tumors was the major cause of discordance in the grading mismatch category in which five low grade astrocytomas were diagnosed as high grade and two high-grade astrocytomas diagnosed as low-grade gliomas. In the tumor versus nontumor category, five out of nine cases were diagnosed as normal or reactive tissues, which were confirmed as tumors on histopathological examination as shown in ►Table 3. In four cases, intraoperative diagnosis was of tumor; however, the final diagnosis turned out to be reactive or inflammatory.
| Category (%) | Intraoperative diagnosis | HPE diagnosis | No. of cases | |
|---|---|---|---|---|
| Misclassification (40.7%) | High-grade glioma | B-NHL | 1 | |
| Poorly differentiated carcinoma | B-NHL | 1 | ||
| Metastasis from poorly differentiated Ca | GBM WHO Gr IV | 1 | ||
| Cystic lesion lined by endothelial cells | Cavernoma | 1 | ||
| High-grade glioma | Medulloblastoma WHO Gr IV | 1 | ||
| High-grade glioma | Psammomatous meningioma WHO Gr I | 1 | ||
| Metastasis from poorly differentiated carcinoma | Metastatic high-grade sarcoma | 1 | ||
| Meningioma | Pituitary adenoma | 1 | ||
| Meningioma | Pleomorphic xanthoastrocytoma WHO Gr II | 1 | ||
| High-grade NHL | Poorly differentiated carcinoma | 1 | ||
| High-grade glioma | Primary CNS Lymphoma | 1 | ||
| Grading mismatch (25.9%) | Low-grade glioma | Anaplastic astrocytoma WHO Gr III | 1 | |
| High-grade glioma | Diffuse astrocytoma WHO Gr II | 4 | ||
| Low-grade glioma | GBM WHO Gr IV | 1 | ||
| High-grade glioma | SEGA WHO Gr I | 1 | ||
| Tumor vs. nontumor (33.3%) | Low-grade glioma | Cavernoma | 1 | |
| Low-grade glioma | Benign vasoformative lesion | 1 | ||
| Metastasis favored over high-grade glioma | Exuberant reactive microglial proliferation | 1 | ||
| Gliosis | Diffuse astrocytoma WHO Gr II | 1 | ||
| Low-grade glioma | Granulomatous lesion | 1 | ||
| Necrotizing lesion | GBM WHO Gr IV | 1 | ||
| Normal brain tissue | Granular cell tumor | 1 | ||
| Normal brain tissue | Hemangioblastoma | 1 | ||
| Normal brain tissue | Pituitary adenoma | 1 | ||
| Total | 27 | |||
Abbreviations: CNS, central nervous system; GBM, glioblastoma; NHL, non-Hodgkin lymphoma; WHO, World Health Organization.
The category of partial correlation lesions included those which were correctly categorized broadly as high grade or low grade but could not be further characterized or those tumors which were categorized broadly as round blue cell tumors or spindle cell neoplasm without further characterization.
Discussion
Intraoperative pathological preparations (cytology/FS) are an integral part of neurosurgical pathology.[11] The ideal intraoperative technique used should be rapid, accurate, and should allow the preservation of tissue for paraffin section study.[9] Although FSs give the best architectural details,[12,13] the innate soft nature and high water content of the brain tissue often render poor-quality FSs with freezing artifacts leading to interpretative errors. Hence, it is often complemented by alternative methods such as squash smears and touch imprint cytology.[14,15]
“Squash smears” are technically easier to prepare without necessity of special equipment and requirement of a very small amount of tissue. Variable smear thickness, crushing artifacts, and difficult smearing are some significant drawbacks of squash smears.[9] Soft lesions such as astrocytomas, oligodendrogliomas, and pituitary adenomas smear well. Lesions such as meningiomas and schwannomas are difficult to smear due to the increased fibrous component or calcification, often rendering smears of poor quality.[14-16] “Frozen sections” are better in diagnosing firmer lesions and offer better architectural features as compared with cytology. In the present study, the drawbacks of one technique were overcome by complementing with the other technique. In this study, squash smears yielded good cytomorphological details in cases of astrocytomas, meningothelial variant of meningioma, oligodendroglioma, medulloblastoma, metastatic carcinomas, and pituitary adenoma, while FSs fared well in fibroblastic meningioma and schwannoma. These findings were in accordance with studies by Rao S et al,[17] Shukla K et al,[18] and Savargaonkar and Farmer.[19] However, various limitations were still noted in both the techniques. Categories of discrepancies were noted in the study.
Problem in Tumor Grading in Astrocytoma
Incorrect assessment of grading of astrocytomas was seen in seven cases. Four cases of diffuse astrocytoma WHO Gr II and one case of subependymal giant cell astrocytoma WHO Gr I were diagnosed as high-grade glioma due to enhanced anisonucleosis, which was possibly a result of freezing artifact incorrectly resembling high-grade tumor. One case of GBM and one case of anaplastic astrocytoma were diagnosed as low-grade glioma probably due to sampling error from the peritumoral areas. This problem of either over-grading or under-grading of astrocytomas has also been reported by other authors as well.[14,20-22] Most of the studies have attributed this to the heterogeneity of astrocytomas. Improper grading of astrocytic neoplasms in cytologic preparations has also been documented in other studies.[15,16,23] According to some, it is inappropriate to attempt to grade CNS neoplasms on small biopsy material whether by squash smear or frozen technique as astrocytomas are known for their heterogeneity. In some studies, paraffin sections showed that in under-graded cases, there were areas of both high- and low-grade types and cytological sampling might have failed to show the anaplastic areas.[15,16]
Differentiating the Coagulative Tumor Necrosis of Glioblastoma from Inflammation-Associated Necrosis
One case of GBM was misdiagnosed as necrotizing lesion due to sampling error. In this case, the necrotic areas were sampled and both the crush smears and FS yielded only necrotic tissue with very few viable cells. The case was diagnosed as GBM in subsequent FFPE sections where extensive areas of coagulative necrosis were seen, along with viable malignant cells. Chand et al, Mitra et al, and Plesec and Prayson also reported similar findings in their studies.[11,24,25]
Misclassification of Tumors
An evaluation of the 11 discordant cases in this category revealed various pitfalls, which led to misdiagnosis. One case of B-cell non-Hodgkin lymphoma (B-NHL) was diagnosed as high-grade glioma. In this case, the squash smear showed cellular clusters of atypical cells (►Fig. 1A) and FS showed cellular tumor with dispersed cell population (►Fig. 1B) thereby rendering a diagnosis of high-grade glioma. However, the final FFPE section revealed cellular tumor with cells arranged in sheets having prominent perivascular arrangement. (►Fig. 1C, D) These tumor cells were CD20-positive (►Fig. 1E) with a Ki 67 index of 60 to 70% (►Fig. 1F), finally confirming a B-NHL. Similarly, one case of primary CNS lymphoma was misinterpreted as high-grade glioma and one case of poorly differentiated carcinoma was reported as high-grade NHL. It may be hard to distinguish high-grade gliomas from lymphomas on FS having artifacts. Cytological smear preparation is recommended when suspecting a case of lymphoma. However, malignant cells in squash smear spread over an abundant gliofibrillary background to simulate a glioma.[26] In such scenarios, the background showing presence of lymphoglandular bodies in squash cytology helps in lymphoma diagnosis.[27] A case of poorly differentiated carcinoma was also interpreted as high-grade glioma as squash smears showed cellular clusters of atypical cells (►Fig. 2A) and FS revealed a cellular tumor with dispersed atypical cell population and increased vascularity. (►Fig. 2B) The final paraffin section showed a cellular tumor in sheets and papillary pattern with fibrovascular cores, thereby confirming a poorly differentiated carcinoma. (►Fig. 2C, D). In this case the cellularity, atypia and vascularity lead to the misdiagnosis of high-grade glioma. Few other analytical errors included a case each of GBM and metastatic sarcoma diagnosed as metastatic deposits of poorly differentiated carcinoma. Medulloblastoma and psammomatous meningioma were diagnosed as high-grade gliomas. These interpretative errors resulting in misdiagnoses were attributed to increased vascularity and nuclear atypia with freezing artifacts compounding difficulties encountered. One case each of pituitary adenoma and pleomorphic xanthoastrocytomas was misinterpreted as meningioma The interpretative errors in these two cases were mainly due to the freezing process that distorted the architecture, with variable section thickness and uneven staining obscuring the cytologic details and influencing subjective judgment.
![Case of primary CNS lymphoma reported as high grade glioma; photomicrographs of squash smear, frozen section, and final paraffin section. (A) Photomicrograph of squash smear showing a cellular cluster of atypical cells (Leishman Giemsa stain, 40×). (B) Photomicrograph of frozen section showing a cellular tumor with dispersed cell population (hematoxylin and eosin [H&E] stain, 40×). (C, D) Photomicrograph of final paraffin section showing cellular tumor with cells arranged in a sheet (H&E stain, 10×) and prominent perivascular arrangement (H&E Stain, 20×). (E) Tumor cells showing diffuse membranous positivity for CD 20 immunostain (IHC, 20×). (F) Ki 67 is high (60–70%); (IHC stain, 20×). CNS, central nervous system; IHC, immunohistochemistry.](/content/164/2022/14/4/img/JLP-14-384-g001.png)
-
Case of primary CNS lymphoma reported as high grade glioma; photomicrographs of squash smear, frozen section, and final paraffin section. (A) Photomicrograph of squash smear showing a cellular cluster of atypical cells (Leishman Giemsa stain, 40×). (B) Photomicrograph of frozen section showing a cellular tumor with dispersed cell population (hematoxylin and eosin [H&E] stain, 40×). (C, D) Photomicrograph of final paraffin section showing cellular tumor with cells arranged in a sheet (H&E stain, 10×) and prominent perivascular arrangement (H&E Stain, 20×). (E) Tumor cells showing diffuse membranous positivity for CD 20 immunostain (IHC, 20×). (F) Ki 67 is high (60–70%); (IHC stain, 20×). CNS, central nervous system; IHC, immunohistochemistry.
![Case of poorly differentiated carcinoma reported as high-grade glioma; photomicrographs of squash smear, frozen section, and final paraffin section. (A) Photomicrograph of squash smear showing a cellular cluster of atypical cells (Leishman Giemsa stain, 40×). (B) Photomicrograph of frozen section showing a cellular tumor with dispersed cell population (hematoxylin and eosin [H&E] stain, 40×). (C) Photomicrograph of final paraffin section showing cellular tumor with cells arranged in a sheet (H&E stain, 10×) and papillary architecture (H&E Stain, 20 ×). (D) High-power view showing a papillary architecture with well-defined fibrovascular core (H&E Stain, 40×).](/content/164/2022/14/4/img/JLP-14-384-g002.png)
-
Case of poorly differentiated carcinoma reported as high-grade glioma; photomicrographs of squash smear, frozen section, and final paraffin section. (A) Photomicrograph of squash smear showing a cellular cluster of atypical cells (Leishman Giemsa stain, 40×). (B) Photomicrograph of frozen section showing a cellular tumor with dispersed cell population (hematoxylin and eosin [H&E] stain, 40×). (C) Photomicrograph of final paraffin section showing cellular tumor with cells arranged in a sheet (H&E stain, 10×) and papillary architecture (H&E Stain, 20 ×). (D) High-power view showing a papillary architecture with well-defined fibrovascular core (H&E Stain, 40×).
Tumor versus Nontumor
One case each of hemangioblastoma, pituitary adenoma, and granular cell tumor were reported as normal brain tissue and one case of diffuse astrocytoma WHO Gr II was diagnosed as gliosis. Sampling error is one of the most common causes of this discrepancy. Thorough sampling and adequate smears and sections are recommended to reduce intraoperative discrepancies.[28] Similarly, exuberant reactive microglial proliferation in one case was misinterpreted as metastasis and a cavernoma and arteriovenous malformation (AVM) were diagnosed as low-grade glioma. The squash smears of AVM showed increased cellularity in a gliofibrillary background (►Fig. 3A) supported by mildly increased cellularity on FS (►Fig. 3B) misinterpreting it as low-grade glioma. The final FFPE section revealed a benign vascular lesion with varying sized blood vessels with surrounding reactive gliosis (►Fig. 3C). The endothelial cells were highlighted by CD34 immunostain (►Fig. 3D). A case of cerebral infarction was interpreted as metastatic deposits. A small cellular cluster of atypical cells with surrounding necrosis (►Fig. 4A) was noted on squash smears with similar observation on FS (►Fig. 4B). The final paraffin section shows sheets of large cells with abundant vacuolated cytoplasm (►Fig. 4C) which were immunonegative for Glial Fibrillary Acid Protein (GFAP) (►Fig. 4D), Epithelial Membrane Antigen (EMA) (►Fig. 4E), and immunopositive for CD68 (►Fig. 4F). A diagnosis of infarction was confirmed. In this case, the most probable explanation was interpreting histiocytes as epithelial cells due to their size and abundant cytoplasm.
![Case of AVM reported as low-grade glioma; photomicrographs of squash smear, frozen section, and final paraffin section. (A) Photomicrograph of squash smear showing cellular cluster of atypical cells arranged in a glial background suggestive of low-grade glioma (Leishman Giemsa stain, 40×). (B) Photomicrograph of frozen section showing a fragment of brain parenchyma with mild increase in cellularity (hematoxylin and eosin [H&E] stain, 40×). (C) Photomicrograph of final paraffin section showing lesion composed of varying sized blood vessels with surrounding reactive gliosis (H&E stain, 10×). (D) CD 34 immunostain highlights endothelial cells of these blood vessels (IHC Stain, 40×). AVM, arteriovenous malformation.](/content/164/2022/14/4/img/JLP-14-384-g003.png)
-
Case of AVM reported as low-grade glioma; photomicrographs of squash smear, frozen section, and final paraffin section. (A) Photomicrograph of squash smear showing cellular cluster of atypical cells arranged in a glial background suggestive of low-grade glioma (Leishman Giemsa stain, 40×). (B) Photomicrograph of frozen section showing a fragment of brain parenchyma with mild increase in cellularity (hematoxylin and eosin [H&E] stain, 40×). (C) Photomicrograph of final paraffin section showing lesion composed of varying sized blood vessels with surrounding reactive gliosis (H&E stain, 10×). (D) CD 34 immunostain highlights endothelial cells of these blood vessels (IHC Stain, 40×). AVM, arteriovenous malformation.
-
A case of cerebral infract reported as metastatic deposit of carcinoma; photomicrographs of squash smear, frozen section, and final paraffin section. (A) Photomicrograph of squash smear showing a small cellular cluster of atypical cells with surrounding necrosis (H&E, 40×). (B) Photomicrograph of frozen section also showing a small cellular cluster of atypical cells with surrounding necrosis (H&E stain, 40×). (C) Photomicrograph of final paraffin section showing cellular tumor with cells arranged in sheets of large cells with abundant vacuolated cytoplasm (H&E stain, 10×). (D–F) These atypical cells were negative on GFAP immunostain (4D, IHC, 20×), negative on EMA immunostain (4E, IHC, 20×), and were positive for CD68 immunostain (4F, IHC, 20×).
Overall, the results of the present study were comparable with those of other studies. In present study intraoperative final diagnosis based on a combination of squash cytology and FS showed overall 95% diagnostic accuracy, which is comparable to other studies by Rao et al[17] and Di Stefano et al[20] where they achieved diagnostic accuracies of 94 and 95.29%, respectively, when both these procedures were used together. Other authors[14,29] also advocated the same and recommended the use of both squash preparation and FS for intraoperative diagnosis. While cytomorphology is preserved in crush smears, the FS, though lacking the cytological details, preserves the tissue architecture.
The strength of the study was the large sample size and the fact that all cases of discordance category were traced and re-evaluated to determine the possible cause for variation in the result or interpretation. The limitation of the study was inherent to its retrospective nature and unavailability of data on the progress of the patients after surgery.
Conclusion
The role of intraoperative consultation is of paramount importance and it remains a rapid and reliable tool for assessing CNS lesions intraoperatively. Communication between the surgeon and pathologist with complete clinical and neuroimaging details is imperative. Our study shows that squash cytology in conjugation with FS shows a high percentage of accuracy in arriving at intraoperative diagnosis for intracranial lesions. Regular audits of discordant cases should be conducted by surgeons and pathologists as part of a quality assurance measure to sensitize themselves with the potential pitfalls minimizing misinterpretation and helping in providing a more conclusive opinion to the operating surgeons.
Contributions
K.R. and P.S.M., specialist and epidemiologist, supervised the manuscript and performed the data analysis. V.S., pathologist, conceptualized and prepared the manuscript. M.Y., junior resident, collected data and prepared the draft manuscript. P.S., pathologist, conceptualized and prepared the manuscript. R.T., pathologist, conceptualized and supervised the manuscript. P.S.M., pathologist, supervised the manuscript.
Conflicts of Interest
The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication.
Funding
None.
References
- WHO Classification of Tumours of the Central Nervous System. Lyon: International Agency for Research on Cancer; 2016.
- [Google Scholar]
- Intraoperative consultation in the diagnosis of posterior fossa brain tumors following the 2016 WHO update. Cytopathology. 2021;32(04):459-471.
- [CrossRef] [PubMed] [Google Scholar]
- Intraoperative consultation on pediatric central nervous system tumors by squash cytology. Cancer Cytopathol. 2015;123(06):331-346.
- [CrossRef] [PubMed] [Google Scholar]
- Frozen section diagnosis of pediatric brain tumors. Surg Pathol Clin. 2010;3(03):769-796.
- [CrossRef] [PubMed] [Google Scholar]
- Accuracy of intraoperative diagnosis in central nervous system tumors: report of 1315 cases [in French] Neurochirurgie. 2003;49(2–3 Pt 1):67-72.
- [Google Scholar]
- High diagnostic accuracy of cytologic smears of central nervous system tumors. A 15-year experience based on 4,172 patients. Acta Cytol. 2002;46(04):667-674.
- [CrossRef] [PubMed] [Google Scholar]
- Utility of squash smear cytology in intraoperative diagnosis of central nervous system tumors. J Cytol. 2016;33(04):205-209.
- [CrossRef] [PubMed] [Google Scholar]
- Comparison of squash smears and frozen sections versus paraffin sections in intraoperative diagnosis of central nervous system lesions. Oncol J India. 2017;1(02):25-30.
- [CrossRef] [Google Scholar]
- Intraoperative diagnosis of central nervous system lesions: comparison of squash smear, touch imprint, and frozen section. J Cytol. 2015;32(03):153-158.
- [CrossRef] [PubMed] [Google Scholar]
- Use of cytological preparations in the frozen section diagnosis of central nervous system neoplasia. Am J Surg Pathol. 1985;9(05):344-354.
- [CrossRef] [PubMed] [Google Scholar]
- Errors, limitations, and pitfalls in the diagnosis of central and peripheral nervous system lesions in intraoperative cytology and frozen sections. J Cytol. 2016;33(02):93-97.
- [CrossRef] [PubMed] [Google Scholar]
- The centennial anniversary of the frozen section technique at the Mayo Clinic. Arch Pathol Lab Med. 2005;129(12):1532-1535.
- [CrossRef] [PubMed] [Google Scholar]
- Central nervous system lesions: correlation of intraoperative and final diagnoses, six year experience at a referral centre in a developing country, Pakistan. Asian Pac J Cancer Prev. 2011;12(06):1435-1437.
- [Google Scholar]
- Smears and frozen sections in the intraoperative diagnosis of central nervous system lesions. Neurosurg Clin N Am. 1994;5(01):1-18.
- [CrossRef] [PubMed] [Google Scholar]
- Role of squash-smear technique for rapid diagnosis of neurosurgical biopsies–a cytomorphological evaluation. Indian J Pathol Microbiol. 1989;32(03):152-160.
- [Google Scholar]
- Challenges in neurosurgical intraoperative consultation. Neurol India. 2009;57(04):464-468.
- [CrossRef] [PubMed] [Google Scholar]
- Accuracy of cytologic diagnosis of central nervous system tumours in crush preparation. Indian J Pathol Microbiol. 2006;49(04):483-486.
- [Google Scholar]
- Utility of intra-operative consultations for the diagnosis of central nervous system lesions. Ann Clin Lab Sci. 2001;31(02):133-139.
- [Google Scholar]
- Intraoperative diagnosis of nervous system lesions. Acta Cytol. 1998;42(02):346-356.
- [CrossRef] [PubMed] [Google Scholar]
- Smear technique in the intra-operative brain tumor diagnosis: its advantages and limitations. Neurol Res. 1999;21(01):121-124.
- [CrossRef] [PubMed] [Google Scholar]
- Evaluation of the squash smear technique in the rapid diagnosis of central nervous system tumors: a cytomorphological study. ISPUB. 2010;11:1.
- [CrossRef] [Google Scholar]
- Intraoperative consultation and smear cytology in the diagnosis of brain tumours. Kathmandu Univ Med J (KUMJ). 2008;6(24):453-457.
- [CrossRef] [PubMed] [Google Scholar]
- Squash preparation: a reliable diagnostic tool in the intraoperative diagnosis of central nervous system tumors. J Cytol. 2010;27(03):81-85.
- [CrossRef] [PubMed] [Google Scholar]
- Frozen section discrepancy in the evaluation of central nervous system tumors. Arch Pathol Lab Med. 2007;131(10):1532-1540.
- [CrossRef] [PubMed] [Google Scholar]
- Cytopathology of the central nervous system. Part I. Utility of crush smear cytology in intraoperative diagnosis of central nervous system lesions. Acta Cytol. 2006;50(06):608-616.
- [CrossRef] [PubMed] [Google Scholar]
- Intraoperative cytology (squash smear) in neurosurgical practice - pitfalls in diagnosis experience based on 3057 samples from a single institution. Cytopathology. 2007;18(05):300-308.
- [CrossRef] [PubMed] [Google Scholar]
- Intraoperative frozen section diagnosis in skull base surgery. Skull Base Surg. 1993;3(03):159-163.
- [CrossRef] [PubMed] [Google Scholar]
- Imprints, smears, and frozen sections of brain tumors. Neurosurgery. 1991;29(04):575-579.
- [CrossRef] [PubMed] [Google Scholar]
