Impact of immediate evaluation of touch imprint cytology from computed tomography guided core needle biopsies of mass lesions: Single institution experience

INTRODUCTION

Core needle biopsy (CNB) has been adopted as a minimally invasive and effective method of tissue sampling of many organs. CNB provides an equal or greater sensitivity and specificity then fine-needle aspiration (FNAs) biopsies and is a less invasive procedure when compared with open surgical biopsies.[12] When compared with FNA, CNB provides more sample material for evaluation and more often leads to a definitive diagnosis.[1] In our experience, CNB has become the diagnostic procedure of choice in many visceral organ samplings. The method of obtaining tissue through CNB involves either ultrasound or computed tomography (CT) to guide the 18-gauge needle.[3] With each pass of the needle through the suspected lesion, small amounts of specimen are collected. To ensure adequate specimen is obtained for analysis, often times multiple passes of the needle are required to collect the sample.

The implementation of touch imprint cytology (TIC) as a method of evaluating sample collected through ultrasound guided or CT-guided CNB has improved the efficacy and expedited results.[4567] TIC in CNB specimens enables immediate reporting on specimen amount and composition without any additional risk to the patient.[278] TIC procedure involves gently pressing the specimen obtained through CNB onto a glass slide, fixing the cells on the slide and examining them under the microscope. This method is useful in establishing morphology of cells quickly, typically requiring 1-2 min before a preliminary diagnosis is made[9] while preserving the original sample for further histopathologic analysis.[7] In this manner, an on-site pathologist is able to determine not only if the core biopsy needle has been placed in to correct position to obtain the lesion in question, but also whether adequate amount of sample is collected.

In the present study, we examined whether on-site evaluation of TIC in CT-guided CNBs minimized the number of passes required to obtain adequate diagnostic tissue and its impact on specimen triaging. Our study took into account CT-guided CNBs performed in our institution, where a pathologist was present for on-site evaluation of TIC from various sites. Each case was evaluated for: (1) The number of passes required before the TIC was interpreted as adequate, (2) the total number of passes performed and (3) the number of cases where on-site evaluation assisted in triaging specimen for appropriate ancillary studies.

MATERIALS AND METHODS

RESULTS

Out of 140 total CNBs performed with TIC and immediate on-site evaluation, 109 (77.9%) cases were diagnosed as malignant and 28 (20%) as benign. In 10 out of 28 cases granulomas were identified on TIC and on the core biopsies. Eight cases had benign changes consistent treatment effect with no residual tumor. Nine cases showed inflammatory process including fat necrosis and mesothelial proliferation and one case was hematoma.

Three cases produced insufficient material from the CNBs and were deemed “Inadequate” both on TIC evaluation and on the final diagnosis.

Eight cases (6%) were interpreted as benign on TIC and malignant on CNB. The TIC were reviewed and the interpretation was confirmed. The causes of this discrepancy were contributed to “sampling errors.” In 9 (8.2%) of the malignant cases the tumor cells were present only on TIC slides and the cores were depleted of the malignant cells. These samples were obtained from liver and lung. These cases deemed adequate based on the cytologic features on the touch imprint. In three cases diagnosis of squamous cell carcinoma and in six cases diagnosis of adenocarcinomas were favored based on the cytological features of the malignant cells obtained on the touch imprint slides. Additional confirmation was reached through immunostaining of the destained touch imprint slides. Three cases of the liver were metastatic colonic adenocarcinoma and the diagnoses were confirmed with assessment of the cytological features of the touch imprint slides and comparison with the original tumor.

Follow-up was available for 27 positive cases. No false positive result was noted. As for the benign lesions histological follow-up were not available however reviewing the charts revealed clinical follow-up in 15 patients and no diagnosis of malignancy was documented.

Of the 137 adequate biopsies, 106 (75.7%) were deemed adequate by TIC on the first pass and in 19 cases (13.6%), adequacy was reached with an additional pass. Seven cases (5%) required a third pass to be performed. Only five of the 140 cases (3.6%) required more than three passes to be performed before an adequate sample was collected and these samples were collected from the lung (3 of 5 or 60%), peritoneum (1 of 5 or 20%) and bone/soft-tissue (1 of 5 or 20%). Table 1 summarizes our findings.

Table 1 Number of CNB passes required to obtain adequate tissue using on-site touch imprint cytology

Of the biopsies that were adequate on the first pass, 71% resulted in termination of the procedure or one additional pass to be performed. In five cases, the prompt on-site TIC examination resulted in a portion of the fresh sample to be sent for ancillary studies, including flow cytometry and cytogenetic studies.

Of the 140 cases evaluated at our institution, adequate tissue was obtained with the use of TIC in 75.7% of cases with just the first pass sample, greatly eliminating the need for additional biopsies in 71% of samples. The liver and lungs were the most common sites of biopsy seen in our study of 140 cases, with 68 of the cases (49%) involving the liver and 54 (39%) the lungs. These tissue sites are oftentimes difficult to biopsy, with a great potential for adverse events such as bleeding, pneumothorax with multiple repeated attempts.[811121314] In our study, 72% of liver biopsies were discontinued after evaluation of a single pass, relieving the patient from unnecessary risk of bleeding which can be caused by repeated attempts. Furthermore, 79.6% of lung biopsies evaluated by onsite TIC, resulted in termination after a single pass. Of the 140 cases evaluated, three cases (2.1%) were terminated before adequate tissue was collected and in these cases the liver and lung were the collection sites.

The size of the lesions was available for 65 cases and it ranged between 0.7 cm and 12 cm. No statistically significant relationship was noted between the size of the lesion and the success of the procedure.

As per radiology reports, complications related to the procedure was only small Pneumothorax in 18 cases (33%). One case of subcapsular hematoma was reported after a liver biopsy. No complication required further work-up.

Our data did not show any relation between the number of the passes and the rate of complication. Our data demonstrates that the use of TIC by on-site pathologist provided an efficient means of assessing sample adequacy while minimizing patient exposure to repeated biopsies and decreasing patient morbidity.

DISCUSSION

TIC is extremely helpful when evaluating malignancies and has been extensively used in evaluation of sentinel lymph nodes in breast cancer, with studies demonstrating that TIC and frozen section histological diagnosis were comparable for detecting metastasis in the sentinel lymph nodes.[7] The use of TIC as an adjunct to CNB, not only improves the accuracy of diagnosis[61115] but decreases patient exposure to unnecessary further intervention and discomfort.[711121314151617] Whereas tissue fixation, sectioning and further evaluation may take 24-48 h,[9] TIC would provide the physician with a preliminary diagnosis in mere 1-2 min, enabling for a more rapid initiation of ancillary testing.

While it has been stated that CT-guided core needle biopsies on average could require 3-7 passes to ensure adequate sampling of the tissue,[9] undergoing less passes could reduce the risk of trauma, hemorrhage and potential for infection.[1213] An on-site evaluation of TIC by a pathologist is useful in assessing if adequate sample has been retrieved with each pass of the CNB needle, eliminating the need for unnecessary passes to be performed.

On the other hand, tumor exhaustion in the tissue as a result of TIC is an important pitfall of the procedure and can be further prevented by a gentle touch of the core biopsy. After confirmation that the right lesion has been sampled, any additional biopsy can be preserved for the permanent evaluation. This can also reduce chances of losing the material. Another problem with TIC arises with sampling of the soft-tissue tumors or recurrent tumors after radiation with extensive fibrotic background. In these cases TIC may not be helpful due to technically inferior material.

During CT-guided CNB, pathologists can direct radiologists in targeting the right lesion, efficiently triaging the specimen for histology and ancillary studies and alleviating patients discomfort by minimizing the number of needle passes. In addition, this approach also decreases the expenses associated with these procedures.[37]

Multiple studies have reported on the high accuracy of TIC as method for diagnosis of suspicious lesions.[29151819] In fact, some of these studies have shown an even greater improvement in diagnostic accuracy when TIC is combined with histopathologic studies indicating that the synergy of two methods yield better results than histopathology alone.[71519202122] Mannweiler et al. addressed multiple reasons for the increased sensitivity seen with the use of TIC such as the increase of surface area, the common problem of cutting and embedding biopsy cylinders, the amount of tissue “wasted” with sectioning of paraffin blocks and the possibility of aspiration effect during the core biopsy. With the use of on-site pathologist, one can attribute a further increase in the accuracy of TIC to improved positioning of the biopsy needle by the interventional radiologist.[11]

The use of TIC in CNB evaluation has potential to be helpful in a number of ways: For physicians, it can help in completing the procedure efficiently while getting the adequate sample that assists diagnosing the lesions with increased accuracy. For patients, TIC can be beneficial in minimizing procedure related complications and discomfort, providing a quick preliminary diagnosis and alleviating uncertainties and avoiding unnecessary costs associated with repeated procedures.

In our study, minimal complications associated with CNBs included small pneumothorax and hemorrhage in which are comparable to the previous study performed by Loh et al.[14]

Our study demonstrates that the use of TIC by an on-site pathologist was an effective and efficient method of obtaining adequate diagnostic material, while expediting ancillary studies and decreasing the risk and potential cost associated with the procedure.

COMPETING INTERESTS STATEMENT BY ALL AUTHORS

The authors declare that they have no competing interests.

AUTHORSHIP STATEMENT BY ALL AUTHORS

All authors of this article declare that we qualify for authorship as defined by the ICJME. Each author has participated sufficiently in the work and takes public responsibility for appropriate portions of the content of the article.

MP and AF are responsible for execution and analysis of the work and drafting the manuscript.

HA and MM are responsible for planning and revising the article and responsible for conceiving and coordinating the whole work, execution and analysis of the work and final approval of the version to be published.

Each author acknowledges that the final version was read and approved.

All authors take the responsibility of maintaining relevant documentation of records, slides and other data used in this study on archival material as per the Institutional policy. All patient identifiers were suppressed in the final analysis of the data.

ETHICS STATEMENT BY ALL AUTHORS

All authors take the responsibility of maintaining relevant documentation of records, slides and other data used in this study on archival material as per the Institutional policy. All patient identifiers were suppressed in the final analysis of the data.

EDITORIAL/PEER-REVIEW STATEMENT

To ensure the integrity and highest quality of CytoJournal publications, the review process of this manuscript was conducted under a double blind model.(authors are blinded for reviewers and vice versa) through automatic online system.