Evaluation of Role of Ultra Sound Guided Fine Needle Aspiration Cytology for Diagnosis of Ovarian Lesions with Particular References to Diagnostic Pitfalls

Introduction: Cytology is a simple and reliable method for diagnosis of a variety of female genital tract lesions.Previous workers also successfully utilized FNAC for assessment of ovarian SOLs. Still gynaecologists all over the world are reluctant to accept this procedure for diagnosis of malignant ovarian lesions due to potential risk of intraperitoneal tumor implantation. However a lot of studies have clearly documented that risk of tumor spreading by needle tract is negligible in comparison to the potential benefits of this simple,quick and effective modality of diagnosis. Geier and Strecker strongly recommended cytology for diagnosis of non-malignant ovarian lesions as well as inoperable, recurrent or metastatic ovarian malignancies. Aspiration of ovarian lesions can be done under the guidance of USG or computed tomography(CT). MRI is considered to be the best method for pre operative assessment of possible nature of ovarian masses. Higher cost limits its use in routine cases and it is also unsuitable for guiding purposes. CT is better regarding assessment of stage of tumors but is not only costly but also associated with significant risk of radiation. USG is relatively cheap, easy to perform and as a guiding method free from radiation

hazards.It provides real time guidance and multiple attempts can easily be made.FNAC under USG guidance effectively enhances the chance of aspirating adequate material from ovarian SOLs without any hazardous effects and at a reasonable expence 8,11 .
In the present study, USG guided FNAC has been utilized for primary assessment of ovarian lesions.

Aims and Objectives:
Aims and Objectives of the present study are -1.To evaluate the role of USG guided FNAC for diagnosis of ovarian mass lesions.2.To compare the cytological evaluation with final confirmatory histopathological diagnosis of available biopsied samples and to discuss possible causes of cytological misdiagnosis in cases with cytohistological diagnostic inconsistency.

Materials and Methods:
This study was done in the Department of Pathology, Bankura Sammilani Medical College, Bankura, WB, for a period of five years (1 st January 2007 to 31st December 2012) in collaboration with department of Radiodiagnosis of the same institute.All patients with USG proved ovarian lesions were included in the study group, only after receiving proper consent.Aspirations were done by 22-23guage needles fitted to 10ml disposable syringes.Lumber puncture needles we utilized for deep seated lesions according to standard recommendation 8 .Multiple passes were attempted in all cases.Smears or aspirated fluids were dealt following recommendations of other workers 8 and examined under microscope for cytological diagnosis.Biopsy samples were obtained in all possible cases and processed routinely.Finally, comparison between cyto and histological diagnosis were made and cases with disparity were identified and evaluated for probable causes of misinterpretations.Pain abdomen was the most frequent complaint in all categories followed by menstrual irregularities and abdominal palpable lump as seen in the table no 3. Biopsy samples were available in 47 out of 116 cytodiagnosed cases.In 41 cases(87.2%)final histopathological diagnosis were consistent with cytological categorization.But 6 cases(12.8%)failed to show correlation.In the non neoplastic group, out of 12 cytodiagnosed cases two cases were proved to be benign neoplasms on histology.Three cases of cytodiagnosed benign lesions were confirmed as malignant tumours on histopathology.Among malignant cases, one case was identified as benign neoplasm during histological evaluation (Table no 4).Table no.5 describes 6 cases with cytohistological diagnostic disparity.2 cases of cytodignosed follicular cysts proved to be serous cystadenomas on histology.Two cases of serous cystadenomas as diagnosed on cytology,were proved to be serous cystadenocarcinoma or border line serous tumour.One case of mucinous cystadenocarcinoma was wrongly interpreted as mucinous cystadenoma on cytology.On the other hand , another case with cytological diagnosis of mucinous cyst adenocarcinoma was diagnosed to be benign mucinous neoplasm after histopathological evaluation.So in the present study 1 false positive and 3 false negative cases were reported regarding cytodetection of malignancy.Sensitivity and specificity of cytodiagnosis for diagnosis of malignancy, as evidenced in our study were 85.71% and 96.55%, respectively.

Discussion:
In the present study, 7.2% cases were excluded due to inadequate aspirates.Even after repeated aspirations under guidance, no diagnosis could be offered in those cases.Failure rate of upto 20% for similar cause is reported by various workers 12 .More than half the cases in our series ,belonged to 21 to 40 years age group, as also the experience of other researchers 12 .

Table 1: Age distribution of cases.
Non-neoplastic and benign lesions were much frequent than malignant neoplasms, as evidenced during cytological evaluations.Among the neoplastic lesions surface epithelial tumours were commonest, quite consistent with the findings of other workers 12,13 .

Table 3: Common clinical presentation of various categories.
During cytohistological correlation, we achieved 87.2% consistent result.Higgin RV et al 14 and Goel S et al 12 reported 90% specificity of cytodoagnosis during assessment of ovarian lesions.
6 cases with faulty cytological interpretation are now discussed in detail.

Table 4: Cytohistological correlation.
Case No-1: 38 year female presented with a small [2cm in diameter] cystic uniloculated lesion of left ovary.
Case No-2: 29 year female with a 1.5 cm diameter small uniloculated cyst involving right ovary.
Aspirates from both cases were scanty fluid containing macrophages, degenerated cells and few round cells in small clusters having scanty to moderate cytoplasm in a clear background .The smears were interpreted as non-neoplastic cysts-possibly follicular cyst, according to criteria reported by Nunez et al 15 .Histopathological examination of both the cysts established diagnosis of serous cystadenoma.Possible cause of cytological misdiagnosis was scanty cellular materials from degenerated atrophic lining epithelium.Similar experiences were reported by a lot of other workers 13,[16][17] .
Case No-3: 39 year female with a right ovarian cyst measuring 6cm in diameter.
Case No-4: 43 year female with a small cyst of 4cm in diameter in right ovary.
Aspirates were straw coloured fluids.Smears were moderately cellular containing loose clusters of epithelial cells with minimum dysplastic changes mixed with histiocytes and foamy macrophages.
Cytodignosis in both cases were serous cystadenoma, as recommended by Ramzy et al 3 .
Histopathologically, lesions were interpreted as low grade serous cystadenocarcinoma and borderline serous tumors, respectively.Sampling errors leading to aspiration from areas of cysts lined by degenerated epithelium were responsible for relatively low cellularity and absence of obvious malignant features.Similar type of misdiagnosis were also reported by various workers 8,12 .
Case No-5: 48 year female presented with a moderately large cystic swelling (6cm in diameter) of left ovary.A drop of mucoid fluid was aspirated.Smears show abundant mucinous background with few clusters of columnar cells with cytoplasmic vacuoles and basally pushed bland appearing nuclei.A diagnosis of mucinous cystadenoma was offered following standard recommendation 12 .But histological diagnosis came out as mucinous cystadenocarcinoma.Aspiration of thick, abundant mucoid material obscuring cellular anaplasia is the possible cause of failure in this case, as also reported by Higgins et al 14 .

Conclusion:
USG guided FNAC is a quick, inexpensive and safe procedure for preliminary assessment of ovarian lesions.Present study also proved the effectivity of this diagnostic tool, achieving almost 90% accuracy in comparison to histopathology.Cases of misdiagnosis were dealt in detail and attempts were made to identify possible underlying causes.We sincerely hope that future workers will try their best for correction of those possible causes of failures.Thus, our knowledge will be enriched and cytology will become more and more effective for evaluation of ovarian lesions.

Table 5 : Inconsistent cases [categories of error: non neoplastic to benign ;non neoplastic to malig- nant (false-ve);benign to malignant(false- ve);benign to nonneoplastic; malignant to benign or nonneoplastic(false positive).]
18se No-6: 52 year female presented with a large multiloculated cyst (7.5cm in diameter) of left ovary.Mucoid material was aspirated.Smears were moderately cellular containing numerous columnar cell clusters and vague glandular structures often with multilayering in a background of thick mucinous material.Few cells show nuclear hypercromasia.Smears were interpreted as mucinous cystadenocar-cinoma.But subsequent histopathology confirmed the lesion as mucinous cystadenoma.Suggestive clinical presentation and presence of small glandular clusters and cell balls, giving impression of multilayering , as seen in smears were the causes behind over diagnosis of malignancy as also observed by Roy et al18.