JHC-2019-Jan-3 (3):27-37-Artile-4-Mahmuda Begum

Intraoperative Imprint Cytology of Ovarian Masses and Association with Histopathological Findings

 

*Begum M,1 Habiba U,2 Mosarrat SS,3 Barua M,4 Jahan M,5 Nasreen S,6Ahamad MS,7 Bhattacharjee P,8 Rahman MZ9

Download PDF

 

Abstract

Objective: The present study was conducted to assess whether imprint cytology during intraoperative procedure can help in rapid diagnosis of ovarian masses and thus facilitate individualized treatment. Aims of our study were to establish the validity and reliability of imprint cytology in intraoperative diagnosis of ovarian lesions and to compare the  cytological diagnosis with histopathological findings.

Methods: This  cross-sectional study was carried out in the Department of Pathology, Chittagong Medical College (CMC) in collaboration with Department of Gynaecology and obstetrics, Chittagong Medical College Hospital (CMCH) from July 2013 to June 2014. A total of 69 cases were selected consecutively. Intraoperative imprint smears were taken peroperatively in the Department of Gynaecology and Obstetrics, Chittagong Medical College Hospital.

Results: Out of total 69 cases, 56 (81.2.%) were benign lesions and 13 (18.8%) were malignant lesions on intraoperative imprint cytology (IOIC). Histopathological diagnosis showed 54 (78.3%) were benign lesions and 15(21.7%) were malignant lesions. There was no borderline malignancy in our study. Association between imprint cytology and histopathology was highly significant (P=0.000). The overall sensitivity was 86.70%, specificity was 100.00%, PPV 100.00%, NPV 96.4% and diagnostic accuracy was 97.10% with 2 false negative (FN) and no false positive case.

Conclusion: Imprint cytology can be used as an adjunct to histopathology for rapid and early diagnosis in the operation theatre, thus helping better management of patients.

 

[Journal of Histopathology and Cytopathology, 2019 Jan; 3 (1):27-37]

 

Keywords: IOIC, Ovarian mass, Histopapathology

 

  1. *Dr. Mahmuda Begum, Clinical Pathologist, Department of Clinical Pathology, Chittagong Medical College Hospital, Chittagong, Bangladesh. drmahmuda1998@gmail.com
  2. Ummeh Habiba, Clinical Pathologist, Department of Clinical Pathology Chittagong Medical College Hospital, Chittagong, Bangladesh.
  3. Shahe Systa Mosarrat, Clinical Pathologist, Department of Pathology Chittagong Medical College, Chittagong, Bangladesh.
  4. Mitasree Baura, Lecturer, Department of Pathology Chittagong Medical College, Chittagong, Bangladesh.
  5. Mahmuda Jahan, Assistant Professor, Department of Pathology, Southern Medical College, Chittagong, Bangladesh.
  6. Sayeeda Nasreen, Assistant Professor, Department of Pathology, Chittagong Medical College, Chittagong, Bangladesh.
  7. M Shahab Uddin Ahamad, Associate Professor, Department of Pathology, Chittagong Medical College, Chittagong, Bangladesh.
  8. Pradip Bhattacharjee, Associate Professor, Department of Pathology, Chittagong Medical College, Chittagong, Bangladesh.
  9. Md. Zillur Rahman, Associate Professor, Department of Pathology, Chittagong Medical College, Chittagong, Bangladesh.

 *For correspondence

 Introduction

Ovarian cancer represents the sixth most commonly diagnosed cancer among women in the world.1 Majority of the ovarian masses are benign (80%) with cystic, solid or mixed characteristics. The remaining (20%) of these masses are malignant in nature and run a fatal prognosis. Furthermore, the lifetime risk of development of ovarian cancer is   5-7%.  Due to the fatal outcome of this disease, early and accurate diagnosis is essential. The detection and assessment of ovarian lesions are an important part of gynaecological practice. Advance knowledge of whether an overt ovarian mass is malignant may be useful for improving the effectiveness of surgical treatment.2

The diagnosis of ovarian neoplasm depends on histopathological examinations as they are inaccessible for cytological techniques except when  approached through image guidance.3

The ovarian tumours manifest with wide spectrum of clinical, morphological and histological features. Screening for ovarian tumours  are improved by various diagnostic modalities, such as Doppler Color flow ultrasonography and transvaginal ultrasonography, measurement of tumour markers such as serum HCG, serum CA-125, Serum alpha-fetoprotein, placental alkaline phosphatase and lactate dehydrogenase, ovarian cancer antigen OVXI and CA 15-3 and many others. But their accessibility to the practicing gynaecologist for rural based poor population remains very limited even today.4

Despite the help from tumour markers, it may not be possible to determine before surgery whether patients presenting with ovarian or adnexal masses have benign or malignant disease. Clinical, serological and radiological findings are not  specific and the gross intraoperative findings may be misleading. Assessing the nature of tumors confined to the ovaries is problematic because benign and malignant neoplasm can have identical gross appearance.  As the optimal management of benign, borderline and malignant ovarian tumours differs and many young patients wish to retain fertility, intraoperative assessment is frequently used to provide a provisional diagnosis and to guide the extent of surgery and  additional staging procedures.5

Imprint cytology is a well-recognised simple technique for preparing a surgical specimen for pathological assessment. A valuable information on morphological details of the cell can be obtain through imprint cytology

Imprint cytology gives intraoperative diagnosis of ovarian mass within 20 minutes.  It is simple, inexpensive procedure. It offers excellent preservation of cellular details, with no loss of tissue as occurs with cryostat sections, and adequacy of surgical margins.6

Fine needle aspiration cytology in the preoperative investigation of ovarian tumours has been discouraged since puncture of a cystic carcinoma might cause intraperitoneal seedling; but intraoperative imprint cytology enables a rapid diagnosis without fear of dissemination of ovarian cancer.7

Intraoperative imprint cytology is being used increasingly and regularly in the diagnosis of disease of certain organs like thyroid and breast neoplasm where it is found to show accuracy and rapidity of diagnosis.8

The present study was carried out to determine the accuracy of imprint cytology in diagnosis of ovarian masses.

 Methods

This  cross-sectional study was carried out in the Department of Pathology, Chittagong Medical College (CMC) in collaboration with Department of Gynaecology and obstetrics, Chittagong Medical College Hospital (CMCH) from July 2013 to June 2014.

It was non probability convenience sampling . Total 69 female patients, irrespective of their age presenting with clinically palpable or radiologically detected ovarian mass attending in the Department of Gynaecology and obstetrics Chittagong Medical College Hospital were the study subjects. Intraoperative imprint smears were taken peroperatively in the Department of gynaecology and obstetrics, Chittagong Medical College Hospital.

 Inclusion Criteria

  • Female patients of any age with clinical & radiological evidence of ovarian mass who gave written informed consent for the study.
  • Patients who underwent

 Exclusion criteria

  • Those patients, who did not give written informed consent for the study.
  • Previously diagnosed patients of ovarian tumour.

 Clinical history, questionnaire, thorough physical examination, and interpretation of  relevant investigations like ultrasonography were recorded in details in all cases

Procedure

After collection of the fresh specimen imprint smears from cut surface of the lesions  were made immediately, applying  glass slides firmly to the surface.  The smears were immediately fixed in 95% ethyl alcohol and then stained with rapid   Haematoxylin and Eosin (H&E) stain. At least two cytology slides were prepared for each case and were examined under light microscope and reported as benign or malignant.

After the cytology slides were prepared with adequate material the specimens were   fixed in 10% formalin.

Gross examination of specimens were  done . During gross examination, special  emphasis was given on content of cystic mass, uni or multi loculated, any visible papillary growth, area of necrosis, and were noted. Blocks were taken from different sites.  Routine tissue processing with paraffin impregnating was done. Sections  prepared from the biopsy specimen were  stained by H&E stain and were examined under light microscope to get a definitive diagnosis of the lesions and their types. All the prepared slides from imprint were examined under light microscope and  compared with the histological diagnosis.

Data Analysis

Data were  processed and analyzed  using the SPSS (Statistic Package for Social Science) version-18 software . Various indices such as sensitivity, specificity, false positive rate, false negative rate, positive predictive value, negative predictive value and accuracy were calculated. The association between imprint cytology and histopathology was tested by Chi square test .

 Results

The age range of the 69 patients was 13-70 years. Mean age was 36.38 years, SD ± 14.1. The lesions were found 5 (7%) in age group 41-50 years and 5 (7%) in 51-60 years respectively. The cases were divided into six age group. It was observed that maximum number of benign lesions 17 (24.6%) were in age group  21-30 years and 15 (21.7%) were in age group 31-40 years. Maximum number of malignant lesions were found 5(7%) in age group 41-50 years and 5(7%) in 51-60 years, respectively. The age distribution which is shown in Table I.

 

Table I: Distribution of ovarian lesions according to age group (n = 69)

 

Age Range Histopathological Diagnosis Total
Benign Lesion Malignant Lesion
1.  ≤ 20 Years 09 00 09 (13.0%)
2.  21 – 30 Years 17 03        20 (29.0%)
3.  31 – 40 Years 15 01 16 (23.2.%)
4.  41 – 50 Years 05 05 10 (14.4%)
5.  51 – 60 Years 07 05 12 (17.4%)
6.  61 – 70 Years 01 01 02 (2.9%)
Total 54 (78.3%) 15 (21.7%) 69 (100.0%)

 

Most of the patients  58 (84.1%) were from average socio-economic condition and incidence of ovarian lesion was more common in multiparous 54 (78.3%) women.

 

Intraoperative imprint cytology (IOIC) impression in ovarian lesions smears

Intraoperative imprint cytology smears were taken from all69cases.Out of these, 56 (81.2%) cases were benign and 13 (18.8%) were malignant. Among benign cases, 23 (33.4%) were serous cyst adenoma, 16 (23.3%) were mucinous cyst adenoma, 4 (5.8%) mature teratoma, 2(2.9%) fibroma, 1(1.4%) benign mesenchymal lesion, 2 (2.9%) inflammatory lesion & 8 (11.6%) were benign cyst. Out of 13 malignant cases, 12 (17.4%) were diagnosed as positive for malignant cells and 1(1.4%) was reported as  suspicious  for malignancy.The distribution  is shown in table- II.

 

Table II: Distribution of intra-operative imprint cytology diagnosis and interpretations among the study subjects (n = 69)

 

Imprint cytology impression Imprint cytology

Diagnosis

 No(%)

 

 Total No (%)
Benign Lesions

 

 Serous cyst adenoma 23 (33.4) 56 (81.2)

 

 

 
Mucinous cyst adenoma 16 (23.3)
Benign cyst 8 (11.6)
Mature teratoma 4 (5.8)
Fibroma 2 (2.9)
Inflammatory lesions 2 (2.9)
Benign mesenchymal lesion 1 (1.4)
 Malignant Lesions

 

 Positive for malignant cell 12 (17.4) 13 (18.8)

 
Suspicious cell 1 (1.4)
  Total 69 (100) 69 (100)

 

Histopathological findings

Histopathological diagnosis showed 54 (78.3%)  benign lesions and 15 (21.7%)  malignant lesions. Among benign lesions,  23 (33.4%) were serous cyst adenoma, 15 (21.6%) were mucinous cyst adenoma, 04 (5.8%) mature teratoma, 02 (2.9%), fibroma, 2 (2.9%) inflammatory lesions and 8 (11.6%) were haemorrhagic cyst. Among malignant lesions, 5 (7.2%) were  serous cyst adenocarcinoma, 4 (5.8%) mucinous cyst adenocarcinoma, 1(1.4%) endometrioid carcinoma, 1 (1.4%) fibrosarcoma, 1(1.4%) yolksac tumour 1(1.4%) clear cell carcinoma, 1(1.4%) struma ovarii, 1(1.4%) granulosa cell tumour. The distribution is shown in table III

 

Table III: Distribution of histopathology diagnosis and interpretations among the study subjects (n = 69)

 

Histopathological impression Histopathological

Diagnosis

 No(%)

 

 Total No (%)
 

Benign Lesions

 

 

 

 

 

 Serous cyst adenoma 23 (33.4) 54 (78.3%)

 

 

 
Mucinous cyst adenoma 15 (21.6)
Haemorrhagic cyst 8 (11.6)
Mature teratoma 4 (5.8)
Fibroma 2 (2.9)
Inflammatory lesions 2 (2.9)
 

 

 

Malignant Lesions

 

 

 Serous cyst adenocarcinoma 5 (7.2) 15 (21.7)

 

 

 

 
Mucinous cyst adenocarcinoma 4 (5.8)
Clear cell carcinoma 1 (1.4)
Endometrioid carcinoma 1 (1.4)
Fibro-sarcoma 1 (1.4)
Follicular variant of Papillary carcinoma within struma ovarii 1 (1.4)
Granulosa cell tumour 1 (1.4)
Yolk sac tumour 1 (1.4)
  Total 69 (100) 69 (100)

Comparison of IOIC impressions and histopathological diagnosis

IOIC impression of 69 cases showed 56 benign and 13 malignant cases. Out of 56 benign IOIC impressions 54 proved to be benign histologically and 2 cases were malignant. Out of 13 malignant IOIC impressions, all were proved to be malignant histologically. Association between imprint cytology and histopathology was shown in Table IV . True positive (TP) cases are 13, false negative (FN) cases are 2, true negative (TN) cases are 54 and no false positive (FP) case.

 

Table  IV: Association between imprint cytology & histopathology

 

Imprint Cytology (IC) Histopathology
Benign Lesion Malignant Lesion Total
Benign Lesions 54 02 56
Malignant Lesions 00 13 13
Total 54 15 69

 

Chi-square (X2) = 57.664; P = 0.000 (Highly Significant)

 

Statistical analytic result of IOIC reports

In Intraoperative imprint cytology the sensitivity was found 86.7%, specificity 100.0%, PPV 100%, NPV 96.40% and accuracy was 97.1%. The result  are shown in Table V.

 

Table V: Evaluation of Imprint Cytology as screening test in respect to histopathology as diagnostic test

 

Validity of IC (Imprint cytology)
Sensitivity 86.70%
Specificity 100.00%
Positive Predictive Value 100.00%
Negative Predictive Value 96.40%
Diagnostic Accuracy 97.10%

 

 

 

 

 

 

 

 

 

 

Fig 1. Serous cyst ademoma (imprint smear H & E stain 400X)

 

 

 

 

 

 

 

 

 

 

Fig 2. Serous cyst ademoma (Histo pathology H & E stain 400X)

 

 

 

 

 

 

 

 

 

Fig 3. Serous cyst adenocarcinoma (imprint smear H & E stain 400X)

 

 

 

 

 

 

 

 

 

Fig 4. Serous cyst adenocarcinoma (Histopathology H & E stain 400X)

 

 

 

 

 

 

 

 

Fig 5. Mucinous cyst adenoma (Histopathology H & E stain 400X)

 

 

 

 

 

 

 

 

 

 

Fig 6. Mucinous cyst adenoma (Imprint smear H & E stain 400X)

 

 

 

 

 

 

 

 

 

 

Fig 7. Mucinous cyst adenocarcinoma (imprint smear H & E stain 400X)

 

 

 

 

 

 

 

 

 

Fig 8. Mucinous cyst adenocarcinoma (Histopathology H & E stain 400X)

 

Discussion

The age range of study patients was 13-70 years. The mean age was 36.38 years. Patients were divided into six age groups each consisting of a decade. Maximum number of patients (20, 29.0%) were in 21-30 year group,   followed by 16 (23.2%)  in 31-40 year age group . Study done by Vaidya et al. (2014) showed highest number of patients (58%) were in 21-40 year age group which is similar to our study.9 In this study were in 21-40 year age group; total number of patients were 36 (52.2%). In a study done by Bhagyalakshmi et al. (2014), patients age ranged from 11-70 years, majority  of benign cases. Malignant cases were between 21 to 40 years and 41 to 60 years respectively.10  Tushar et al. (2005) also showed age range  from 16 to 70 years with  benign and malignant lesions  in 21 to 40 years and 41 to 60 years respectively, which was comparable with this study.11 Intraoperative imprint smears in this study collected from 69 cases revealed 56 (81.2%)  benign lesions and 13 (18.8%)  malignant lesions. Among the benign cases, 23 (33.4%) cases were serous cyst adenoma, 16 (23.3%) cases were mucinous cyst adenoma, 8 (11.6%) were benign cyst, 4 (5.8%) were mature cystic teratoma, 2 (2.9%) were fibroma, 2 (2.9%) were inflammatory lesions and 01 (1.4%) case was benign mesenchymal lesion. Among the malignant cases, 12 (17.4%) cases were diagnosed as positive for malignant cells and 1 (1.4%) case was suspicious  for malignancy. On histological examination, among 15 malignant cases, 13 cases were proved correctly diagnosed  by IOIC. 2 cases that were diagnosed by IOIC as mucinous cyst adenoma and  benign mesenchymal lesion, were subsequently confirmed histopathologically as mucinous cyst adenocarcinoma and fibrosarcoma respectively. Fibrosarcoma  was missed due to scanty cellular material in imprint smear. Mucinous neoplasms are particularly problematic because they are often large and frequently show benign, border line and malignant features  within a single tumour. Thus the limited sample examined in the intraoperative cytology procedure may not be  representative of the highest grade component and there is  chance of false negative diagnosis.12  There was no false positive diagnosis in imprint smear, similar findings were seen by Michael et al (1996) in their study.13 All 54 cases that were correctly diagnosed benign by IOIC,were also found benign by histopathology .

 

Incidence of benign lesions were (78.3%) and malignant lesions were (21.7%) and there was no borderline malignancy histologically in the present study . Study done by Yogambal et al, Mali et al and  Pilli et al showed incidence of benign  lesion  more than malignant lesions.14,15,16 Study done by Yasmine et al showed  no borderline malignancy.17 In this study percentages of benign and malignant lesions were nearer to that of our study.

 

Statistical analysis of the study cases was done considering the imprint cytology as screening test and histopathology as gold standard. Diagnostic accuracy of imprint cytology was 97.10% with 13  true positive cases, no false positive  case, 54 true negative cases and 2 false negative cases.  The sensitivity of IOIC was found 86.70%, specificity 100.00%, PPV 100.00%, NVP 96.40% and accuracy was 97.10%.

 

In this study the sensitivity, specificity and accuracy are statistically significantly comparable to those reported in many studies. Study of Saito et al showed sensitivity, specificity, accuracy of IOIC were 90.52%, 100%, 93.95%, respectively.18 study by Santin et al  showed sensitivity, specificity, accuracy were 93%,98%, 97% respectively.19 Sahid et al study showed  Sensitivity, specificity, accuracy were 95.8%,96%, 95.8% respectively.6

 

This study has evaluated the accuracy of intraoperative imprint smears of ovarian lesions both benign and malignant. Accuracy, sensitivity and specificity of the test show overall success of the study. Sensitivity measures how well a test identifies truely ill people and specificity measures how well a test identifies truely well people. Sensitivity will be high if false negative report is low. Specificity will be high if false positive report will be low. The PPV and NPV for IOIC of ovarian lesions were 100% and 96.40% respectively. The PPV and NPV for IOIC in this study reflect a chance to missing cancer of 3.60% ,and no chance of over diagnosis of a cancer. The comparable study between IOIC and histopathology was highly significant (P=0.000).

 

However, different studies have compared diagnostic efficacy of imprint cytology with that of frozen sections (FS). Issam et al showed sensitivity, specificity and accuracy for IOIC were 95.8%, 100%, 96.8% respectively and for FS these were 91.7%, 100% and 90.3% respectively.20  Succhi et al showed sensitivity, specificity and accuracy for IOIC were 94.9%, 96.8%, 96% respectively and for FS  were 89.9%, 97.9% and  94.9% respectively.21 IOIC is an acceptable alternative in hospital with limited technical, financial and human supplies for frozen section with available experienced cytopathologist.

 

Intraoperative imprint smears do not require any specialized equipment. It is cost effective, less time consuming, with at per accuracy as compared to frozen sections. It has its own inherent advantages. Thus intraoperative imprint smear is safe, reliable, rapid and economic diagnostic aid for intraoperative consultation.

 

Conclusion

Intraoperative imprint smears like frozen sections help in on table diagnosis, help in taking proper decisions with prior counseling of the patients, so that a repeat surgery can be  avoided. Imprint cytology is an acceptable method for diagnosis of ovarian lesions. It is  simple, inexpensive and rapid diagnostic tool and can be used as alternative to frozen section where facilities for frozen section are not available.

 

References

  1. Permuth WJ and SellersTA. (2009) Epidemiology of ovarian cancer, 492: pp1-20.
  2. Pelusi G, TaronivB and Flamigni C. (1996) Benign ovarian tumours, Frontiers in bioscience;1: pp 861-70.
  3. Vijayakumar A. The diagnostic utility of intraoperative cytology in the management of ovarian tumour. J Clin Diagn Res 2013; 7(6): 047-1050.
  4. Chandra Shekhar T, Govindraja E and Chandana N. The diagnostic accuracy of intraoperative imprint cytology in ovarian tumour. Journal of Dental and Medical sciences .2014; 3( 43):10614 – 10625.
  5. Coffey Kaplan AI and Ramzy I. Intraoperative consultation in gynecologic Pathology. J Arch Pathol 2005;129: 1544-1547.
  6. Sahid M, Zahee S, Mubeen A, Rahman K and Sherwani RK. The role of intraopereaive cytology in the diagnostic evaluation of ovarian neoplasms.Acta cytological 2012; 56:467-973.
  7. Dey S., Vatsala Misra V, Singh PA, Sanjay Mishra S and  Sharma N. Role of intraoperative imprmt cytology in diagnosis of suspected ovarian neoplasms. Asian Pacific Journal of Cancer Prevention 2010;11:1389-1391.
  8. Veneti S, Loannidou-Mouzaka L, Toufixi H, Xenitides J and Anastasiadis P. Imprint Cytology, A rapid, Reliable method of diagnosing brest malignancy. Acta Cytol 1996; 40:649-652.
  9. Vadiya S, Sharma P, KC S and Vaidya SA Spectrum of ovarian tumours in a referral hospital in Nepal, Journal of pathology of Nepal 2014; 4:539-543.
  10. Bhagyalakshmi A, Sreelekha A, Sridev S, Chandralekha J, Pravathi G and Venkatalakshmi, A. Prospective study of histopathological patterns of ovarian tumours in a tertiary care center. Int J Res and Sci 2014; 2 (2):448-956.
  11. Tushar K, Asaranti K and Mohapatra PC. Intraoperative cytology of ovarian tumours. J Obstet Gynecol 2005; (55) 4:345-349.
  12. Stewart CJR, Brennan BA, Koay E,  Nuren A and Ruba S. Value of cytology in the intraoperative assessment of ovarian tumors. Cancer Cytopathology 2010; 118: pp 127-136.
  13. Michael CW, Lawrence WD and Bedrossian CW. Intraoperative consultation in ovarian lesions a comparison between cytology and frozen section. Diagn-Cytopathol 1996; 15(5): 387-394.
  14. Yogambal M, Arunalatha P, Chandramoulesswari K and Palaniappan V. ovarian tumours incidence and distribution in a tertiary referral centre in south India, IOSR Journal of Dental and Medical sciences 2014; 13 (2):74-80.
  15. Mali M, Vyas B, Gupta S and Deasi H. Ahistological study of ovarian tumours in different age group. Int J Med Sci public health 2012; 3(3): 338 341.
  16. Pilli GS, Sunitha KP, Dhaded AV and Yenni W. Ovarian tumour a study of 282 cases. J Indian Med Associ 2002; 100(7):420-424.
  17. Yasmine S, Yasmine A and Asif M. Clinichistological pattern of ovarian tumours in Peshwar region J Ayub Med Coll Abbottabad 2008; 20 (4):pp11.
  18. Saito M. An application of cytology for ovarian tumors, Rinsho-Byori 1992;40(2): pp156-162.

 

  1. Santin A C, Sica A, Melesi S, Feijo A, Garrido G.and Rodriguez et al. contribution of intraoperetive cytology to the diagnosis of ovarian lesion. Acta Cytol 2011;55(1): 85-91.
  2. Issam M, Francis K and Dilip D. Role of Fine Needle Aspiration, Intraoperative Imprint Cytology and Frozen Section in the Diagnosis of Breast Lumps and Thyroid Lesions. Med Principles Pract 2009; 8: 173-182.
  3. Scucchi LF, Stefano DD, Cosentino L and Vecchino A. Value of cytology as an adjunctive intraoperative diagnosis method. Acta Cytol 1997; 41:1489-1496.