- Email: [email protected]
The accuracy of frozen section (intraoperative consultation) in the diagnosis of ovarian masses
Gynecologic Oncology 97 (2005) 395 – 399 www.elsevier.com/locate/ygyno
The accuracy of frozen section (intraoperative consultation) in the diagnosis of ovarian masses Sennur Ilvana,T, Rana Ramazanoglua, Elif Ulker Akyildizb, Zerrin Calaya, Tugan Besec, Nafi Oruca a
Department of Pathology, Cerrahpasa School of Medicine, Istanbul University, Istanbul, Turkey b The Institute of Forensic Medicine, Ministry of Justice, Turkey c Department of Gynecology, Cerrahpasa School of Medicine, Istanbul University, Istanbul, Turkey Received 7 August 2004 Available online 16 March 2005
Abstract Objective. Frozen section is an important and helpful adjunct in the intraoperative diagnosis of ovarian tumors. This retrospective study was undertaken to determine the accuracy of frozen section diagnosis of ovarian masses and the reasons of discordance. Methods. From January 1995 to December 2003, 1494 ovarian specimens were received for histopathological evaluation, and 617 of them were submitted for frozen section examination. Results. The final paraffin section diagnoses of these 617 cases were a nonneoplastic lesion in 18.3% of the cases, benign tumor in 56.1%, borderline tumor in 6.2%, and malignant tumor in 19.4%. The overall accuracy was 97%. Twenty-one cases were incorrectly diagnosed by frozen section. All of them were false negatives. There were no deferred cases. The majority of the cases of disagreement were mucinous and borderline tumors. The sensitivity for benign, borderline, and malignant tumors were 100%, 87%, and 87%, respectively. The specificity for benign tumors was 97%; for borderline tumors 98%; and for malignant tumors 100%. Conclusion. Our data confirm that frozen section diagnosis is a reliable method for the surgical management of patients with an ovarian mass. However, diagnostic problems can occur in mucinous and borderline tumors during frozen section examination. The clinicians and pathologists must be aware of the pitfalls of this method; therefore, a good communication established between them is necessary to obtain more accurate results and to minimize the number of deferred cases. D 2005 Elsevier Inc. All rights reserved. Keywords: Ovary; Frozen section; Intraoperative consultation
Introduction Preoperative evaluation of patients with an ovarian mass is usually made by imaging studies and determination of serum levels of tumor markers. Since these methods have limited value for the recognition of ovarian cancer [1,2], the diagnosis and the course of the surgery are usually determined by frozen section examination during the operation. Therefore, the accuracy of frozen section diagnosis
of ovarian tumors is important especially in young women who may be managed conservatively with preservation of fertility. In this study, we aimed to determine the agreement between frozen and final paraffin section diagnosis of ovarian lesions in our department and to discuss the reasons of discordance.
Materials and methods T Corresponding author. Tibbiye Caddesi, GATA Haydarpasa Askeri Lojmanlari A-blok Daire: 10, 34660, Uskudar, Istanbul, Turkey. Fax: +90 216 3278513. E-mail address: [email protected] (S. Ilvan). 0090-8258/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2005.01.037
In this retrospective study, we reviewed the pathology reports of 617 ovarian lesions evaluated according to frozen and paraffin section diagnosis in our department between
396
S. Ilvan et al. / Gynecologic Oncology 97 (2005) 395–399
January 1, 1995 and December 31, 2003. The frozen section diagnosis was compared with the final paraffin section diagnosis in terms of whether it was a nonneoplastic lesion or a benign, borderline, and malignant tumor. The frozen and paraffin section diagnoses were reported by one of the two pathologists (SI and ZC) experienced in gynecological pathology. In 590 out of 617 cases, the whole of the tumor or the cyst was removed, and in the remaining 27 cases, frozen section examination was made from incisional biopsy material. After careful macroscopic evaluation of the specimens by the one of the two pathologists, touch imprints were prepared from the most suspicious areas, and one to four of the most representative samples were frozen in a cryostat and 5-Am sections were taken and stained with hematoxylin and eosin. In all cases, at least one sample was obtained for every 1 cm of maximal tumor diameter for paraffin section examination. Furthermore, even the whole of the specimen was sampled in suspicious lesions. Depending on the confidence of the pathologist, 46 cases (7.5%) were diagnosed on gross inspection alone. If there were any discrepancies between the frozen section diagnosis and final diagnosis, the slides were reassessed by the pathologists to ascertain whether the errors were due to gross sampling or interpretation. Mucinous tumors were diagnosed using the criteria defined by Hart and Norris [3]. The diagnosis of other tumors was made according to the guidelines of Scully [4].
Results Between the given years above, a total of 1494 ovarian masses was assessed in our pathology department, out of which 617 (41.3%) was sent for frozen section examination. The mean age of these patients was 50.8 (range, 16 to 83 years). On paraffin block examination, 113 (18.3%) of these 617 cases were nonneoplastic lesions, 346 (56.1%) benign tumors, 38 (6.2%) borderline tumors, and 120 (19.4%) malignant tumors (Table 1). Intraoperative consultation was performed in 22.2% of nonneoplastic lesions, 60.7% of benign tumors, 84.4% of borderline tumors, and 32.5% of Table 1 Final histological classification of all cases, in which frozen section examination was requested or not Classification
FS requested n (%)
FS not requested n (%)
Nonneoplastic lesions Benign tumors Borderline tumors Malignant tumors Total
113 (22.2)
397 (77.8)
346 38 120 617
224 7 249 877
FS: frozen section.
(60.7) (84.4) (32.5) (41.3)
(39.3) (15.6) (67.5) (58.7)
malignant tumors. The exact histologic types of all the cases, to which frozen section examination requested or not, was given in Table 2. Frozen section diagnosis agreed with final paraffin section diagnosis in 596 cases (Table 3). The overall accuracy was found as 97%. There was no incorrect diagnosis among the nonneoplastic lesions. The accuracy rates for benign, borderline, and malignant tumors were 99%, 97%, and 97%, respectively. The sensitivity for benign tumors was 100%, specificity was 97%, and positive and negative predictive values were 98% and 100%, respectively. For borderline tumors, the corresponding values were 87%, 98%, 72%, and 99%, and for malignant tumors, 87%, 100%, 100%, and 97%, respectively. At intraoperative consultation, 21 cases were diagnosed incorrectly. All of them were false negatives. There were no false positive cases. Comparison of frozen section diagnoses with final paraffin diagnoses of 21 cases, in which there was disagreement, revealed the following (Table 4): 13 cases with intraoperative diagnosis of borderline tumor were later classified as mucinous (11 cases) and serous (2 cases) carcinomasgrade I. In 5 cases, the frozen section diagnosis of benign tumor turned out to be mucinous borderline tumor in 4 cases and serous in 1 case in paraffin sections. The two cases were misdiagnosed as mature teratoma on frozen sections, but later immature glial elements were seen focally, and final diagnoses of these two cases were reported as immature teratoma-grade I. The reason for discordance in these cases was due to sampling error. In one case, because of the solid macroscopic appearance and sparsity of signet-ring cells in microscopic examination, a metastatic tumor from the stomach was misdiagnosed as fibroma. The disagreement in this case was attributed to an interpretational error. There were no incorrect diagnoses in 46 cases evaluated by gross inspection alone during intraoperative consultation. Twenty-two of these cases were endometriotic cysts, 3 follicle cysts, 12 mature cystic teratomas, and 9 benign serous cystadenomas. Twenty of the 120 malignant ovarian tumors were metastatic, 8 from colon, 7 stomach, and 5 breast. Nineteen of 20 metastatic tumors were correctly diagnosed as metastatic during frozen section examination. In eight of these 20 cases, only after the frozen section diagnosis was given as metastatic, was the surgeon then able to detect a primary tumor in the abdomen during the operation.
Total (n) 510 570 45 369 1494
Discussion Despite its limitations, frozen section diagnosis is an important and reliable tool in the clinical management of patients with ovarian tumors. However, little information has been published concerning the utilization ratio of
S. Ilvan et al. / Gynecologic Oncology 97 (2005) 395–399
397
Table 2 Distribution of histological types of all cases according to final paraffin diagnosis
Table 4 Cases with disagreement between frozen section and final paraffin diagnosis
Histological type
FS requested (n)
FS not requested (n)
Frozen section diagnosis
Final paraffin diagnosis
No. of cases
Nonneoplastic lesions Endometriotic cyst Follicle cyst Corpus luteum cyst Abscess Others Epithelial tumors Benign Borderline Malignant Sex cord-stromal tumors Fibroma and fibrothecoma Granulosa cell Sertoli cell Germ cell tumors Mature teratoma Immature teratoma Dysgerminoma Yolk sac tumor Metastatic tumors Total
113 71 9 18 13 2 331 217 38 76 65 53
397 271 31 74 20 1 363 135 7 221 17 9
510 342 40 92 33 3 694 352 45 297 82 62
Borderline mucinous tumor Borderline serous tumor Benign mucinous tumor Benign serous tumor Mature teratoma Fibroma
Mucinous carcinoma, grade I Serous carcinoma, grade I Borderline mucinous tumor Borderline serous tumor Immature teratoma Metastatic tumor
11 2 4 1 2 1
12 – 88 76 4 5 3 20 617
7 1 81 80 – 1 – 19 877
19 1 169 156 4 6 3 39 1494
Total (n)
frequently (92.3%) to decide the course of the surgery, since these patients are usually young and fertile. With the exception of one case (dysgerminoma originating from a dysgenetic gonad), whole of the malignant germ cell tumors was sent for frozen section examination. Frozen section diagnosis should not be seen merely as a microscopic examination of the tissue. Rather, it is an intraoperative consultation method in which other diagnostic tests such as gross examination, touch imprints, and fine-needle aspiration cytology are used in combination. Hence, some have preferred that the term bfrozen section examinationQ be eliminated in favor of the term bintraoperative consultationQ [10]. Careful gross examination of the specimen is of utmost importance for both correct sampling and arriving at the correct diagnosis. Purely cystic lesions are almost certain to be benign and frozen section does not add significant accuracy to their differentiation from malignant and borderline tumors. Frozen section appears most helpful when the lesions are solid/cystic or completely solid [2]. However, it is noteworthy that malignant or borderline tumors and also metastatic tumors may show completely cystic features on gross examination. For this reason, the macroscopic evaluation of inner and outer surfaces of cyst must be done in great scrutiny. In our study, 46 cystic lesions were diagnosed intraoperatively by gross inspection alone and there was no misdiagnosis among these cases. Endometriotic cysts and mature cystic teratomas were sufficiently characteristic that a presumptive diagnosis could be offered based on the macroscopic appearance alone and the other cystic lesions had extremely thin walls with smooth inner and outer surfaces.
FS: frozen section.
frozen section in ovarian lesions. The recommended utilization ratio of frozen section in general surgical practice is between 5% and 15% [5,6]. Since the preoperative diagnosis of ovarian masses is limited [1,2], the surgeons need more frequent frozen section examination in these lesions. In the literature, the reported frozen section utilization ratios for ovarian lesions range from 7.4% to 47% [7–9]. In our study, 41.3% of 1494 ovarian lesions were examined intraoperatively. We established that surgeons did not need frozen section examination in 77.8% of nonneoplastic lesions. A majority of these were endometriotic cysts (271 cases). However, frozen section examination was performed in 60.7% of benign tumors and 84.4% of borderline tumors. This ratio decreased to 25.6% in malignant epithelial tumors, because obvious extraovarian dissemination was seen in majority of these cases during the operation. While, in malignant germ cell tumor group, the surgeons needed intraoperative consultation more Table 3 Comparison of frozen section diagnosis vs. final paraffin diagnosis Frozen section diagnosis
Final paraffin diagnosis Nonneoplastic lesion
Nonneoplastic lesion Benign tumor Borderline tumor Malignant tumor Total
113 – – – 113
Tumor
Total
Benign
Borderline
Malignant
– 346 – – 346
– 5 33 – 38
– 3 13 104 120
113 354 46 104 617
398
S. Ilvan et al. / Gynecologic Oncology 97 (2005) 395–399
In the literature, frozen section examination has a high accuracy rate in ovarian tumors, reported at greater than 90% [2,11–19]. In our study, the overall accuracy was 97%. The sensitivity rates for benign and malignant tumors were found as 100% and 87%, respectively, which correlate with the other series [12,14,16–18]. The sensitivity for borderline tumors was 87%. This ratio seems to be better than published rates in the literature, 45% to 78% [12,15–19]. We established the positive predictive value of frozen section evaluation in the diagnosis of ovarian malignancies to be 100%. This finding is consistent with the other studies which reported positive predictive values of 99.1% to 100%, making overtreatment an unlikely event [2,7,11,12,15]. In our study, diagnostic problems especially occurred in mucinous and borderline tumor groups as seen in previously reported studies [7,12,15,17,19]. Fifteen of our 21 cases with disagreement were mucinous tumors. In 11 cases, the frozen section diagnosis of borderline mucinous tumor turned out to be mucinous carcinoma-grade I. On paraffin sections, 5 of these cases showed no stromal invasion but contained epithelial multilayering of four or more cell thick and back-to-back glandular arrangement with no intervening stroma, these being the criteria defined by Hart and Norris [3]. Four cases of borderline mucinous tumors were misdiagnosed as benign mucinous tumor on frozen sections. We decided that the reason for discordance in all these cases was due to sampling error. Since mucinous ovarian tumors frequently contain benign, borderline, and malignant components together and have relatively larger dimensions, they are more likely to be underdiagnosed than serous ones [9,17,20,21]. In cases where the signet-ring cells are sparse, the stroma of Krukenberg tumor may mimic fibroma or fibrothecoma on frozen sections [22], as seen in our study. The causes of discordance and deferred cases are due to sampling error, misinterpretation, lack of communication with the surgeons, and technical problems [14]. The deferral rates reported in the literature range from zero to 6.3% [2,7,11–15]. We had no deferred case. We believe that the deferred cases were eliminated by good communication with the surgical team. Additionally, in our department, frozen and paraffin section examination of all gynecologic material is done by only two pathologists. During the intraoperative consultation, the pathologists are able to communicate with a responsible surgeon from the surgical team and also receive the clinical information, preoperative imaging, and intraoperative findings. The management of some suspicious cases was even planned together. Some authors mentioned that the experience of the pathologist is of crucial importance for the success of frozen section diagnosis [17,21]. However, Menzin et al. [23] stated that the level of expertise in gynecological pathology did not influence the accuracy of the frozen section diagnosis. In conclusion, frozen section evaluation of the ovarian masses provides mostly accurate diagnosis and guides the
surgeon planning the management of the operation. However, like the other diagnostic methods in medicine, frozen section examination also has some pitfalls. Pathologists and surgeons must be aware of the limitations of this procedure, and a good communication between them is required to obtain more accurate results.
References [1] Wakahara F, Kikkawa F, Nawa A, Tamakoshi K, Ino K, Maeda O, et al. Diagnostic efficacy of tumor markers, sonography, and intraoperative frozen section for ovarian tumors. Gynecol Obstet Invest 2001;52:147 – 52. [2] Lim FK, Yeoh CL, Chong SM, Arulkumaran S. Pre and intraoperative diagnosis of ovarian tumours: how accurate are we? Aust NZ J Obstet Gynaecol 1997;37:223 – 7. [3] Hart WR, Norris HJ. Borderline and malignant mucinous tumors of the ovary. Histologic criteria and clinical behavior. Cancer 1973;31: 1031 – 45. [4] Scully RE. Atlas of tumor pathology. Tumors of the Ovary and Maldeveloped Gonads. Washington7 Armed Forces Institute of Pathology; 1979. [5] Kindschi GW. Frozen sections. Their use and abuse. JAMA 1984;251:2559 – 60. [6] Kern WH. Frozen sections. Letter to the editor. JAMA 1985;253:512. [7] Twaalfhoven FC, Peters AA, Trimbos JB, Hermans J, Fleuren GJ. The accuracy of frozen section diagnosis of ovarian tumors. Gynecol Oncol 1991;41:189 – 92. [8] Maiman M, Seltzer V, Boyce J. Laparoscopic excision of ovarian neoplasms subsequently found to be malignant. Obstet Gynecol 1991;77:563 – 5. [9] Puls L, Heidtman E, Hunter JE, Crane M, Stafford J. The accuracy of frozen section by tumor weight for ovarian epithelial neoplasms. Gynecol Oncol 1997;67:16 – 9. [10] Fechner RE. Frozen section (intraoperative consultation). Hum Pathol 1988;19:999 – 1000. [11] Obiakor I, Maiman M, Mittal K, Awobuluyi M, DiMaio T, Demopoulos R. The accuracy of frozen section in the diagnosis of ovarian neoplasms. Gynecol Oncol 1991;43:61 – 3. [12] Rose PG, Rubin RB, Nelson BE, Hunter RE, Reale FR. Accuracy of frozen-section (intraoperative consultation) diagnosis of ovarian tumors. Am J Obstet Gynecol 1994;171:823 – 6. [13] Spann CO, Kennedy JE, Musoke E. Intraoperative consultation of ovarian neoplasms. J Natl Med Assoc 1994;86:141 – 4. [14] Usubutun A, Altinok G, Kucukali T. The value of intraoperative consultation (frozen section) in the diagnosis of ovarian neoplasms. Acta Obstet Gynecol Scand 1998;77:1013 – 6. [15] Wang KG, Chen TC, Wang TY, Yang YC, Su TH. Accuracy of frozen section diagnosis in gynecology. Gynecol Oncol 1998; 70:105 – 10. [16] Yeo EL, Yu KM, Poddar NC, Hui PK, Tang LC. The accuracy of intraoperative frozen section in the diagnosis of ovarian tumors. J Obstet Gynaecol Res 1998;24:189 – 95. [17] Gol M, Baloglu A, Yigit S, Dogan M, AydVn C, Yensel U. Accuracy of frozen section diagnosis in ovarian tumors: is there a change in the course of time? Int J Gynecol Cancer 2003;13:593 – 7. [18] Pinto PB, Andrade LA, Derchain SF. Accuracy of intraoperative frozen section diagnosis of ovarian tumors. Gynecol Oncol 2001;81: 230 – 2. [19] Tangjitgamol S, Jesadapatrakul S, Manusirivithaya S, Sheanakul C. Accuracy of frozen section in diagnosis of ovarian mass. Int J Gynecol Cancer 2004;14:212 – 9. [20] Houck K, Nikrui N, Duska L, Chang Y, Fuller AF, Bell D, et al. Borderline tumors of the ovary: correlation of frozen and
S. Ilvan et al. / Gynecologic Oncology 97 (2005) 395–399 permanent histopathologic diagnosis. Obstet Gynecol 2000;95: 839 – 43. [21] Kayikcioglu F, Pata O, Cengiz S, Tulunay G, Boran N, Yalvac S, et al. Accuracy of frozen section diagnosis in borderline ovarian malignancy. Gynecol Obstet Invest 2000;49:187 – 9.
399
[22] Acs G. Intraoperative consultation in gynecologic pathology. Semin Diagn Pathol 2002;19:237 – 54. [23] Menzin AW, Rubin SC, Noumoff JS, LiVolsi VA. The accuracy of a frozen section diagnosis of borderline ovarian malignancy. Gynecol Oncol 1995;59:183 – 5.
The accuracy of frozen section (intraoperative consultation) in the diagnosis of ovarian masses Sennur Ilvana,T, Rana Ramazanoglua, Elif Ulker Akyildizb, Zerrin Calaya, Tugan Besec, Nafi Oruca a
Department of Pathology, Cerrahpasa School of Medicine, Istanbul University, Istanbul, Turkey b The Institute of Forensic Medicine, Ministry of Justice, Turkey c Department of Gynecology, Cerrahpasa School of Medicine, Istanbul University, Istanbul, Turkey Received 7 August 2004 Available online 16 March 2005
Abstract Objective. Frozen section is an important and helpful adjunct in the intraoperative diagnosis of ovarian tumors. This retrospective study was undertaken to determine the accuracy of frozen section diagnosis of ovarian masses and the reasons of discordance. Methods. From January 1995 to December 2003, 1494 ovarian specimens were received for histopathological evaluation, and 617 of them were submitted for frozen section examination. Results. The final paraffin section diagnoses of these 617 cases were a nonneoplastic lesion in 18.3% of the cases, benign tumor in 56.1%, borderline tumor in 6.2%, and malignant tumor in 19.4%. The overall accuracy was 97%. Twenty-one cases were incorrectly diagnosed by frozen section. All of them were false negatives. There were no deferred cases. The majority of the cases of disagreement were mucinous and borderline tumors. The sensitivity for benign, borderline, and malignant tumors were 100%, 87%, and 87%, respectively. The specificity for benign tumors was 97%; for borderline tumors 98%; and for malignant tumors 100%. Conclusion. Our data confirm that frozen section diagnosis is a reliable method for the surgical management of patients with an ovarian mass. However, diagnostic problems can occur in mucinous and borderline tumors during frozen section examination. The clinicians and pathologists must be aware of the pitfalls of this method; therefore, a good communication established between them is necessary to obtain more accurate results and to minimize the number of deferred cases. D 2005 Elsevier Inc. All rights reserved. Keywords: Ovary; Frozen section; Intraoperative consultation
Introduction Preoperative evaluation of patients with an ovarian mass is usually made by imaging studies and determination of serum levels of tumor markers. Since these methods have limited value for the recognition of ovarian cancer [1,2], the diagnosis and the course of the surgery are usually determined by frozen section examination during the operation. Therefore, the accuracy of frozen section diagnosis
of ovarian tumors is important especially in young women who may be managed conservatively with preservation of fertility. In this study, we aimed to determine the agreement between frozen and final paraffin section diagnosis of ovarian lesions in our department and to discuss the reasons of discordance.
Materials and methods T Corresponding author. Tibbiye Caddesi, GATA Haydarpasa Askeri Lojmanlari A-blok Daire: 10, 34660, Uskudar, Istanbul, Turkey. Fax: +90 216 3278513. E-mail address: [email protected] (S. Ilvan). 0090-8258/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2005.01.037
In this retrospective study, we reviewed the pathology reports of 617 ovarian lesions evaluated according to frozen and paraffin section diagnosis in our department between
396
S. Ilvan et al. / Gynecologic Oncology 97 (2005) 395–399
January 1, 1995 and December 31, 2003. The frozen section diagnosis was compared with the final paraffin section diagnosis in terms of whether it was a nonneoplastic lesion or a benign, borderline, and malignant tumor. The frozen and paraffin section diagnoses were reported by one of the two pathologists (SI and ZC) experienced in gynecological pathology. In 590 out of 617 cases, the whole of the tumor or the cyst was removed, and in the remaining 27 cases, frozen section examination was made from incisional biopsy material. After careful macroscopic evaluation of the specimens by the one of the two pathologists, touch imprints were prepared from the most suspicious areas, and one to four of the most representative samples were frozen in a cryostat and 5-Am sections were taken and stained with hematoxylin and eosin. In all cases, at least one sample was obtained for every 1 cm of maximal tumor diameter for paraffin section examination. Furthermore, even the whole of the specimen was sampled in suspicious lesions. Depending on the confidence of the pathologist, 46 cases (7.5%) were diagnosed on gross inspection alone. If there were any discrepancies between the frozen section diagnosis and final diagnosis, the slides were reassessed by the pathologists to ascertain whether the errors were due to gross sampling or interpretation. Mucinous tumors were diagnosed using the criteria defined by Hart and Norris [3]. The diagnosis of other tumors was made according to the guidelines of Scully [4].
Results Between the given years above, a total of 1494 ovarian masses was assessed in our pathology department, out of which 617 (41.3%) was sent for frozen section examination. The mean age of these patients was 50.8 (range, 16 to 83 years). On paraffin block examination, 113 (18.3%) of these 617 cases were nonneoplastic lesions, 346 (56.1%) benign tumors, 38 (6.2%) borderline tumors, and 120 (19.4%) malignant tumors (Table 1). Intraoperative consultation was performed in 22.2% of nonneoplastic lesions, 60.7% of benign tumors, 84.4% of borderline tumors, and 32.5% of Table 1 Final histological classification of all cases, in which frozen section examination was requested or not Classification
FS requested n (%)
FS not requested n (%)
Nonneoplastic lesions Benign tumors Borderline tumors Malignant tumors Total
113 (22.2)
397 (77.8)
346 38 120 617
224 7 249 877
FS: frozen section.
(60.7) (84.4) (32.5) (41.3)
(39.3) (15.6) (67.5) (58.7)
malignant tumors. The exact histologic types of all the cases, to which frozen section examination requested or not, was given in Table 2. Frozen section diagnosis agreed with final paraffin section diagnosis in 596 cases (Table 3). The overall accuracy was found as 97%. There was no incorrect diagnosis among the nonneoplastic lesions. The accuracy rates for benign, borderline, and malignant tumors were 99%, 97%, and 97%, respectively. The sensitivity for benign tumors was 100%, specificity was 97%, and positive and negative predictive values were 98% and 100%, respectively. For borderline tumors, the corresponding values were 87%, 98%, 72%, and 99%, and for malignant tumors, 87%, 100%, 100%, and 97%, respectively. At intraoperative consultation, 21 cases were diagnosed incorrectly. All of them were false negatives. There were no false positive cases. Comparison of frozen section diagnoses with final paraffin diagnoses of 21 cases, in which there was disagreement, revealed the following (Table 4): 13 cases with intraoperative diagnosis of borderline tumor were later classified as mucinous (11 cases) and serous (2 cases) carcinomasgrade I. In 5 cases, the frozen section diagnosis of benign tumor turned out to be mucinous borderline tumor in 4 cases and serous in 1 case in paraffin sections. The two cases were misdiagnosed as mature teratoma on frozen sections, but later immature glial elements were seen focally, and final diagnoses of these two cases were reported as immature teratoma-grade I. The reason for discordance in these cases was due to sampling error. In one case, because of the solid macroscopic appearance and sparsity of signet-ring cells in microscopic examination, a metastatic tumor from the stomach was misdiagnosed as fibroma. The disagreement in this case was attributed to an interpretational error. There were no incorrect diagnoses in 46 cases evaluated by gross inspection alone during intraoperative consultation. Twenty-two of these cases were endometriotic cysts, 3 follicle cysts, 12 mature cystic teratomas, and 9 benign serous cystadenomas. Twenty of the 120 malignant ovarian tumors were metastatic, 8 from colon, 7 stomach, and 5 breast. Nineteen of 20 metastatic tumors were correctly diagnosed as metastatic during frozen section examination. In eight of these 20 cases, only after the frozen section diagnosis was given as metastatic, was the surgeon then able to detect a primary tumor in the abdomen during the operation.
Total (n) 510 570 45 369 1494
Discussion Despite its limitations, frozen section diagnosis is an important and reliable tool in the clinical management of patients with ovarian tumors. However, little information has been published concerning the utilization ratio of
S. Ilvan et al. / Gynecologic Oncology 97 (2005) 395–399
397
Table 2 Distribution of histological types of all cases according to final paraffin diagnosis
Table 4 Cases with disagreement between frozen section and final paraffin diagnosis
Histological type
FS requested (n)
FS not requested (n)
Frozen section diagnosis
Final paraffin diagnosis
No. of cases
Nonneoplastic lesions Endometriotic cyst Follicle cyst Corpus luteum cyst Abscess Others Epithelial tumors Benign Borderline Malignant Sex cord-stromal tumors Fibroma and fibrothecoma Granulosa cell Sertoli cell Germ cell tumors Mature teratoma Immature teratoma Dysgerminoma Yolk sac tumor Metastatic tumors Total
113 71 9 18 13 2 331 217 38 76 65 53
397 271 31 74 20 1 363 135 7 221 17 9
510 342 40 92 33 3 694 352 45 297 82 62
Borderline mucinous tumor Borderline serous tumor Benign mucinous tumor Benign serous tumor Mature teratoma Fibroma
Mucinous carcinoma, grade I Serous carcinoma, grade I Borderline mucinous tumor Borderline serous tumor Immature teratoma Metastatic tumor
11 2 4 1 2 1
12 – 88 76 4 5 3 20 617
7 1 81 80 – 1 – 19 877
19 1 169 156 4 6 3 39 1494
Total (n)
frequently (92.3%) to decide the course of the surgery, since these patients are usually young and fertile. With the exception of one case (dysgerminoma originating from a dysgenetic gonad), whole of the malignant germ cell tumors was sent for frozen section examination. Frozen section diagnosis should not be seen merely as a microscopic examination of the tissue. Rather, it is an intraoperative consultation method in which other diagnostic tests such as gross examination, touch imprints, and fine-needle aspiration cytology are used in combination. Hence, some have preferred that the term bfrozen section examinationQ be eliminated in favor of the term bintraoperative consultationQ [10]. Careful gross examination of the specimen is of utmost importance for both correct sampling and arriving at the correct diagnosis. Purely cystic lesions are almost certain to be benign and frozen section does not add significant accuracy to their differentiation from malignant and borderline tumors. Frozen section appears most helpful when the lesions are solid/cystic or completely solid [2]. However, it is noteworthy that malignant or borderline tumors and also metastatic tumors may show completely cystic features on gross examination. For this reason, the macroscopic evaluation of inner and outer surfaces of cyst must be done in great scrutiny. In our study, 46 cystic lesions were diagnosed intraoperatively by gross inspection alone and there was no misdiagnosis among these cases. Endometriotic cysts and mature cystic teratomas were sufficiently characteristic that a presumptive diagnosis could be offered based on the macroscopic appearance alone and the other cystic lesions had extremely thin walls with smooth inner and outer surfaces.
FS: frozen section.
frozen section in ovarian lesions. The recommended utilization ratio of frozen section in general surgical practice is between 5% and 15% [5,6]. Since the preoperative diagnosis of ovarian masses is limited [1,2], the surgeons need more frequent frozen section examination in these lesions. In the literature, the reported frozen section utilization ratios for ovarian lesions range from 7.4% to 47% [7–9]. In our study, 41.3% of 1494 ovarian lesions were examined intraoperatively. We established that surgeons did not need frozen section examination in 77.8% of nonneoplastic lesions. A majority of these were endometriotic cysts (271 cases). However, frozen section examination was performed in 60.7% of benign tumors and 84.4% of borderline tumors. This ratio decreased to 25.6% in malignant epithelial tumors, because obvious extraovarian dissemination was seen in majority of these cases during the operation. While, in malignant germ cell tumor group, the surgeons needed intraoperative consultation more Table 3 Comparison of frozen section diagnosis vs. final paraffin diagnosis Frozen section diagnosis
Final paraffin diagnosis Nonneoplastic lesion
Nonneoplastic lesion Benign tumor Borderline tumor Malignant tumor Total
113 – – – 113
Tumor
Total
Benign
Borderline
Malignant
– 346 – – 346
– 5 33 – 38
– 3 13 104 120
113 354 46 104 617
398
S. Ilvan et al. / Gynecologic Oncology 97 (2005) 395–399
In the literature, frozen section examination has a high accuracy rate in ovarian tumors, reported at greater than 90% [2,11–19]. In our study, the overall accuracy was 97%. The sensitivity rates for benign and malignant tumors were found as 100% and 87%, respectively, which correlate with the other series [12,14,16–18]. The sensitivity for borderline tumors was 87%. This ratio seems to be better than published rates in the literature, 45% to 78% [12,15–19]. We established the positive predictive value of frozen section evaluation in the diagnosis of ovarian malignancies to be 100%. This finding is consistent with the other studies which reported positive predictive values of 99.1% to 100%, making overtreatment an unlikely event [2,7,11,12,15]. In our study, diagnostic problems especially occurred in mucinous and borderline tumor groups as seen in previously reported studies [7,12,15,17,19]. Fifteen of our 21 cases with disagreement were mucinous tumors. In 11 cases, the frozen section diagnosis of borderline mucinous tumor turned out to be mucinous carcinoma-grade I. On paraffin sections, 5 of these cases showed no stromal invasion but contained epithelial multilayering of four or more cell thick and back-to-back glandular arrangement with no intervening stroma, these being the criteria defined by Hart and Norris [3]. Four cases of borderline mucinous tumors were misdiagnosed as benign mucinous tumor on frozen sections. We decided that the reason for discordance in all these cases was due to sampling error. Since mucinous ovarian tumors frequently contain benign, borderline, and malignant components together and have relatively larger dimensions, they are more likely to be underdiagnosed than serous ones [9,17,20,21]. In cases where the signet-ring cells are sparse, the stroma of Krukenberg tumor may mimic fibroma or fibrothecoma on frozen sections [22], as seen in our study. The causes of discordance and deferred cases are due to sampling error, misinterpretation, lack of communication with the surgeons, and technical problems [14]. The deferral rates reported in the literature range from zero to 6.3% [2,7,11–15]. We had no deferred case. We believe that the deferred cases were eliminated by good communication with the surgical team. Additionally, in our department, frozen and paraffin section examination of all gynecologic material is done by only two pathologists. During the intraoperative consultation, the pathologists are able to communicate with a responsible surgeon from the surgical team and also receive the clinical information, preoperative imaging, and intraoperative findings. The management of some suspicious cases was even planned together. Some authors mentioned that the experience of the pathologist is of crucial importance for the success of frozen section diagnosis [17,21]. However, Menzin et al. [23] stated that the level of expertise in gynecological pathology did not influence the accuracy of the frozen section diagnosis. In conclusion, frozen section evaluation of the ovarian masses provides mostly accurate diagnosis and guides the
surgeon planning the management of the operation. However, like the other diagnostic methods in medicine, frozen section examination also has some pitfalls. Pathologists and surgeons must be aware of the limitations of this procedure, and a good communication between them is required to obtain more accurate results.
References [1] Wakahara F, Kikkawa F, Nawa A, Tamakoshi K, Ino K, Maeda O, et al. Diagnostic efficacy of tumor markers, sonography, and intraoperative frozen section for ovarian tumors. Gynecol Obstet Invest 2001;52:147 – 52. [2] Lim FK, Yeoh CL, Chong SM, Arulkumaran S. Pre and intraoperative diagnosis of ovarian tumours: how accurate are we? Aust NZ J Obstet Gynaecol 1997;37:223 – 7. [3] Hart WR, Norris HJ. Borderline and malignant mucinous tumors of the ovary. Histologic criteria and clinical behavior. Cancer 1973;31: 1031 – 45. [4] Scully RE. Atlas of tumor pathology. Tumors of the Ovary and Maldeveloped Gonads. Washington7 Armed Forces Institute of Pathology; 1979. [5] Kindschi GW. Frozen sections. Their use and abuse. JAMA 1984;251:2559 – 60. [6] Kern WH. Frozen sections. Letter to the editor. JAMA 1985;253:512. [7] Twaalfhoven FC, Peters AA, Trimbos JB, Hermans J, Fleuren GJ. The accuracy of frozen section diagnosis of ovarian tumors. Gynecol Oncol 1991;41:189 – 92. [8] Maiman M, Seltzer V, Boyce J. Laparoscopic excision of ovarian neoplasms subsequently found to be malignant. Obstet Gynecol 1991;77:563 – 5. [9] Puls L, Heidtman E, Hunter JE, Crane M, Stafford J. The accuracy of frozen section by tumor weight for ovarian epithelial neoplasms. Gynecol Oncol 1997;67:16 – 9. [10] Fechner RE. Frozen section (intraoperative consultation). Hum Pathol 1988;19:999 – 1000. [11] Obiakor I, Maiman M, Mittal K, Awobuluyi M, DiMaio T, Demopoulos R. The accuracy of frozen section in the diagnosis of ovarian neoplasms. Gynecol Oncol 1991;43:61 – 3. [12] Rose PG, Rubin RB, Nelson BE, Hunter RE, Reale FR. Accuracy of frozen-section (intraoperative consultation) diagnosis of ovarian tumors. Am J Obstet Gynecol 1994;171:823 – 6. [13] Spann CO, Kennedy JE, Musoke E. Intraoperative consultation of ovarian neoplasms. J Natl Med Assoc 1994;86:141 – 4. [14] Usubutun A, Altinok G, Kucukali T. The value of intraoperative consultation (frozen section) in the diagnosis of ovarian neoplasms. Acta Obstet Gynecol Scand 1998;77:1013 – 6. [15] Wang KG, Chen TC, Wang TY, Yang YC, Su TH. Accuracy of frozen section diagnosis in gynecology. Gynecol Oncol 1998; 70:105 – 10. [16] Yeo EL, Yu KM, Poddar NC, Hui PK, Tang LC. The accuracy of intraoperative frozen section in the diagnosis of ovarian tumors. J Obstet Gynaecol Res 1998;24:189 – 95. [17] Gol M, Baloglu A, Yigit S, Dogan M, AydVn C, Yensel U. Accuracy of frozen section diagnosis in ovarian tumors: is there a change in the course of time? Int J Gynecol Cancer 2003;13:593 – 7. [18] Pinto PB, Andrade LA, Derchain SF. Accuracy of intraoperative frozen section diagnosis of ovarian tumors. Gynecol Oncol 2001;81: 230 – 2. [19] Tangjitgamol S, Jesadapatrakul S, Manusirivithaya S, Sheanakul C. Accuracy of frozen section in diagnosis of ovarian mass. Int J Gynecol Cancer 2004;14:212 – 9. [20] Houck K, Nikrui N, Duska L, Chang Y, Fuller AF, Bell D, et al. Borderline tumors of the ovary: correlation of frozen and
S. Ilvan et al. / Gynecologic Oncology 97 (2005) 395–399 permanent histopathologic diagnosis. Obstet Gynecol 2000;95: 839 – 43. [21] Kayikcioglu F, Pata O, Cengiz S, Tulunay G, Boran N, Yalvac S, et al. Accuracy of frozen section diagnosis in borderline ovarian malignancy. Gynecol Obstet Invest 2000;49:187 – 9.
399
[22] Acs G. Intraoperative consultation in gynecologic pathology. Semin Diagn Pathol 2002;19:237 – 54. [23] Menzin AW, Rubin SC, Noumoff JS, LiVolsi VA. The accuracy of a frozen section diagnosis of borderline ovarian malignancy. Gynecol Oncol 1995;59:183 – 5.