What is the role of intra-operative assessment of the uterus in low and high risk apparent early stage endometrial cancer?
What is the role of intra-operative assessment of the uterus in low and high risk apparent early stage endometrial cancer?
In Australia, and indeed, worldwide, there is no consistent approach to the surgical management of endometrial carcinoma particularly in regard to the performance of pelvic and para-aortic lymph node dissection and/or sampling as a standard procedure. In some centres full surgical staging including nodal dissection is performed in all cases regardless of tumour characteristics unless contraindicated by factors such as medical co-morbidities or technical limitations. Conversely, some surgeons do not perform routine nodal sampling at all, with considerations for adjuvant therapy being based upon the metastatic risk determined from the final surgical-pathological findings. In general, neither of these approaches requires intra-operative assessment (IOA) of the endometrial tumour since the operative procedures would not be influenced by the results.
A third more flexible approach to surgical management restricts node dissection to those cases that are considered ‘high-risk’ (HR) for metastases. In this context HR designation may be based upon pre-operative findings, in particular the diagnosis of high-grade endometrioid adenocarcinoma or high-risk histologic types on endometrial sampling, or on the operative disclosure of obvious extra-uterine disease. Again IOA is usually not required in this setting since full staging would be performed unless otherwise contra-indicated. Generally, therefore, IOA is used to identify those patients with (apparent) low-stage and low-grade endometrioid adenocarcinomas who have adverse prognostic features identified only at operation.
Factors that have been used to assign tumours to the high risk category include high-grade histology or tumour subtype, deep (>50%) myometrial invasion, cervical invasion, larger tumour size (>2cm), and/or the presence of lympho-vascular space invasion (LVSI).
Basis of current review
Studies of IOA of endometrial carcinoma were initially identified from the literature search. Studies available only in abstract form were excluded. In total, 21 published studies that included from 31 to 403 patients were assessed.[4-24]
The overall quality of the studies was poor with most studies being retrospective, non-consecutive, non blinded (to reference standard) reviews (ie level III-2 evidence). Only three studies were level II evidence. Of those, one had an independent blinded comparison to a reference standard using consecutive patients, but patients with complex atypical hyperplasia (CAH) were included. A second study reported on 64 consecutive patients with endometrial cancer in whom depth of invasion was assessed by a pathologist blinded to the frozen section result. The third study was a retrospective review, however, it was reported that all patients were included, blinding was used and information was available on all patients.
Methods of intra-operative assessment
Intra-operative assessment of HR features in endometrial carcinoma may be based upon either gross or histological (frozen section) examination of the hysterectomy specimen. Cervical involvement and depth of myometrial invasion may be assessed naked eye (by the surgeon and/or the pathologist) or microscopically, whereas assessment of tumour type, grade and LVSI require frozen section. The use of different IOA methods (gross or histological) in different centres probably partly reflects the availability of pathology services locally.
Methods of intra-op assessment:
- Gross visual inspection (GVI) of tumour invasion
- Microscopic (frozen section) assessment of high risk features (tumour invasion, grade of tumour, histologic type or lymphatic space invasion)
Gross visual inspection (GVI)
Twelve groups have reported their experience with gross visual inspection (GVI) of myometrial invasion in a total of 2083 patients.          Inclusion criteria varied with respect to histologic tumour type and grade as well as stage of disease. Accuracy in predicting myometrial invasion varied from 80% to 90%, with a sensitivity of 65% to 80%. The depth of invasion was underestimated in 15% to 27% of patients and overestimated in 2% to 17% of patients.
Accuracy of GVI of myometrial invasion is influenced by grade, size and histologic variant of the tumour. Accuracy is best with low grade tumours and tumours less than 2 cm and worse with grade 3 tumours, aggressive histologic variants or tumours with multiple foci.   Accuracy for grade 3 tumours was reported as less than 60% in one study . Overestimation of depth of invasion occurred most often in patients with adenomyosis and leiomyomas.
Several groups have also reported their experience with GVI of cervical involvement with tumour.   Accuracy ranged from 79% to 97% but sensitivity was as low as 32% in one study. The extent of involvement was underestimated in 11% to 67% of patients. The clinical implications of determining cervical invasion intra-operatively is unclear given that optimal therapy for stage 2 disease has yet to be defined.
Frumovitz combined pre-operative tumour grade and intra-op assessment of gross tumour invasion in order to try and predict those who were at increased risk for lymph node metastases and therefore requiring lymphadenectomy. One hundred and fifty three patients with either grade 1 or 2 endometrioid tumours and intra-operative assessment of myometrial invasion of <50% were compared to the final pathology using a predictor score. Pre-operative tumour grade was upgraded in over 20% of patients on final pathology. The depth of invasion was greater than 50% in 21% of grade 1 and 32% of grade 2 tumours. The authors concluded that the combination of pre-operative grade of tumour and intra-operative gross myometrial invasion was a poor predictor for extrauterine disease.
Microscopic (frozen section) assessment
Nine authors have reported on their experience with frozen section assessment of myometrial invasion in a total of 1035 patients. As with the studies of GVI, these studies were also mostly retrospective reviews of varying sizes (between 31 and 318 patients).
In the best designed study, Case prospectively evaluated in a blinded fashion, 60 consecutive patients with either complex atypical hyperplasia or endometrial cancer on whom frozen section for myometrial invasion was performed. Accuracy for frozen section was 67% with clinically significant upgrading in 18% of patients. Similarly, Ozdemir reported on 64 consecutive patients with endometrial cancer who had intra-operative frozen section assessment of myometrial invasion in a blinded fashion. Accuracy of frozen section for deep myometrial invasion was 80%.
Others have conducted retrospective, non blinded, non-consecutive reviews and have reported an accuracy of 80% to 95%. Myometrial invasion was underestimated in 3% to 19% and overestimated in 2% to 8%, (usually because of the presence of adenomyosis or deep lymphatic space tumour emboli).
In addition to frozen section assessment of myometrial invasion, some authors have reported on frozen section assessment of grade of tumour, as compared to final pathology. The accuracy of frozen section grading of tumours ranges from 58% to 86%. Grade of tumour was underestimated in 8% to 38% of patients and overestimated in 4% to 6 % of patients. Case reported that it was grade 1 tumours that were most often upgraded whilst Kucera reported that accuracy of grading was poorest for grade 2 and 3 tumours.
One study has attempted to assess all high risk parameters (depth of invasion, tumour grade, cervical invasion, histological subtype and lymphatic space invasion) in a retrospective review of 318 pts with endometrial adenocarcinoma who underwent frozen section of both myometrium and cervix. Based on specific parameters, the patients were divided into low risk and high risk groups. They found that intra op frozen section results corresponded with final pathology in 95% of cases, giving a PPV of 99% and a NPV of 92% and concluded that intra-operative frozen section was a reliable and applicable tool in assessing risk. However identification of high risk patients who then subsequently underwent lymphadenectomy did not translate into improved survival.
Summary of findings
The depth of myometrial invasion, presence of cervical invasion, and histological grade/ type are the factors that have been assessed most commonly intra-operatively with overall accuracies of approximately 80% to 90%, 84% to 99%, and 85% to 90% respectively (compared to final histological assessment deemed to be the ‘gold standard’).[4-24] The reported accuracies for macroscopic and microscopic assessment of myometrial invasion are similar. Several studies have noted that the depth of myometrial invasion is more accurately determined in low-grade tumours where such assessment is likely to be more relevant. Most investigators have interpreted their findings to be supportive of the practice of IOA in endometrial carcinoma. However, others have expressed concern that the diagnostic accuracy is not sufficient to influence operative management. Since most intra-operative errors lead to under-grading or under-staging (compared with final assessment), it has been argued that complete surgical staging should be performed in all endometrial carcinomas.
Intra-operative assessment may be used to identify those patients with (apparent) low-stage and low-grade endometrioid adenocarcinomas who have adverse prognostic features identified only at operation.
Difficulties in assessment of the literature and evaluation of the role of intra-operative assessment in endometrial carcinoma
A fundamental problem in making recommendations about IOA of endometrial tumours is that the optimal surgical management remains controversial (see section on Lymphadenectomy). Thus, the impact of IOA depends on individual surgeons own threshold for performing lymphadenectomy or complete surgical staging.
Even in situations where IOA is used, there appears to be variation regarding its principal role, and this influences the interpretation (and potential significance) of diagnostic errors. In those cases where it is primarily used to determine which patients could avoid lymphadenectomy, the false negative rate could be regarded as a more significant error than the over-diagnosis of myometrial invasion or tumour grade (‘false positive’ error) since the default surgical procedure would include nodal dissection. Conversely, in those situations where it is used to determine who should have a full surgical staging (the default position being to avoid lymphadenectomy), an over-diagnosis (‘false positive’ error) could be considered to be more serious than under-diagnosis. Thus, while details such as the overall accuracy, specificity and sensitivity, and positive and negative predictive values of IOA are provided in many studies, few authors have specifically addressed the implications of erroneous diagnoses in their own practice.
There are major differences in case selection in the published series with a variable proportion of tumour subtypes and grades. Only one study is restricted to endometrioid adenocarcinoma (and its variants); three studies do not comment upon tumour type. The proportion of non-endometrioid carcinomas, most of which will fall into the high-grade category, will influence the accuracy and relevance of the IOA findings.
The reviewed studies were published between 1996 and 2010 and, not surprisingly, most use the previous FIGO staging system. Therefore usually it is not possible, based upon the available data, to extrapolate the findings to the 2009 FIGO staging system.
The macroscopic and histological features used to designate HR tumours (deep myometrial invasion, high-grade or aggressive histological subtype, gross cervical invasion, and LVSI) are inter-related and thus many tumours will be positive for more than one of these factors. However, this has seldom been taken into account so that in most studies it is not clear whether the apparent false negative or false positive rate using one particular feature (for example, depth of myometrial invasion) was clinically relevant. This point is illustrated in the study by Quinliven and colleagues who did assess the clinical implications of their erroneous IOA findings. They describe four patients in whom IOA falsely suggested deep myoinvasion but since all of these tumours were also high-grade histological subtypes, lymphadenectomy was appropriately performed according to local guidelines.
Logistic problems of intra-operative assessment
Apart from the potential diagnostic limitations of the type of IOA, the procedure is also associated with some operative delay, estimated to be 3-5 minutes for macroscopic assessment alone, and from 10-16 minutes or up to 30 minutes when frozen section is employed. As with all frozen section procedures, the use of IOA in endometrial carcinoma also has logistic implications for histopathology departments in terms of pathologist/ scientist’s time, and the potential disruption and delay to the routine diagnostic services.
Evidence summary and recommendations
|The depth of myometrial invasion, presence of cervical invasion, and histological grade/ type are the factors that have been assessed most commonly intra-operatively with overall accuracies of approximately 80% to 90%, 84% to 99%, and 85% to 90% respectively (compared to final histological assessment deemed to be the ‘gold standard’)||III-2||, , , , , , , , , , , , , , , , , , , , |
|However, the quality of the studies are poor with only three studies using consecutive patients with an independent blinded comparison to final histopathology.||II||, , |
|Caution should be exercised in relying on intra-operative assessment of depth of invasion, involvement of cervix and histological grade as a means to determine extent of surgical staging||C|
|Measurement of depth of myometrial invasion using either gross visual assessment or frozen section is less accurate when dealing with high grade, histological aggressive or larger tumours||III-2||, , , |
|Patients with high grade, histologically aggressive or large tumours are unlikely to benefit from intra-operative assessment.||D|
- Creasman WT, Morrow CP, Bundy BN, Homesley HD, Graham JE, Heller PB. Surgical pathologic spread patterns of endometrial cancer. A Gynecologic Oncology Group Study. Cancer 1987 Oct 15;60(8 Suppl):2035-41 Available from: http://www.ncbi.nlm.nih.gov/pubmed/3652025.
- Schink JC, Lurain JR, Wallemark CB, Chmiel JS. Tumor size in endometrial cancer: a prognostic factor for lymph node metastasis. Obstet Gynecol 1987 Aug;70(2):216-9 Available from: http://www.ncbi.nlm.nih.gov/pubmed/3601286.
- Morrow CP, Bundy BN, Kurman RJ, Creasman WT, Heller P, Homesley HD, et al. Relationship between surgical-pathological risk factors and outcome in clinical stage I and II carcinoma of the endometrium: a Gynecologic Oncology Group study. Gynecol Oncol 1991 Jan;40(1):55-65 Available from: http://www.ncbi.nlm.nih.gov/pubmed/1989916.
- Case AS, Rocconi RP, Straughn JM Jr, Conner M, Novak L, Wang W, et al. A prospective blinded evaluation of the accuracy of frozen section for the surgical management of endometrial cancer. Obstet Gynecol 2006 Dec;108(6):1375-9 Available from: http://www.ncbi.nlm.nih.gov/pubmed/17138769.
- Ozdemir S, Celik C, Emlik D, Kiresi D, Esen H. Assessment of myometrial invasion in endometrial cancer by transvaginal sonography, Doppler ultrasonography, magnetic resonance imaging and frozen section. Int J Gynecol Cancer 2009 Aug;19(6):1085-90 Available from: http://www.ncbi.nlm.nih.gov/pubmed/19820373.
- Lampe B, Kürzl R, Dimpfl T, Fawzi H. Accuracy of preoperative histology and macroscopic assessment of cervical involvement in endometrial carcinoma. Eur J Obstet Gynecol Reprod Biol 1997 Aug;74(2):205-9 Available from: http://www.ncbi.nlm.nih.gov/pubmed/9306120.
- Fotiou S, Vlahos N, Kondi-Pafiti A, Zarganis P, Papakonstantinou K, Creatsas G. Intraoperative gross assessment of myometrial invasion and cervical involvement in endometrial cancer: Role of tumor grade and size. Gynecol Oncol 2009 Mar;112(3):517-20 Available from: http://www.ncbi.nlm.nih.gov/pubmed/19117598.
- Franchi M, Ghezzi F, Melpignano M, Cherchi PL, Scarabelli C, Apolloni C, et al. Clinical value of intraoperative gross examination in endometrial cancer. Gynecol Oncol 2000 Mar;76(3):357-61 Available from: http://www.ncbi.nlm.nih.gov/pubmed/10684710.
- Frumovitz M, Singh DK, Meyer L, Smith DH, Wertheim I, Resnik E, et al. Predictors of final histology in patients with endometrial cancer. Gynecol Oncol 2004 Dec;95(3):463-8 Available from: http://www.ncbi.nlm.nih.gov/pubmed/15581947.
- Larson DM, Connor GP, Broste SK, Krawisz BR, Johnson KK. Prognostic significance of gross myometrial invasion with endometrial cancer. Obstet Gynecol 1996 Sep;88(3):394-8 Available from: http://www.ncbi.nlm.nih.gov/pubmed/8752246.
- Mao Y, Wan X, Chen Y, Lv W, Xie X. Evaluation of the accuracy of intra-operative gross examination for the surgical management of endometrial cancer. Eur J Obstet Gynecol Reprod Biol 2008 Dec;141(2):179-82 Available from: http://www.ncbi.nlm.nih.gov/pubmed/18845374.
- Obrzut B, Obrzut M, Skret-Magierło J, Skret A, Ulman D, Król P, Zmuda M.. Value of the intraoperative assessment of the depth of myometrial invasion in endometrial carcinoma. Ginekol Pol 2008;79(6):404-9. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18652127.
- Sato S, Itamochi H, Shimada M, Fujii S, Naniwa J, Uegaki K, et al. Preoperative and intraoperative assessments of depth of myometrial invasion in endometrial cancer. Int J Gynecol Cancer 2009 Jul;19(5):884-7 Available from: http://www.ncbi.nlm.nih.gov/pubmed/19574778.
- Vorgias G, Hintipas E, Katsoulis M, Kalinoglou N, Dertimas B, Akrivos T. Intraoperative gross examination of myometrial invasion and cervical infiltration in patients with endometrial cancer: decision-making accuracy. Gynecol Oncol 2002 Jun;85(3):483-6 Available from: http://www.ncbi.nlm.nih.gov/pubmed/12051878.
- Berretta R, Merisio C, Piantelli G, Rolla M, Giordano G, Melpignano M, et al. Preoperative transvaginal ultrasonography and intraoperative gross examination for assessing myometrial invasion by endometrial cancer. J Ultrasound Med 2008 Mar;27(3):349-55 Available from: http://www.ncbi.nlm.nih.gov/pubmed/18314512.
- Ghaemmaghami F, Aminimoghaddam S, Modares-Gilani M, Mousavi A, Khazaeipour Z, Fereidoni F. Assessment of gross examination and frozen section of uterine specimen in endometrial cancer patients. Arch Gynecol Obstet 2010 Dec;282(6):685-9 Available from: http://www.ncbi.nlm.nih.gov/pubmed/20213133.
- Homesley HD, Boike G, Spiegel GW. Feasibility of laparoscopic management of presumed stage I endometrial carcinoma and assessment of accuracy of myoinvasion estimates by frozen section: a gynecologic oncology group study. Int J Gynecol Cancer 2004;14(2):341-7 Available from: http://www.ncbi.nlm.nih.gov/pubmed/15086735.
- Cunha TM, Félix A, Cabral I. Preoperative assessment of deep myometrial and cervical invasion in endometrial carcinoma: comparison of magnetic resonance imaging and gross visual inspection. Int J Gynecol Cancer 2001;11(2):130-6 Available from: http://www.ncbi.nlm.nih.gov/pubmed/11328411.
- Furukawa N, Takekuma M, Takahashi N, Hirashima Y. Intraoperative evaluation of myometrial invasion and histological type and grade in endometrial cancer: diagnostic value of frozen section. Arch Gynecol Obstet 2010 May;281(5):913-7 Available from: http://www.ncbi.nlm.nih.gov/pubmed/19862538.
- Kayikçioğlu F, Boran N, Meydanli MM, Tulunay G, Köse FM, Bülbül D. Is frozen-section diagnosis a reliable guide in surgical treatment of stage I endometrial carcinoma? Acta Oncol 2002;41(5):444-6. Available from: http://www.ncbi.nlm.nih.gov/pubmed/12442920.
- Papadia A, Azioni G, Brusacà B, Fulcheri E, Nishida K, Menoni S, et al. Frozen section underestimates the need for surgical staging in endometrial cancer patients. Int J Gynecol Cancer 2009 Dec;19(9):1570-3 Available from: http://www.ncbi.nlm.nih.gov/pubmed/19955939.
- Quinlivan JA, Petersen RW, Nicklin JL. Accuracy of frozen section for the operative management of endometrial cancer. BJOG 2001 Aug;108(8):798-803 Available from: http://www.ncbi.nlm.nih.gov/pubmed/11510702.
- Kucera E, Kainz C, Reinthaller A, Sliutz G, Leodolter S, Kucera H, et al. Accuracy of intraoperative frozen-section diagnosis in stage I endometrial adenocarcinoma. Gynecol Obstet Invest 2000;49(1):62-6 Available from: http://www.ncbi.nlm.nih.gov/pubmed/10629376.
- Egle D, Grissemann B, Zeimet AG, Müller-Holzner E, Marth C. Validation of intraoperative risk assessment on frozen section for surgical management of endometrial carcinoma. Gynecol Oncol 2008 Sep;110(3):286-92 Available from: http://www.ncbi.nlm.nih.gov/pubmed/18653219.