J Clin Gynecol Obstet

Journal of Clinical Gynecology and Obstetrics, ISSN 1927-1271 print, 1927-128X online, Open Access

Article copyright, the authors; Journal compilation copyright, J Clin Gynecol Obstet and Elmer Press Inc

Journal website http://www.jcgo.org

Original Article

Volume 8, Number 1, March 2019, pages 9-16

Is an Intraoperative Frozen Section Useful for Judging the Necessity of Lymphadenectomy in Patients With Endometrial Cancer?

**Table 1.** Characteristics of the 106 Patients With Endometrial Carcinoma
G1: endometrioid G1; G2: endometrioid G2; High-grade: endometrioid G3, clear cell, serous; Low-risk: G1 or G2, no muscular invasion; Intermediate-risk: G1 or G2, < 1/2 invasion; High-risk: high-grade or ≥ 1/2 invasion.
Age	56 (51 - 64)
Stage (n)
IA	64
IB	24
II	1
IIIA	6
IIIB	1
IIIC1	3
IIIC2	7
Grade (n)
G1	63
G2	32
Risk classification (n)
High-grade	11
Low-risk	34
Intermediate-risk	34
High-risk	38

Table 1. Characteristics of the 106 Patients With Endometrial Carcinoma

G1: endometrioid G1; G2: endometrioid G2; High-grade: endometrioid G3, clear cell, serous; Low-risk: G1 or G2, no muscular invasion; Intermediate-risk: G1 or G2, < 1/2 invasion; High-risk: high-grade or ≥ 1/2 invasion.

Age

56 (51 - 64)

Stage (n)

IIIA

IIIB

IIIC1

IIIC2

Grade (n)

Risk classification (n)

High-grade

Low-risk

Intermediate-risk

High-risk

**Table 2.** Comparison of the Tumor Grade Between the Preoperative and Permanent-Section Diagnoses in the 106 Patients With Endometrial Carcinoma
Permanent-section diagnosis	Preoperative diagnosis	Total	Accuracy
G1	5	3	52	3	0	63	83%
G2	6	2	14	10	0	32	44%
High-grade	1	1	3	1	5	11	45%
Total	12	6	69	14	5	106	63%

Table 2. Comparison of the Tumor Grade Between the Preoperative and Permanent-Section Diagnoses in the 106 Patients With Endometrial Carcinoma

Permanent-section diagnosis

Preoperative diagnosis

Total

Accuracy

Unknown

No tumor

High- grade

83%

44%

High-grade

45%

Total

106

63%

**Table 3.** Comparison of the Tumor Grade Between the Frozen-Section and Permanent-Section Diagnoses in the 106 Patients With Endometrial Carcinoma
Permanent-section diagnosis	Frozen-section diagnosis	Total	Accuracy
G1: endometrioid G1; G2: endometrioid G2; High-grade: endometrioid G3, clear cell, serous. There was not a significant difference between the frozen-section and preoperative diagnoses (P = 0.391).
G1	1	4	54	4	0	63	86%
G2	1	0	15	15	1	32	47%
High-grade	0	0	3	4	4	11	36%
Total	2	4	72	23	5	106	69%

Table 3. Comparison of the Tumor Grade Between the Frozen-Section and Permanent-Section Diagnoses in the 106 Patients With Endometrial Carcinoma

Permanent-section diagnosis

Frozen-section diagnosis

Total

Accuracy

Unknown

No tumor

High- grade

G1: endometrioid G1; G2: endometrioid G2; High-grade: endometrioid G3, clear cell, serous. There was not a significant difference between the frozen-section and preoperative diagnoses (P = 0.391).

86%

47%

High-grade

36%

Total

106

69%

**Table 4.** Comparison of the Grade and Muscular Invasion Between the Preoperative and Frozen-Section Diagnoses in the 106 Patients With Endometrial Carcinoma
Diagnosis	Test target	Accuracy	Sensitivity	Specificity	PPV	NPV	P value
PPV: positive predictive value; NNP: negative predictive value.
Preoperative	G2 and high-grade	67%	37%	87%	84%	73%
Frozen-section	G2 and high-grade	77%	56%	92%	86%	76%	0.054
Preoperative	High-grade	84%	45%	88%	100%	94%
Frozen-section	High-grade	91%	36%	97%	80%	93%	0.070
Preoperative	Invasion present	75%	81%	65%	83%	67%
Frozen-section	Invasion present	89%	88%	91%	97%	78%	0.007
Preoperative	≥ 1/2 invasion	81%	67%	89%	89%	89%
Frozen-section	≥ 1/2 invasion	85%	61%	97%	96%	84%	0.522

Table 4. Comparison of the Grade and Muscular Invasion Between the Preoperative and Frozen-Section Diagnoses in the 106 Patients With Endometrial Carcinoma

Diagnosis

Test target

Accuracy

Sensitivity

Specificity

PPV

NPV

P value

PPV: positive predictive value; NNP: negative predictive value.

Preoperative

G2 and high-grade

67%

37%

87%

84%

73%

Frozen-section

G2 and high-grade

77%

56%

92%

86%

76%

0.054

Preoperative

High-grade

84%

45%

88%

100%

94%

Frozen-section

High-grade

91%

36%

97%

80%

93%

0.070

Preoperative

Invasion present

75%

81%

65%

83%

67%

Frozen-section

Invasion present

89%

88%

91%

97%

78%

0.007

Preoperative

≥ 1/2 invasion

81%

67%

89%

Frozen-section

≥ 1/2 invasion

85%

61%

97%

96%

84%

0.522

**Table 5.** Comparison of Muscular Invasion Between the Preoperative (MRI) and Permanent-Section Diagnoses in the 106 Patients With Endometrial Carcinoma
Permanent-section diagnosis	Preoperative (MRI) diagnosis	Total	Accuracy
No invasion	0	1	22	10	1	34	65%
< 1/2 invasion	1	3	9	21	2	36	58%
≥ 1/2 invasion	2	2	2	6	24	36	67%
Total	3	6	33	37	27	106	63%

Table 5. Comparison of Muscular Invasion Between the Preoperative (MRI) and Permanent-Section Diagnoses in the 106 Patients With Endometrial Carcinoma

Permanent-section diagnosis

Preoperative (MRI) diagnosis

Total

Accuracy

Unknown

Invasion present (depth unknown)

No invasion

< 1/2

≥ 1/2

No invasion

65%

< 1/2 invasion

58%

≥ 1/2 invasion

67%

Total

106

63%

**Table 6.** Comparison of Muscular Invasion Between the Frozen-Section and Permanent-Section Diagnoses in the 106 Patients With Endometrial Carcinoma
Permanent-section diagnosis	Frozen-section diagnosis	Total	Accuracy
There was a significant difference between the frozen-section and preoperative diagnoses (P = 0.018).
No invasion	1	0	31	2	0	34	91%
< 1/2 invasion	0	0	7	28	1	36	78%
≥ 1/2 invasion	0	1	2	11	22	36	61%
Total	1	1	40	41	23	106	76%

Table 6. Comparison of Muscular Invasion Between the Frozen-Section and Permanent-Section Diagnoses in the 106 Patients With Endometrial Carcinoma

Permanent-section diagnosis

Frozen-section diagnosis

Total

Accuracy

Unknown

Invasion present (depth unknown)

No invasion

< 1/2

≥ 1/2

There was a significant difference between the frozen-section and preoperative diagnoses (P = 0.018).

No invasion

91%

< 1/2 invasion

78%

≥ 1/2 invasion

61%

Total

106

76%

**Table 7.** Comparison of the Risk of Lymph-Node Metastasis Between the Preoperative and Permanent-Section Diagnoses in the 106 Patients With Endometrial Carcinoma
Permanent-section diagnosis	Preoperative diagnosis	Total	Accuracy
Low-risk	3	18	8	1	30	60%
Intermediate-risk	7	9	20	3	39	51%
High-risk	8	5	4	20	37	54%
Total	18	32	32	24	106	55%

Table 7. Comparison of the Risk of Lymph-Node Metastasis Between the Preoperative and Permanent-Section Diagnoses in the 106 Patients With Endometrial Carcinoma

Permanent-section diagnosis

Preoperative diagnosis

Total

Accuracy

Unknown

Low-risk

Intermediate-risk

High-risk

Low-risk

60%

Intermediate-risk

51%

High-risk

54%

Total

106

55%

**Table 8.** Comparison of the Risk of Lymph-Node Metastasis Between the Frozen-Section and Permanent-Section Diagnoses in the 106 Patients With Endometrial Carcinoma
Permanent-section diagnosis	Frozen-section diagnosis	Total	Accuracy
Low-risk: G1 or G2, no muscular invasion; Intermediate-risk: G1 or G2, < 1/2 invasion; High-risk: high grade or ≥ 1/2 invasion; High-grade: endometrioid G3, clear cell, serous. There was a significant difference between the preoperative diagnosis and the frozen-section diagnosis (P = 0.002).
Low-risk	1	27	2	0	30	90%
Intermediate-risk	0	8	29	2	39	74%
High-risk	2	2	10	23	37	62%
Total	3	37	41	25	106	75%

Table 8. Comparison of the Risk of Lymph-Node Metastasis Between the Frozen-Section and Permanent-Section Diagnoses in the 106 Patients With Endometrial Carcinoma

Permanent-section diagnosis

Frozen-section diagnosis

Total

Accuracy

Unknown

Low-risk

Intermediate-risk

High-risk

Low-risk: G1 or G2, no muscular invasion; Intermediate-risk: G1 or G2, < 1/2 invasion; High-risk: high grade or ≥ 1/2 invasion; High-grade: endometrioid G3, clear cell, serous. There was a significant difference between the preoperative diagnosis and the frozen-section diagnosis (P = 0.002).

Low-risk

90%

Intermediate-risk

74%

High-risk

62%

Total

106

75%

**Table 9.** Accuracy of the Grade and Muscular Invasion Using Frozen-Section Diagnosis According to Past Studies
Paper	Year	Number of cases	Accuracy of grade (three groups)	Accuracy of grade (two groups)	Accuracy of invasion (three groups)	Accuracy of invasion (two groups)	Conclusion of the paper
Accuracy of grade (three groups): endometrioid G1, endometrioid G2, high-grade. Accuracy of grade (two groups): endometrioid G1 or G2, high-grade. Accuracy of invasion (three groups): no invasion, < 1/2 invasion, ≥ 1/2 invasion. Accuracy of invasion (two groups): no invasion or < 1/2 invasion, ≥ 1/2 invasion.
This study		106	69	91	76	85	Useful
Wang et al [40]	2016	112	89	-	-	97	Useful
Acikalin et al [41]	2015	291	84	-	92	-	Useful
Karabagli et al [42]	2015	79	90	-	89	-	Useful
Gallego et al [22]	2014	51	-	-	-	90	Unuseful
Stephan et al [43]	2014	116	88	-	-	98	Useful
Turan et al [44]	2013	756	89	-	85	-	Useful
Kisu et al [21]	2013	111	-	-	-	94	Useful
Kumar et al [45]	2012	784	99	-	99	-	Useful
Ozturk et al [25]	2012	220	-	90	-	92	Useful
Savelli et al [26]	2012	131	-	-	-	92	Useful
Ugaki et al [20]	2011	303	71	-	77	87	Useful
Kumar et al [46]	2011	146	65	-	72	-	Unuseful
Furukawa et al [18]	2010	168	85	-	86	-	Useful
Ozdemir et al [16]	2009	64	-	-	-	91	Useful
Kucera et al [47]	2009	63	86	-	87	-	Useful
Wang et al [37]	2009	218	69	-	-	87	-
Maneschi et al [34]	2008	78	-	-	-	95	-
Montalto et al [48]	2008	87	84	-	94	-	Useful
Sanjuan et al [14]	2006	89	-	87	-	89	Useful
Case et al [49]	2006	77	58	-	-	67	Unuseful
Quilivan et al [50]	2001	209	89	-	-	95	Useful
Kucera et al [51]	2000	70	84	-	80	-	Useful
Shim et al [52]	1992	199	-	-	-	91	Useful
Fanning et al [38]	1990	216	-	96	-	95	Useful

Table 9. Accuracy of the Grade and Muscular Invasion Using Frozen-Section Diagnosis According to Past Studies

Paper

Year

Number of cases

Accuracy of grade (three groups)

Accuracy of grade (two groups)

Accuracy of invasion (three groups)

Accuracy of invasion (two groups)

Conclusion of the paper

Accuracy of grade (three groups): endometrioid G1, endometrioid G2, high-grade. Accuracy of grade (two groups): endometrioid G1 or G2, high-grade. Accuracy of invasion (three groups): no invasion, < 1/2 invasion, ≥ 1/2 invasion. Accuracy of invasion (two groups): no invasion or < 1/2 invasion, ≥ 1/2 invasion.

This study

106

Useful

Wang et al [40]

2016

112

Useful

Acikalin et al [41]

2015

291

Useful

Karabagli et al [42]

2015

Useful

Gallego et al [22]

2014

Unuseful

Stephan et al [43]

2014

116

Useful

Turan et al [44]

2013

756

Useful

Kisu et al [21]

2013

111

Useful

Kumar et al [45]

2012

784

Useful

Ozturk et al [25]

2012

220

Useful

Savelli et al [26]

2012

131

Useful

Ugaki et al [20]

2011

303

Useful

Kumar et al [46]

2011

146

Unuseful

Furukawa et al [18]

2010

168

Useful

Ozdemir et al [16]

2009

Useful

Kucera et al [47]

2009

Useful

Wang et al [37]

2009

218

Maneschi et al [34]

2008

Montalto et al [48]

2008

Useful

Sanjuan et al [14]

2006

Useful

Case et al [49]

2006

Unuseful

Quilivan et al [50]

2001

209

Useful

Kucera et al [51]

2000

Useful

Shim et al [52]

1992

199

Useful

Fanning et al [38]

1990

216

Useful