Baseline characteristics of 918 patients suffering LGIN diagnosed by EFBs
Table 1 summarizes baseline characteristics of 918 included patients suffering LGIN diagnosed by EFBs. Of the 918 patients, 584 were male and 332 were female, with a median age of 58y (IQR: 50–66). 34.7% patients (319/918) did not have adequate EFB number. The predominant macroscopic shapes were Yamada III or IV (490/918, 53.4%). 42 (4.6%) lesions were LSTs. Most lesions (721/918, 78.5%) were located in the distal large intestine. Lobulated, erythema, erosion and surface unevenness patterns were observed in 192 (20.9%), 461 (50.2%), 102 (11.1%), 149 (16.2%) patients, respectively. And 553 cases (60.2%) just suffered single colorectal polyp. Further, the median maximum diameter of the target adenoma is 1.5 cm (IQR: 1, 2). <1.0 cm, 1.0-1.9 cm, 2.0-2.9 cm and ≥ 3.0 cm maximum diameters of target adenoma were found in 92 (10%), 487 (53.1%), 233 (25.4%), 106 (11.5%) patients.
Table 1
Baseline characteristics of 918 patients with colorectal LGIN diagnosed by EFB
Baseline Characteristics
|
Value
|
Median Age, y (IQR)
|
58 (50, 66)
|
Gender, n (%)
|
|
Male
|
586 (63.8)
|
Female
|
332 (36.2)
|
Lack of standardization of EFB number, n (%)
|
|
Yes
|
319 (34.7)
|
No
|
599 (65.3)
|
Shape pattern of the target adenoma, n (%)
|
|
Yamada I + II
|
428 (46.6)
|
Yamada III + IV
|
490 (53.4)
|
LST
|
42 (4.6)
|
Location, n (%)
|
|
Proximal
|
197 (21.5)
|
Distal
|
721 (78.5)
|
Lobulated shape, n (%)
|
192 (20.9)
|
Erythema, n (%)
|
461 (50.2)
|
Erosion, n (%)
|
102 (11.1)
|
Surface unevenness, n (%)
|
149 (16.2)
|
Number of colorectal polyps
|
|
Single
|
553 (60.2)
|
Multiple
|
366 (39.8)
|
Median largest diameter of the target, cm (IQR)
|
1.5 (1, 2)
|
Largest diameter of the target adenoma, n (%)
|
|
<1.0 cm
|
92 (10)
|
1.0-1.9 cm
|
487 (53.1)
|
2.0-2.9 cm
|
233 (25.4)
|
≥3.0 cm
|
106 (11.5)
|
LGIN, low-grade intraepithelial neoplasia; IQR, inter-quartile range; EFB, endoscopic fragment biopsy; LST, laterally spreading tumor |
Clinicopathological and endoscopic characteristics of patients in the histopathologically upgrade group and the histopathologically concordant group
Clinicopathologic characteristics of 918 patients were presented in Table 2 between the histopathologically upgrade and concordant groups. Clinical details of 15 patients who have an initial EFB diagnosis of LGIN and ER specimen histopathology of the adenocarcinoma was shown in Supplement Table 1. Figure 2 shows a typical case with upgrade histopathologically discrepancies between the histological results of EFBs and the ER specimen. This patient was initially diagnosed as LGIN by EFBs and finally confirmed as HGIN and adenocarcinoma by ER, respectively.
Table 2
Baseline characteristics of the concordant group and the upgrade group disease after endoscopic resection.
Characteristics
|
Upgrade group (n = 162)
|
Concordant group (n = 756)
|
P value
|
Age/y, median (IQR)
|
59 (48.75, 67)
|
58 (50, 65)
|
0.449
|
Female, n (%)
|
69 (42.6)
|
263 (34.8)
|
0.061
|
Past medical history, n (%)
|
|
|
|
HT
|
48 (29.6)
|
156 (20.6)
|
0.012
|
DM
|
21 (13.0)
|
71 (9.4)
|
0.170
|
NSAID
|
22 (13.6)
|
29 (3.8)
|
< 0.001
|
Colorectal polyps
|
5 (3.1)
|
27 (3.6)
|
0.76
|
FAP
|
1 (0.6)
|
2 (0.3)
|
0.442
|
CRC
|
8 (4.9)
|
48 (6.3)
|
0.496
|
Lack of standardization of EFB number, n (%)
|
102 (63.0)
|
217 (28.7)
|
0.001
|
EFB diagnosis, n (%)
|
|
|
< 0.001
|
Tubular
|
126 (77.8)
|
663 (87.7)
|
|
Tubulovillous
|
25 (15.4)
|
77 (10.2)
|
|
Villous
|
11 (6.8)
|
16 (2.1)
|
|
Postoperative pathological diagnosis, n (%)
|
|
|
|
Tubular adenoma
|
-
|
712 (94.2)
|
|
Tubulovillous adenoma
|
-
|
34 (4.5)
|
|
Villous adenoma
|
-
|
10 (1.3)
|
|
HGD
|
114 (70.4)
|
-
|
|
Intramucosal carcinoma
|
33 (20.3)
|
-
|
|
Submucosal adenocarcinoma
|
10 (6.2)
|
-
|
|
Advanced adenocarcinoma
|
5 (3.1)
|
|
|
Maximum diameter of the target adenoma/cm, median (IQR)
|
2.44 (1.5, 3)
|
1.59 (1, 2)
|
< 0.001
|
Lobulated pattern, n (%)
|
85 (52.5)
|
107 (14.2)
|
< 0.001
|
Erythema, n (%)
|
127 (78.4)
|
334 (44.2)
|
< 0.001
|
Erosion, n (%)
|
69 (42.6)
|
33 (4.4)
|
< 0.001
|
Surface unevenness, n (%)
|
80 (49.4)
|
69 (9.1)
|
< 0.001
|
Blood biochemistry, median (IQR)
|
|
|
|
TG, mmol/L
|
1.14 (0.80, 1.63)
|
1.27 (0.89, 1.84)
|
0.025
|
TC, mmol/L
|
4.33 (3.73, 5.08)
|
4.6 (4.03, 5.24)
|
0.002
|
HDL, mmol/L
|
1.27 (1.03, 1.49)
|
1.25 (1.05, 1.52)
|
0.742
|
LDL, mmol/L
|
2.56 (2.04, 3.10)
|
2.79 (2.25, 3.30)
|
0.001
|
Shape pattern, n (%)
|
|
|
0.547
|
Yamada I + II
|
79 (48.8)
|
349 (46.2)
|
|
Yamada III + IV
|
83 (51.2)
|
407 (53.8)
|
|
LST, n (%)
|
5 (3.1)
|
37 (4.9)
|
0.318
|
Multiple colorectal polyps
|
73 (45.1)
|
293 (38.8)
|
0.137
|
Distal location of the target adenoma, n (%)
|
143 (88.3)
|
578 (76.5)
|
< 0.001
|
HT, hypertension; DM, diabetes metillus; NSAID, nonsteroidal anti-inflammatory drug; FAP, familial adenomatous polyposis; CRC, colorectal cancer; HGD, high-grade dysplasia; TG, triglyceride; TC, total cholesterol; HDL, high-density lipoprotein; LDL, low-density lipoprotein |
The predominant LGIN upgrade discrepancies by EFB diagnosis was upgrades to high-grade dysplasia (n = 114, 70.4%), followed by upgrades to intramucosal carcinoma (n = 33, 20.3%), submucosal adenocarcinoma (n = 10, 6.2%) and advanced adenocarcinoma (n = 5, 3.1%). The proportion of patients with HT (29.6 vs 20.6%, P = 0.012), the proportion of patients with NSAID (13.6 vs 3.8%, P < 0.001), lack of standardization of EFB number (63.0 vs 28.7%, P = 0.001), maximum diameter of the target adenoma (2.44 vs 1.59 cm, P < 0.001), lobulated pattern (52.5 vs 14.2%, P < 0.001), erythema (78.4 vs 44.2%, P < 0.001), erosion (42.6 vs 4.4%, P < 0.001), surface unevenness (49.4 vs 9.1%, P < 0.001) and distal location of the target adenoma (88.3 vs 76.5%, P < 0.001) were significantly greater in histopathologically upgrade group. TC (4.33 vs 4.6 mmol/L, P = 0.002) and LDL (2.56 vs 2.79 mmol/L, P = 0.001) were significantly lower in upgrade group. There was no significant difference in the proportion of patients with DM, colorectal polyps, FAP and CRC history between the groups. Gender, age, TC, HDL, the number of Yamada III + IV lesions, the number of LSTs and the number of patients with multiple colorectal polyps did not differ between two groups.
There are also significantly different in EFB result (P < 0.001) between the group. Based on Post hoc testing, the values of Cramer’s V in EFB diagnosis is 0.127 (P = 0.001), which indicated that they had a small magnitude of effect size. In histopathlogically upgrade group, the adjusted standardized residual of tubular, tubulovillous and villous EFB diagnosis were − 3.3, + 1.9, + 3.2, respectively.
Predictive risk factors for an initial EFB diagnosis of LGIN being upgraded to a higher-grade histopathologic diagnosis by ER specimen.
Univariate analysis was taken to screen a series of variates including female, age ≥ 60, HT history, NSAID history, lack of standardization of EFB number, ≥ 25% Villous in EFB diagnosis, ≥ 1.0 cm maximum diameter of the target adenoma, lobulated shape, erythema, erosion, surface unevenness, LST, multiple colorectal polyps, distal location of the target adenoma. Then our multivariate analysis showed that NSAID history (OR 4.83; 95% CI, 2.27–10.27; P < 0.001), lack of standardization of EFB number (OR 2.99; 95% CI, 1.91–4.68; P < 0.001), ≥ 1.0 cm maximum diameter of the target adenoma (OR 6.18; 95% CI, 1.32–28.99; P = 0.021), lobulated shape (OR 2.68; 95% CI, 1.65–4.36; P < 0.001), erythema (OR 2.42; 95% CI, 1.50–3.91; P < 0.001), erosion (OR 7.12; 95% CI, 3.91–12.94; P < 0.001), surface unevenness (OR 2.31; 95% CI, 1.33–4.01; P = 0.003) and distal location of the target adenoma (OR 3.29; 95% CI, 1.68–6.41; P < 0.001) were significantly associated with the histologically upgrade discrepancies (Table 3).
Table 3
Multivariate analyses for predictive factors demonstrating histopathologically upgrade discrepancies between pretreatment EFB and endoscopic resection specimens
|
Univariate analysis
|
Multivariable analysis
|
|
OR (95%CI)
|
P value
|
OR (95%CI)
|
P value
|
Female
|
1.39 (0.98,1.97)
|
0.061
|
|
0.054
|
Age ≥ 60
Past medical history
|
1.18 (0.84, 1.66)
|
0.337
|
|
0.256
|
NSAID
|
3.94 (2.20, 7.06)
|
< 0.001
|
4.83 (2.27, 10.27)
|
< 0.001
|
Colorectal polyps
|
0.86 (0.33,2.27)
|
0.760
|
|
|
FAP
|
2.34 (0.21,25.98)
|
0.488
|
|
|
CRC
|
0.77 (0.36,1.65)
|
0.497
|
|
|
Lack of standardization of EFB number
|
4.22 (2.96, 6.03)
|
< 0.001
|
2.99 (1.91, 4.68)
|
< 0.001
|
≥ 25% Villous in EFB diagnosis
|
2.04 (1.33, 3.13)
|
0.001
|
|
0.845
|
Maximum diameter ≥ 1.0 cm
|
10.81 (2.64, 44.36)
|
0.001
|
6.18 (1.32, 28.99)
|
0.021
|
Lobulated shape
|
6.70 (4.63, 9.69)
|
< 0.001
|
2.68 (1.65, 4.36)
|
< 0.001
|
Erythema
|
4.59 (3.07, 6.85)
|
< 0.001
|
2.42 (1.50, 3.91)
|
< 0.001
|
Erosion
|
16.26 (10.18, 25.95)
|
< 0.001
|
7.12 (3.91, 12.94)
|
< 0.001
|
Surface unevenness
|
9.71 (6.54, 14.42)
|
< 0.001
|
2.31 (1.33, 4.01)
|
0.003
|
LST
|
0.62 (0.24, 1.60)
|
0.322
|
|
|
Multiple colorectal polyps
|
1.30 (0.92, 1.83)
|
0.138
|
|
|
Distal location of the target adenoma
|
2.32 (1.40, 3.85)
|
< 0.001
|
3.29 (1.68, 6.41)
|
< 0.001
|