Copyright
©The Author(s) 2025.
World J Gastrointest Surg. Nov 27, 2025; 17(11): 111619
Published online Nov 27, 2025. doi: 10.4240/wjgs.v17.i11.111619
Published online Nov 27, 2025. doi: 10.4240/wjgs.v17.i11.111619
Table 1 Comparison of clinical data between the two groups of patients
| Factors | Bleeding group (n = 29) | Non-bleeding group (n = 149) | t/χ2 | P value |
| Age (years) | 61.21 ± 12.35 | 60.55 ± 11.56 | 0.277 | 0.782 |
| Sex, n (%) | 0.035 | 0.852 | ||
| Male | 16 (55.17) | 85 (57.15) | ||
| Female | 13 (44.83) | 6442.95 | ||
| Diabetes, n (%) | 1.074 | 0.300 | ||
| Yes | 6 (20.69) | 45 (30.20) | ||
| No | 23 (79.31) | 104 (69.80) | ||
| History of chronic kidney disease, n (%) | 1.542 | 0.214 | ||
| Yes | 6 (20.69) | 18 (12.08) | ||
| No | 23 (79.31) | 131 (87.92) | ||
| Preoperative anemia, n (%) | 0.013 | 0.910 | ||
| Yes | 5 (17.24) | 27 (18.12) | ||
| No | 24 (82.76) | 122 (81.88) | ||
| Tumor size [cm; n (%)] | 12.010 | < 0.001 | ||
| < 3 | 10 (34.48) | 47 (31.54) | ||
| ≥ 3 | 19 (65.52) | 102 (68.46) | ||
| Infiltration depth, n (%) | 1.717 | 0.424 | ||
| M1 | 26 (89.66) | 122 (81.88) | ||
| M2 | 3 (10.34) | 20 (13.42) | ||
| M3 | 0 (0.00) | 74.70 | ||
| Tumor location, n (%) | 0.656 | 0.720 | ||
| Upper esophagus | 10 (34.48) | 41 (27.52) | ||
| Middle esophagus | 5 (17.24) | 32 (21.48) | ||
| Lower esophagus | 14 (48.28) | 76 (51.00) | ||
| Tumor morphology, n (%) | 2.675 | 0.263 | ||
| Uplift | 13 (44.83) | 84 (56.38) | ||
| Depression | 13 (44.83) | 59 (39.60) | ||
| Flat | 3 (10.34) | 6 (4.03) | ||
| Intraoperative bleeding, n (%) | 6.622 | 0.010 | ||
| Yes | 8 (27.59) | 17 (11.41) | ||
| No | 21 (72.41) | 132 (88.59) | ||
| Operation time (minute) | 50.10 ± 3.32 | 49.22 ± 2.95 | ||
| Postoperative anticoagulant use, n (%) | 0.543 | 0.461 | ||
| No | 17 (58.62) | 98 (65.77) | ||
| Yes | 12 (41.38) | 51 (34.23) |
Table 2 Comparison of preoperative laboratory examination indexes between the two groups of patients
| Indicators | Bleeding group (n = 29) | Non-bleeding group (n = 149) | t value | P value |
| CRP (mg/L) | 52.14 ± 6.60 | 48.77 ± 5.37 | 2.981 | 0.003 |
| Platelet count (× 109/L) | 193.32 ± 37.55 | 207.55 ± 39.21 | 1.800 | 0.074 |
| Neutrophils (× 109/L) | 6.33 ± 1.55 | 5.89 ± 1.43 | 1.534 | 0.127 |
| Lymphocytes (× 109/L) | 1.88 ± 0.46 | 1.91 ± 0.49 | 0.203 | 0.839 |
| Hematocrit (%) | 38.56 ± 3.22 | 39.10 ± 3.52 | 0.765 | 0.446 |
| Hemoglobin (g/L) | 125.68 ± 16.22 | 128.31 ± 17.88 | 0.735 | 0.463 |
| White blood cell count (× 109/L) | 7.88 ± 1.90 | 7.25 ± 1.63 | 1.877 | 0.062 |
| PT (second) | 12.50 ± 2.32 | 12.22 ± 2.45 | 0.569 | 0.570 |
| APTT (second) | 35.65 ± 3.67 | 34.85 ± 3.23 | 1.195 | 0.234 |
| Fibrinogen (g/L) | 3.53 ± 0.52 | 3.61 ± 0.60 | 0.653 | 0.515 |
Table 3 Comparison of bleeding after endoscopic submucosal dissection trend from Japan scores between bleeding group and non-bleeding group in patients with early esophageal cancer
| Item | Bleeding group (n = 29) | Non-bleeding group (n = 149) | t/χ2 | P value |
| BEST-J score (points) | 1.97 ± 1.12 | 1.55 ± 0.93 | 2.829 | 0.005 |
| BEST-J score distribution, n (%) | 11.494 | 0.042 | ||
| 0 | 1 (3.45) | 12 (8.05) | ||
| 1 | 8 (27.59) | 77 (51.68) | ||
| 2 | 15 (51.72) | 47 (31.54) | ||
| 3 | 2 (6.90) | 10 (6.71) | ||
| 4 | 2 (6.90) | 2 (1.34) | ||
| 5 | 1 (3.45) | 1 (0.67) | ||
| BEST-J risk, n (%) | 9.025 | 0.029 | ||
| Low risk | 9 (31.03) | 89 (59.73) | ||
| Medium risk | 15 (51.72) | 47 (31.54) | ||
| High risk | 4 (13.79) | 12 (8.05) | ||
| Extremely high risk | 1 (3.45) | 1 (0.67) |
Table 4 Multivariate analysis of factors affecting bleeding after endoscopic submucosal dissection for early esophageal cancer
| Factors | β | SE | Wals χ2 | P value | OR | 95%CI |
| Tumor size | 1.544 | 0.456 | 11.455 | < 0.001 | 4.685 | 1.916-11.458 |
| Surgical bleeding | 1.341 | 0.556 | 5.824 | 0.016 | 3.825 | 1.287-11.370 |
| CRP | 0.114 | 0.040 | 8.116 | 0.004 | 1.120 | 1.036-1.211 |
| Constant | -8.332 | 2.107 | 15.630 | < 0.001 | 0.000 | - |
Table 5 Comparison of the predictive effects of the gradient boosting machine model and bleeding after endoscopic submucosal dissection trend from Japan score on bleeding after endoscopic submucosal dissection in patients with early esophageal cancer
| Project | AUC (90%CI) | Sensitivity | Specificity | Cut-off value | P value | 95%CI |
| BEST-J score | 0.653 | 0.759 | 0.738 | 0.166 | < 0.001 | 0.545-0.762 |
| GBM prediction model | 0.818 | 0.724 | 0.792 | 0.191 | 0.009 | 0.729-0.906 |
- Citation: Liu SN, Chen Z. Clinical value of predicting bleeding after endoscopic submucosal dissection for early esophageal cancer. World J Gastrointest Surg 2025; 17(11): 111619
- URL: https://www.wjgnet.com/1948-9366/full/v17/i11/111619.htm
- DOI: https://dx.doi.org/10.4240/wjgs.v17.i11.111619
