Background

Lymph node (LN) metastasis portends a worse prognosis following resection of intrahepatic cholangiocarcinoma (ICC); however, lymphadenectomy is not routinely performed, as its role remains controversial. Herein, we developed a risk model for LN metastasis by identifying its predictive factors and assessed a subset of patients who might not benefit from LN dissection (LND).

Methods

210 patients who underwent curative-intent surgery for ICC were retrospectively reviewed. A preoperative risk model for LN metastasis was developed following identification of its preoperative predictive factors using the recursive partitioning method.

Results

In the multivariable analysis, CA 19-9 level of >120 U/mL, an enlarged LN on computed tomography, and a tumor location abutting the Glissonean pedicles were independent predictors of LN metastasis. The preoperative risk model classified the patients according to their risk: high, intermediate, and low risks at a rate of LN metastasis on final pathology of 60.9%, 35%, and 2.3%, respectively. In the subgroup analysis among the low-risk patients, performance of LND had no survival advantage over non-performance of LND.

Conclusion

Routine LND for preoperatively diagnosed ICC should be recommended to patients at an intermediate and a high risk of developing LN metastasis but may be omitted for low-risk patients.

中文翻译：

---

可切除肝内胆管癌淋巴结转移的预后意义和风险预测模型：都需要淋巴结清扫吗？

背景

淋巴结（LN）转移预示着肝内胆管癌（ICC）切除后预后较差；然而，淋巴结切除术并不常规进行，因为它的作用仍然存在争议。在此，我们通过确定其预测因素开发了 LN 转移的风险模型，并评估了可能无法从 LN 剥离 (LND) 中受益的一部分患者。

方法

回顾性分析了 210 名因 ICC 接受根治性手术的患者。在使用递归分区方法识别其术前预测因素后，开发了 LN 转移的术前风险模型。

结果

在多变量分析中，CA 19-9 水平 > 120 U/mL、计算机断层扫描上扩大的 LN 和毗邻 Glissonean 椎弓根的肿瘤位置是 LN 转移的独立预测因子。术前风险模型根据患者的风险对患者进行分类：高风险、中风险和低风险，最终病理学上的 LN 转移率分别为 60.9%、35% 和 2.3%。在低风险患者的亚组分析中，进行 LND 与未进行 LND 相比，没有生存优势。

结论

术前诊断为 ICC 的常规 LND 应推荐给发生 LN 转移的中高风险患者，但对于低风险患者可以省略。