[Rule of acupoint selection for cancer pain based on association rule and entropy clustering].

Objective: To analyze the rule of acupoint selection for cancer pain based on data mining.

Methods: The published literature regarding acupuncture for cancer pain in the recent 10 years was searched in PubMed, CNKI, VIP and Wanfang database. The acupoint selection was summarized and analyzed by TCMISS V2.5.

Results: Totally 68 literature was collected and 73 acupoint prescriptions were included, involving 117 acupoints. These acupoints were mainly in bladder meridian, stomach meridian, liver meridian and spleen meridian. Among them, 40 acupoints used more than 4 times were identified, and the top three acupoints were Zusanli (ST 36, 65 times), Neiguan (PC 6, 55 times) and Taichong (LR 3, 50 times). A total of 68 acupoint combinations used more than 19 times were identified, and the top three acupoint combinations were Zusanli (ST 36)-Neiguan (PC 6), Taichong (LR 3)-Zusanli (ST 36) and Zusanli (ST 36)-Sanyinjiao (SP 6). There were 103 acupoint combinations with strong association; based on the entropy clustering algorithm, 20 new acupoint combinations and 10 new acupoint prescriptions were obtained.

Conclusion: The main meridians for cancer pain are bladder meridian, stomach meridian, liver meridian and spleen meridian, with Zusanli (ST 36), Neiguan (PC 6), Taichong (LR 3), Hegu (LI 4), Sanyinjiao (SP 6) and points as core acupoints, and regulating spleen-stomach and treating -blood are the main principles.