Phantom Limb Pain and Sensations in Chinese Malignant Tumor Amputees: A Retrospective Epidemiological Study

Objective: Phantom limb pain (PLP) was a common problem in malignant tumor amputees that can cause considerable suffering. The purposes of this study were to determine the incidence and factors associated with the occurrence of post-operation PLP, stump limb pain (SLP), and phantom limb sensations (PLS) in tumor amputees within the first month after surgery. Additionally, differences in phantom phenomena between upper and lower extremities were investigated. Methods: In total, 162 amputees participated in this retrospective study who underwent malignant limb amputation between 2012 and 2019. Clinical characteristics were collected from medical records and reconfirmed by telephone interviews. A numerical rating scale (NRS) was used to quantitate phantom phenomena. We used analysis of variance and non parametric statistics for categorical variables and ordinal variables separately. Results: In the first month after malignant amputation, the incidence of PLP was 54.3%, that of PLS was 65.4%, and that of SLP was 32.7%. The duration of preoperative pain and amputation level was significantly different for the incidence of acute PLP. Further subgroup analysis of amputation level showed that patients whose amputation level was below the wrist and ankle joints had a significantly reduced incidence of PLP (p<0.0083 in Bonferroni test). Binary logistics regression analysis determined that amputation level was the primary risk factor for the incidence of PLP. Factors related to the severity of postoperative PLP also included amputation level, preoperative pain, and amputation times. By comparing the differences between upper and lower limbs after amputation, we found that the incidence of PLS was higher after lower limb amputation, but there was no significant difference in the incidence of PLP and SLP. Preoperative experience of chemotherapy was not a risk factor for PLP. Conclusion: Proximal amputation and long-term preoperative pain seemed to count more for PLP incidence. Further research may be required to individually determine factors associated with the occurrence and chronicity of phantom phenomena.