Exploring Factors Influencing Low Back Pain in People With Nondysvascular Lower Limb Amputation: A National Survey

Background: Chronic low back pain (LBP) is a common musculoskeletal impairment in people with lower limb amputation. Given the multifactorial nature of LBP, exploring the factors influencing the presence and intensity of LBP is warranted. Objective: To investigate which physical, personal, and amputee-specific factors predicted the presence and intensity of LBP in persons with nondysvascular transfemoral amputation (TFA) and transtibial amputation (TTA). Design: A retrospective cross-sectional survey. Setting: A national random sample of people with nondysvascular TFA and TTA. Participants: Participants (N = 526) with unilateral TFA and TTA due to nondysvascular etiology (ie, trauma, tumors, and congenital causes) and a minimum prosthesis use of 1 year since amputation were invited to participate in the survey. The data from 208 participants (43.4% response rate) were used for multivariate regression analysis. Methods (Independent Variables): Personal (ie, age, body mass, gender, work status, and presence of comorbid conditions), amputee-specific (ie, level of amputation, years of prosthesis use, presence of phantom-limb pain, residual-limb problems, and nonamputated limb pain), and physical factors (ie, pain-provoking postures including standing, bending, lifting, walking, sitting, sit-to-stand, and climbing stairs). Main Outcome Measures (Dependent Variables): LBP presence and intensity. Results: A multivariate logistic regression model showed that the presence of 2 or more comorbid conditions (prevalence odds ratio [POR] = 4.34, P = .01), residual-limb problems (POR = 3.76, P < .01), and phantom-limb pain (POR = 2.46, P = .01) influenced the presence of LBP. Given the high LBP prevalence (63%) in the study, there is a tendency for overestimation of POR, and the results must be interpreted with caution. In those with LBP, the presence of residual-limb problems (beta = 0.21, P = .01) and experiencing LBP symptoms during sit-to-stand task (beta = 0.22, P = .03) were positively associated with LBP intensity, whereas being employed demonstrated a negative association (beta = 0.18, P = .03) in the multivariate linear regression model. Conclusions: Rehabilitation professionals should be cognizant of the influence that comorbid conditions, residual-limb problems, and phantom pain have on the presence of LBP in people with nondysvascular lower limb amputation. Further prospective studies could investigate the underlying causal mechanisms of LBP.