Lumbopelvic Muscle Performance and Morphology Differences: Implications for Low Back Pain in Individuals with Unilateral Above-The-Knee Amputation

Background and Purpose: The number of people living with an amputation in the United States is rising at a rate of almost 200,000 cases a year [1]. The incidence of low back pain (LBP) in individuals with lower limb amputation (LLA) has been shown to be much greater than in the general population at 52%- 72% [2-4]. High amputation level, disuse, and abnormal movement patterns associated with prosthetic gait have been theorized to lead to decreased lower back muscle performance, contributing to the high prevalence of LBP in individuals with LLA [5-7]. However, lumbar muscle morphology and performance have not been examined in those with comorbid LLA and LBP. Therefore, the purpose of this study was to investigate the anatomical and functional characteristics of the lumbar muscles in individuals with unilateral LLA with and without LBP, and age-matched non-amputee controls. We compared the lumbar multifidus muscle cross-sectional area (CSA) and thickness, and spinal extensor muscle strength and endurance between amputees with and without histories of LBP. Participants: This study included 11 participants, with 3 participants in each of the following groups: LLA without LBP (3 male, age=43.67±7.77 years, height=171.87±5.29 cm, weight=81.04±22.23 kg), LLA with LBP (3 male, age=64±13.08 years, height=180.34±5.08 cm, weight=76.66±8.02 kg), and 5 age-matched participants without LLA or LBP serving as the control group (5 male, age=45.4±7.47 years, height=178.82±8.15 cm, weight=87.82±12.11 kg). Participants were classified into the LBP group if they had experienced at least one episode of activity limiting LBP within the previous 12 months. Methods: On the day of testing, participants were asked to complete the following 5 questionnaires: the Oswestry Low Back Pain Disability Questionnaire, SF-36, PROMIS-29, Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), and Houghton Scale of Prosthetic Use (Houghton Scale) [8- 13]. In addition to these questionnaires, a visual analogue score (VAS) was used to identify current, best, and worst pain levels. Ultrasound imaging was used to measure multifidus cross-sectional area (MF CSA) and thickness. During imaging, participants assumed a prone position on a table with 0-5° of lumbar lordosis (confirmed with inclinometry). For MF CSA, the imaging transducer was placed in the transverse plane with images captured bilaterally at L3-5 spinous process levels. For MF thickness, the transducer was placed in the longitudinal plane with images captured bilaterally at the level of L4 only. All images were analyzed using Image J software (ImageJ, 1.51m9, National Institutes of Health, USA). Test-retest reliability of the imaging and analysis procedures was established prior to data collection. Participants then completed a maximal effort back extension test using a dynamometer followed by a lumbar paraspinal endurance test following the Biering-Sorensen Protocol. Results: Average test time for the LLA with LBP group (32.66±52.50 sec) was significantly shorter than the LLA without LBP group (102.83±32.49 sec p=0.040). The LLA with LBP group reported significantly higher pain levels according to VAS scores over the previous year than both the control group and the LLA without LBP group (p=0.008 and p=0.006 respectively). Patient reported disability and functioning were significantly different when comparing LLA with LBP to healthy controls for the SF-36 Physical Functioning sub-set (p=0.033), Promis-29 Physical Function sub-set (p=0.036), and WOMAC (p=0.017). Discussion: Our findings suggested that individuals with LLA with LBP exhibited decreased spinal muscle endurance when compared to those with LLA but without LBP and the controls. This could be a contributing factor in the higher incidence of LBP among people with amputation as the MF are important spinal stabilizers during activity [14]. Our results agree with previous studies of in individuals with LBP that the rate of neuromuscular fatigue is significantly higher than those without LBP, leading to shorter Sorensen\u27s times [15]. However, the causality of LBP and decreased lumbopelvic muscle performance could not be determined due to the cross-sectional nature of this study. Taken together we found that the LBP group had significantly more pain, decreased physical function and increased disability compared to the no-LBP and control groups