, 2008; Min et al., 2009). The muscle-fat-index (MFI) is another method for interindividual comparison of intramuscular fatty infiltration, involving the calculation of the ratio of the mean SI in a region of muscle tissue relative to the SI in a homogenous

region of fat (Elliott et al., 2005, 2008b; Cagnie et al., 2009; Elliott et al., 2010). Combining the measures total, lean muscle and fat CSA and MFI with MRI provides a quantitative and multifaceted view, to investigate whether lumbar muscle morphometry and composition Fulvestrant differs during remission of unilateral recurrent LBP compared to a healthy control group, and whether this is pain-side related. We hypothezised that lumbar muscle degeneration would be present in participants with a history of LBP, and being most prominent on the previously painful side. Thirteen individuals with recurrent non-specific LBP were recruited via advertisement in the local community and university. Inclusion criteria were a history of at least 2 previous episodes of LBP (onset >6 months) that interfered with activities of daily living and/or required treatment (LBP characteristics: Table 1). Episodes were defined as bouts of LBP for a minimum of 24 h, preceded and followed by a period of minimum 1 month without symptoms (de Vet et al., 2002). Testing was scheduled at least 1 month after the end of the previous episode (time since last episode: 64 ± 33,6 days). Thirteen individuals without a history of LBP, comparable

for gender, age, weight, length and level of physical activity, formed a healthy Rapamycin datasheet control group (demographic characteristics: Table 2). Participants were excluded from either group if they reported: central, bilateral or variable localization of LBP; pain elsewhere in the body; lumbar muscle training in the past year; spinal deformities or surgery; task-limiting medical conditions or contra-indications for MRI. After notification of the study procedures, which were approved by the local Ethics Committee, participants Demeclocycline provided written informed consent. T1-weighted images were acquired using a 3-T MRI-scanner (Magnetom Trio-Tim,

SyngoMR VB15 software, Siemens AG®, Erlangen Germany). Participants were placed supine with a foam wedge supporting the legs (∼30° hip flexion). A flexible 6-element body-matrix coil, centered ventrally on L4, was combined with the standard phased-array spine coil dorsally as a receiver–coil combination. On a sagittal localizing scan, 3 slices were positioned as axially as possible along the upper endplate of L3 and L4 and lower endplate of L4, visualizing lumbar MF, erector spinae (ES) and PS. These levels were selected as paraspinal and PS muscle mass is at or near maximal, enhancing the possibility to demonstrate CSA differences (Danneels et al., 2000; Lee et al., 2008). Level L4 lower endplate was used as a substitute for L5, because the inclination of L5 is often too large to visualize the muscles’ cross-section appropriately.