The objective of this study is to investigate the prevalence of Andersson lesions (AL) in ankylosing spondylitis (AS) patients who will start anti-tumor necrosis factor (TNF) treatment. Radiographs and magnetic resonance imaging (MRI) of the spine were performed before therapy with anti-TNF. ALs were defined as discovertebral endplate destructions on MRI, associated with bone marrow edema and fat replacement or sclerosis, a decreased signal on T1, enhancement after contrast administration (gadolinium diethylenetriamine pentaacetic acid (Gd-DTPA)), and increased signal on T2 and short tau inversion recovery (STIR). Additionally, conventional radiography showed a fracture line, irregular endplates, and increased sclerosis of adjacent vertebral bodies. Fifty-six AS patients were included, 68% males, mean age of 43 years, and mean disease duration of 11 years. The mean bath ankylosing spondylitis disease activity index was 6.4, and 24% of all patients had ankylosis. Only one patient showed a discovertebral abnormality with bone marrow edema of more than 50% of the vertebral bodies adjacent to the intervertebral disk of T7/T8 and T9/T10, a hypodense signal area on T1, and a high signal on STIR. Irregular endplates were depicted, and T1 after Gd-DTPA demonstrated high signal intensity around the disk margins. However, no fracture line was visible on conventional radiology, and therefore, this case was not considered to be an AL. No AL was detected in our AS patients, who were candidates for anti-TNF treatment. One patient showed a discovertebral abnormality on MRI, without a fracture line on conventional radiology. The relative small proportion of patients with a long-established disease might explain this finding for, particularly, an ankylosed spine is prone to develop an AL.