Prevalence, distribution, and morphology of ossification of the ligamentum flavum: a population study of one thousand seven hundred thirty-six magnetic resonance imaging scans

Abstract

Study design: Large scale, cross-sectional imaging study of a general population.

Objectives: To evaluate the prevalence, morphology, and distribution of ossification of the ligamentum flavum (OLF) in a population, and synthesize the scientific literature on the prevalence of OLF and some factors associated with its occurrence.

Summary of background data: OLF is a rare disease in which the pathogenesis has not been conclusively established. Little is known about its epidemiology. To date, there is no study that comprehensively assessed the distribution and prevalence of OLF in the whole spine using magnetic resonance imaging (MRI).

Methods: A total of 1736 southern Chinese volunteers (1068 women; 668 men) between 8 and 88 years of age (mean, 38 years) were recruited by open invitation. MRI was administered to all the participants. T2-weighted, 5-mm spin-echo MRI sequences of the whole spine were obtained. Presence of OLF was identified as an area of low signal intensity in the T2 sagittal sequence located in the posterior part of the spinal canal, and subsequently confirmed by computed tomography scans showing areas of ossification within the ligamentum flavum. The distribution of OLF was classified into 3 types: the isolated type, continuous type, and noncontinuous type. While the morphology of the lesion was classified into triangular, round, and beak shapes based on the pattern of ossification on T2-weighted sagittal MRIs.

Results: OLF was identified in a total of 66 subjects or 3.8% of the population (52 women and 14 men). In 45(68.2%) cases, OLF was present at a single-level (isolated type), whereas in 21 (31.8%) cases OLF was present at multiple levels. The isolated type was found in 45 (68.2%) cases, continuous type in 11 (16.7%), and noncontinuous type in 10 (15.2%). The most common site of involvement is the lower thoracic spine, but they can also occur in the upper thoracic spine. The majority of the segments had a round morphology (n = 75: 81.5%), while 17 (18.5%) segments were triangular in shape. A literature review of the past 26 years showed only 4 reports on the prevalence of OLF, all were in special patient groups.

Conclusion: Case reports have described postoperative paraplegia from failure to identify and decompress all stenotic segments of OLF. This study demonstrated that OLF is not uncommon, and that some 15% of the lesions are noncontinuous, and therefore could be missed. The authors recommend that for patients undergoing surgical decompression for 1 level of OLF, the whole spine should be routinely screened for other stenotic segments. Failure to do so could result in paraplegia from the nondecompressed levels.