Characteristics of Spinal Morphology According to the Global Alignment and Proportion (GAP) Score in a Diverse,...

Multi-Ethnic Alignment Normative Study (MEANS) cohort: prospective, cross-sectional, multicenter.

To analyze the distribution of GAP scores in the MEANS cohort and compare the spinal shape via stratification by GAP alignment category, age, and country.

The GAP score has been used to categorize spinal morphology and prognosticate adult spinal deformity surgical outcomes and mechanical complications. We analyzed a large, multiethnic, asymptomatic cohort to assess the distribution of GAP scores.

Four hundred sixty-seven healthy volunteers without spinal disorders were recruited in five countries. Sagittal radiographic parameters were measured via the EOS imaging system. The GAP total and constituent factor scores were calculated for each patient. Kruskal-Wallis rank sum test was performed to compare variables across groups, followed by the post hoc Games-Howell test. Fisher exact test was used to compare categorical variables. The significance level was set to P <0.05.

In the MEANS cohort, 13.7% (64/467) of volunteers were ≥60 years old, and 86.3% (403/467) were <60 years old. 76.9% (359/467) was proportioned, 19.5% (91/467) was moderately disproportioned, and 3.6% (17/467) was severely disproportioned. There was no significant difference in the frequency of proportioned, moderately, or severely disproportioned GAP between subjects from different countries ( P =0.060). Those with severely disproportioned GAP alignment were on average 14.5 years older ( P =0.016), had 23.1° lower magnitude lumbar lordosis (LL) ( P <0.001), 14.2° higher pelvic tilt ( P <0.001), 13.3° lower sacral slope ( P <0.001), and 24.1° higher pelvic-incidence (PI)-LL mismatch ( P <0.001), 18.2° higher global tilt ( P <0.001) than those with proportioned GAP; thoracic kyphosis and PI were not significantly different ( P >0.05).

The GAP system applies to a large, multiethnic, asymptomatic cohort. Spinal alignment should be considered on a spectrum, as 19.5% of the asymptomatic volunteers were classified as moderately disproportioned and 3.6% severely disproportioned. Radiographic malalignment does not always indicate symptoms or pathology.

3.