J Manipulative Physiol Ther. 1997 Sep; 20(7): 459-67.

Effect of anterior wedging of L1 on the measurement of lumbar lordosis: comparison of two roentgenological methods.

Worrill NA, Peterson CK.

Department Chair, Radiology, Anglo-European College of Chiropractic, Bournemouth, Dorset, UK.

OBJECTIVE: To determine whether the choice of either the superior or inferior endplate of the L1 vertebra as the proximal landmark for the measurement of lumbar lordosis could significantly affect the categorization (i.e., hypo- or hyperlordotic or normal) of that lordosis in subjects where the L1 vertebra is wedged anteriorly. DESIGN: Concurrent validity. SETTING: The Anglo-European College of Chiropractic teaching clinic. SELECTION: A total of 260 files were screened from new patient files at the clinic dating from the year 1980 onward. A total of 96 films fulfilled the inclusion criteria, which were: the patient was 20-50 yr old at the time of presentation, and had no evidence of severe degeneration, scoliosis or bony deformity, and the quality of the radiograph was not poor. Wedging of the L1 vertebra in excess of 2 mm was measured in 70 radiographs; these films were used for the two measurements of lumbar lordosis. OUTCOME MEASURES: The lumbar lordosis was roentgenometrically measured on lateral lumbar radiographs. RESULTS: Seventy-three per cent of the X-rays meeting the inclusion criteria showed wedging of the L1 vertebra in excess of 2 mm. Using the Student's t test, the mean lumbar lordosis, measured using the superior endplate of L1 as the proximal boundary for measurement of the lordotic angle, was 52.0 +/- 11.82 degrees and was significantly lower (p < .0001) than the mean lumbar lordosis measured using the inferior endplate of L1, which was 59.06 +/- 12.01 degrees. The Kappa analysis performed to determine the level of agreement in the categorization of the lumbar lordosis using these two methods indicated that, statistically (K = .394), there was poor agreement in the categorization of the lumbar lordosis between the two methods. An intraexaminer reliability study indicated significant intraexaminer agreement using both methods. CONCLUSION: The choice of landmark (i.e., superior or inferior endplate of L1) can significantly affect the value of the lumbar lordosis angle and its subsequent categorization as hypo-/hyperlordotic or normal in patients where this vertebra is wedged anteriorly. In view of the high incidence of this anatomical variant, L1 might not necessarily represent the best choice of landmark for the measurement of the lumbar lordosis. However, further work is necessary to determine which method is the most accurate and to investigate the validity of the suggestion that L2 may be a reasonable alternative.