PRODIGY Quantitative Lateral Vertebral Assessment (LVAq) – Accuracy and Precision

XI Zjazd Polskiego Towarzystwa Osteoartrologii i Polskiej Fundacji Osteoporozy
V Krakowskie Sympozjum Osteoporozy
Kraków 27-29.09.2001

Streszczenia:
Materiały kongresowe: STRESZCZENIA, s34.
Druk: Drukarnia Skinder, ISBN – 83-904008-5-5
oraz
Osteoporosis International 2001; vol. 12 (Suppl 1), s5.

L10
PRODIGY Quantitative Lateral Vertebral Assessment (LVAq) – Accuracy and Precision
(Ilościowa ocena projekcji bocznej kręgosłupa (LVAq) aparatem prodigy – dokładność i precyzja)
Dave Pankratz (1), Joe Bisek (1), Howard Barden (1), Rob Washenko (1), Kenneth Faulker (1), Jef Van Dam (2)

¹GE Lunar Corporation, Madison, Wisconsin, USA,
²LUNAR Europe, Belgium, Europe, GE Lunar Corporation, 726 Heartland Trail, Madison, WI, USA 53717

Low bone density and prevalent vertebral fractures, the classic hallmark of osteoporosis, are associated independently with a dramatically increased risk of future vertebral and non-vertebral fractures. The advent of the GE LUNAR PRODIGY bone densitometer has made accurate measurement of vertebral deformities a reality through its quantitative assessment application, LVAq. The PRODIGY utilizes a narrow-angle (4^o) fan-beam to acquire high-resolution dual-energy images, which can be used for qualitative and quantitative determination of deformation as well as lateral spine BMD. Scans of the lumbar spine can be acquired in less than 90 seconds and the lumbar and thoracic spine in about 180 seconds. Accuracy of the LVAq measurements were determined by comparing LVA results with biomechanical measurement of vertebral heights in vitro. Error was found to be less than 1 mm. We also compared PRODIGY LVAq with GE LUNAR EXPERT-XL morphometry data using the European COMAC phantom, and a GE LUNAR lateral spine phantom. Excellent comparability was found (r = 0.99, p = 0.3), which indicates the large EXPERT-XL morphometry reference data can be used. Vertebral height precision in vivo was determined from repeat observations of 5 patients with repositioning. Semi-automated analysis tools were used to calculate intra- and inter-individual variation using four different operators (untrained, following written procedures) measuring ten vertebral bodies multiple times.

These results indicate the precision error is comparable to that of repeat measurements using standard lateral radiographs (%CV=3-4%).