PT  - JOURNAL ARTICLE
AU  - Rogers, Mason
TI  - <strong>Investigating the Prospect of Cross Calibrating Bone Density Scanners across a Medical Enterprise</strong>
DP  - 2024 Jun 01
TA  - Journal of Nuclear Medicine
PG  - 2429--2429
VI  - 65
IP  - supplement 2
4099  - http://jnm.snmjournals.org/content/65/supplement_2/2429.short
4100  - http://jnm.snmjournals.org/content/65/supplement_2/2429.full
SO  - J Nucl Med2024 Jun 01; 65
AB  - 2429 Introduction: Bone density measurements are crucial in diagnosing and monitoring the progression of a patient’s bone loss. The ability to quickly measure and quantify the density of a patient’s bones guides doctors in designing treatment plans for diseases like osteoporosis. This institution’s current bone density scanner protocol requires that if a reading is done on one scanner, subsequent scans must be done on the same scanner. If they are done on different scanners, the measurements are not compared. If scanners across an institution are not cross-calibrated, diagnostic quality values can only be attained from one scanner for each patient. This research examines the validity of that protocol and whether we need to continue requiring patients to return to the same scanner- even if one is available closer to home. This policy change, if possible, will increase patient satisfaction and help decrease the backlog of patients at the main campus site. Methods: Data was collected at five locations within this hospital system that have a bone density scanner. Four of the locations have a GE Lunar Idxa scanner while one site has a GE Lunar Prodigy scanner. With one site having a different model scanner, the initial assumption was that cross-calibration was not feasible, but data was still collected. The same Lunar Spine phantom (Lunar 9278) was scanned using the spine protocol on each scanner 10 times, recording the bone mineral density value (g/cm2) for each scan. The phantom was also removed from the table and repositioned for each scan to mimic the scanning of real patients. The five sites will be designated as A, B, C, D, and E. Results: Standard deviations were calculated to find the average variation that each scanner had. The average BMD from each site was compared to all other sites individually to find the percent differences. These variations are marked in the following table (Table 1). Each percentage only attributes the difference between those two respected sites, and the possibility for a patient to have medically comparable scans at either location. Conclusions: The International Society of Clinical Densitometry (ISCD) sets the standards of practice for bone density and accredits health systems and technologists to practice bone density. The ISCD also states that when conducting cross-calibration testing, scans should be done on at least 20 separate days to collect BMD means, which was unattainable in the timeframe of this research. Their standards on cross-calibration state that if there is a difference of less than 0.5% between the average BMD, then values taken on either scanner can be compared and used for diagnosis. The Prodigy scanner, at site E, was in the ISCD standard range of variation for two of the four comparisons. Cross-calibration would not be able to be completed using this scanner, but the comparisons are worth noting. The ISCD also states that when cross-calibrating across a health system, the use of an index device is the best practice in ensuring precision between all scanners within the organization. All scanners will be calibrated to match the selected index device, along with any new scanners added to the system. Site A would have the index scanner as it is the main campus of the health system and has the highest patient load. The data collected and calculated shows that 50% of the comparisons have less than 0.5% difference without any previous work done to cross-calibrate the scanners. The sites with a variation greater than 0.5% would need manufacturer calibration to be within the accepted range before values can be compared between sites. Half of the site combinations are already within the accepted range, but more research needs to be conducted to support cross-calibration or the use of a calibration factor in this system.