Background: B-mode ultrasound measurements of the echolucency of the carotid intima-media may hold information on cardiovascular risk. The information obtained from this measurement may depend on which gain settings are used. We studied the effect of gain settings on echolucency measurements and its consequences on risk factor relations and treatment effects.
Methods: We used two approaches. In the first, we examined the relationship between calibration, gain and common grey-scale median (GSM) from repeated ultrasound images obtained from four healthy individuals at gain settings ranging from -20 to 20 dB. In the second, we evaluated the effect of gain settings on the relation of risk factors and statin treatment with common GSM, using images from 325 participants of the Measuring Effects on Intima-Media Thickness: an Evaluation of Rosuvastatin (METEOR) study with documented gain settings. Echolucency of the carotid intima-media was measured from ultrasound images using PaintShop Pro and Artery Measurement Software and expressed as GSM.
Results: In healthy individuals, common GSM increased with increments in gain setting, primarily when the measurements were not calibrated. In the METEOR study sample, age and sex were significantly related to gain setting. The risk factor relations with common GSM were of the same magnitude and direction after adjustment for gain setting. Furthermore, adjustment for gain setting did not alter the rates of GSM change over time.
Conclusion: Extreme variability in gain settings has a major impact on the echolucency measurements of the far wall common carotid intima-media. Calibration should be used to adjust for these effects of gain settings. Variability in gain settings, however, seems limited in real practice and did not change the direction and magnitude of the relations under study. However, as age and sex are major determinants of gain settings, adjustment for or stratification by age and sex is recommended in studies into echolucency of the carotid intima-media in situations in which gain settings are unknown.
Trial registration: ClinicalTrials.gov NCT00225589.