Carotid intima-media thickness measurements in intervention studies: Design options, progression rates, and sample size considerations: A point of view

• Published in: Stroke (1970) 2003-12, Vol.34 (12), p.2985-2994
• Main Authors: BOTS, Michiel L, EVANS, Gregory W, RILEY, Ward A, GROBBEE, Diederick E
• Format: Article
• Language: English
• Subjects: Atherosclerosis (general aspects, experimental research); Biological and medical sciences; Blood and lymphatic vessels; Cardiology. Vascular system; Carotid Arteries - diagnostic imaging; Carotid Artery Diseases - diagnosis; Carotid Artery Diseases - physiopathology; Disease Progression; Humans; Medical sciences; Randomized Controlled Trials as Topic - methods; Reproducibility of Results; Research Design; Sample Size; Sensitivity and Specificity; Tunica Intima - diagnostic imaging; Tunica Media - diagnostic imaging; Ultrasonography - instrumentation; Ultrasonography - standards
• ISSN: 0039-2499; 1524-4628
• DOI: 10.1161/01.STR.0000102044.27905.B5

Carotid intima-media thickness (CIMT) measurements are currently widely used in randomized controlled trials (RCTs) to study the efficacy of interventions. In designing a RCT with CIMT as a primary outcome, several ultrasound options may be considered. We discuss the various options and provide a pooled estimate of CIMT progression. In addition, we quantify the effect of these choices on the sample size for a RCT. To estimate the average CIMT progression rate, we performed a pooled analysis using CIMT progression rates of control groups from published RCTs. The pros and cons of the following ultrasound options are discussed: which arterial segments may be studied; whether near and far wall CIMT measurements should be performed; whether a single image (1 angle of interrogation) or multiple images (more angles of interrogation) should be used; whether a manual or an automated edge detection reading system should be used; and whether images should be read in a random fashion or in batches. The pooled analysis showed an annual rate of change in mean common CIMT of 0.0147 mm (95% CI, 0.0122 to 0.0173) and in mean maximum CIMT of 0.0176 mm (95% CI, 0.0149 to 0.0203). Given the current evidence together with our experience with recently developed ultrasound protocols, we favor the use of mean maximum CIMT rather than mean common CIMT as the primary outcome measure in RCTs designed to evaluate the efficacy of pharmacological and nonpharmacological interventions in carotid artery atherosclerosis.