Inter-rater and intra-rater reliability of quantitative ultrasound in preterm infants: an observational cohort study.

Extremely premature infants are at high risk of metabolic bone disease of prematurity. Current diagnostic tools require radiation exposure, additional risks to patients, and likely provide late diagnosis. Serum markers of minerals are the current standard of care, but there are no universal diagnostic definitions. QUS is a potential diagnostic method that is non-invasive, provides immediate results, can be performed in the incubator, and has no radiation exposure. QUS has been studied in preterm infants, but few studies have evaluated the intra-rater and inter-rater reliability of QUS in preterm infants. Our study evaluated the inter-rater and intra-rater reliability in infants less than 28 wk or <1000 g at birth at a single level IV academic center. Three operators performed QUS on the tibia in infants with 3 separate measurements per operator at one month post birth (n = 73) and at 36 wk postmenstrual age (PMA) (n = 81). The MiniOmni was the QUS device, and the smallest probe, CS probe (length 24 mm, range 2200-4700 m/s, and max depth 4.5 mm), was used for all subjects. Measurements were taken at the mid-tibia location. Inter-rater reliability was calculated using 2-way random-effects intraclass correlation coefficients (ICC); intra-rater reliability was calculated using 2-way mixed-effects ICC. 95% CI were calculated for each. Value of ICC estimates less than 0.5, between 0.5 and 0.75, between 0.75 and 0.9, and greater than 0.90 are indicative of poor, moderate, good, and excellent reliability, respectively. Intra-rater reliability at 1 mo post-birth and 36 wk PMA was excellent across all operators. Inter-rater reliability was more variable with good reliability at 1-mo measurements (0.84, 0.78-0.90, 95% CI) and excellent reliability at 36 wk PMA (0.94, 0.90-0.96, 95% CI). Our study demonstrated that QUS is a reliable diagnostic tool and a feasible potential alternative to current diagnostic methodologies for MBPD in extremely preterm infants.