Serial head ultrasound studies in preterm infants: how many normal studies does one infant need to exclude significant abnormalities?

OBJECTIVE

We hypothesized that preterm infants with two normal head ultrasound (HUS) screening studies > or = 7 days apart would have subsequently normal follow-up studies.

POPULATION

We reviewed reports of all HUS studies performed in preterm infants < or = 32 weeks gestation admitted to our nursery between January 1998 and July 2000.

SETTING

Regional perinatal referral center.

DESIGN

A normal HUS screening study was defined as either no findings; or grade I intraventricular hemorrhage (IVH) (Papile classification), germinal matrix irregularity or cyst, or normal but unequal ventricular size. An abnormal study was defined as any with IVH > or = grade II, periventricular leukomalacia (PVL), ventriculomegaly (VM), or periventricular echogenicity (PVE).

RESULTS

Of 98 infants, 92 infants (94%) who had two normal HUS studies > or = 7 days apart had normal repeat studies subsequently, and six (6%) were abnormal. Four of the six abnormal infants were <25 weeks gestation at birth. One infant (27 weeks) became abnormal after culture-positive bacterial sepsis and necrotizing enterocolitis with bowel perforation requiring surgery. The remaining infant (29 weeks) had a question of PVE, and a normal repeat study. The positive predictive value for having a normal HUS after two previously normal studies > or = 7 days apart was 94% with a specificity of 86%.

CONCLUSION

Stable premature infants > or = 25 weeks gestation without intervening deterioration may not need repeat screening HUSs after having had two normal studies > or = 7 days apart. Unstable or extremely premature infants <25 weeks gestation may be subject to late severe IVH, VM, and PVL, and therefore need a repeat study before hospital discharge, even if two initial studies > or = 7 days apart were normal.