Effect of prophylactic indomethacin administration and early feeding on spontaneous intestinal perforation in extremely low-birth-weight infants.

OBJECTIVE: To determine the effect of concomitant administration of prophylactic indomethacin and early enteral feeds on the risk of spontaneous intestinal perforation (SIP) in extremely low-birth-weight (ELBW) infants, and to describe the variation in prophylactic indomethacin use in Canada. STUDY DESIGN: A retrospective cohort study of 4268 ELBW infants born at <30 weeks' gestation admitted to Canadian neonatal units between 2010 and 2014 was conducted. Prophylactic indomethacin (I+ or I-, administered within 24 h) and early feeding (E+ or E-, initiated in the first 2 days) exposures were studied concurrently and independently. The primary outcomes were SIP and death before discharge. Adjusted odds ratios (aORs) and 95% confidence intervals (CIs) were calculated. RESULTS: Compared with the I-/E+ reference group (n=1829), infants in I+/E+ (n=285; aOR 2.92, 95% CI 1.41 to 6.08) and I+/E- (n=213; aOR 2.84, 95% CI 1.35 to 5.98) groups had higher odds of SIP, whereas those in the I-/E- group had similar odds (n=1941; aOR 1.37, 95% CI 0.88 to 2.14). Odds of SIP were higher in the indomethacin exposed group (I+) compared with the unexposed (I-) group when controlled for early feeding (aOR 2.43, 95% CI 1.41 to 4.19), but not in the early feeding group when controlled for indomethacin. The use of prophylactic indomethacin ranged from 0% usage in 13 sites to 78% use in one site. CONCLUSION: Prophylactic indomethacin was associated with increased odds of SIP independently from early feeding in this cohort; however, early enteral feeding was not associated with SIP. Marked variation in the use of prophylactic indomethacin was identified.

The measurement of knee-heel length in newborn infants using a simple vernier calipers.

OBJECTIVE: To assess the usefulness of a simple vernier calipers for measuring knee-heel length in neonates. SUBJECTS AND METHODS: Using a simple vernier calipers, knee-heel length was measured five times by 2 observers in 50 babies (29M, 21F; mean birthweight 1597 g; median gestational age at birth 29 weeks) at a median postnatal age of 11 days. A subgroup of 20 babies had knee-heel length measured similarly at weekly intervals for 3 weeks. Corrected gestational age and weight were simultaneously recorded. One observer was experienced in using the vernier calipers. The precision of the calipers was established using 4 steel gauge blocks of varying length (7.62-10.17 cm). RESULTS: The calipers were very precise when measuring steel gauge blocks. In babies, there was a downward trend across the first 2 measurements for both observers, the measurements stabilizing over the last three. Using the final three measurements per baby (n = 50), the experienced observer had a mean standard deviation of 0.023 cm and mean coefficient of variation 0.23% when measuring an average knee-heel length of 9.99 cm. The inexperienced observer had a mean standard deviation of 0.057 cm and a mean coefficient of variation of 0.56%, when measuring an average knee-heel length of 10.14 cm. The inter-observer reliability, measured by the intra-class correlation coefficient, was 0.99. The agreement between observers was such that one observer measured knee-heel length consistently less (0.15 cm, SD 0.18 cm) than the other. The reliability for knee-heel length velocity was lower (R = 0.85), but agreement between observers was high with an average difference of 0.016 cm/week. Knee-heel length was significantly correlated (p < 0.001) with corrected gestational age (r = 0.85) and with weight (r = 0.96). There was a weaker but significant correlation (r = 0.47, p < 0.001) between knee-heel length velocity and rate of weight gain (g/day), indicating that weight gain may not always be accompanied by an increase in linear growth. CONCLUSION: The measurement of knee-heel length by a simple vernier calipers is an accurate, reproducible and non-invasive method of assessing short-term linear growth in neonates. However, it is recommended that measurements of knee-heel length in a individual baby should be made by a single experienced observer.