Early, Fast-Paced Growth Benefits Short-Term Health of Underweight Infants
Rapid catch-up growth in the first 20 months of life appears to confer short-term health benefits on babies who are born small for their gestational age. According to data from a cohort of babies born in Brazil in 1982,1 those who were born small but gained weight rapidly had a significantly lower rate of hospitalization in 1985 than did similar infants who did not gain weight as quickly (6% vs. 16%). These fast-growing small babies had nearly the same rate of hospitalization as fast-growing infants whose birth weight was appropriate for their gestational age.
The data come from a population-based cohort study of babies born in 1982 in the southern Brazilian city of Pelotas. These infants were followed up twice--at age 20 months and at age 42 months. Complete records were available for 3,582 infants; these records included information on the infant's birth characteristics (length of gestation, weight, and size for gestational age) and on maternal characteristics (age, education and family income), 20-month follow-up data on infant hospital admissions and 42-month follow-up data on mortality.
For weight gain from birth to age 20 months, the investigators calculated z-scores, a measure that assesses the pace of growth (i.e., the mean value of the change in weight-for-age); for infants who are small for their gestational age, a z-score of .66 or higher signals that their weight is catching up with that of infants whose size at birth was appropriate for their gestational age. Logistic regression techniques were used to determine the odds of hospitalization for any cause in 1985, the odds of hospitalization for diarrhea and lower respiratory tract infections in 1985, and the odds of dying by 1987. These analyses controlled for family income and for maternal age and education. The investigators also conducted one-sided tests for linear trends in the proportions who experienced each outcome; these tests compared the fastest growing babies to those who were growing at a moderate or slow pace, both for babies whose birth weight was appropriate for their gestational age and for those whose birth weight was not.
Of the infants in the analysis, 6% were low-birth-weight and 14% were born small for their gestational age. By the first follow-up interview, however, 59% of small-for-gestational-age infants had gained enough weight (change in z-score above the mean) to catch up with babies whose birth size had been appropriate for their gestational age.
By 1985, the small-for-gestational-age infants who had gained weight rapidly in the first 20 months of life had significantly lower rates of hospitalization than similar infants who had gained weight less quickly (6% vs. 16%, p<.001). Moreover, infants whose weight at birth had been appropriate for their gestational age benefited from rapid weight gain as well: Only 5% of those with the most rapid weight gain were hospitalized in 1985, compared with 9% who had gained weight slowly or at a moderate pace.
There were significant linear trends by pace of growth in the proportions hospitalized; that is, the rate of hospitalization for any cause decreased linearly with increasing pace of weight gain among all infants. Linear trends in the proportions hospitalized for diarrhea and lower respiratory tract infections were also significant (i.e., rates for these specific causes decreased linearly with increasing pace of weight gain among all infants, regardless of birth weight). According to results of the logistic regression analyses, moderately paced weight gain (as opposed to rapid weight gain) was significantly associated with higher odds of hospitalization for any cause (adjusted odds ratio, 2.8).
Only 10 of the infants in the analysis had died before their fifth birthday. The mortality rate by 1987 was highest by far among small-for-gestational-age babies whose weight gain was below the mean (13 deaths per 1,000 live-born infants), while the rate among other small-for-gestational-age infants who caught up quickly was similar to that of infants whose birth weight was appropriate for their gestational age (three deaths per 1,000 vs. 1-2 per 1,000). Tests for linear trends by growth pace in the proportion of infants who died were significant--these proportions decreased linearly with increasingly fast-paced growth. Moreover, multivariate analyses suggest that compared with small infants who caught up in weight early in life, those who did not had significantly higher odds of dying (adjusted odds ratio, 8.1).
The investigators acknowledge that their study has several limitations, including an overall loss to follow-up of 15%, the determination of gestational age by the mother's recollection of her last menstrual period (with 20% of women unable to provide even that information) and the possibility that small infants with chronic diseases who could not catch up in weight were hospitalized at disproportionate rates. The researchers further note that their study did not consider whether fast catch-up growth might have negative health effects later in life (i.e., possible increased incidence of coronary disease and obesity), as has been suggested in studies conducted in industrialized countries. They assert that even if further research reveals that quick catch-up growth in infancy can have negative effects in adulthood, the country's level of development has to be taken into account. In their view, encouraging early catch-up growth for short-term health benefits appears to make sense in developing countries, where high child mortality rates make efforts to enhance child health crucial. The authors thus conclude that their results "support the efforts of the international pediatric community to promote fast growth among children who are born small."--L. Remez
1. Victora CG et al., Short-term benefits of catch-up growth for small-for-gestational-age infants, International Journal of Epidemiology, 2001, 30(6)1325-1330.
*Although abortion is illegal in Bangladesh, pregnancies may be terminated in the early weeks of gestation through menstrual regulation.