Malnutrition, zinc supplementation and catch-up growth: changes in insulin-like growth factor I, its binding proteins, bone formation and collagen turnover.

OBJECTIVE: Zinc may be a limiting factor in restricting catch-up growth in severely malnourished children. This study had two aims: (i) to examine the effect of different zinc supplementation regimens on IGF-I, its binding proteins and on markers of bone and collagen turnover in severely malnourished children and (ii) to investigate mechanisms underlying catch-up growth by examining changes in these markers during nutritional rehabilitation, their inter-relationships and their relationships with ponderal and linear growth. DESIGN: Double-blind randomized intervention study of three regimens of oral zinc supplementation. PATIENTS: One hundred and forty-one children, aged 6-36 months, mean (SD) age 15.4 (8.7) months, with day 1 weight-for-height SD score (whz) -2.6 (0.93) and height-for-age SD score (haz) -3.79 (1.29). MEASUREMENTS: Weight, height, lower leg length (by knemometry) at 15-day intervals from day 1 to day 90 of nutritional rehabilitation. Blood collection on days 1, 15 and 30 for IGF-I, IGFBP3, IGFBP2, bone alkaline phosphatase (BAP, osteoblast marker), procollagen type I C-terminal propeptide (PICP, marker of type I collagen synthesis), procollagen type III N-terminal propeptide (P3NP, marker of soft tissue type III collagen synthesis) and type I collagen telopeptide (ICTP, marker of type I collagen breakdown). RESULTS: There was early rapid weight gain during refeeding, whereas height gain occurred later in the trial. IGF-I, IGFBP3, BAP, PICP and P3NP were low or very low on day 1 compared to well-nourished age-matched European children, and all increased within 15 days (P < 0.001), with PICP and P3NP reaching levels higher than European norms. IGFBP2 and ICTP were high on day 1 and decreased over the same period (P < 0.001). There were no differences in anthropometric outcome or marker responses among zinc regimens. Day 1 whz was correlated with BAP, PICP and P3NP (P < 0.001). Changes in IGF-I, IGFBP3, BAP, PICP and P3NP over 30 days correlated with ponderal growth (whz change) over the same period (all P < 0.01). However, changes in these markers over 30 days correlated better with lower leg growth (all P < 0.01) and linear growth (haz change, P < 0.01 for PICP and P3NP, P < 0.05 for IGFBP3) measured over 90 compared with 30 days. At most time points, there were strong positive correlations (i) among IGF-I, IGFBP3, BAP, PICP and P3NP (P < 0.01) and (ii) between IGFBP2 and ICTP (P < 0.01). Conversely, IGFBP2 was negatively correlated with IGF-I, IGFBP3, BAP, PICP and P3NP at most time points (P < 0.01). CONCLUSIONS: We found no difference among zinc regimens in growth, IGF-I and its binding proteins or markers of bone and collagen turnover. Severe malnutrition was associated with low rates of bone and collagen synthesis and high rates of collagen degradation, and nutritional rehabilitation was associated with full or partial 'normalization' of the markers studied. Early weight gain and subsequent linear growth were associated with early increments in IGF-I, IGFBP3 and markers of bone and collagen formation. The study of these markers has provided additional insights into the mechanisms of the effects of malnutrition and refeeding on growth.

Linear and knemometric growth in the early phase of rehabilitation from severe malnutrition.

The relationship between ponderal, linear and lower leg growth in children recovering from severe malnutrition remains unclear. We report on the early growth of 141 severely malnourished Bangladeshi children aged 6 to 36 months of age who were followed for 90 d. Mean (sd) weight for height (WHZ) and height for age (HAZ) catch-up growth z scores over the 90 d were 1.6 (0.85) and 0.47 (0.325) respectively. Mean (sd) lower leg length growth was 10.35 (4.5) mm. Change in HAZ was significantly associated with initial WHZ, but linear growth occurred in the presence of severe wasting and no threshold WHZ score was identified. Lower leg length gain correlated throughout with ponderal indices but with change in HAZ score only after day 45. Only initial WHZ score and maternal height predicted for linear growth and only accounted for 20 % of total variance. We conclude that linear growth occurs early in severely malnourished children but that knemometry behaves as a ponderal index acutely.

Zinc and rehabilitation from severe protein-energy malnutrition: higher-dose regimens are associated with increased mortality.

A randomized, double-blind trial was undertaken to measure the effects of zinc supplementation on catch-up growth in severe protein-energy malnutrition, with particular reference to linear growth. One hundred forty-one children between the ages of 6 mo and 3 y were enrolled after admission to a nutritional rehabilitation unit in Dhaka, Bangladesh, and randomly assigned to receive elemental zinc by mouth, 1.5 mg/kg for 15 d, 6.0 mg/kg for 15 d, or 6.0 mg/kg for 30 d, and thereafter they were followed for a total of 90 d. Anthropometric outcome measures included change in knee-heel length, midupper arm circumference, subscapular and triceps skinfold thicknesses, and change in height-for-age, weight-for-age, and weight-for-height z scores. Higher zinc doses were not associated with significant change in any anthropometric measurement, but mortality was significantly greater in children who received high-dose zinc (6.0 mg/kg) initially as opposed to those who received low-dose zinc supplementation (1.5 mg/kg) (Yates-corrected chi-square P value of 0.033 and a risk ratio of 4.53; 95% CI: 1.09 < risk ratio < 18.8). We conclude that there is no benefit to using high-dose zinc supplementation regimens and that they could contribute to increased mortality in severely malnourished children.