Institutional members access full text with Ovid®

Share this article on:

Creamatocrit Analysis of Human Milk Overestimates Fat and Energy Content When Compared to a Human Milk Analyzer Using Mid-infrared Spectroscopy

O’Neill, Edward F.; Radmacher, Paula G.; Sparks, Blake; Adamkin, David H.

Journal of Pediatric Gastroenterology & Nutrition: May 2013 - Volume 56 - Issue 5 - p 569–572
doi: 10.1097/MPG.0b013e31828390e4
Original Articles: Hepatology and Nutrition

Background and Objective: Human milk (HM) is the preferred feeding for human infants but may be inadequate to support the rapid growth of the very-low-birth-weight infant. The creamatocrit (CMCT) has been widely used to guide health care professionals as they analyze HM fortification; however, the CMCT method is based on an equation using assumptions for protein and carbohydrate with fat as the only measured variable. The aim of the present study was to test the hypothesis that a human milk analyzer (HMA) would provide more accurate data for fat and energy content than analysis by CMCT.

Methods: Fifty-one well-mixed samples of previously frozen expressed HM were obtained after thawing. Previously assayed “control” milk samples were thawed and also run with unknowns. All milk samples were prewarmed at 40°C and then analyzed by both CMCT and HMA. CMCT fat results were substituted in the CMCT equation to reach a value for energy (kcal/oz). Fat results from HMA were entered into a computer model to reach a value for energy (kcal/oz). Fat and energy results were compared by paired t test with statistical significance set at P < 0.05. An additional 10 samples were analyzed locally by both methods and then sent to a certified laboratory for quantitative analysis. Results for fat and energy were analyzed by 1-way analysis of variance with statistical significance set at P < 0.05.

Results: Mean fat content by CMCT (5.8 ± 1.9 g/dL) was significantly higher than by HMA (3.2 ± 1.1 g/dL, P < 0.001). Mean energy by CMCT (21.8 ± 3.4 kcal/oz) was also significantly higher than by HMA (17.1 ± 2.9, P < 0.001). Comparison of biochemical analysis with HMA of the subset of milk samples showed no statistical difference for fat and energy, whereas CMCT was significantly higher than for both fat (P < 0.001) and energy (P = 0.002).

Conclusions: The CMCT method appears to overestimate fat and energy content of HM samples when compared with HMA and biochemical methods.

Neonatal Nutrition Research Laboratory, University of Louisville, Louisville, Kentucky.

Address correspondence and reprint requests to Paula G. Radmacher, PhD, Neonatal Nutrition Research Laboratory, 511 S Floyd St, Room 107, University of Louisville, Louisville, KY 40292 (e-mail:

Received 16 August, 2012

Accepted 17 December, 2012

The authors report no conflicts of interest.

© 2013 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology,