Reference distributions for the positive acute phase serum proteins, alpha1-acid glycoprotein (orosomucoid), alpha1-antitrypsin, and haptoglobin: a practical, simple, and clinically relevant approach in a large cohort.

Most clinical conditions are accompanied by corresponding changes in serum levels of some, if not all, of the acute phase proteins. While conditions that affect the acute phase proteins are usually inflammatory in nature, non-inflammatory conditions also can cause changes (e.g., malnutrition, some malignancies without secondary inflammation, and genetic polymorphism). Only after the confounding effects of non-inflammatory conditions are taken into account can these measurements be used to detect and stage the inflammatory process and to evaluate the impact of treatment. In this third article in a series, reference ranges for serum levels for three of the acute phase proteins that increase during inflammation are examined: alpha1-acid glycoprotein (orosomucoid), alpha-antitrypsin, and haptoglobin. The study is based on a cohort of 55,199 Caucasian individuals from northern New England, tested in our laboratory between 1994 and 1999. Measurements were standardized against CRM 470 (RPPHS) and analyzed using a previously described statistical approach. Individuals with unequivocal laboratory evidence of inflammation (C-reactive protein of 10 mg/l or higher) were excluded. Levels of a,-acid glycoprotein changed little during life and between the sexes. Levels of alpha1-antitrypsin varied somewhat by age, rising slightly beyond age 55; males followed a pattern similar to that for females. For this protein, it was necessary to apply two equations to describe the lower levels associated with certain phenotypes. Haptoglobin levels fell significantly during the first decade of life for both males and females and climbed thereafter. Males and females displayed a similar pattern. When values were expressed as multiples of the age- and gender-specific median levels, the resulting distributions fitted a log-Gaussian distribution well over a broad range. When patient data are normalized in this manner, the distribution parameters can be used to assign a centile corresponding to an individual's measurement, thus simplifying interpretation.