HOMA: Often Mentioned, Rarely Defined

HOMA: Often Mentioned, Rarely Defined
Alexander Julian Doctor of Pharmacy Candidate
University of Florida College of Pharmacy

Homeostatic model assessment (HOMA) is commonly cited in the literature but many clinicians and health professionals are unfamiliar with what it is, what it is used for, and most importantly, why it is used.  A pubmed search of the term “HOMA” results in over 3600 hits; many of these are studies that use HOMA to analyze their data and draw conclusions.  Simply put, HOMA is a method to quantify insulin resistance and beta-cell function.  Originally described in 1985, the original model (HOMA1) was derived mathematically by looking at the interaction between beta-cell function (HOMA1-%B) in response to glucose concentrations, and insulin resistance (HOMA1-IR) in which insulin has diminished suppressive effect on hepatic glucose production.  Because a feedback loop between the liver and beta-cells balances insulin secretion and hepatic glucose output, theoretically, it makes it possible to compute steady-state insulin and glucose concentrations.  The original model was calibrated to give normal beta-cell function of 100% and normal insulin resistance of 1.  Higher insulin resistance is characterized by higher steady state insulin levels and reduced beta-cell function characterized by a decreased compensation to increase glucose levels. By utilizing this interrelationship, any pair of fasting glucose and insulin concentrations can be used to quantify beta-cell function and insulin resistance using the following formulas.

HOMA1-IR = (FPI (mU/L) x FPG (mmol/L))/22.5

HOMA1-%B = (20 x FPI (mU/L))/(FPG (mmol/L) – 3.5)

Where FPI is fasting plasma insulin and FBG is fasting plasma glucose; to convert mg/dL to mmol/L, simply divide by 18.

The original model did not account for differences between hepatic and peripheral insulin sensitivity, increases in insulin secretion or decreases in hepatic glucose production for plasma glucose concentrations above 180 mg/dL, renal glucose losses, or the contribution of circulating proinsulin.  An updated HOMA model (HOMA2) has since been created, however it is a computer model and has no simple equation but it adjusted to account for these variations.  It models insulin sensitivity (HOMA2-%S) where 100% is normal which is the reciprocal of insulin resistance (100/S%).  In addition, the original HOMA model uses equations that were calibrated to insulin assays used in the 1970’s which result in underestimation of %-S and overestimation of %-B.  With knowledge of these differences, it is therefore important when evaluating studies to determine whether the HOMA formula or the computer-based HOMA was used to quantify insulin resistance and beta-cell function.

When determining insulin concentrations to compute HOMA, there are some important considerations that should be addressed.  C-peptide is a common marker of insulin secretion as it is released in equal amounts as insulin from beta-cells in response to elevated glucose.  Therefore, c-peptide may be most useful only in assessing beta-cell function (%B) while specific insulin concentrations are will more useful in determining insulin action or sensitivity (%S).  This is especially true if patients use exogenous insulin as part of their diabetes management.

A HOMA2 calculator can be found at the following website: http://www.dtu.ox.ac.uk/index.php?maindoc=/homa/

Although the gold standard for assessing insulin resistance remains that of the Hyperinsulinemic Glucose Clamp, it is often too laborious and not feasible for large studies, leaving HOMA to be a method likely used in studies we will read today as well as those in the future.  I am hopeful that this column gives basic understanding of the HOMA model, some limitations, and how it’s use in the diabetic literature.

References:

Wallace TM, Levy JC, et al.  Use and Abuse of HOMA modeling.  Diabetes Care (2004); 27(6): 1487-95

Matthews DR.  Insulin resistance and beta-cell function – a clinical perspective.  Diabetes, Obesity and Metabolism (2001); 3 (Suppl. 1): S28-S33

Muniyappa R, Lee S, et al.  Current approaches for assessing insulin sensitivity and resistance in vivo: advantages, limitations, and appropriate usage.  Am J Physiol Endocrinol Metab 294: E15-E26, 2008

Alexander Julian
3rd Year Pharm D. Candidate
UF College of Pharmacy

Provided by www.diabetesincontrol.com