ABSTRACT
Background: Reference intervals are important in the analysis of results obtained for patients in the Hospital. These intervals tend to change as population demographics change. Globally; there has been an increase in the prevalence of diabetes and a change in the criteria set in diagnosing this disease. Since we have used the same reference intervals since 1984; it became imperative to review the reference intervals in use in our Hospital; the University of Port Harcourt Teaching Hospital; Port Harcourt; Nigeria. Methods : A total of 605 patients from the Out-Patient department and the Hospital community randomly following informed consent. Five milliliters (5mls) of blood was collected into fluorides oxalate bottles between 9.00-11.00 a.m. after an overnight fast for the determination of plasma glucose. The samples were assayed daily within two hours of collection. Serum glucose was determined using the glucose oxidase method (Randox Laboratories-Antrim UK). Results: The reference interval covering the central 95th centile as determined by non-parametric methods was 3.0-5.7 mmol/L (95confidence interval-2.7-3.2 and 5.6-5.8 mmol/L) for the entire study population; while that for males was 2.9-6.3 mmol/L and that for females was 3.3-5.6 mmol/L. Conclusion : It is essential to review reference intervals periodically as changes may occur due to changes in population demographics