Delta check and precision standards for sodium.

Q What would you consider a significant delta regarding sodium on repeated analysis of a patient's specimen? We have noticed that when we repeat low sodiums (128 mmol/L to 134 mmol/L), we will see an increase on the repeated analysis of anywhere from 4 mmol/L to 10mmol/L. We are repeating these sodiums because we are seeing low anion gaps (1 to 5). QC and calibration are okay. Our specimens are plasma; we have not noted anything floating in the specimen that might cause the interference. At what point would you consider the delta significant on a repeated sodium on the same specimen?

A The determination of what to consider a significant change in an analyte has been widely studied. Unfortunately, there are almost as many recommendations as there are publications devoted to this particular topic. Your question brings up two different issues; delta checks and within-run precision. Delta checks, defined as "the comparison of laboratory test results obtained on a current specimen with those values obtained from the same patient on a previous specimen," is one technique employed by laboratories to detect specimen mix-up errors. The expected variability between test results obtained on the two specimens depends on both the analyte and time interval between determinations. Delta check limits that have been recommended for sodium are 5% and 10 mmol/L. (1)

The question you ask, "What would one consider to be a significant change after repeating the same specimen," refers to within-run precision rather than true delta checking. The easiest way to evaluate what the expected variation observed with repeated measurements would be to perform a replication experiment. A replication experiment is typically performed by obtaining test results on 20 samples of the same material and calculating the mean, standard deviation, and coefficient of variation. The purpose of this experiment is to observe the variation expected in a test result under similar operating conditions. Various factors, however, such as differences in matrices between samples, and the concentration of analyte within the sample may result in greater than normal imprecision. In our laboratory, within-run imprecision for sodium is less than 2 mmol/L between repeated measurements of this analyte.

Another way to look at this problem is to establish the maximum difference in repeated measurements that can be tolerated before the difference is considered to be clinically significant. This method takes into account the analytical and within-subject variation of an analyte and is calculated, for P [less than or equal to]0.05, as 2.77 ([[CV.sup.2].sub.A] + [[CV.sup.2].sub.I])[.sup.1/2], where [[CV.sup.2].sub.A] is the square of the analytical CV for sodium and [[CV.sup.2].sub.I] is the square of the normal intraindividual variability of sodium. Using an analytical CV of 0.75 mmol/L and an intraindividual variability of 0.86 mmol/L for sodium, a critical difference of up to 3.2% between repeated measures for sodium are acceptable. (2)

--Steven C. Kazmierczak, PhD, DABCC 

Department of Pathology 

Oregon Health and Sciences University

Portland, OR

References

1. Westgard JO, Klee GG. Quality Management. In: Burtis CA, Ashwood ER, eds. Tietz Textbook of Clinical Chemistry. Philadelphia, PA: W.B. Saunders Co; 1999;407.

2. Fraser CG, Cummings ST, Wilkinson SP, Neville RG, Knox JDE, Ho 0, et al. Biological variability of 26 clinical chemistry analytes in elderly people. Clin Chem. 1989;35:783-786.