to the editor: In the article, “Lactose Intolerance,”¹ Dr. Swagerty and colleagues acknowledge that the lactose tolerance test has false-positive or false-negative results in 20 percent of patients. They also state that the lactose breath hydrogen test has been found positive in 90 percent of patients with lactose malabsorption.

We compared three different methods of diagnosing lactose malabsorption: (1) lactose load with determination of blood glucose concentration; (2) lactose load with urinary galactose measurement; and (3) duodenal lactase activity assay. Our studies show a higher false-negative rate when using blood glucose measurements, causing lactose malabsorption to remain undiagnosed in many patients. The sensitivity and specificity in our studies was only 60 percent and 96 percent, respectively. The specificity is comparable with the use of breath hydrogen or urinary galactose measurement tests.^2,3 Therefore, the lactose test with blood glucose measurement should not be used as a screening test when high sensitivity is important.

When assessing the effectiveness of these tests in the clinical practice setting, one must consider predictive values as well as false-positive and false-negative results. Although the sensitivity and specificity of the test can be the same, positive and negative predictive values depend on the pretest probability or disease prevalence. For example, our studies produced an increase in the positive predictive value from 82 percent among a general Estonian population with a lower lactose malabsorption prevalence of 23 percent, to 95 percent among a Russian population in Estonia with a high lactose malabsorption prevalence of 57 percent, to 99.5 percent among a Khanty population in Siberia with a lactose malabsorption prevalence of 83 percent.^4,5 If the test is used among patients with an increased pretest probability of lactose malabsorption (e.g., patients reporting milk intolerance), the positive predictive value of the test is higher than in the general population.

In clinical practice, it is important to accept that clinical intolerance to lactose may not be synonymous with low lactase activity. When the diagnosis is confirmed by lactose testing, we prefer to speak about lactose malabsorption because our experience from hundreds of tests among the general population shows that approximately 30 percent of patients with lactose malabsorption can clinically tolerate lactose without any symptoms.