Correspondence

Zidovudine and Red-Cell Distribution Width

N Engl J Med 2005; 352:2141-2142May 19, 2005

Article

To the Editor:

Zidovudine was the first drug approved for the treatment of HIV infection and is still a mainstay of antiretroviral therapy. Patients taking zidovudine have an increase in the mean corpuscular volume, a variable that is often used to evaluate patients' adherence to treatment.1-3 However, changes in the mean corpuscular volume are seen only several months after the initiation of zidovudine therapy. Red-cell distribution width (calculated as [standard deviation of red-cell volume ÷ mean cell volume] × 100) is a measure of anisocytosis and is determined by an automated, laboratory-based technique. We speculated that changes in the red-cell distribution width might be useful as an early marker of adherence to zidovudine treatment.

We compared hematologic variables among 47 HIV-infected patients who had initiated treatment for the first time with zidovudine, lamivudine, and lopinavir–ritonavir and whose rate of adherence was higher than 90 percent. Patients with anemia or macrocytosis were excluded. Erythrocyte-related values were determined after one, two, and four months of treatment and every four months thereafter. Most of the patients were men with fewer than 200 CD4 cells per cubic millimeter. The median baseline hemoglobin level was 13.7 g per deciliter (interquartile range, 12.1 to 14.6), the median mean corpuscular volume 87 μm³ (interquartile range, 85.2 to 91.5; normal range, 80 to 100), and the median red-cell distribution width 13.8 percent (interquartile range, 13.1 to 15.2; normal range, 11 to 15). After one month and two months of treatment, 100 percent and 63 percent of the patients, respectively, had a mean corpuscular volume within the normal range (80 to 100 μm³). Meanwhile, the red-cell distribution width was above the upper limit of normal in 100 percent of the patients at these time points. The red-cell distribution width increased significantly, to 17.7 percent (interquartile range, 16.3 to 19.6) and 19.8 percent (interquartile range, 18.7 to 21.6) at one month and two months, respectively (P<0.001 for the comparison of both values with baseline). After four months, the mean corpuscular volume was higher than 100 μm³ in all but one patient, and the red-cell distribution width returned to values similar to those at baseline (Figure 1Figure 1Median Red-Cell Distribution Width and Median Mean Corpuscular Volume during Treatment with a Zidovudine-Containing Regimen.).

Poor adherence to treatment is an important factor in treatment failure.4 Traditional methods of monitoring adherence, such as interviewing patients, counting pills, reviewing pharmacy refill logs, and monitoring serum drug levels, are often inaccurate and expensive. During the first two months of zidovudine therapy, the mean corpuscular volume is not useful as a measure of adherence because in most patients the values remain within the normal range. However, during this early period, two populations of erythrocytes coexist: one of normal size and another that has increased in size as a result of the zidovudine therapy. This variability can be measured by determining the red-cell distribution width, and in all the patients in our study, the red-cell distribution width was above the upper limit of normal at these two time points. After four months, there is a new, macrocytic population of erythrocytes, and most patients have a mean corpuscular volume above 100 μm³ and red-cell distribution width that has returned to normal. The red-cell distribution width may be a simple and cheap-to-measure marker of early adherence to zidovudine therapy. In addition, zidovudine therapy should be included in the differential diagnosis of an elevated red-cell distribution width.

Pablo Rivas, M.D.
Miguel Górgolas, M.D.
Manuel L. Fernández-Guerrero, M.D.
Fundación Jiménez Díaz, 28012 Madrid, Spain
pablorivasg@hotmail.com

4 References

1
Snower DP, Weil SC. Changing etiology of macrocytosis: zidovudine as a frequent causative factor. Am J Clin Pathol 1993;99:57-60
Web of Science | Medline

2
Romanelli F, Empey K, Pomeroy C. Macrocytosis as an indicator of medication (zidovudine) adherence in patients with HIV infection. AIDS Patient Care STDS 2002;16:405-411
CrossRef | Web of Science | Medline

3
Steele RH, Keogh GL, Quin J, Fernando SL, Stojkova V. Mean cell volume (MCV) changes in HIV-positive patients taking nucleoside reverse transcriptase inhibitors (NRTIs): a surrogate marker for adherence. Int J STD AIDS 2002;13:748-754
CrossRef | Web of Science | Medline