To the Editor:

We appreciate the comments by Dr. Hietaharju and Dr. Auvinen and apologize for not citing their initial report (1) on validation of the criteria in 19 NP syndromes in NPSLE as recommended by the ACR ad hoc committee in 1997 (2). Although both articles refer to evaluating the validity of the ACR battery of tests for NPSLE, the studies are quite different in terms of their goals, methods, and conclusions. Our study was directed at 1 of the 19 NP syndromes, the cognitive dysfunction syndrome. In contrast, the study by Ainiala et al (1) aimed to validate the use of the criteria in all 19 NPSLE syndromes.

Several differences between our 2 studies are worth highlighting. First, although both studies used neuropsychological tests that were recommended by the ACR ad hoc committee, the 2 batteries of tests are very different. In comparison with the 1-hour proposed ACR-SLE battery used in our study, Ainiala et al (1) followed the broad ACR recommendations for a comprehensive clinical neuropsychology evaluation. This included multiple cognitive domains (administered over 3 to 4 hours). Ainiala et al (1) reported higher cognitive impairment in patients with SLE compared with controls; but more than 25% of the controls were classified as being cognitively impaired. Ainiala et al interpret this finding to mean that minor central nervous system (CNS) dysfunction (measured by cognitive impairment) leads to poor specificity and a lack of distinction between the patients with SLE and controls. It is also possible that some methodologic differences (i.e., overly inclusive criteria for impairment) exist, because cognitive differences between patients with SLE and controls have been shown to exist in multiple prior studies (3–8).

The entire goal of our study was to establish the reliability and validity of the 1-hour proposed ACR-SLE battery and to ensure sensitivity in detecting cognitive dysfunction in SLE. To achieve our goal, we administered the proposed research battery as well as a comprehensive 4-hour battery that had previously shown sensitivity in documenting cognitive impairment in patients with SLE compared with matched controls. Results showed high correspondence between the short battery and the larger, more comprehensive battery in detecting cognitive dysfunction. Interestingly, agreement between the batteries was higher in patients without overt NP syndromes such as seizure or stroke. Thus, the main hypothesis, that mild cognitive dysfunction in SLE can be evaluated in a brief 1-hour neuropsychological battery, was established.

We acknowledge the important first step made by Ainiala et al (1) toward investigating the overall ACR criteria for the 19 NPSLE syndromes, but suggest that a more detailed and careful investigation of neuropsychological tools and impairment criteria, as introduced by our study, will better classify the mild CNS symptoms.

Elizabeth Kozora PhD, ABPP/CN*, Sterling West MD†, * National Jewish Medical and Research Center Denver, Colorado, † University of Colorado Health Sciences Center Denver, Colorado.