CNS Vital Signs^(TM)

A Comparison of Simple Office Screening and a Computer-Based Test Batteryto Quantify elements of Cognitive Dysfunction in Persons withChronic Fatigue Syndrome (CFS) and Fibromyalgia (FM)

Charles W. Lapp^*, MD, Rebekah S. Smith*, Wendy Fallick*, Tom Gualtieri, MD†

* from the Hunter-Hopkins Center, P.A., Charlotte, North Carolina
†  from North Carolina Neuropsychiatry, P.A.,  and CNS Vital Signs®, Chapel Hill, North Carolina

Chronic Fatigue Syndrome (CFS) is a disorder characterized by debilitating fatigue, recurrent flu-like symptoms, sleep disruption, and neurocognitive difficulties, most commonly memory loss, reduced attention, and slow processing [Ref 1]. Fibromyalgia (FM) is a related disorder characterized by chronic widespread pain, but also fatigue, sleep disruption, and cognitive difficulties.  It is difficult for medical practitioners to quantify cognitive dysfunction unless they are trained in neurocognitive testing.

In 2003, a computer-based program called CNS Vital Signs®  was released, and purports to measure these parameters using adaptations of simple, standardized, psychometric tests [Ref 2].  The program is inexpensive and easy for untrained assistants to administer. It generally requires less than 30 minutes to complete, and provides immediate on-site reporting of results. It is, therefore, well suited for screening, obtaining baseline levels, or following cognitive changes over time.

The CNS Vital Signs battery addresses the most important cognitive domains: Attention, Memory, Motor Control, Psychomotor Speed, Reaction Time & Information Processing Speed.  The screening battery includes seven selectable tests (Verbal Memory Test, Visual Memory Test, Symbol Digit Coding Test, Finger Tapping Test, Stroop Test, Continuous Performance Test and the Shifting Performance Test) covering five cognitive domains. All of these tests are publicly available and widely accepted measures of their respective domains.  CNS Vital Signs simply provides a convenient computerized platform for administering and grading these tests.  For ease of interpretation, CNS Vital Signs also calculates five parameters based on the results of testing  (Memory, Mental Speed, Reaction Time, Attention, and Cognitive Flexibility) and grades the subject in each category as Above Average, Average, Below Average, or Well Below Average  [Table 1, Overview of Domain Scores].

In order to minimize patient fatigue, we limited this study to 4 tests:  Verbal Memory Test, Visual Memory Test, Symbol Digit Coding Test, Finger Tapping Test [ Table 2, CNS VS Domains].

The purpose of this study was to compare results from this computer-based program in patients with CFS and/or FM (CFS/FM) to two simple office screening tools (Serial 7 Subtraction and the Digit Span Test), normative data, the DSM-IV based General Assessment of Functioning, the Modified Karnofsky Score, and three standardized instruments (the Medical Outcome Survey Short Form-36 Scores, Hospital Depression and Anxiety Scale, Fatigue Scale, and a Visual Analog Pain Scale) [Table 3, Screening Tests and Instruments].

Methods

Subjects were derived from new and established patients of our tertiary medical clinic, which specializes in CFS and FM. Subjects were not selected or excluded in any way except that they benefited from cognitive testing as part of their management.  The data was retrospectively compiled from questionnaires and instruments that are routinely collected from patients in our practice, and each subject provided a general release for the use of anonymous data in the chart.

Age and gender-matched controls were derived from a database of anonymous normal healthy subjects maintained by Dr. Gualtieri.

Statistical analysis was performed using standard statistical software [Ref 3].

Results

Using the t-test for each independent sample, subjects with CFS / FM fared much worse in all of the CNS Vital Signs survey (verbal memory  (VBM), Visual Memory (VIM), finger tapping (FTT), and Symbol Digit Coding (SDC) ) than normal controls [see Table 4, Subjects versus Controls].

A number of conclusions can be drawn from examining the intercorrelation of variables using ANOVA analysis of the data [Table 5, Intercorrelations of Variables].

Serial 7 Subtraction (SSS)

Serial 7 Subtraction correlated moderately well with the Finger Tapping Test, Symbol Digit Coding correct, and Mental Speed on CNS VS.

Poor performance on Serial 7 Subtraction was moderately correlated with a low Karnofsky Score, but was not affected by anxiety or depression as measured by the Hospital Depression and Anxiety Scale.

The Karnofsky score correlated well with the Energy Subscale of the Medical Outcome Survey as well as the Finger Tapping Test and Mental Speed.

Digit Span

Digit Span forward and backward correlated moderately with Verbal Memory and overall Memory as expected, since Digit Span is thought to reflect both memory and attention.

Digit Span did not correlate with the Karnofsky Score nor to anxiety and depression as measured by the Hospital Depression and Anxiety Scale, suggesting that it is relatively  independent of physical illness and psychological stresses.

Instruments

Depression and anxiety (from the Hospital Depression and Anxiety Scale) correlated moderately well with  Symbol Digit Coding, the Finger Tapping Test, and Mental Speed.  This is understandable since these psychological disorders are associated with both mental and psychomotor slowing.

The Medical Outcome Survey Short Form 36 correlated well with GAF, the Fatigue Scale, and the Karnofsky Scale, thus distinguishing itself as an excellent measure of impairment due to psychological factors, fatigue, and general wellness.  The use of the Medical Outcome Survey for predicting impairment in Chronic Fatigue Syndrome has been previously established by Komoroff, and confirmed by our data as well [Ref 4]. 

None of the instruments (MOS SF-36, GAF, Fatigue Scale,Karnofsky) except the Pain Scale correlate at all with cognitive function as measured by CNS VS.  Thus, CNS VS provides another and unique dimension for categorizing persons with CFS / FM.

The Pain Scale correlates inversely with Symbol Digit Coding, Mental Speed, and the general Karnofsky Score as expected, because pain is well known to interfere with concentration or attention (Symbol Digit Coding and Mental Speed),  mobility, sleep, and general well-being (Karnofsky).  Interestingly, increased pain does not seem to interfere with fine motor activity (Finger Tapping Test).

Conclusions

Simple office screening tests such as Serial 7 Subtraction and the Digit Span test correlate with and are indicative of cognitive dysfunction, based on this study.   In cases where these screening tests are abnormal, CNS Vital Signs is able to capture and quantify abnormalities in specific domains such as verbal memory, visual memory, psychomotor slowing, and mental speed.  Furthermore, this study shows that CNS Vital Signs measures elements of cognitive dysfunction that are not reflected in other frequently used instruments such as the Karnofsky Scale, Fatigue Scale, Medical Outcome Survey, or even the Global Assessment of Function.

CNS Vital Signs is inexpensive and simple to administer in the office, so that it can provide objective evidence of neurocognitive impairment for disability purposes, and serial testing may be valuable for following the neurocognitive response to therapy.   The authors suggest that using all seven tests available on CNS Vital Signs allows the calculation of Reaction Time and Attention scores in addition to the domains measured in this paper.  Based on previous studies of neurocognitive dysfunction in CFS / FM, it is suspected that these results would define more deficits in persons with Chronic Fatigue Syndrome and  Fibromyalgia. 

References

1. Busichio K, Tiersky LA, Deluca J, Natelson BH, “Neuropsychological deficits in patients with Chronic Fatigue Syndrome,” J Int Neuropsychol Soc. 2004 Mar;10(2):278-85
2. Gualtieri, T, et al., “Reliability and validity of a new computerized cognitive screening battery,” presented to the International Neuropsychological Society Annual Meeting, Baltimore MD, February 4-7, 2004
3. Software for statistics
4. Komaroff AL, Fagioli LR, Doolittle TH, Gandek B, Gleit MA, Guerriero RT, Kornish RJ 2nd, Ware NC, Ware JE Jr, Bates DW, “Health status in patients with chronic fatigue syndrome and in general population and disease comparison groups,” Am J Med. 1996 Sep;101(3):281-90.

This was presented as a poster at the 7^th Annual Research and Clinical Conference of the American Association of Chronic Fatigue Syndrome held October 8-10, 2004, in Madison, Wisconsin,

© Charles W. Lapp, MD, October 2004

Table 1, Overview of Domain Scores

Table 2, CNS VS Domains

Table 3, Explanation of Screening Tests and Instruments

Table 4, Subjects versus Controls

Table 5, Intercorrelations of Variables

All shaded figures indicate a correlation of r < 0.5.  Dark shading indicates stronger correlation.

Key:

Updated March 2005