Managing Medical Complications in MS
Jodi Haselkorn, MD; Moderator
Jill Fischer, PhD
Gerald Goldstein, MD
Joan Robbins, PhD
Joseph Maino, OD
Stephen Fausti, PhD
The best data on prevalence of cognitive impairment in MS that we have comes from two large-scale studies. The first study was a clinic-based study done at the Rocky Mountain MS Center in the mid-80s. The second study was a community-based study. These two studies differed in terms of how the samples were derived, but not in terms of the specific neuro-psychological measures that were used as well as the criteria for defining cognitive impairment. Both resulted in similar estimates that roughly half of all patients with MS are going to experience some measurable changes in their cognitive function.
Drawing on these same studies, the natural question for people to ask is "How do these cognitive changes relate to some of the physical changes that we see in MS?'' Neuro-psychological test performance, which is our indicator of cognitive impairment, correlates relatively weakly with physical impairment in MS traditionally assessed by the Kurtzke EDSS. We get correlations on the order of .17 and .25, which, although it is statistically significant, is a relatively weak relationship and certainly not one that we would want to use to make predictions in the cases of individual patients.
Neuro-psychological impairment in MS does correlate more strongly with what we see on MRI. If you look at correlations between specific test performance and total lesion area (but not at any regional distributions), you get moderate correlations in terms of neuro-psychological test performance and total lesion load on MR.
If you look for specific regional correlations in order to make hypotheses about which tests should be related to which areas, you get stronger correlations. For example, we know that measures of abstract problem-solving performance on those measures correlate relatively highly with lesions in the left-frontal lobe. So, some of these findings that have been demonstrated in other neurologic populations have also been demonstrated in MS.
Neuro-psychological impairment in MS is not uniform. What we are seeing in studies is something that we've known: that the domain in which we see the greatest rate of significant impairment is memory. Between 22 and 31 percent of patients who are tested using these measures will show memory impairment. However, deficits in concentration (or what is sometimes called "information processing"), is nearly as common as impairment, with anywhere between 22 and 25 percent of samples showing some degree of impairment in this area.
The next tier of cognitive domains, or next most prevalent deficits, is in the area of visual spatial perception and problem solving, where we see from 12 to 19 percent of the patients showing impairment, and problems with language or attention span are relative.
A review of 4 case studies shows how varied the appearance of cognitive impairments are within the population of patients with MS.
First, we look at a 38-year-old woman who had MS for 12 years when we first saw her. She was complaining of cognitive problems, but when we tested her, all of her scores were within what was considered technically the "average'' range or within normal limits. She had a couple of areas of what might be termed "relative impairment'' where her performance wasn't as strong as it was in the other domains. In this case, for her it was visual spatial abilities in sustaining attention. Even though she complained of impairment, she had relatively little impairment on formal testing.
We have a 47-year-old woman who had MS for about 5 years when we saw her, who shows what I consider the classic MS-related cognitive impairment. This would be deficits in verbal memory, and processing speed, or concentration, but normal performance in other cognitive domains.
The next case is a 52-year-old male former engineer who had MS for a relatively long period, 25 years, benign from a physical standpoint. However, we found that he not only had impairment in processing speed and memory, but his performance in terms of visual spatial abilities, even calculations, problem-solving, and planning are well below what this very premorbidly, high-functioning man had done.
Finally, we have a 39-year-old woman who had MS for 10 years, who is impaired in virtually all cognitive domains and yet is the primary caregiver for two young children.
We really don't know a lot about what happens in terms of patient's cognitive functions over time. Some studies show relatively little change in the patient's performance over four years. An Italian group, which is investigating a very mild sample, reported that new deficits were evident after four and one-half years in the sample as a whole. It wasn't broken down into subgroups, but there was deterioration in overall performance of the MS group.
A Finnish group has stratified people based on whether they were intact at the initial observation or whether they were impaired. They found that the group that was intact (again this is a group who had the disease for a relatively long period of time) remained intact over a two-to-four year follow-up period. Whereas, the members of the group that was impaired continued to get more impaired, although at variable rates. Patients are always asking me "Well, what's going to happen to me over time?'' The bottom line is much the same answer we have to give them in terms of their physical progression over time. We really can't say with an individual patient. We know that in some patients the cognitive impairment does progress, but we don't know exactly for whom at this point.
Recent clinical trials have studied the impact of the disease-modifying medications on cognitive impairment. A small single center Methotrexate trial looked at Methotrexate in chronic-progressive MS patients. These are patients who had documented disease progression for at least 6 months, but no exacerbations for at least 8 months prior to study entry with entry EDSS of 3 to 6.5. Overall trial outcome was positive on this composite measure with Methotrexate slowing time to progression of disability.
With this study, a comprehensive neuro-psychological battery and multi-varied analysis allowed us to look at overall change on five different measures of neuro-psychological functions, looking at five different domains. The overall multi-varied analysis approached statistical significance. So, we looked further at performance on the individual measures. We found on the Pace Dototory Serial Addition Test, a measure of information processing and to some extent calculation ability, that this was quite sensitive to the effects of Methotrexate over time. We saw that even at 6 weeks, the group started to separate on this particular measure, which was surprising. We did not see the distributions overlap again until approximately the two year marker. So, in a relatively limited way, Methotrexate may be useful to neuro-pyschological performance.
The second clinical-trial data that I am going be presenting to you comes from our Avonex Relapsing Trial. This was a four-center study in which Larry Jacobs was the principal investigator.
This is quite a different sample from what we had in the Methotrexate sample. Also, it was a much larger sample. There were 301 in the overall study, and 276 on whom we had complete neuro-psychological data.
Therefore, what we have here is the overall analysis in this case. Since we had so many measures, we initially did a factor analysis, which is a statistical procedure to allow us to group the measures into cognitive domains and to compare that with our assumptions about what these tests were measuring. Then we used that to select measures that would be included in the final analysis. This was all done based on the baseline data.
The measures were then grouped into sets based on the likelihood of that domain being impaired in MS. The data was examined based on domains where impairment was most likely, with information processing and memory the most common and followed by visual, spatial and executive problems. We rarely saw any problems with verbal abilities through auditory attention span. This is again looking at two-year change and, again, expressed in Z-scores. The Interferon group was significantly different from the placebo group in this multi-varied analysis of measures of information processing and memory.
We also got statistically significant differences in visual spatial abilities and executive functions. However, there is quite a bit of overlap in the measures. When you're looking at abilities that typically don't change in MS, you don't see change over a two-year period in either group and you certainly see no separation between the groups even though we know this is an effective treatment.
Looking a little more closely at measures of information processing and memory, we formed a neuro-psychological composite measure, consisting of a measure of verbal learning; a measure of visual learning; and the pace of the processing rate. Therefore, on this composite measure, you are looking at Z-scores. The difference was not statistically significant until a year and a half of treatment had occurred; then this was maintained over time. Again, very different medications, very different patient groups, and what we're seeing is that these neuro-psychological measures can be useful as outcome measures.
So what does all this mean? Rayo reported that cognitively impaired MS patients are about three times as likely to be unemployed as a comparable group of patients who have comparable physical disability, but no cognitive impairment. Therefore, it seems to be an independent factor contributing to unemployment. They needed significantly more help in performing activities of daily living such as doing basic household tasks, and they were perceived by others as being more emotionally volatile and more confused.
We have also noticed clinically that these cognitively impaired patients seemed to have more difficulty with complicated, multiple-step activities; things like cooking, doing home repairs, managing finance, and doing a lot of the treatment regimens, be it taking medications on a particular schedule or following a prescribed exercise program. They clearly have more accidents, both at home and in driving. For these reasons, they are less likely to adhere to our treatment recommendations. Not necessarily because they don't want to, but because they forget about them, or they can't do the steps.
Finally, there have been four published screening batteries. I can't recommend any one of them and say that this one will really help you in the clinic. All of these take from 30 to 50 minutes long to administer. Many of them include tests that have to be administered by a neuro-psychologist or a psychologist. They are protected tests, but we certainly can do better with various screening batteries than we can in making a guess based on interactions in the neurologic examination. If I were going to choose, I would include a measure of learning in memory, a measure of information processing, and probably either Rayo's or another battery to be as close as we're going to come, but these batteries clearly need measures of executive functions as well.
Go to TOP.
|
|
|