Increased Intensity of Physiotherapy After Stroke
To the Editor:
The article by Lincoln et al1 in the March issue of Stroke raises some questions regarding their method of data analysis. The authors evaluated the effects of variations in the intensity of physical therapy on arm function after stroke. The patients were randomized into 1 of 3 treatment groups: routine physical therapy (RPT) or 10 hours of additional physical therapy with either a qualified senior research therapist (QPT) or an assistant physical therapist (APT). There was no significant benefit reported from the additional therapy. The authors acknowledge that almost 50% of the treatment group did not complete the additional treatment. The patients who did not complete treatment pose some interesting questions that warrant further explanation by the authors.
The authors report on page 576 that “20% of QPT and 14% of APT patients” were excluded from the final analysis because they were unable to tolerate the additional therapy. Could the authors provide information on how many patients in the RPT group were unable to tolerate routine therapy? If the percentages are similar, one must consider the possibility that a certain percentage of patients simply cannot tolerate any level of therapy provided in an acute inpatient rehabilitation setting. Although it may not be feasible to give more therapy, it may be just as difficult to give regular therapy.
Larger concerns are raised when looking at the group of patients with the second-most-common reason for noncompletion. The authors report that “10% of QPT and 13% of APT patients” were not included in the final analysis because they recovered to “minimal arm impairment” during the intervention period. The major goal of any therapy program is the recovery of function; therefore, why was this group excluded from the analysis of potential benefits from additional therapy? Would not recovery to this level be a desired outcome? Also, what percentage of patients in the RPT group achieved minimal arm impairment during the intervention time frame? If the percentage was low, a significant difference may actually have been achieved with increased physical therapy.
Many challenges are encountered in the rehabilitation of patients after a stroke, and any treatment program must reflect the diversity of impairments. We must continue the efforts to identify which patients can benefit from and tolerate intensive rehabilitation versus those who will respond better to lower-level, subacute programs. The emphasis should be on meeting the individual needs of each patient. This article adds to the literature that assists with these decisions, but we must be cautious that we do not overlook potential interventions by excluding the very group of patients that may demonstrate the most benefit.
- Copyright © 1999 by American Heart Association
Lincoln NB, Parry RH, Vass CD. Randomized, controlled trial to evaluate increased intensity of physiotherapy treatment of arm function after stroke. Stroke. 1999;30:573–579.
We would like to thank Drs Shutter and Whyte for their comments and requests for further information about our trial, and particularly about the patients who did not complete treatment.
Before providing further information, we must first clarify that, as described in the footnotes to Table 2 (page 576) in our article,R1 the only patients excluded from our “final” comparison of outcomes were those who had been lost to follow-up or had died prior to the particular outcome point. This was an “intention-to-treat” analysis. For results to be applied in clinical practice, information is needed about the effectiveness of providing the service to all patients, even though some may withdraw from it. We did conduct an analysis wherein we excluded noncompleters of treatment. The differences in outcome between the groups remained nonsignificant. The information that we provide about noncompleters of treatment on page 576 and in Table 3 is provided for completeness of reporting compliance with the intervention, because this is important information for clinicians in their consideration of both the feasibility and effectiveness of providing additional therapy.
Having clarified which patients were included in our analysis, we would like to make some further points and provide what information we can. Shutter and Whyte suggest that a certain percentage of patients simply cannot tolerate any level of therapy in an acute inpatient rehabilitation setting. We would agree with them on this point. As we note on page 575, we excluded a considerable proportion of the 1265 patients who were admitted to this large general hospital, including 181 (14%) who were unable to tolerate a half-hour session of physiotherapy per day even by 5 weeks after stroke. Besides this 14%, a further 64 (5%) were significantly physically or mentally disabled premorbidly such as to rule out “typical” routine therapy.
We are unable to answer completely the request for information on the number of patients in the routine therapy group who were unable to tolerate routine therapy. To be recruited into the study, all patients had to be able to tolerate 30 minutes of physiotherapy daily. As reported, 20% of QPT and 14% of APT patients were unable to tolerate the full amount of additional therapy. Some of these patients deteriorated after recruitment; others could tolerate their 30 minutes of routine treatment but no more. We have no reason to expect that this would have been different in the RPT group, particularly as we collected information on the routine treatment in order to check for differences between the groups and found none.
We provide further information on the routine treatment another article,R2 which reports a post hoc subgroup analysis examining the effects of severity of arm impairment on the response to additional therapy.
The second concern of Shutter and Whyte, regarding patients who recovered to minimal impairment, should be allayed by our clarification that these patients were indeed included in the “final analysis.” Again, we cannot completely answer their question regarding the numbers in the RPT group who achieved minimal arm impairment during the intervention time frame. However, we have no reason to expect that the proportion in the RPT group differed from that in the intervention groups, particularly in light of the similar medians across the groups for the various arm measures at the postintervention assessment (Table 2).
We appreciate the call by Shutter and Whyte for continued efforts to identify which patients can benefit from intensive rehabilitation. We also agree that there should be an emphasis on meeting the individual needs of each patient. Randomized controlled trials answer important questions, but we would certainly acknowledge that the translation of evidence from group studies to individual patients always requires clinical judgment. We would again refer to our report of the subgroup analysis.R2
Lincoln NB, Parry RH, Vass CD. Randomized, controlled trial to evaluate increased intensity of physiotherapy treatment of arm function after stroke. Stroke.. 1999;30:573–579.
Parry RH, Lincoln NB, Vass CD. Effect of severity of arm impairment on response to additional physiotherapy early after stroke. Clin Rehabil.. 1999;13:187–198.