Re: Randomized Controlled Study of Stroke Unit Versus Stroke Team Care in Different Stroke Subtypes
To the Editor:
We have read with interest the article by Evans et al,1 which presents a post hoc analysis of a previously published randomized controlled trial comparing a stroke unit, a stroke team, and domiciliary care.2 The demonstration that the stroke team could not provide sufficient stroke management can be considered the most important conclusion of this trial. This finding confirms our previously reported study comparing a stroke unit with a stroke team within the same neurological department, demonstrating that a stroke unit is better than the stroke team in terms of reduction of mean hospital stay, systemic and neurological complications, and hospital costs, in addition to an improvement of functional state at discharge with an increase in the number of independent patients and a decrease in long-term hospitalization. Besides, we also demonstrated that cerebral hemorrhage and territorial infarction patients (both atherosclerotic and cardioembolic stroke subtypes) get more benefit from a neurological stroke unit care than lacunar or transient ischemic attack patients,3 a finding that could help us to optimize the stroke unit results, taking into account the limited stroke unit resources, and the preliminary data of this analysis were communicated in the 10th European Stroke Conference that took place in Lisbon, Portugal, in May 2001.4 In this report Evans et al, in a post hoc analysis of a randomized trial, found no differences in outcome in the stroke unit versus the stroke team in terms of mortality, mortality or institutionalization, level of neurological recovery, or dependence, confirming our observations about lacunar stroke.
In our study, the data of 285 lacunar stroke patients admitted in a neurological department during 3 consecutive years, 78 attended by a stroke team (first year) and 206 in a stroke unit (second and third years), were analyzed. There were no differences in vascular risk factors or comorbidity between the 2 samples. Stroke unit care was associated with a decrease in both systemic and neurological complications, 54.7% and 84.4% reductions, respectively, as well as a decrease of 61.8% in the percentage of lacunar stroke patients with any complication. Only 1 patient in the stroke unit (0.49%) and no patients in the stroke team group died. No patient required institutionalization. There was a trend to an improvement in functional status (modified Rankin Scale).
Moreover, we found that lacunar stroke patients in the stroke unit had a significantly lower length of stay than in the stroke team group (7.17 versus 12.83 days; P=0.000 Bonferroni post hoc test). However, in the report by Evans et al, lacunar stroke patients had a longer length of stay in the stroke unit (27.4 versus 18.5 days; P<0.01). The shorter length of stay in our study (whether stroke team or stroke unit) is easily explained by the different stroke unit design as compared with the design by Evans et al. Our stroke unit is an acute neurological stroke unit, attending patients within the first 48 hours from stroke onset, and does not include the rehabilitation ward stay. In addition, although early physiotherapy and rehabilitation are important features of our stroke unit, when a stroke patient has been clinically stabilized, the diagnostic procedures have been finalized, and the patient has a modified Rankin Scale >2, the rehabilitation treatment can continue in an appropriate poststroke rehabilitation ward. On the other hand, Evans et al’s stroke unit is a combined acute and rehabilitation stroke unit settled in a geriatric medicine ward, with no limit of time in unit, although with a program of discharge, usually to final placement (home or institution).5
It has been suggested that an early start of physical therapy and stabilization of blood pressure are probably the most significant aspects of care at the stroke units, in addition to a reduction in the number of patients with a temperature >38°C in the first 5 days, that contribute to the positive results of stroke units as compared with a general medicine ward.6 A possible explanation for the lesser benefit of a stroke unit in lacunar stroke is that in this stroke subtype, located in subcortical white matter, the prognosis is probably not as sensitive to changes in blood pressure, temperature, or glucose blood levels than in large vessel infarctions with cortical damage.
In our opinion, all stroke patients should be admitted to an acute stroke unit. However, the data of Evans et al’s study as well as ours, showing that large vessels infarctions get more benefit from a stroke unit than lacunar strokes, are useful to improve stroke unit resources, taking into account that there are not always enough available beds.
Finally, although we agree with Hacke7 that it is difficult to understand the design of this randomized study on stroke units, considering the amount of evidence about their benefit from previous randomized studies at the time of the beginning of this study when stroke unit care was a level of evidence I, grade A recommendation, it is true that an adequately powered prospective, randomized, controlled trial in patients with lacunar syndromes to compare the benefits and cost-effectiveness of stroke unit care with organized care in other settings could have no ethical problems, because of the lack of evidence in this stroke subtype both on outcome and in costs. Although this study observed a longer length of stay in the stroke unit group, in our study we found benefits in lacunar stroke in terms of reduction of length of stay and consequently a decrease in hospital costs, and better clinical outcome. However, a randomized trial would clarify the efficacy of the stroke unit in lacunar stroke patients.
- ↵Evans A, Farzaneh H, Donaldson N, Kalra L. Randomized controlled study of stroke unit versus stroke team care in different stroke subtypes. Stroke. 2002; 33: 449–455.
- ↵Fuentes B, Dìez Tejedor E, Lara M, Frank A, Barreiro P. Stroke unit: which stroke subtype benefit the most? Cerebrovascular Dis. 2001; 11 (suppl 4): 121.Abstract.
- ↵Langhorne P, Dennis M. Stroke Units: An Evidence Based Approach. London, UK: BMJ Books; 1998.
- ↵Indredavik B, Bakke F, Slordahl SA, Rokseth R, Haheim LL. Treatment in a combined acute and rehabilitation stroke unit. Stroke. 1999; 30: 917–923.
Thank you for forwarding the letter by Fuentes and Dìez Tejedor. Their studies support our findings on the differences between stroke unit and stroke team care, both in the acute and the postacute phases of stroke management. Their experience, similar to ours, is that patients with large-vessel stroke benefit more than those with lacunar stroke, probably because of a lower probability of death and fewer rehabilitation needs.
The difference in the length of stay between the study by Dìez Tejedor and Fuentes1 and our study are of interest and are probably due to differences between healthcare systems in Spain and the United Kingdom. As highlighted in the letter, the unit in their study was an acute neurological unit rather than the combined acute and rehabilitation stroke unit described in our study. It is likely that the process of comprehensive multidisciplinary patient assessment by medical, nursing, therapy, and social service professionals and assessment of home environment, social support, and caregiver needs prior to discharge were responsible for longer stays. The practice of discharging patients from the stroke unit only when community support and continuing rehabilitation systems were in place may also have contributed. Many of these processes were not undertaken in patients on the general wards managed by the stroke team.2
We agree with Fuentes and Dìez Tejedor that all suspected stroke patients should be admitted to a specialist stroke unit for comprehensive assessments, investigations, and acute management, regardless of the type of stroke. It is probable that patients with lacunar infarcts may not be disadvantaged by postacute management in other settings, as long as they receive specialist stroke care. However, only mortality and basic functional abilities were assessed in our study; it is possible that patients managed on the stroke unit may have had better outcome if assessments for higher functional abilities and quality of life were included. The evidence presented in our article is not robust enough to change practice because of the post hoc nature of the analysis, small sample size, and limited outcome measurement. However, it does suggest that an adequately powered definitive trial on this aspect of stroke care may be justified.