To the Editor:
In the February issue of Stroke, Jørgensen and coworkers1 report that treatment of stroke patients in a stroke unit (SU) compared with a general ward (GW) offers benefits on the order of odds ratios of 0.17 to 0.66. Consistently, their conclusion is that all patients with acute stroke should be treated in a dedicated SU. Moreover, their findings raise serious considerations of whether further treatment of stroke patients in GWs is ethically justifiable.
In an earlier publication of the same investigation,2 Jørgensen and coworkers described an inhospital mortality of 23% in patients treated in the SU, which was 21% lower than that in patients treated in a GW. That is in fair agreement with the findings of the meta-analysis presented by the SU Trialists,3 which depended on studies predominantly performed in the 1980s.
Logistic multiple regression analysis blew up this benefit to an odds ratio for poor outcome of 0.50 for the total stroke population and 0.17 for patients with severe stroke treated in the SU. However, these figures depend heavily on the reliability of cormorbidity assessment. Why should the comorbidity of patients treated in Bispebjerg, where the SU was located, have been much greater than in Frederiksberg? The only hint given is the lower income of people in Bispebjerg. However, very little information is provided: how and by whom comorbidity was assessed and whether any control of these data was performed. Evaluation of previous myocardial infarctions and ischemic heart disease may be difficult in stroke patients. Diagnosis of hypertension and possibly also of diabetes depends on corresponding treatment. Especially in hypertension, decision to treat blood pressure may have changed to lower levels between 1989 to 1990, when GW patients were evaluated, and 1991 to 1993, when SU patients were studied.
Furthermore, neither treatment in the SU nor in the GW is detailed. The only information provided about the GW is that traditional treatment was given. Both Bispebjerg and Frederiksberg hospitals reportedly possessed all major medical and surgical specialities, but again we are left in uncertainty of whether that included a neurosurgery department that might provide life-saving services for at least a small portion of patients with severe strokes.
The main question, however, is whether the findings reported by Jørgensen et al1 still hold true, since at least those for the GW date back for more than 10 years. As pointed out by the authors1 and others,4 5 the mainstay of SU treatment appears to be team approach, earlier and more intensive mobilization and rehabilitation of patients, and prompt and determined correction of fever, infections, hyperglycemia, and other medical problems. It is the undisputable merit of SUs to have elaborated these treatment principles, but they may be easily adopted by general wards. That this has happened is suggested by more recent findings from Sweden and from Austria. A multicenter observational study6 of over 10 000 stroke patients admitted to Swedish hospitals in 1996 shows mortality and other outcome benefits of SU treatment only for patients who were fully conscious on admission. The relative risk in case fatality for treatment in an SU of this collective was a modest 0.91. In contrast to the findings reported by Jøorgensen et al,1 outcome of patients with more severe strokes (impaired consciousness at admission) did not show a significant difference in any of 15 outcome measurements, whether treated in an SU or a GW. Furthermore, 27-day case fatality of all patients treated in GWs was 17% (B. Stegmayr, MD, PhD, written communication, 2000), which is considerably lower than the rate of 23% reported in the SU in Copenhagen some years earlier.
Preliminary unpublished results (J. Slany, MD, 2000) of 300 totally unselected stroke patients of an ongoing Austrian stroke registry conducted in general medical wards show an inhospital mortality of 17%, which is well in accordance with the Swedish data.
It is reassuring to see that improvements in the management of acute stroke seem to have leaked out of the boundaries of SUs and to have caught hold also in general medical and neurological wards. Thus, the results and conclusions of Jørgensen and coworkers seem to reflect past glories rather than present-day facts.
- Copyright © 2000 by American Heart Association
Jørgensen HS, Kammersgaard LP, Houth J, Nakayama H, Raaschou HO, Larsen K, Hübbe P, Olsen TS. Who benefits from treatment and rehabilitation in a stroke unit? Stroke. 2000;31;434–439.
Jørgensen HS, Nakayama H, Raaschou H, Larsen K, Hübbe P, Olsen TS. The effect of a stroke unit: reductions in mortality, discharge rate to nursing home, length of hospital stay, and cost. Stroke.. 1995;26:1178–1182.
Stroke Unit Trialists’ Collaboration. Collaborative systematic review of the randomised trials of organised inpatient (stroke unit) care after stroke. BMJ.. 1997;314:1151–1159.
Stroke Unit Trialist’s Collaboration. How do stroke units improve patient outcomes? A collaborative systematic review of the randomized trials. Stroke.. 1997;28:2139–2144.
Indredavik B, Bakke F, Slørdahl SA, Rokseth R, Håheim LL. Treatment in a combined acute and rehabilitation stroke unit: which aspects are most important? Stroke.. 1999;30:917–923.
Stegmayr B, Asplund K, Hulter-Åsberg K, Norrving B, Peltonen M, Terént A, Wester PO. Stroke units in their natural habitat: can results of randomized trials be reproduced in routine clinical practice? Stroke.. 1999;30:709–714.
Jörg Slany questions the validity of the assessment of comorbidity in our study. Comorbidity was assessed by the same person (the principal investigator of the study, Henrik Stig Jørgensen) in both study groups. Definitions of comorbidities were identical in the 2 study populations. There is, therefore, no reason to believe that the difference found in comorbidity between the 2 study groups is an artifact. The higher comorbidity in patients treated on the stroke unit probably reflects that the catchment area for the stroke unit is a working-class district, whereas the catchment area for the hospital offering treatment on general wards is an upper class district.
Throughout the 80s and 90s, treatment and rehabilitation on dedicated stroke units have consistently been proved superior to treatment on general wards.R1 R2 R3 This is true even in completely unselected patients with stroke.R4 R5 R6 Slany argues that this may no longer be true. It would be interesting to see scientific proof for this statement.
Indredavik B, Slørdahl SA, Bakke F, Rokseth R, Håheim LL. Stroke unit treatment: long-term effects. Stroke.. 1997;28:1861–1866.
Rønning OM, Guldvog B. Outcome of subacute stroke rehabilitation: a randomized controlled trial. Stroke.. 1998;29:779–784.
The Stroke Unit Trialists’ Collaboration. Collaborative systematic review of the randomised trials of organised inpatient (stroke unit) care after stroke. BMJ. 1997;1151–1159.
Jørgensen HS, Nakayama H, Raaschou HO, Larsen K, Hübbe P, Olsen TS. The effect of a stroke unit: reductions in mortality, discharge rate to nursing home, length of hospital stay, and cost. Stroke.. 1995;26:1178–1182.
Jørgensen HS, Kammersgaard LP, Nakayama H, Raaschou HO, Larsen K, Hübbe P, Olsen TS. Treatment and rehabilitation on a stroke unit improves 5-year survival: a community-based study. Stroke.. 1999;30:930–933.
Jørgensen HS, Kammersgaard LP, Houth J, Nakayama H, Raaschou HO, Larsen K, Hübbe P, Olsen TS. Who benefits from treatment and rehabilitation in a stroke unit? A community-based study. Stroke.. 2000;31:434–439.