Effect of Functional Status on Survival in Patients With Stroke: Is Independent Ambulation a Key Determinant?

Transcription

1 527 ORIGINAL ARTICLE Effect of Functional Status on Survival in Patients With Stroke: Is Independent Ambulation a Key Determinant? Hsi-Ting Chiu, MD, Yen-Ho Wang, MD, Jiann-Shing Jeng, MD, PhD, Bang-Bin Chen, MD, Shin-Liang Pan, MD, PhD ABSTRACT. Chiu H-T, Wang Y-H, Jeng J-S, Chen B-B, Pan S-L. Effect of functional status on survival in patients with stroke: is independent ambulation a key determinant? Arch Phys Med Rehabil 2012;93: Objective: To investigate the effect of functional status, measured using the Modified Rankin Scale (MRS), at 3 months after stroke on survival in patients with stroke. Design: Cohort study. Setting: Referral medical center. Participants: Patients with stroke (N 1032). Interventions: Not applicable. Main Outcome Measure: Survival after stroke. Results: The Kaplan-Meier survival curves stratified by the 3-month MRS score showed 2 clear groups of patients with 3-month MRS scores of 0 to 3 (able to walk without assistance) and 4 or 5 (unable to walk without assistance). Accordingly, we grouped the patients into a high function (HF) group (3-month MRS 3) and a low function (LF) group (3-month MRS 4). Multiple Cox regression analysis showed that the LF group had significantly poorer survival (adjusted hazard ratio 4.69; 95% confidence interval [CI], ; P.001) than the HF group. Other significant risk factors of higher mortality were older age, history of diabetes mellitus, and heart disease. Conclusions: This study showed a significant influence of the 3-month MRS score on stroke survival. Moreover, independent ambulation may be a major determinant of a favorable survival prognosis. This finding suggests a potential role of rehabilitation in promoting stroke survival by maximizing ambulation function. Key Words: Rehabilitation; Risk factors; Stroke; Survival by the American Congress of Rehabilitation Medicine From the Departments of Physical Medicine and Rehabilitation (Chiu, Wang, Pan), Neurology (Jeng), and Department of Medical Imaging and Radiology (Chen), National Taiwan University Hospital, Taipei; Department of Physical Medicine and Rehabilitation, Keelung Hospital, Department of Health, Executive Yuan, Keelung (Chiu); and College of Medicine, National Taiwan University, Taipei, Taiwan (Wang, Jeng, Chen, Pan). Supported by the National Science Council of Taiwan, Republic of China (grant no. NSC B ). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Correspondence to Shin-Liang Pan, MD, PhD, Clinical Assistant Professor, Dept of Physical Medicine and Rehabilitation, National Taiwan University Hospital and National Taiwan University College of Medicine, No 7, Chun-San S Rd, Taipei, 100, Taiwan, ROC, panslcb@gmail.com. Reprints are not available from the author /12/ $36.00/0 doi: /j.apmr STROKE IS A MAJOR cause of disability and death worldwide, 1-3 resulting in an enormous economic and social burden. 4 The exploration of prognostic factors can assist clinicians in determining treatment options, setting goals, and allocating resources. 4,5 It is therefore very important to investigate prognostic factors of survival after stroke. 6 Higher poststroke functional performance has been linked to lower mortality in patients with stroke. 2,7 Previous studies have shown that a higher score on the Barthel Index (BI), a scale for measuring functional performance of basic activities of daily living, 8 correlates with better survival after stroke. 5,9 However, the BI is susceptible to ceiling effects, 10 which may limit its application to the evaluation in stroke patients with mild to moderate severity. The Modified Rankin Scale (MRS), which is a global outcome measure consisting of 7 grades ranged from 0to6 11 (table 1), has been used for functional evaluation in patients with stroke and is considered to have less of a ceiling effect. 12 However, few studies have investigated the prognostic role of MRS in long-term survival after stroke. Prencipe et al 7 showed that the MRS measured during the first month after minor stroke is a significant predictor of 10-year mortality. 7 Nevertheless, this study only included patients with minor stroke (MRS scores 2), and therefore the prognostic value of MRS in patients with more severe stroke (eg, MRS score 3) was not assessed. In addition, Slot et al 13 evaluated the prognostic role of the MRS measured at 6 months after stroke and found that it was a significant predictor of survival. 13 However, this study did not include an adjustment for the potential influence of the initial neurologic deficit in the analysis, which may affect the interpretation of the prognostic role of the MRS, because prior studies have indicated that initial neurologic impairment is a major predictor of survival after stroke. 14,15 The purpose of the present prospective cohort study was to evaluate the prognostic role of the 3-month MRS in long-term survival after stroke, including an adjustment for initial stroke severity in the analysis. We chose the 3-month MRS as the predictor of interest because it has been suggested that functional recovery in stroke patients typically reaches a plateau at about 3 months after stroke. 16 Moreover, we used a study population with a broader spectrum of stroke severity so as to extend the applicability of our results. METHODS Participants The data for this study were collected prospectively from stroke patients consecutively admitted to our medical center AIC BI CI HF HR LF MRS NIHSS List of Abbreviations Akaike information criterion Barthel Index confidence interval high function hazard ratio low function Modified Rankin Scale National Institutes of Health Stroke Scale

2 528 INDEPENDENT AMBULATION AND STROKE SURVIVAL, Chiu Score Table 1: The MRS 11 Description 0 No symptoms at all 1 No significant disability despite symptoms; able to carry out all usual duties and activities 2 Slight disability; unable to carry out all previous activities, but able to look after own affairs without assistance 3 Moderate disability; requiring some help, but able to walk without assistance 4 Moderately severe disability; unable to walk without assistance and unable to attend to own bodily needs without assistance 5 Severe disability; bedridden, incontinent, and requiring constant nursing care and attention 6 Dead between January 2006 and December Patients who were admitted within 10 days after the onset of stroke during this period were considered for inclusion in this study. The exclusion criteria for the recruitment of subjects were: (1) traumatic intracranial hemorrhage, subdural hematoma; (2) no brain images available for the diagnosis of stroke; and (3) subsequent tests showing noncerebrovascular causes for the clinical presentations. Because this study enrolled patients without an age limit, age was treated as a continuous variable and was included in the analysis. The study was approved by the human subject research ethics committee of the medical center. The diagnosis of stroke was made on the basis of the clinical features, neurologic examination, and brain imaging (computed tomography and/or magnetic resonance imaging). We classified the type of stroke into cerebral infarction and intracerebral hemorrhage. The diagnostic criteria for cerebral infarction were focal neurologic deficits with sudden onset and lasting more than 24 hours or evidence of cerebral infarction on brain imaging corresponding to the clinical manifestations. Intracerebral hemorrhage was diagnosed by a brain parenchymal hemorrhage or subarachnoid hemorrhage identified through brain imaging corresponding to the clinical features. Clinical Parameters Demographic and clinical factors of the stroke patients were collected. Initial severity of stroke was assessed at admission using the National Institutes of Health Stroke Scale (NIHSS). 17 The hematologic and biochemical laboratory tests were performed in our medical center. A history of diabetes mellitus was determined by any 1 of the following 3 criteria: (1) fasting plasma glucose concentration greater than 126mg/dl; (2) a random plasma glucose level more than 230mg/dl; and (3) patients diagnosed with diabetes by doctors or already taking hypoglycemic agents before stroke. Presence of hypertension was defined as any of the following 3 criteria: (1) use of antihypertensive agents before stroke; (2) resting systolic blood pressure greater than 140mmHg; or (3) diastolic blood pressure greater than 90mmHg after 7 days of stroke onset. Hypercholesterolemia was defined as any of the following: (1) fasting serum total cholesterol level 200mg/dl; (2) low-density lipoprotein 130mg/dl and high-density lipoprotein 40mg/dl; or (3) use of hypocholesterolemic agents. Presence of heart disease was determined by a history of any of the following diseases: atrial fibrillation, sick sinus syndrome, endocarditis, heart failure, or valvular or ischemic heart disease. Smoking status was defined as smoking more than 1 cigarette a day for longer than 6 months. Alcohol use was defined as drinking alcohol at least once daily for more than 1 year. Each subject followed a regular rehabilitation training program during hospitalization, including physical therapy, occupational therapy, speech therapy, and nursing support. The MRS scores were measured by research nurses at 3 months after stroke. To determine the date of death, the data were linked to the National Mortality Registry until December 31, Because the present study was designed to investigate the prognostic role of the 3-month MRS score on subsequent survival, only subjects surviving beyond 3 months after stroke were included in the analysis. The survival time was calculated from 3 months after stroke. Statistical Analysis The Kaplan-Meier method was used to plot survival curves stratified by MRS scores. The Cox regression model was used to estimate the hazard ratios (HRs) of the prognostic factors on survival after stroke. The effects of the following independent variables on survival were assessed: (1) demographic factors, including age and sex; (2) 3-month MRS; (3) initial NIHSS score; (4) medical comorbidities, including diabetes mellitus, heart disease, hypercholesterolemia, hypertension, and previous stroke; and (5) life style factors, including alcohol use and smoking. Univariate analysis was initially performed for each variable, followed by stepwise multiple regression analysis. A variable had to be significant at a P value of.25 to be entered in the stepwise regression model, while a variable in the model has to be significant at the.15 level for it to remain in the model. 18 The initial NIHSS score was always included in the multiple regression analysis because we sought to adjust for the initial neurologic deficit when evaluating the prognostic value of 3-month MRS. In addition, age and sex were considered as basic elements and were always included in the multiple regression analysis. A P value of less than.05 was considered statistically significant. The analyses were performed using SAS 9.2 software. a Classification of Functional Groups In the preliminary analysis, we first plotted the Kaplan- Meier survival curves stratified by the 3-month MRS score, that is, 6 curves corresponding to MRS scores from 0 to 5 (fig 1A), and found that these 6 curves could be visually grouped into 2 clusters (upper and lower). The upper cluster consisted of the 4 curves for MRS scores 0, 1, 2, and 3, whereas the lower cluster consisted of the 2 curves for MRS scores 4 and 5. Because the main difference between these 2 clusters (ie, MRS score 3 and MRS score 4) was whether the patients could walk independently (MRS score 3) or not (MRS score 4), such clustering in the survival curves implies that independent ambulation may play a role in survival after stroke. We therefore carried out 2 sets of exploratory Cox regression analyses 1 that used the original 6 categories of MRS, and 1 that used the dichotomized MRS with the cut point between MRS score 3 and MRS score 4. Both sets of analyses showed that a lower MRS score was significantly associated with better survival. In addition, we compared the values of the Akaike information criterion (AIC), a measure of the relative goodness of fit of a statistical model, between the model that includes the original 6 categories of MRS and the model that consists of only 2 aggregated groups with 1 group for MRS score 3 and the other group for MRS score 4. Note that a lower value of AIC indicates a better fit of the model. The model with only 2 groups showed lower AIC values than the original model with full 6 categories of MRS ( vs ), suggesting that

3 INDEPENDENT AMBULATION AND STROKE SURVIVAL, Chiu 529 Table 2: Demographic and Clinical Characteristics of the Participants (N 1032) Variable Value Age (y) (40 98) Men 612 (59.3) Intracerebral hemorrhage 199 (19.3) Diabetes mellitus 355 (34.4) Heart disease 393 (38.1) Hypertension 830 (80.4) Hypercholesterolemia 359 (34.8) Previous stroke 268 (29.6) Alcohol use 136 (13.2) Smoking 321 (31.1) Initial NIHSS score (0 38) 3-mo MRS score (14.1) (33.3) (12.3) (13.5) (10.9) (16.0) NOTE. Values are expressed as no. (%) or mean SD (range). Fig 1. (A) Kaplan-Meier survival curves stratified by the 3-month MRS score. (B) Kaplan-Meier survival curves for the HF (MRS score 0, 1, 2, or 3) and LF (MRS score 4 or 5) groups. the model with only 2 groups provides better model fitting and is more parsimonious than the original model. Accordingly, we classified the subjects into 2 groups, the high function (HF) group consisting of patients with an MRS score from 0 to 3 (n 755), and the low function (LF) group consisting of patients with an MRS score of 4 or 5 (n 277). RESULTS Descriptive Findings A total of 1032 patients with stroke who survived for at least 3 months after stroke were enrolled in the study. Table 2 shows the demographic and clinical characteristics of the participants. The distribution of the 3-month MRS scores is shown in the lower part of table 2. The majority (755/1032, 73.2%) of the subjects were able to walk independently (MRS score 0, 1, 2, or 3). The mean follow-up time SD, calculated from 3 months after stroke, was months, ranging from 0.03 to 35.7 months. One hundred and sixteen deaths were seen in the cohort (116/1032, 11.2%) during 15,207.1 person-years of follow-up, yielding a crude mortality rate of 7.6 deaths per 1000 person-years. Survival Curves As mentioned above, we classified the subjects into the HF and LF groups. We plotted the survival curves stratified by these 2 groups (fig 1B). Of the 755 patients in the HF group, 37 (4.9%) died during 11,225.9 person-years of follow-up, yielding a crude mortality rate of 3.3 deaths per 1000 person-years, while of the 277 patients in the LF group, 79 (28.5%) died during person-years of follow-up, yielding a crude mortality rate of 19.8 deaths per 1000 person-years. The HF group therefore had a significantly higher survival rate than the LF group (P.001, log-rank test). Results of Cox Regression Analysis The results of univariate analysis for each variable are listed in the left part of table 3. For age and the initial NIHSS score, the estimated HR indicates the change in hazard for each 1-unit change in these variables. The covariates with a P value less than.25 were age, the 3-month MRS (LF vs HF group), the initial NIHSS score, diabetes mellitus, heart disease, hypercholesterolemia, and history of prior stroke, and these were then included in the subsequent multiple regression analysis. Age, sex, and the initial NIHSS score were considered as basic elements and were always retained in the model. The results of the multiple regression analysis are presented in the right part of table 3 and show that the significant prognostic factors with P.05 were the 3-month MRS score (LF vs HF group), age, diabetes mellitus, and heart disease. The LF group had a significantly higher mortality rate (adjusted HR 4.69; 95% confidence interval [CI], ; P.001) compared with the HF group. We also found a significantly higher risk of mortality in subjects of older age (adjusted HR 1.03; 95% CI, ; P.001), with a history of diabetes mellitus (adjusted HR 1.64; 95% CI, ; P.009), and with heart disease (adjusted HR 1.67; 95% CI, ; P.010). It should be noted that, although the initial NIHSS score and hypercholesterolemia had a significant effect in the univariate analysis, they were not significant predictors in the multiple regression model.

4 530 INDEPENDENT AMBULATION AND STROKE SURVIVAL, Chiu Table 3: Univariate and Multiple Cox Regression Analyses for Survival After Stroke Univariate Analysis Multiple Regression Variable HR (95% CI) P HR (95% CI) P Sex (men/women) 1.03 ( ) ( ).290 Age (y) 1.06 ( ) ( ) mo MRS score (LF/HF group) 6.06 ( ) ( ).001 Initial NIHSS score 1.05 ( ) ( ).430 Intracerebral hemorrhage (yes/no) 0.78 ( ).313 ND Diabetes mellitus (yes/no) 1.87 ( ) ( ).009 Heart disease (yes/no) 2.76 ( ) ( ).010 Hypercholesterolemia (yes/no) 0.63 ( ) ( ).066 Hypertension (yes/no) 0.98 ( ).943 ND Previous stroke (yes/no) 1.27 ( ).229 ND Alcohol use (yes/no) 0.94 ( ).812 ND Smoking (yes/no) 0.84 ( ).398 ND Abbreviation: ND, no data. DISCUSSION The main finding of our study was that the 3-month MRS score was an independent predictor for survival in stroke patients. Patients with a MRS score 3 (HF group) showed a significantly higher survival rate than those with a MRS score 4 (LF group). Because the major difference between the HF and LF groups was whether the patient could walk independently (MRS 3) or not (MRS 4), this indicates that independent ambulation is a key determinant for survival in patients with stroke. One possible explanation is that an inability to walk independently may be associated with a higher risk of immobilization-related complications, which then result in higher mortality. Wang et al 2 found that mobility impairment was associated with poor survival after stroke. Dighe et al 19 also found that stroke patients with severe activity limitations had a higher 2-year mortality rate. These patients are also at higher risk of a fall, which may cause fractures, such as hip fractures, and further increase the risk of immobilization-related complications, such as pneumonia, 20 pressure ulcers, 21 deep vein thrombosis, pulmonary embolism, 20 and poor cardiopulmonary function. 22,23 In addition, regular exercise has been correlated with higher long-term survival after stroke. 24 This beneficial effect of exercise may also explain why independent ambulation at 3 months after stroke is a key determinant of long-term survival. Theoretically, rehabilitation training can enhance functional capability beyond the degree of neurologic recovery in patients with stroke. Roth et al 25 found that the functional performance gained during rehabilitation could not be totally explained by the extent of neurologic recovery at follow-up. Moreover, early stroke rehabilitation has been linked to a better long-term functional outcome. 21 Because ambulation can be promoted, at least in part, through rehabilitation training, our findings highlight the potential benefit of rehabilitation for stroke survival by maximizing ambulation function. Furthermore, Slot et al 13 reported that stroke patients who are functionally dependent at 6 months after onset were at higher risk of death, and suggested that reducing dependency as well as secondary prevention of stroke should be at high priority in the management of stroke patients. This may also support a potential beneficial role of rehabilitation in improving the survival of stroke patients through reducing dependency. Initial stroke severity had been suggested as a predictor for survival. 14,15 We also found that both the initial NIHSS score and the 3-month MRS correlated with stroke survival in the univariate analysis. However, in the multiple regression analysis, there was no significant effect of the NIHSS score, while the 3-month MRS remained a significant predictor (adjusted HR 4.69; 95% CI, ). This suggests that the 3-month functional status after stroke is a more informative predictor of long-term stroke survival than the initial stroke severity. In the present study, the multiple regression analysis results showed that preexisting heart disease, diabetes, and old age were significant prognostic factors, consistent with previous studies. 9,26 Study Limitations In this study, we dichotomized the MRS scores into the LF and HF groups based on the observation of the clustered survival curves (see fig 1A). It should be noted that the choice of the cut point (MRS 3 vsmrs 4) was made on the basis of exploratory analysis of the survival curves. Because this study only recruited patients from a single medical center, it limits the generalizability of our findings. Further studies are required to evaluate whether such dichotomization of MRS scores is appropriate for predicting survival after stroke. In addition, because most of the patients are of Chinese ethnicity, it is uncertain whether our findings can be generalized to other ethnic groups. CONCLUSIONS The present study showed that the 3-month MRS is a significant predictor of long-term survival in patients with stroke after controlling for initial stroke severity and various clinical parameters. Moreover, capability of independent ambulation may be a key determinant for a favorable prognosis for stroke survival. This finding implies a potentially beneficial role of rehabilitation in promoting stroke survival by improving ambulation function. Further studies are required to determine the benefit of maximizing ambulation function with the aim of increasing survival after stroke. References 1. Warlow CP. Epidemiology of stroke. Lancet 1998;352 (Suppl 3): S Wang SL, Pan WH, Lee MC, Cheng SP, Chang MC. Predictors of survival among elders suffering strokes in Taiwan: observation from a nationally representative sample. Stroke 2000;31: Bonita R. Epidemiology of stroke. Lancet 1992;339: Huybrechts KF, Caro JJ. The Barthel index and modified Rankin scale as prognostic tools for long-term outcomes after stroke: a

5 INDEPENDENT AMBULATION AND STROKE SURVIVAL, Chiu 531 qualitative review of the literature. Curr Med Res Opin 2007;23: Hankey GJ, Jamrozik K, Broadhurst RJ, et al. Five-year survival after first-ever stroke and related prognostic factors in the Perth community stroke study. Stroke 2000;31: Pan SL, Wu SC, Lee TK, Chen TH. Reduction of disability after stroke is a more informative predictor of long-time survival than initial disability status. Disabil Rehabil 2007;29: Prencipe M, Culasso F, Rasura M, et al. Long-term prognosis after a minor stroke: 10-year mortality and major stroke recurrence rates in a hospital-based cohort. Stroke 1998;29: Mahoney FI, Barthel DW. Functional evaluation: the Barthel Index. Md State Med J 1965;14: Salgado AV, Ferro JM, Gouveia-Oliveira A. Long-term prognosis of first-ever lacunar strokes. A hospital-based study. Stroke 1996; 27: Kwon S, Hartzema AG, Duncan PW, Min-Lai S. Disability measures in stroke: relationship among the Barthel Index, the Functional Independence Measure, and the Modified Rankin Scale. Stroke 2004;35: Bonita R, Beaglehole R. Recovery of motor function after stroke. Stroke 1988;19: Lees KR, Zivin JA, Ashwood T, et al. Nxy-059 for acute ischemic stroke. N Engl J Med 2006;354: Slot KB, Berge E, Sandercock P, Lewis SC, Dorman P, Dennis M. Causes of death by level of dependency at 6 months after ischemic stroke in 3 large cohorts. Stroke 2009;40: Appelros P, Nydevik I, Viitanen M. Poor outcome after first-ever stroke: predictors for death, dependency, and recurrent stroke within the first year. Stroke 2003;34: Weimar C, Ziegler A, Konig IR, Diener HC. Predicting functional outcome and survival after acute ischemic stroke. J Neurol 2002; 249: Jørgensen HS, Nakayama H, Raaschou HO, Vive-Larsen J, Støier M, Olsen TS. Outcome and time course of recovery in stroke. Part II: Time course of recovery. The Copenhagen Stroke Study. Arch Phys Med Rehabil 1995;76: Brott T, Adams HP Jr, Olinger CP, et al. Measurements of acute cerebral infarction: a clinical examination scale. Stroke 1989;20: Krall JM, Uthoff VA, Harley JB. A step-up procedure for selecting variables associated with survival. Biometrics 1975;31: Dighe MS, Aparasu RR, Rappaport HM. Factors predicting survival, changes in activity limitations, and disability in a geriatric post-stroke population. Gerontologist 1997;37: Dennis MS, Burn JP, Sandercock PA, Bamford JM, Wade DT, Warlow CP. Long-term survival after first-ever stroke: the Oxfordshire Community Stroke Project. Stroke 1993;24: Musicco M, Emberti L, Nappi G, Caltagirone C. Early and longterm outcome of rehabilitation in stroke patients: the role of patient characteristics, time of initiation, and duration of interventions. Arch Phys Med Rehabil 2003;84: Suesada MM, Martins MA, Carvalho CR. Effect of short-term hospitalization on functional capacity in patients not restricted to bed. Am J Phys Med Rehabil 2007;86: Convertino VA. Cardiovascular consequences of bed rest: effect on maximal oxygen uptake. Med Sci Sports Exerc 1997;29: Gordon NF, Gulanick M, Costa F, et al. Physical activity and exercise recommendations for stroke survivors: an American Heart Association scientific statement from the Council on Clinical Cardiology, Subcommittee on Exercise, Cardiac Rehabilitation, and Prevention; the Council on Cardiovascular Nursing; the Council on Nutrition, Physical Activity, and Metabolism; and the Stroke Council. Stroke 2004;35: Roth EJ, Heinemann AW, Lovell LL, Harvey RL, McGuire JR, Diaz S. Impairment and disability: their relation during stroke rehabilitation. Arch Phys Med Rehabil 1998;79: Elneihoum AM, Goransson M, Falke P, Janzon L. Three-year survival and recurrence after stroke in Malmo, Sweden: an analysis of stroke registry data. Stroke 1998;29: Supplier a. SAS Institute, 100 SAS Campus Dr, Cary, NC