Functional Outcomes and Long-term Survival in Patients After Stroke Rehabilitation

背景與目標：腦中風復健患者之功能預後對其長期存活之影響屆今仍不清楚。本研究之目的是想探討影響腦中風復健患者功能預後之預測因子，並藉由存活分析探討腦中風患者之功能預後是否為其長期存活之獨立預測因子。本研究乃針對腦中風患者復健後之「步行能力」及「管灌飲食依賴性」等兩項功能預後進行探討。 研究對象與方法：本研究的第一部份以西元1990-1995年間，曾於臺大醫院復健部接受住院復健的590位45歲以上腦中風患者為研究世代，依據病歷記載將患者出院時之步行能力分成「適當」及「不良」兩組，並根據追蹤至西元2000年底患者之存活狀態，利用考克斯迴歸模式進行存活分析，以探討影響腦中風住院復健患者出院後長期存活之預測因子。第二部分以西元1994-1999年間，曾於臺大醫院影像醫學部接受電視螢光錄影吞嚥檢查的182位腦中風合併吞嚥障礙患者為研究世代，記錄其亞急性期之電視螢光錄影吞嚥檢查結果及管灌飲食依賴性，並根據追蹤至西元2000年底患者之存活狀態，利用考克斯迴歸模式進行存活分析，以探討影響腦中風合併吞嚥障礙患者之長期存活預測因子。第三部分以西元1994-1999年間，曾於臺大醫院影像醫學部接受電視螢光錄影吞嚥檢查的181位需使用管灌飲食之腦中風住院患者為研究對象，利用羅吉斯迴歸預測模式探討患者出院管灌飲食依賴性之相關因子。 結果：第一部份：在出院後追蹤71.1個月（中位數）的情況下，出院時仍為不良步行能力的腦中風復健患者，相較於適當步行能力的患者，其追蹤期間之死亡危險比（hazard ratio）為2.01（95%信賴區間：1.55-2.61）。僅於65歲以上老年腦中風患者中，方可發現患者年齡與之其出院時步行能力及長期存活具有統計學上顯著意義的相關性（P < 0.05）。第二部份：在接受電視螢光錄影吞嚥檢查後追蹤30.8個月（中位數）的情況下，年齡（每增加10歲之死亡危險比為1.46，95%信賴區間：1.13-1.89）、多次腦中風（死亡危險比為1.74，95%信賴區間：1.06-2.85）、出院時仍依賴管灌飲食（死亡危險比為2.07，95%信賴區間：1.19-3.59）、以及出院時僅能以輪椅活動無法行走（死亡危險比為2.83，95%信賴區間：1.54-5.19）等為長期存活之獨立預測因子。第三部份：年齡（每增加1歲之勝算比(Odds ratio)為1.07，95%信賴區間：1.02-1.12）、多次腦中風（勝算比為2.85，95%信賴區間：1.15-7.07）、出院時僅能以輪椅活動無法行走（勝算比為8.54，95%信賴區間：3.29-22.15）、以及電視螢光錄影吞嚥檢查結果（吸入現象、谿域或梨狀窩滯留現象）等為患者出院時仍依賴管灌飲食之獨立相關因子。依據羅吉斯迴歸模式之結果，我們提出一個初步的預測公式，以預測腦中風合併吞嚥困難患者其出院時之管灌飲食依賴性。 結論：腦中風住院復健患者出院時之步行能力為其長期存活之獨立預測因子，出院時仍依賴管灌飲食為腦中風合併吞嚥困難患者其長期存活之獨立預測因子，電視螢光錄影吞嚥檢查中之吸入現象則與腦中風合併吞嚥困難患者之長期存活無關。然而電視螢光錄影吞嚥檢查中之吸入現象及滯積現象，則為腦中風合併吞嚥困難患者出院時仍依賴管灌飲食之獨立相關因子。本研究證實腦中風患者之復健功能預後（包括步行能力與管灌飲食依賴性）為其長期存活之獨立預測因子，腦中風復健不僅能增進腦中風患者之生活品質，也能改善其長期存活。Background and Objectives: The effect of functional outcomes after rehabilitation on the long-term survival of patients with stroke is unclear. The aims of the study are to explore the predictors of functional outcomes after rehabilitation, and to determine the effect of functional outcomes after rehabilitation on long-term survival in patients with stroke. The functional outcomes of interest in the study are the patient’s walking ability and tube feeding dependency due to stroke-related dysphagia after rehabilitation. Materials and Methods: Part I: A cohort of 590 patients, aged 45 years and older, admitted for inpatient stroke rehabilitation from 1990 to 1995 was constructed. The walking ability of patients after rehabilitation was classified as either adequate or poor based on chart records. Cox proportional hazards regression analysis was used to identify independent predictors of survival until the end of 2000. Part II: A cohort of 182 patients with stroke-related dysphagia referred for videofluoroscopic swallowing study (VSS) from 1994 to 1999 was constructed. VSS findings and clinical features in the post-acute phase of stroke were recorded. Cox proportional hazards regression analysis was used to identify independent predictors of survival until the end of 2000. Part III: A total of 181 tube-fed inpatients referred for VSS from 1994 to 1999 due to stroke-related dysphagia were enrolled. Multivariate logistic regression analysis was used to identify independent predictors of tube feeding dependency at discharge. Results: Part I: In the median follow-up time of 71.1 months after discharge, poor walking ability was independently associated with long-term survival, with a hazard ratio (HR) of 2.01 (95% confidence intervals (CI) = 1.55-2.61). The associations of increasing age with poor long-term survival and poor walking ability after stroke rehabilitation were observed in the older group (age ≥ 65), but not in the middle-aged group (age 45-64). Part II: In the median follow-up time of 30.8 months after VSS, advanced age (HR = 1.46, 95% CI = 1.13-1.89, every 10 years of increase), recurrent stroke (HR = 1.74, 95% CI = 1.06-2.85), the need of tube-feeding (HR = 2.07, 95% CI = 1.19-3.59), and being wheelchair-confined (HR = 2.83, 95% CI = 1.54-5.19) during follow-up were independent predictors of long-term survival. Part III: Advanced age (Odds ratio (OR) = 1.07, 95% CI = 1.02-1.12, every 1 year of increase), recurrent stroke (OR = 2.85, 95% CI = 1.15-7.07), wheelchair-confined at discharge (OR = 8.54, 95% CI = 3.29-22.15), long duration from stroke onset to VSS (OR = 1.02, 95% CI = 1.01-1.04, every 1 day of increase), and findings of VSS (stasis in valleculae or pyriform sinus, and aspiration) were independently associated with tube feeding dependency at discharge. Based on the final regression model, a preliminary equation was proposed to help clinicians better predict the tube feeding dependency at discharge in stroke patients with dysphagia. Conclusions: The patient’s walking ability after inpatient rehabilitation independently predicts long-term survival in patients with stroke. Tube feeding dependency independently predicts long-term survival in stroke patients with dysphagia. VSS-detected aspiration during the post-acute phase of stroke was not predictive for the long-term survival in stroke patients with dysphagia. However, findings of VSS are associated with tube feeding dependency at discharge in stroke patients with dysphagia. The functional outcomes after rehabilitation, such as walking ability and tube feeding dependency, predict long-term survival of patients with stroke. Stroke rehabilitation not only adds quality of life but also improves long-term survival of patients with stroke.圖目錄 （List of Figures） XI 表目錄 （List of Tables） XII 縮寫目錄 （List of Abbreviations） XIV Chapter 1 Introduction 1 1.1 DEFINITION OF STROKE 1 1.2 METHODOLOGICAL ASPECTS OF EPIDEMIOLOGICAL STUDIES OF STROKE 1 1.3 REVIEW OF LONG-TERM SURVIVAL STUDIES OF STROKE 3 1.3.1 Definitions of survival and mortality 3 1.3.2 Predictors of long-term survival in patients with stroke 3 1.3.2.1 Age 3 1.3.2.2 Gender 4 1.3.2.3 Heart diseases 4 1.3.2.4 Hypertension 5 1.3.2.5 Diabetes mellitus 5 1.3.2.6 Stroke types and subtypes 5 1.3.2.7 Laboratory findings 6 1.3.2.8 Status of physical function 6 1.3.2.9 Status of mental function and social function 8 1.3.2.10 Miscellaneous 9 1.4 SURVIVAL STUDIES OF STROKE IN TAIWAN 9 1.4.1 Short-term survival studies of stroke in Taiwan 9 1.4.5 Long-term survival studies of stroke in Taiwan 10 1.5 PREDICTORS OF FUNCTIONAL OUTCOMES AFTER STROKE REHABILITATION 12 1.6 FUNCTIONAL OUTCOMES AFTER REHABILITATION AND LONG-TERM SURVIVAL OF PATIENTS WITH STROKE 13 Chapter 2 Aim and Hypothesis of the Study 15 2.1 AIM OF THE STUDY 15 2.2 HYPOTHESIS OF THE STUDY 16 Chapter 3 Subjects and Methods 17 3.1 WALKING ABILITY AFTER INPATIENT REHABILITATION: ITS DETERMINANTS AND EFFECT ON LONG-TERM SURVIVAL OF PATIENTS WITH STROKE 17 3.1.1 Study population 17 3.1.2 Assessment of clinical features and risk factors 18 3.1.3 Assessment of end point 19 3.1.4 Statistical analyses 19 3.2 THE EFFECTS OF ASPIRATION DETECTED BY VIDEOFLUOROSCOPIC SWALLOWING STUDIES AND TUBE FEEDING DEPENDENCY ON LONG-TERM SURVIVAL OF STROKE PATIENTS WITH DYSPHAGIA 22 3.2.1 Study population 22 3.2.2 Assessment of clinical features and risk factors 23 3.2.3 Assessment of dysphagia 24 3.2.4 Assessment of end point 25 3.2.5 Statistical analyses 25 3.3 FACTORS ASSOCIATED WITH TUBE FEEDING DEPENDENCY AT DISCHARGE IN STROKE PATIENTS WITH DYSPHAGIA 27 3.3.1 Study population 27 3.3.2 Assessment of clinical features and risk factors 28 3.3.3 Assessment of dysphagia 28 3.3.4 Assessment of outcome 30 3.3.5 Statistical analyses 30 Chapter 4 Results 32 4.1 WALKING ABILITY AFTER INPATIENT REHABILITATION: ITS DETERMINANTS AND EFFECT ON LONG-TERM SURVIVAL OF PATIENTS WITH STROKE 32 4.1.1 Demographic and clinical characteristics 32 4.1.2 Walking ability after inpatient stroke rehabilitation and long-term survival 32 4.1.3 Factors associated with walking ability after inpatient stroke rehabilitation 34 4.2 THE EFFECTS OF ASPIRATION DETECTED BY VIDEOFLUOROSCOPIC SWALLOWING STUDIES AND TUBE FEEDING DEPENDENCY ON LONG-TERM SURVIVAL OF STROKE PATIENTS WITH DYSPHAGIA 35 4.2.1 Demographic and clinical characteristics 35 4.2.2 Factors associated with long-term survival in stroke patients with dysphagia 36 4.3 FACTORS ASSOCIATED WITH TUBE FEEDING DEPENDENCY AT DISCHARGE IN STROKE PATIENTS WITH DYSPHAGIA 37 4.3.1 Demographic and clinical characteristics 37 4.3.2 Predictive equation for tube feeding dependency at discharge 38 Chapter 5 Discussion 40 5.1 WALKING ABILITY AFTER INPATIENT REHABILITATION: ITS DETERMINANTS AND EFFECT ON LONG-TERM SURVIVAL OF PATIENTS WITH STROKE 40 5.1.1 Main findings of this work 40 5.1.2 Argument for the findings of this work 40 5.1.3 Study limitation of this work 42 5.1.4 Conclusions and future work 43 5.2 THE EFFECTS OF ASPIRATION DETECTED BY VIDEOFLUOROSCOPIC SWALLOWING STUDIES AND TUBE FEEDING DEPENDENCY ON LONG-TERM SURVIVAL OF STROKE PATIENTS WITH DYSPHAGIA 44 5.2.1 Main findings of this work 44 5.2.2 Argument for the findings of this work 44 5.2.3 Study limitation of this work 46 5.2.4 Conclusions and future work 47 5.3 FACTORS ASSOCIATED WITH TUBE FEEDING DEPENDENCY AT DISCHARGE IN STROKE PATIENTS WITH DYSPHAGIA 48 5.3.1 Main findings of this work 48 5.3.2 Argument for the findings of this work 48 5.3.3 Study limitation of this work 51 5.3.4 Conclusions and future work 52 5.4 PERSPECTIVE OF THE STUDY 53 Tables 54 Figures 70 Appendix 74 APPENDIX A: SUMMARY OF STUDIES OF LONG-TERM SURVIVAL IN PATIENTS WITH STROKE 74 APPENDIX B: DEFINITIONS OF SELECTED RISK FACTORS OF STROKE 86 APPENDIX C: BRUNNSTROM STAGES OF MOTOR IMPAIRMENT 89 APPENDIX D: DEFINITIONS OF ABNORMAL FINDINGS IN VIDEOFLUOROSCOPIC SWALLOWING STUDY 90 References 91 Publications 105 Journal papers related to dissertation 11