Brief research report 317 Stroke rehabilitation therapy in a patient with a cardiac pacemaker for chronic atrial fibrillation Hikaru Muramatsu and Kimiko Koike A 65-year-old man was implanted with an artificial pacemaker for chronic bradycardic atrial fibrillation associated with hypertensive heart disease. Five years after the pacemaker implantation, he suffered from a cerebral embolism. Approximately 4.5 months after the ictus, he was transferred to the rehabilitation ward. He had flaccid left hemiplegia and severe disuse syndrome. He could not sit and could tilt his head up for only two minutes because of severe orthostatic hypotension. By modulating the rateresponsive mode of the pacemaker every 2–4 weeks, we were able to rehabilitate the patient. Thus, the patient could sit in a wheelchair for more than three hours. This case emphasizes the importance of examining the mode and function of a previously implanted artificial pacemaker. In accord with varying rehabilitation programs and gradual Introduction Atrial fibrillation (AF) is the most common sustained arrhythmia encountered in clinical practice and accounts for 35% of arrhythmia diagnoses (Kottkamp et al., 1998). Non-valvular (NV) AF has an overall prevalence of 0.4% in adults (Kannel et al., 1982). The prevalence of NVAF shows a strong age-dependency with a substantial increase to 2–4% after the age of 60 (Kannel et al., 1983). Between 20 and 26% of patients with NVAF have congestive heart failure (CHF) associated with underlying diseases and/or loss of booster pump action, i.e., asynchronous atrioventricular (AV) sequential contraction, which contributes 10–30% of the cardiac ejection fraction at rest or during exercise (Benjamin et al., 1994, Atwood et al., 1997). improvement in a patient’s physical activities, periodic modulation of a programmable pacemaker can lead to a better functional outcome during rehabilitation therapy. International Journal of Rehabilitation Research 26:317–321 c 2003 Lippincott Williams & Wilkins. International Journal of Rehabilitation Research 2003, 26:317–321 Keywords: cerebral embolism, chronic atrial fibrillation, orthostatic hypotension, rate-responsive pacemaker, rehabilitation Department of Internal Medicine, Kasugai Rehabilitation Hospital, Yamanashi, Japan. Correspondence to: Hikaru Muramatsu, 7-3 Daiwa, Koufu, Yamanashi 400-0072, Japan. Tel/fax: + 81 55 251 4828; e-mail: m-hikaru@rainbow.plala.or.jp the progression of routine rehabilitation programs for stroke patients. In the clinical practice of stroke rehabilitation therapy, patients with an artificial pacemaker for chronic AF are not uncommon. With these patients, we are inclined to recognize the importance of pacing failure, but not necessarily the functional mechanism of the pacemaker. This case report demonstrates the valuable experience of successful stroke rehabilitation therapy in a patient with bradycardic AF and whose programmable pacemaker was regularly and appropriately modulated to resolve orthostatic intolerance/hypotension in the course of his rehabilitation therapy. Case report Chronic NVAF is not an entirely benign arrhythmia because it is associated with significant morbidity, including an increased susceptibility to cerebral embolism (Wolf et al., 1998). The annual incidence of cerebral embolism among elderly patients with chronic NVAF reaches 3–9% (6% on average) (Cerebral Embolism Task Force, 1986; Hinchey et al., 1998). Ischaemic stroke lesions associated with AF are mostly large in the total area of occluded arteries (‘territorial infarct’) because of their sudden onset and little time to develop sufficient collateral circulation (Hinchey et al., 1998, Warlow et al., 2001). Functional disability from the ictus is usually severe (Karataş et al., 2000). In addition, chronic AF restricts the endurance of exercise and (Atwood et al., 1997), therefore, can limit or even prevent 0342-5282 c 2003 Lippincott Williams & Wilkins A 65-year-old, right-handed man suffered from CHF (New York Heart Association: NYHA functional class II) (Harvey et al., 1974) due to chronic NVAF and sick sinus syndrome associated with hypertensive heart disease. Although he had no episodes of syncope attack associated with low cardiac output, i.e., Adams-Stokes syndrome, examinations by ambulance dynamic electrocardiogram (ECG) revealed frequent episodes of heart rate lower than 40 beats/min. A single chamber pacemaker (Guidant VIGOR SR, Model 1130, Indianapolis, IN, USA) in VVI-R (ventricular pacing, ventricular sensing, inhibition response, rate adaptation) mode was implanted to treat the bradycardia and CHF (Gregoratos et al., 1998). The minimum pacing rate had been set at 70 pulses per minute, and the rate-adaptive pacing function had not DOI: 10.1097/01.mrr.0000102056.48781.4c Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 318 International Journal of Rehabilitation Research 2003, Vol 26 No 4 been turned on. After implantation of the pacemaker, he functioned well as an office worker and could play a round of golf (CHF in NYHA class I). On November 15, 2000, approximately five years after the pacemaker implantation, he had a sudden stroke with left hemiplegia and unconsciousness, and was admitted to an emergency care unit. A computed tomography (CT) scan revealed hemorrhagic infarction in a large area of the brain perfused by the right middle cerebral artery. Because an angiographical examination denoted complete occlusion of the right internal cerebral artery, he was diagnosed as having a cardiogenic cerebral embolism associated with chronic AF. Although he was treated conservatively, he had serious complications of pneumonia and CHF, which caused severe disuse syndrome. Approximately 4.5 months after the ictus, he was transferred to our rehabilitation ward for intensive therapy. On admission, his chest X-ray showed CHF with a cardiothoracic ratio (CTR) of 62% (Figure 1A). A head CT scan denoted large, low-density areas with partial high-density areas in the right hemisphere of the brain (Figure 1B). An ECG revealed chronic AF, poor AV conduction, and a heart rate of 70 beats/min, which was imposed almost all of the time by the artificial pacemaker beats (Figure 1C). His activities of daily living (ADL) were totally dependent (Barthel index = 0). Clinical problems at the start of rehabilitation therapy were: (1) severe disuse syndrome after prolonged immobilization on a bed; (2) flaccid severe left hemiplegia (Brunnstrom’s stage: all I for arm, hand and leg); (3) higher cortical dysfunction; (4) totally dependent status in ADL only on a bed; and (5) complication of CHF (NYHA class IV). The patient could not sit or even tilt his head up for two minutes, because of severe orthostatic hypotension with dizziness and presyncope, and flaccid left hemiplegia. The pulse rate was 70 beats/min, and the blood pressure was 122/76 mmHg in the supine position and it fell to 92/ 54 mmHg in the sitting position with chronotropic incompetence, i.e., heart rate of 70 beats/min. At the start of intensive rehabilitation therapy, he was treated not only by continuous medications, including digoxin (0.125 mg/day), torasemide (8 mg/day) and denopamine (40 mg/day), but also by fine modulation of the programmable pacemaker. Its function was reset every 2–4 weeks, depending on the various rehabilitation programs that were initiated and the levels of physical activity that were required. The functions that were modulated were: (1) Fig. 1 Chest X-ray (A), head CT scan (B) and electrocardiogram (C) on admission. Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Stroke rehabilitation therapy in a patient with a pacemaker Muramatsu and Koike 319 Figure 2 shows typical histograms of paced or sensed (intrinsic) beats during each periodic check-up. The physiological rate-responsive pacing in response to physical activity, i.e., exposure to orthostatic stature, gradually and substantially increased during the course of hospitalisation for rehabilitation. The fraction of rateresponsive pacing to the total pacing in every interval of pacemaker check (Fig. 2B, C and D) obviously and gradually increased as compared to that by the initial check (Fig. 2A) in the rehabilitation ward. The time in which the heart was paced decreased to approximately 30%. Meanwhile, the fraction of rate-response-dependent pacing increased from zero to as much as 12% of the total number of heart beats (Figure 3). Especially with restoration of chronotropic competence by sensed (intrinsic) as well as paced heart beats, CHF and poor AV conduction were improved (CTR = 54%, NYHA class III). After months of rehabilitation and paceFig. 3 80 Pacing Lower Rate (ppm) 70 Fraction of Pacing (%) minimum (optimal) pacing lower rate; (2) rate-adaptive pacing rate in response to motion/activity, including upper sensor rate, activity threshold, activity acceleration time, response factor and recovery time. Firstly, we adopted physiological rate-responsive function to attain chronotropic competence by paced heart beats, which could successfully treat the orthostatic hypotension upon orthostatic stress. The fall in blood pressure was not significant; his blood pressure was 124/78 mmHg in the supine position and 120/80 mmHg in the sitting position for 10 minutes with chronotropic competence (heart rate of 80 beat/min). 60 70 65 60 50 40 30 Pacing Mode: WI to WI-R 20 "Rate-response"-Dependent Pacing 10 0 1 2 3 2001 4 5 6 7 8 9 10 11 12 1 2 2002 Date (month) Fraction of pacing during the course of rehabilitation. After the ‘rateresponse’ mode of pacemaker function was enabled (VVI to VVI-R) and regulated, the total fraction of paced heart beats (empty circles) by the implanted pacemaker gradually decreased. Meanwhile, the fraction of rate-responsive paced heart beats (filled circles) substantially increased. Filled squares denote the programmed pacing lower rates, i.e., initially 70 pulses/min, next 65 and finally 60. Fig. 2 (A) (B) 100 80 Fraction (%) 100 Jun 22, 2001 Paced 60 Sensed Paced Paced 60 Sensed 40 40 20 20 0 0 −70 −65 71−85 86−100 101−130 131− 66−75 76−95 96−115 116− (D) (C) 100 Fraction (%) Sept. 26, 2001 80 100 Dec 3, 2001 80 Paced 80 60 Paced 60 Jan 22, 2002 Paced Paced Sensed 40 40 20 20 Sensed 0 0 −60 61−70 71−90 91−110 111− Heart Rate (beats/min) −60 61−70 71−90 91−110 111− Heart Rate (beats/min) Histograms of paced and sensed (intrinsic) QRS complexes in each periodic check of pacemaker function in VVI-R mode. The pacemaker had a readable memory. Fractions of paced rhythm (meshed or shaded columns) or sensed rhythm (blank columns) are shown in each range of heart rate. The paced rhythms (shaded columns) above the programmed pacing lower rate, i.e., > 70 pulses/min in A, > 65 in B, > 60 in C and D, indicate ‘rateresponsive’ pacing to increased physical motion/activity. Chi-square analysis revealed that the fraction of rate-responsive pacing to the total pacing in every interval (B, C, D) significantly (p < 0.001) increased as compared to that in the initial check (A). Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 320 International Journal of Rehabilitation Research 2003, Vol 26 No 4 maker modulation, the patient could sit in a wheelchair for more than three hours without significant orthostatic hypotension, and his ADL had improved from bed rest to the wheelchair level with moderate assistance. Discussion In developed countries where rheumatic fever is rare, the prevalence of NVAF increases as the patient population continues to age: it is approximately 1% of the general population and doubles with each advancing decade of age, from 0.5% at 50–59 years of age to almost 9% at 80– 89 years (Feinberg et al., 1995, Kannel et al., 1998). In addition, sinus node and AV node functions are reduced with age, accompanying atrophy of the nodal cells and fatty infiltration (Chida et al., 1999). This phenomenon is related to the increased prevalence of sick sinus syndrome, AV block and chronic AF with aging (Sutton et al., 1986). Therefore, the number of elderly patients with chronic bradycardic AF, in which a permanent pacemaker is implanted, is reasonably large. A VVI(-R) mode pacemaker is generally employed for the treatment of bradyarrhythmias, especially permanent bradycardic AF (Gregoratos et al., 1998). Because its implantation is technically simple, with a single wire and established procedures, it is the most popular pacemaker in clinical treatments at present. However, because the VVI mode fails to produce sequential and synchronous contractions of the atria and ventricles, this mode has disadvantages when one tries to improve cardiac stroke volume and exercise endurance in patients with low cardiac reserve (Rosenqvist et al., 1991). Physiological rare-adaptive (-R) pacemaker function is designated to provide a change in the pacing rate proportional to the level of metabolic demand (‘chronotropic competence’), and therefore to increase the cardiac output in response to increased physical activity. The most widely used are vibration- or motion-sensing devices, either a piezoelectric crystal activity sensor fixed to the inside of the pacemaker case or an accelerometer. The crystal detects vibration generated by body motion, and the accelerometer senses motion directly. Other rate-responsive pacemakers use the right ventricular blood temperature, minute ventilation, the QT interval in ECG, mixed venous oxygen saturation, and other physiological parameters as variables to modulate the pacing rate (Schweikert et al., 1999). The rateresponse to body motion could be most effective to the orthostatic intolerance associated with bradycardic chronic AF and the disuse syndrome in the present patient, especially at the beginning of the rehabilitation therapy. In other words, if we had not adopted the rateresponsive pacing function, we could not start to rehabilitate the patient and to develop the program of the rehabilitation therapy. Chronic AF and CHF patients implanted with VVI mode pacemakers belong to a group at high-risk for cerebral embolism (Andersen et al., 1994). Once the ictus occurs, these patients exhibit critically severe hemiplegia and/or higher cortical dysfunction, disability, and a serious disuse syndrome associated with the prolonged initial therapy in acute care units. Serious orthostatic intolerance/hypotension is incidental to those patients with flaccid hemiplegia and marked disuse syndrome. In the present case, the orthostatic intolerance could have been due to: (1) severe disuse syndrome and the plegic lower extremity, as well as deteriorated autonomic regulation of the peripheral circulation and increased venous compliance; (2) the absence of a physiological Bainbridge reflex and chronotropic incompetence associated with the underlying sick sinus syndrome; (3) poor AV node conduction; and (4) almost all ( > 70%) of the heart beats dependent on the VVI pacing without the booster pump effect. Orthostatic intolerance after prolonged bed rest can be attributed, at least in part, to the depletion of intravascular volume, which might have been compounded by an increase in venous pooling in the lower extremities because of increased venous compliance after bed rest (Downey et al., 2001). The veins of the lower extremities are part of the post-capillary blood pool, which makes up approximately 70% of the overall intravascular volume, and approximately 80% of the blood pooled in the lower limbs is contained in the upper legs (Smit et al., 1999). One half to one litre of thoracic blood is transferred to the lower body upon orthostatic stress. The bulk of venous pooling occurs within the first 10 seconds, and the total transfer is almost complete within 3–5 minutes of the orthostatic stress (Smit et al., 1999). This time course and the hemodynamic changes are well correlated with the appearance of serious orthostatic hypotension in the present case that had severe disuse syndrome and the severe plegic left lower extremity. In summary, modulation of a VVI-R programmable pacemaker, especially its rate-responsive pacing that enabled chronotropic competence, was an effective countermeasure to serious orthostatic intolerance in a patient with bradycardic chronic NVAF. We argue that manipulations of the physiological rate-responsive function of the pacemaker were, indeed, primarily responsible for the initial start and subsequent advancement of the rehabilitation programs. Periodic adjustments of the pacemaker, especially increased rate-responsive function to physical activities, significantly improved his physical endurance, ADL and quality of life. Conclusions This case reminds us of the necessity to examine the mode, type, function and characteristics of previously implanted cardiac pacemakers in stroke patients with Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Stroke rehabilitation therapy in a patient with a pacemaker Muramatsu and Koike 321 arrhythmias, and confirms the merits of regularly modulating pacemaker function during routine stroke rehabilitation therapy to obtain a better functional outcome. In bradycardic chronic AF and stroke patients who have been implanted with an artificial pacemaker, the VVI mode is most common but might not improve cardiac output and physical activity/tolerance because of the AV asynchrony. Thus, an additional rate-adaptive pacing mode, i.e., VVI-R, is essential for effective stroke rehabilitation therapy by chronotropic competence to increase cardiac output and to treat orthostatic hypotension. Acknowledgement The authors gratefully acknowledge Dr. Richard D. Nathan, Texas Tech University Health Sciences Center, for editing the manuscript. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. References Andersen HR, Thuesen L, et al. (1994). Prospective randomised trial of atrial versus ventricular pacing in sick-sinus syndrome. Lancet 344:1523–1528. Atwood JE, Myers J (1997). Exercise hemodynamics of atrial fibrillation. In: Falk RH, Podrid PJ (Eds). Atrial Fibrillation: Mechanisms and Management. 2nd Edition. New York: Lippincott Raven Press; pp. 219–239. Benjamin EJ, Levy D, et al. (1994). Independent risk factors for atrial fibrillation in a population-based cohort: The Framingham Heart Study. Journal of American Medical Association 271:840–844. Cerebral Embolism Task Force. (1986). Cardiogenic brain embolism. Archive of Neurology 43:71–84. Chida K, Imai T, et al. (1999). Implications of marked fatty infiltration around and in the atrophic atrioventricular node in elderly patients with permanent pacemaker implantation for symptomatic sick sinus syndrome. Japanese Journal of Circulation 63:343–349. Downey RJ, Weissman C (2001). Physiological changes associated with bed rest and major body Injury. In: Gonzalez EG, Myers SJ, Edelstein JE, Lieberman JS, Downey JA (Eds). Downey & Darling’s Physiological Basis of Rehabilitation Medicine. 3rd Edition. Boston: Butterworth-Heinemann; pp. 449–484. Feinberg WM, Blackshear JL, et al. (1995). Prevalence, age, distribution and gender of patients with atrial fibrillation. Archives of Internal Medicine 155:469–473. Gregoratos G, Cheitlin MD, et al. (1998). ACC/AHA Guidelines for implantation of cardiac pacemakers and anti-arrhythmia devices. Journal of American College of Cardiology 31:1175–1209. Harvey RM, Doyle EF, et al. (1974). Major changes made by Criteria Committee of the New York Heart Association. Circulation 49:390. Hinchey JA, Furlan AJ, et al. (1998). Cardiogenic brain embolism: incidence, varieties, and treatment. In: Barnett HJM, Mohr JP, Stein BM, Yatsu FM (Eds). Stroke: Pathophysiology, Diagnosis, and Management. 3rd Edition. Pennsylvania: Churchill Livingstone; pp. 1089–1119. Kannel WB, Abbott RD, et al. (1982). Epidemiologic features of chronic atrial fibrillation. New England Journal of Medicine 306:1018–1022. Kannel WB, Abbott RD, et al. (1983). Coronary heart disease and atrial fibrillation: Farmingham Study. American Heart Journal 106:389–396. Kannel WB, Wolf PA, et al. (1998). Prevalence incidence prognosis and predisposing conditions for atrial fibrillation: population-based estimates. American Journal of Cardiology 82:2N–9N. Karataş M, Dilek A, et al. (2000). Functional outcome in stroke patients with atrial fibrillation. Archives Physical Medicine and Rehabilitation 81: 1025–1029. Kottkamp H, Hindricks G, et al. (1998). Atrial fibrillation: epidemiology, etiology, and symptoms. In: Saoudi N, Schoes W, El-Sherif N (Eds). Atrial Flutter and Fibrillation: From Basic to Clinical Application. New York: Futura Publishing Company Inc. pp. 135–151. Rosenqvist M, Isaaz K, et al. (1991). Relative importance of activation sequence compared to atrioventricular synchrony in left ventricular function. American Journal of Cardiology 67:148–156. Schweikert RA, Pashkow FJ, et al. (1999). Rehabilitation of patients with arrhythmias, pacemakers, and defibrillators. In: Pashkow FJ, Dafoe WA (Eds). Clinical Cardiac Rehabilitation: A Cardiologist’s Guide. 2nd Edition. Baltimore: Williams & Wilkins; pp. 192–203. Smit AA, Halliwill JR, et al. (1999). Pathophysiological basis of orthostatic hypotension in autonomic failure. Journal of Physiology (London) 519:1–10. Sutton R, Kenny RA (1986). The natural history of sick sinus syndrome. Pacing and Clinical Electrophysiology 9:1110–1114. Warlow CP, Dennis MS, et al. (2001). What caused this transient or persisting ischaemic event? In: Warlow CP, Dennis MS, van Gijn J, Hankey GJ, Sandercock PAG, Bamford JM, Wardlaw JM (Eds). Stroke: A Practical Guide to Management. 2nd Edition. Oxford: Blackwell Science Ltd; pp. 223–300. Wolf PA, Mitchell JB et al. (1998). Impact of atrial fibrillation on mortality, stroke, and medical costs. Archives of Internal Medicine 158:229–234. Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.