Inspiratory Muscle Training in Patients With Chronic Heart Failure Awaiting Cardiac Transplantation: Results of a Pilot Clinical Trial Background and Purpose. Persons with chronic heart failure (HF) have
poor ventilatory muscle strength, and this weakness is associated with dyspnea. The purpose of this study was to examine the effects of inspiratory muscle training (IMT) on ventilatory muscle strength and dyspnea in patients with chronic HF. Subjects. Fourteen patients (mean age [2SD] =52 +8.5 years) with end-stage cardiomyopathy and chronic HF (mean left ventricular ejection fraction = 23% 13% and New York Heart Association class=3.6+0.6) participated in the study. Methods. Inspiratory muscle training was performed at 20% of maximal inspiratory pressure (MIP) for 5 to 15 minutes, three times a day, for 8 weeks. Dyspnea was evaluated at rest and during exercise. Results. Both MIP and maximal expiratory pressure (MEP) were greater after 2 weeks of IMT (51221 to 63+23 cm H 2 0 and 85?22 to 96519 cm H 2 0 , representing 24% and 13% improvement). Dyspnea scores at rest and during exercise decreased after 2 weeks (2.020.7 to 1.320.5 and 3.650.5 to 2.6+0.6, representing 29% and 28% improvement) and plateaued throughout the remainder of IMT. Baseline MEP was related to the percentage of change in MEP after IMT ( r = - .72), and several measures of pulmonary function were related to the degree of improvement in dyspnea after IMT ( r = - .57 to -.82) and in MIP after IMT ( r = .71). Conclusion and Discussion. Improvements in MIP, MEP, and dyspnea were found after 2 weeks of IMT. Greater pulmonary function was associated with greater improvement in dyspnea and ventilatory muscle strength after IMT. These improvements may decrease the dependency and impairment associated with chronic HF. [Cahalin LP, Semigran MJ, Dec GW. Inspiratory muscle training in patients with chronic heart failure awaiting cardiac transplantation: results of a pilot clinical trial. Phys Ther. 1997;77:830-838.1
+
Key Words: Breathing exercise, Cardiac transplantation, Exercise, Heart failure. Lawrence P Cahalin
Marc J Sernigran G W'illiam Dec 830
Physical Therapy
. Volume 77 . Number 8 . August 1997
nspiratory muscle training (IMT) benefits patients witti pulmonary disease by increasing their ventilatory muscle strength and endurance and by decreasing their dyspnea, need for medications, emergency department visits, and number of hospitalizations. 1-7 Individuals with chronic heart failure (HF) have been found to have poor ventilatory muscle Poor ventilatory muscle strength and enduran~e.~-lO strength has also been found to be correlated to the Only one occurrence of dyspnea during daily activi~y.~JO recent study1] has evaluated IMT in patients with chronic HF. Eight patients with a mean (2SD) New York Heart Association (NYHA) classification of 2.8%1.0 performed an elaborate ventilatory muscle training program (via isocapnic hyperpnea, pressure-gauge strength training, breathing calisthenics, abdominal strengthening exercises, and a Threshold inspiratory muscle trainer") three to seven times a week for 12 weeks and improved their ventilatory muscle strength and endurance as well as their submaximal and maximal exercise capacities.ll
Hcalthscan Products Inc, 908 Pompton Ave, Cedar Grove, NJ 07009.
Improvement in ventilatory muscle strength and symptoms may decrease the medical and functional dependence, impairment, and possibly even costs associated with chronic HF.8-12 Increased ventilatory muscle strength may also enhance early postoperative recovery in patients undergoing cardiac transplantation or thoBecause J* of these potential effects racic s ~ r g e r y . ~ , l ~ from IMT and the need for rapid improvement in ventilatory muscle strength prior to cardiac transplantation or thoracic surgery, the effects of a short-term standard IMT program (via targeted inspiratory muscle training alone) on ventilatory muscle strength and dyspnea in patients with chronic HF awaiting cardiac transplantation were investigated.
Method Subjects The study subjects were 14 patients (12 men and 2 women) with chronic HF (HF of rl year) for an average of 4 years who were randomly selected from patients admitted to Massachusetts General Hospital for precardiac transplantation evaluation. The subjects had a mean age of 52 years (SD=8.5, range=32-64). The
LP Cahalin, PT, CCS, is Physical Therapist, Heart Failure and Transplantation Service, Massachusetts General Hospital, and Clinical Associate Professor, Sargent College of Allied Health Professions, Boston University, Boston, Mass. Address all correspondence to Mr Cahalin at Physical Therapy Department, Sargent College of Allied Health Professions, Boston University, 635 Commonwealth Ave, Boston, MA 02215 (USA) (
[email protected]).
MJ Semigran, MD, is Physician, Heart Failure and Transplantation Service, Massachusetts General Hospital, and Assistant Professor of Medicine, Harvard Medical School, Cambridge, Mass. GW Dec, MD, is Medical Director, Heart Failure and Transplantation Senrice, and Acting Chief Cardiac Unit, Massachusetts General Hospital, and Associate Professor of Medicine, Hamard Medical School. This study was approved by the Human Subjects Review Committee of Massachusetts General Hospital This research was supported in part by a grant from the Ionta Fund and the Foundation for Physical Therapy Inc.
This article was submitted July 9, 1996, and was accepted March 10, 1997. Physical Therapy . Volume 77 . Number 8 . August 1997
Cahalin et a1 . 83 1
Table 1. Patient Characteristicsa
1 Patient No. 1
2
I
4 5 6
8
7 8 9 10 11 12 13 14 X SD
Disease
Gender
Age (y)
Height (cm)
Weight (kg)
NYHA (1-4)
LVEF (%I
Duration of CHF (y]
IDCM lsch CM IDCM lsch CM RCM lsch CM lsch CM IschCM IDCM IschCM IDCM IDCM IDCM IDCM
M M M M M M M M M M M M F F
32 48 64 51 49 62 52 51 40 54 57 61 56 49
173 178 180 185 175 168 175 175 175 185 180 175 160 160
84.0 71 .O 98.6 106.8 83.2 81.8 76.4 81.8 68.2 85.4 84.5 78.2 63.6 52.3
4 3 3 4 4 3 2 4 4 3 4 4 4 4
20 12 54 18 48 13 18 25 10 20 26 17 23 12
1.O 1 .O 4.0 1 .O 10.0 4.0 2.0 2 .O 3.0 3 .O 5.0 10.0 9.0 1.O
5258.5
17528
3.6?.6
23?13
80%14
453
'NTMA=New York Heart Association classificdtion, LVEF=left ventricular ejection Fraction, CHF=congestive heart failure, IDCM=idiopdthic dilated rardiomvopathy, Isch CM=ischemic cardiomvopathy, RCM=restrictive cardiomvopathy.
mean left ventricular ejection fraction for these subjects was 23% 213% (normal=60%-80%), and the mean NYHA classification was 3.620.6 (patients in class 1 have no limitation in ordinary physical activity, whereas patients in class 4 are unable to perform physical activity without discomfort). Twelve of the subjects were hospitalized during the IMT period while awaiting or undergoing evaluation for cardiac transplantation.
subjects who underwent optimization of medical therapy using invasive hemodynamic monitoring included a goal cardiac index of greater than 2.2 ~ / m i n / m ' and a pulmonary capillary wedge pressure of less than 18 mm Hg, which we have accepted as signs of optimal cardiac performance in this patient population.16
The etiology of chronic HF included idiopathic dilated cardiomyopathy (n = 7), ischemic cardiomyopathy (n =6), and restrictive cardiomyopathy (n= 1). All subjects underwent pulmonary function testing, and none had severe pulmonary disease (as defined by both a forced expiratory volume in 1 second [FEV,] and a forced vital capacity [FVC] of less than 50% of the predicted value, as suggested by the American Thoracic Society15),which may limit the effectiveness of IMT. The mean FEV, was 2.52?0.78 L, and the mean FVC was 3.2321.05 L (67%216% of the predicted values). Informed consent was obtained prior to participation in the study. The patient characteristics are listed in Table 1.
Ventilatory muscle strength. Maximal inspiratory pressure (MIP) and maximal expiratory pressure (MEP) were measured (in centimeters of water) with a device consisting of a Magnehelic gauget and one-way valve that was constructed similarly to that described by Black and Hyatt.17 The device has an accuracy of 20.5% and a resolution of 5 cm H,O. A 0.5cm hole was made at the base of the one-way valve to prevent the generation of mouth pressure during testing of ventilatory muscle strength.
Medical therapy was optimized prior to IMT and included the use of digoxin, diuretics, and angiotensinconverting enzyme inhibitor agents by all subjects; intravenous dobutamine was administered to eight subjects. Ventilatory muscle testing and training were initiated after a plateau in clinical status had been achieved (identified by signs and symptoms of stable compensated HF and lack of improvement in, or ability to perform, further exercise conditioning). Criteria for the initiation of ventilatory muscle testing and training in eight of the 832 . Cahalin et al
Measurements
All measurements were made in the manner described by Black and Hyatt,17 with subjects in a seated position and the trunk at a 90-degree angle to the hips. Measurements of MIP were obtained after a maximal expiration (near residual volume), and measurements of MEP were obtained after a maximal inspiration (total lung capacity). A noseclip was worn during ventilatory muscle testing. Measurements of MIP and MEP were repeated until consistent measurements were obtained. These measure-
+
D y e r Instruments Inc. Indiana Hwy 212 and US 12. Michigan City, IN 46360.
Physical Therapy . Volume 77 . Number 8 . August 1997
ments resulted in three to six MIP and MEP determinations, separated by several minutes of rest between measurements. The highest attained MIP and MEP were used and were compared with normal age- and genderspecific MIP and MEP values that were predicted using regression equations developed by Black and Hyatt.17 The MIP was measured first, followed by the MEP. Measurements of MIP and MEP were repeated each week so that the IMT prescription could be updated. Data for test-retest reliability assessment were collected twice on each of two separate consecutive days during the initial test period in the manner described, with a +hour rest period between daily trials. The MIP and MEP data obtained on the first day of testing were compared with the MIP and MEP data obtained on the second day of testing. Intraclass correlation coefficients (ICC[l,:L]) were calculated to evaluate the reliability of the MIP and MEP measurements. Test-retest reliability of the MIP and MEP measurements was essentially identical within and between sessions (ICC=.98). An F test was used to test the null hypothesis that ICC=O. The level of significance was PC.05. Inspiratory muscle training. Targeted IMT was performed with a Threshold inspiratory muscle trainer at 20% c)f MIP for 5 to 15 minutes, three times every day, for 8 weeks. Inspiratory muscle training was initiated only after a plateau in medical and exercise therapy (upright cycle ergometry for 10-20 minutes daily at an rating of perceived exertion of 2-3/10) had been achieved. All subjects were observed during the initial IMT session for adverse signs or symptoms and to verify proper IMT technique. The initial IMT session duration was limited to 5 minutes, three times per day. When subjects were free of overt anxiety, dyspnea, fatigue, and respiratory muscle discomfort, the IMT duration was progressed by 2 to 5 minutes until 15 minutes of IMT was achieved. Subjects were instructed to record all IMT sessiorls on a log sheet. Twice-weekly evaluations of the IMT technique and IMT log sheet were performed. Inspiratory muscle training prescriptions were progressed based on weekly measurements of MIP to maintain IMT at 20% of MIP. To avoid impaired inspiration due to a full stomach, subjects were instructed to perform IMT prior to meals or 1 hour after eating and to wear a noseclip during IMT sessions. All other exercise and functional/recreational activities were kept constant during the study period. Adherence with IMT was calculated as a ratio of the number of documented and obsened IMT sessions to that prescribed. Symptoms. Dyspnea was evaluated using a modified Borg scale of 0 to 10.lHSubjects documented the degree of dyspnea at rest and during submaximal exercise (cycle ergometry at 0-10 W for a mean of 15 minutes).
Data Analysis
Statistical analysis was performed with the SkSTAT statistical packagex and included the calculation of means, standard deviations, and Pearson product-moment correlations; linear regression analysis; repeated-measures analysis of variance (ANOVA), and Friedman's test with post hoc analysis using Dunnett's procedure and all comparisons with a control procedure. The repeatedmeasures ANOVA and Dunnett's procedure were performed on the baseline and week 2 , 4 , 6 , and 8 MIP and MEP measurements and dyspnea scores at rest of the eight subjects who completed the 8-week study. Friedman's test and all comparisons with a control procedure were performed on the baseline and week 2, 4, 6, and 8 dyspnea scores during exercise of the same subjects. Univariate linear regression analysis of the data for these subjects was performed to evaluate the relationships between a variety of patient characteristics (age, height, weight, m 4 classification, left ventricular ejection fraction, duration of HF, and IMT adherence) and MIP and MEP before and after IMT, percentage of change in MIP and MEP after IMT, degree of dyspnea at rest and during submaximal exercise before and after IMT, and percentage of change in dyspnea at rest and during exercise after IMT.
Results Ventilatory Muscle Training
All subjects tolerated IMT without major complaint, complication, or change in medical therapy. The majority of subjects complained of mild abdominal discomfort after initial IMT, and several subjects had slight alterations ( < l o % ) in body weight and diuretic and dobutamine dosages. Thirteen of the 14 subjects completed 4 weeks of IMT (1 subject underwent cardiac transplantation during week 3). Two subjects underwent cardiac transplantation, 1 subject underwent coronary artery bypass surgery, and 1 subject required intra-aortic balloon pump placement during week 5 of IMT, which prevented further IMT or testing of ventilatory muscle strength. One subject expired from sudden death during week 5 of IMT. The 8 remaining subjects completed 8 weeks of IMT. Adherence to IMT was 63% 224%, with the majority of subjects completing three sessions of IMT per day for a mean of 1 0 5 3 minutes each session. Seven subjects performed 50% or less of the prescribed IMT. Five of the seven subjects with adherence of 50% or less were in the cohort of subjects who completed the full 8 weeks of IMT. Nonetheless, the adherence of the eight subjects
SETAT lnc, 1800 Sherman Ave, Emnston, IL 60201.
Physical Therapy . Volume 77 . Number 8 . August 1997
Cohalin et al . 833
cm H 2 0 to 63523 cm H,O; P=.0001) and an additional 8% by week 6 of IMT. The initial and final MIPS were 44% 2 15% and 55%2 15%, respectively, of age- and gender-predicted MIP. Mean MEP increased 13% after 2 weeks of IMT (85222 cm H,O to 9 6 5 19 cm H,O; F.OOO1) and remained at a plateau throughout the remainder of IMT. The initial and final MEPs were 39%?8% and 44% 5696, respectively, of age- and gender-predicted MEP.
110 I
+
100
W
90 UJ W
J
3
-%
80
=
5 g
G-
70
4
60
5
50
F
>
Symptoms . .
40 0
2
4
6
WEEKS
5
4
! i a
4z s a
-
8
Dyspnea scores at rest and during exercise decreased (29% and 28%, respectively) within 2 weeks after the initiation of IMT and remained at a plateau throughout the remainder of IMT (Fig. 1). Dyspnea scores at rest decreased from a mean of 2.050.7 to 1.320.05 ( P .0001; a rating change from slight to very slight), and dyspnea scores during exercise decreased from a mean of 3.620.5 to 2.62 0.6 ( F . 0 0 3 ; a rating change from somewhat severe to moderate).
3-
Correlation Analyses
Correlation analyses identified relationtn > 4 2ships between MIP after IMT and per0 * * * centage of predicted FVC (r=.71, (3 X K F . 0 4 ) , dyspnea score at rest after IMT and FVC (r=-.77, F . 0 2 ) and total 1lung capacity ( r = - .82, P=.01), baseline MEP and percentage of change in I I I I I 0 MEP after IMT ( r = - .72, F .04), and 0 2 4 6 8 adherence to IMT and FEV, (r=.79, WEEKS F . 0 2 ) . There was no correlation between NYHA classification, left venFigure 1. tricular ejection fraction, duration of [Top) Maximal inspiratory pressure [r) and maximal expiratory pressure (0)throughout the HF, or baseline measure of MIP and the ventilatory muscle training period. [Bottom] Dyspnea at rest (r) and during submaximal percentage of change in MIP, MEP, or exercise (0)throughout the ventilatory muscle training period. Asterisk ('1 indicates value is dyspnea score at rest or during exercise different from baseline [P c.05). after IMT. Near-significant correlations, however, were found between baseline who completed 8 weeks of IMT was not different from MIP and percentage of change in dyspnea scores at rest that of the 6 subjects who were unable to complete 8 and during exercise (r=.67, F . 0 7 and r=.69, P=.06, weeks of IMT (60%526% versus 66%522%). The level respectively), level of adherence to IMT and MIP and of adherence for each subject was consistent during the MEP after IMT (r=.65, P . 0 8 and r =.68, P=.06, respecIMT period. The mean intensity of IMT during the tively), and measures of pulmonary function and MIP %week study period increased from a baseline value of and MEP after IMT (FEV, and MIP: r =.68, P=.06 and 7 2 3 cm H,O to 1224 c n ~ H,O. FEV, and MEP: r=.69, P=.05) and level of adherence to IMT (FVC and adherence: r = .66, P= .07). 7
2
Ventilatory Muscle Strength
Both the MIP and MEP increased after IMT (Fig. 1). Mean MIP increased 24% after 2 weeks of IMT (51221 834 . Cahalin et a1
Physical Therapy
. Volume 77 . Number 8 . August 1997
Table 2. Pearson Product-Moment Correlations Between Dyspnea and Pulmonary Function"
I "
Pearson r Dyspnea Variable
FEV, (%)
FVC (%)
Pre-IMT dyspnea at rest (N=14) Pre-IMT dyspnea during exercise (N= 14) Post-IMT dyspnea at rest (n=8)
-.60 - .57
...
FVC (1)
TLC (1)
-.59
...
-.67
...
... ...
...
-.77
-.82
IMT=inspiratory muscle training, FE\',=forced expiratory volurnp in 1 second, FVC=forced vital capacity, TLC=total lung capacity (P
