The Effects of Refeeding on Peripheral and Respiratory Muscle Function in Malnourished Chronic Obstructive Pulmonary Disease Patients1- 3
J. SCOTT WHITTAKER, C. FRANCIS RYAN, PATRICIA A. BUCKLEY, and JEREMY D. ROAD Introduction
Malnutrition has been estimated to occur in as many as 40 to 50"70 of hospitalized emphysematous patients (1, 2). Several studies have reported the association of malnutrition with worsened morbidity and mortality in patients with chronic obstructive lung disease. Malnutrition has been associated with decreased diaphragmatic mass in patients with chronic obstructive pulmonary disease (COPD) (3, 4). In addition, recent studies have shown abnormalities of peripheral muscle function in malnourished patients (5, 6), including increased fatiguability. Limited data in humans suggest that patients with COPD may be susceptible to low frequency fatigue (7, 8). Refeeding studies in patients who showed altered peripheral muscle function have demonstrated improvement within 2 to 4 wk, including marked reduction in fatigability (6). Limited information in intensive care units has shown some improvement with respect to weaning in patients who have been given total parenteral nutrition (9-11). Against this background, there has been increasing interest in the role of malnutrition in the respiratory muscle function of patients with chronic obstructive pulmonary disease, particularly emphysema. Two recent outpatient studies randomly divided COPD patients into a supplemented group and an unsupplemented group (12, 13). The most striking finding in one study was an inability to refeed these subjects, resulting in inadequate weight gain and failure to show improvement in respiratory muscle function parameters (12). In the second study, a small improvement in respiratory muscle strength was observed (13). There have also been a variety of uncontrolled inpatient studies (14-16), some of which have reported improvement in respiratory parameters. In one study, respiratory muscle function testing was performed,
SUMMARY We carried out a prospective randomized controHed trial to Investigate the effects of short-term refeeding (16 days) in 10 malnourished Inpatients with chronic obstructive pulmonary dI.._ (COPD). Six patients were randomized to receive sufficient nasoenterlcally admlnlstanld calories to provide a total caloric int8Ice efI'NIl to 1,000 IlCIliabove their ueuallntake. The other four ptItIeftts were sm- ted, receiving only 100 1liCal1IIOf'8......urements of nutritional status, respiratory muscle atrength and endurance, adductor poH\cis function, and pulmonary function were performed initially and at atuGly end. The reted group gained Significantly more _Ight and showed significant i n c _ In maximal expiratory " . . . , . and mean sustained iftapIratory pressure. There were no significant changes In the maximal Inspiratory pressure or In adductor pollicls function. In malnourished Inpatients with COPO, short-term refeedlng leads to Improvement In respiratory muscle enGIurance and In some parameters of respiratory muscle strength In the absence of demonAM REV RESPIR DIS 1890; 142:283-288 strable changes In peripheral muscle function.
showing a slight improvement. We therefore studied the effects of inpatient refeeding on a sample of COPD patients, using a randomized control group that underwent sham feeding. In addition to the previously reported parameters of respiratory muscle strength testing, we measured respiratory muscle endurance. Methods Study Population Ten patients with moderate malnutrition « 85"10 ideal body weight) and clinically stable COPD who had an FEY,/FYC ratio of 35 to 70% of predicted and a response to bronchodilators of less than 15% were selected for the study. Patients who were clinically unstable, or who had congestive heart failure, an active respiratory infection, malabsorption, or diabetes mellitus, were excluded. Patients underwent an initial workup consisting of standard nutritional assessment, including body weights and anthropometry, metabolic measurements, dietetic consultation, pulmonary function tests,arterial blood gases, adductor pollicis muscle function, and respiratory muscle testing. Patients consented to the passage of a nasoenteric feeding tube prior to randomization.
Study Design Before admission to the hospital, each patient completed a 1-wk dietary diary as an outpatient and was encouraged to maintain that level of food intake while in the hospital. Having given informed consent, each patient was
admitted to the hospital, initial evaluations were performed, and a nasoenteric tube was placed in the distal duodenum or jejunum under fluoroscopic visualization. The patients were randomly divided between a fed group and a control group in a blinded fashion by one of the investigators (JDR). Food intake while in the hospital was carefully monitored by a dietitian (PAB) who weighed foodstuffs in order to ensure that an accurate evaluation ·of caloric intake was performed. The group that was to be refed randomly then received supplemental calories in the form of Isocal (Mead-Johnson, Belleville, Ontario, Canada) given nocturnally to equal at least 1,000 kcal above their usual intake, or 1.7 times the resting energy expenditure (REE), whichever was greater. Adjustments ofthe nasoenteric feeds were made by the dietitian. The control group received an equivalent volume of a dilute solution of Isocal containing less than 100 kcal per night. The patients were fed (Received in original form August 8, 1989 and in revised form January 26, 1990) 1 From the Divisions of Gastroenterology and Respirology, Department of Medicine, University of British Columbia, Vancouver, British Columbia, Canada. 1 Supported by grants from the British Columbia Lung Association, the Volunteer Services Association (University Hospital, UBC Site), and Bristol-Myers (Mead-Johnson Division). 3 Correspondence and requests for reprints should be addressed to J. Scott Whittaker, M.D., Department of Medicine, St. Paul's Hospital, 1081 Burrard Street, Vancouver, BC, Canada V6Z IY6.
283
284
WHITTAKER, RYAN, BUCKLEY, AND ROAD
nasoenterically for 16 ± 3 days, following which the initial parameters were repeated. Both the patients and the observers (CFR, JSW) performing the respiratory and adductor pollicis muscle function tests were blinded as to whether the patients were being refed. The dosages of medications such as theophylline-containing compounds, beta,-agonists, or other treatments were not altered during the study. Three patients who were receiving theophylline preparations, two in the refed group and one in the control group, had serum levels within the therapeutic range.
Nutritional Assessment Nutritional assessment included standard anthropometric measurements (body weight, midarm muscle circumference, triceps skinfold thickness) and serum levels of albumin, transferrin, and absolute lymphocyte count. Resting energy expenditure was measured in the postabsorptive state using the Sensormedics MMC Horizon metabolic cart and canopy system (Sensormedics, Anaheim, CAl.
Adductor Pollicis Function The adductor pollicis function was measured as previously described (6). Each patient's right hand was warmed to 37° C using warmed moist towels, and the thumb was placed in a sling connected to a pressure transducer. A thermal chart recorder (Model 3000; Gould Inc., Cleveland, OH) was used to measure the forces generated at 10, 20, 30, 50, and 100 Hz. In addition, the maximal relaxation rate at 30 Hz was measured (expressed as percent force lossl1O ms). Each measurement was repeated. The force generated at 10 Hz was compared with the maximal force generated (FlO/Fmax). The average of the values of FlO/Fmax and maximal relaxation rate was calculated for each patient. These values have been previously shown to be highly reproducible (6).
Pulmonary Function Testing Spirometry, lung volumes, and carbon monoxide diffusing capacity were measured using the Cybermedic Pulmonary Function System (Spinnaker, 1987; Software 3.04). Lung volumes were measured using the helium dilution technique and diffusing capacity was determined by the single breath method. Blood gas samples were taken from the radial artery (Flo,: room air) and analyzed (ABL 300 Blood Gas Analyzer; Radiometry, Copenhagen).
Respiratory Muscle Performance Both strength and endurance of the respiratory muscles were measured. Measurements were performed while the patients were seated and wearing noseclips. Respiratory muscle strength was assessed as maximum static inspiratory (MIP) and expiratory (MEP) pressures. Patients performed the maneuvers through an occluded valve (Model 2700; Hans Rudolph Inc., Kansas City, MO) containing a small air leak (internal diameter 0.6 mm) to minimize artifact resulting from contrac-
tion of facial muscles. Pressures were measured using a differential pressure transducer (±350 cm H 2 0) (Model DP45-16; Validyne Co., Northridge, CAl, recorded on a thermal chart recorder (Model 3000; Gould Inc., Cleveland, OH) and read directly from the analog tracing. The system was calibrated before each test using a water manometer. MIP was measured near residual volume (VR) and MEP near total lung capacity. Pressures were sustained for greater than 1 s. Repeated measurements were obtained until variability was less than 50/0 over three measurements. The highest value achieved was reported. Normal values for MIP and MEP were taken from Black and Hyatt (17). Inspiratory muscle endurance was assessed by measuring peak (PM peak) and mean (PM) inspiratory mouth pressures during an incremental threshold loading test (18). Patients inspired through a twoway valve (Hans Rudolph Inc.) attached via the inspiratory limb to a device consisting of a weighted plunger and an inspiratory port with an orifice of 6.6 cm'. Weights were added externally and increasing weight on the plunger required increased pressure to open the inspiratory port when' the stopper was in place. Once a threshold inspiratory mouth pressure (PM) was generated, flow was largely independent of PM. Starting with no additional weight, weights were added in 50-g increments at 2-min intervals and the PM peak and PM corresponding to the maximum load tolerated for at least 1 min were measured. Measurements were recorded in an identical fashion to that outlined above for MIP and MEP. PM peak was averaged over 5 breaths. PM was obtained on line by passing the PM signal through a second-order, low-pass filter with a time constant of 20 s. No training was given for this test, as the 2-min incremental format was designed to allow learning of appropriate breathing patterns at low weights. This simple test of respiratory muscle endurance has been shown to be reproducible (18, 19).
Statistical Analysis For each variable, the change from baseline
to the end of the study was calculated. The differences between the groups were then tested using Student's unpaired t test. A level of 5% was chosen as being significant. Since all multiple comparison correction procedures assume complete independence of the variables, which was not the case in this study, no correction was made. Results
Baseline Data Baseline data, including anthropometric measurements, are shown in table 1. For all 10 patients, the average percentage of ideal body weight was 78.5 (using Metropolitan Life Tables, 1959), indicating a modest degree of malnutrition. All patients were at least 151l,70 below ideal body weight as required by the entrance criteria. No patient was above the 25th percentile for midarm muscle circumference or triceps skin fold thickness. There were six patients in the refed group (Group A) and four patients in the sham-fed group (Group B). There were no differences between the two groups with respect to baseline data. The blood values drawn (including electrolytes, calcium, magnesium, phosphorous, creatinine, and urea) were normal in both groups. These values did not change significantly over the period of the study. Pulmonary Function Data These data are presented in table 2 for Groups A and B. There were no differences between the groups at entry with respect to the pulmonary function data, and there were no differences over time in either the refed or the control group. Similarly, baseline arterial blood gases were not different between the groups (table 3).
TABLE 1 COMPARISON OF BASELINE DETAILS AND NUTRITIONAL STATUS BETWEEN THE GROUPS OF PATIENTS ENTERED INTO THE STUDY
Number Sex, M:F Age, yr Weight, kg Height, cm Ideal body weight, % MAMC:J:, cm Triceps skinfold, mm Albumin, giL Transferrin, giL Lymphocyte count, x 10·/L Creatinine-height index, % • Group A = refed group. Group B = sham·fed group. MAMC = midarm muscle circumference.
t
*
Group A'
Group Bt
6
4 1:3 64 ± 17 45.6 ± 9.9 157 ± 10
71 50.3 169 76 21.4 7.2 39 2.5 2.099 74
± 7 ± 9.2 ± 8 ±
5
± ± ± ± ± ±
2.3 2.1 3 0.4 1 .246 17
82 ± 3 19.4 6.6 39 3.0 1.750 53
± 3.7 ± 0.8 ± 2 ± 0.3
± 0.532 ± 15
285
EFFECTS OF REFEEOING IN COPO
TABLE 2
3000
COMPARISON OF PULMONARY FUNCTION DATA BETWEEN REFED AND SHAM-FED PATIENTS BEFORE AND AFTER REFEEDING 2000
Group A* Post§
Pre:!:
Percent Predicted
Group Bt Pre
Post 1000
FEV, FVC FEV,/FVC
40 67 56 163 109 151 52
VR TLC VRlTLC DLeoNA
± ± ± ± ± ± ±
27 27 13 88 13 29 28
44 73 56 146 107 143 49
± ± ± ± ± ± ±
27 27 14 28 14 23 29
47 83 56 121 106 124 46
± ± ± ± ± ± ±
27 41 20 63 26 31 12
48 ± 93 ± 56 ± 113 ± 100 ± 119 ± 48 ±
29 54 15 47 35 26 26
* Group A = refed group. = sham-fed group.
t Group B
*
Harne
Hospital
Group A
Home
Hospital
Group B
Fig. 1. Daily caloric intake. The refed group received approximately 1,000 kcal more than the control group while they were in the hospital.
Pre = value before feeding.
§ Post = value after feeding.
Caloric Intake The caloric intakes of the two groups are shown in figure 1. It is evident that there was significantly greater intake in the group that was refed and that the control group did not eat more than their baseline home values, as determined by the dietary diaries. In all cases, the refed patients received 1,000 kcal above their baseline intake, since in every case this was greater than the increment necessary to achieve 1.7 times the resting energy expenditure as determined by indirect calorimetry using the Sensormedics metabolic cart. Refed patients gained weight over the refeeding period (Aweight = 2.4 kg), whereas the control patients failed to gain weight (Aweight = -0.6 kg). Adductor Pollicis Muscle Function The adductor pollicis function is shown in figure 2. When compared with the results of previous studies (6), adductor pollicis function was normal at baseline in this study and neither the FlO/Fmax ratio nor the maximal relaxation rate changed following refeeding. Respiratory Muscle Function The respiratory muscle function data are shown in table 4 and figure 3. Refed patients, when compared with control subjects, showed significant increases in MEP (AMEP refed group 34 ± 10 cm H 2 0, control group 4 ± 6 em H 2 0) and TABLE 3 BASELINE ARTERIAL BLOOD GASES Group A* pH Peo, Po, Heo,
7.40 43 71 26
± 0.04 ± 4
± 13 ± 1
• Group A = reled group. t Group B = sham·led group.
Group Bt 7.40 46 56 26
± 0.03 ± 9
± g ± 5
PM (APM refed group 11.4 ± 2.9 em H 2 0, control group 0.4 ± 0.1 em H 2 0). MIP and PM peak did not change significantly. Discussion
This study shows that stable malnourished COPO patients can gain weight during inpatient enteral feeding and that the improvement in nutritional status is associated with improved respiratory muscle performance. These changes take place over a relatively short refeeding period (16 ± 3 days). The refed patients consumed 2,489 ± 249 kcal/day while in the hospital, representing 1750Je of their home caloric intake or 220OJo of their measured REE. Refeeding was generally well tolerated apart from transient pharyngeal discomfort in three patients and one episode of pharyngitis, which responded promptly to an oral broad spectrum antibiotic without the necessity of removing the enteral tube. Our patients gained weight and had significant improvements in expiratory muscle strength and inspiratory muscle endurance. Furthermore, linear regression analysis showed significant correlations between weight gain and changes inMIPandMEP(r = 0.67andr = 0.73, respectively). Of the recent studies examining refeeding in COPO patients, it is interesting that only in those studies in which weight gain was achieved has an associated improvement in respiratory muscle performance been demonstrated. In a controlled trial by Efthimiou and colleagues (13), 3 months of supplemental oral nutrition produced significant improvements in nutritional status including a weight gain of 4.2 kg and significant increases in respiratory muscle strength, hand grip strength, and decreased sternocleidomastoid fatigability. Wilson and coworkers (16), in an uncontrolled inpatient study, also noted im-
proved respiratory and peripheral muscle function in association with weight gain. Other refeeding studies, where little or no weight gain resulted, have failed to demonstrate significant changes in respiratory muscle function (12, 20). In the latter studies, difficulties in achieving weight gain may have related to poor tolerance for the orally administered nutritional supplement or a tendency for some patients to reduce their normal oral intake when receiving an oral supplement. None of our patients had intolerance for the supplemental feed. We believe that this related to the method of administration of the feed, whereby patients were given a slow overnight infusion through a nasojejunal tube. Although we also noted a small reduction
40
30
~
20
u:
10
Pre
Pas1
Group A
a a:
"'
'"'"
.2
E
10
0
~
8
a: ~
6
)(
0;
Pas1
Group B
12
c:
,g
Pre
.l!! .E
g
:::> ~ 4 :::;:
2
Pre
Pas1
Group A Fig. 2. Adductor pollicis function.
Pre
Post
Group B
286
WHITTAKER, RYAN, BUCKLEY, AND ROAD
TABLE 4 INDIVIDUAL RESPIRATORY MUSCLE FUNCTION DATA IN REFED AND SHAM·FED PATIENTS BEFORE AND AFTER REFEEDING Patient Number Group A:j::j: 1 2 3 4 5 6 Mean ± SE Group B§§ 7 8 9 10 Mean ± SE P Value"
Weight t. (kg)
MEPt
MIP' Pre'
Post"
2.7 1.9 3.0 3.3 -1.0 4.7 2.4 0.8
62 91 93
84
43 72 68 9
107 78 40 113 81 11
0.5 -0.7 -1.2 -1.0 -0.6 0.4 0.02
34 102 27 20 46 19
27 105 27 30 47 19
44
t.tt
62
0 -7 14 34 -3 41 13 8 -7 3 0 10 1.5 3.5 0.31
PM Peak:j:
PM Mean§
Pre
Post
t.
Pre
Post
t.
Pre
Post
t.
98 99 71 107 91 92 5
104 114 134 129 123 150 126 6
6 15 51 58 16 59 34 10
30 48 70 21 13 30 35 9
54 69 91 42 30 36 54 9
24 21 21 21
12.1 16.8 18.1 5.4 4.0 11.4 11.3 2.3
18.8 33.5 45.6 16.1 11.4 15.4 23.5 5.4
6.7 16.7 27.5 10.7 7.4 4.0 11.4 2.9
48 129 54 72 76 16
39 139 52 88 80 23
-9 10 -2 16 4
35 30 19 19 26 4
13.4 12.7 6.7 4.4 9.3 2.2
12.7 16.1 6.0 4.0 9.7 2.9
-0.7 3.4 -0.7 -0.4 0.4 0.1 0.03
83
6
27 62
17 15 30 11
0.05
17 6 18 2 -8 32 -2 -4 5 10 0.14
• MIP: maximal inspiratory pressure. t MEP: maximal expiratory pressure. PM Peak: peak inspiratory pressure during threshold loading test. § PM Mean: mean inspiratory pressure during threshold loading test. , Pre: value before feeding . •• Post: value alter feeding. tt .
J. SCOTT WHITTAKER, C. FRANCIS RYAN, PATRICIA A. BUCKLEY, and JEREMY D. ROAD Introduction
Malnutrition has been estimated to occur in as many as 40 to 50"70 of hospitalized emphysematous patients (1, 2). Several studies have reported the association of malnutrition with worsened morbidity and mortality in patients with chronic obstructive lung disease. Malnutrition has been associated with decreased diaphragmatic mass in patients with chronic obstructive pulmonary disease (COPD) (3, 4). In addition, recent studies have shown abnormalities of peripheral muscle function in malnourished patients (5, 6), including increased fatiguability. Limited data in humans suggest that patients with COPD may be susceptible to low frequency fatigue (7, 8). Refeeding studies in patients who showed altered peripheral muscle function have demonstrated improvement within 2 to 4 wk, including marked reduction in fatigability (6). Limited information in intensive care units has shown some improvement with respect to weaning in patients who have been given total parenteral nutrition (9-11). Against this background, there has been increasing interest in the role of malnutrition in the respiratory muscle function of patients with chronic obstructive pulmonary disease, particularly emphysema. Two recent outpatient studies randomly divided COPD patients into a supplemented group and an unsupplemented group (12, 13). The most striking finding in one study was an inability to refeed these subjects, resulting in inadequate weight gain and failure to show improvement in respiratory muscle function parameters (12). In the second study, a small improvement in respiratory muscle strength was observed (13). There have also been a variety of uncontrolled inpatient studies (14-16), some of which have reported improvement in respiratory parameters. In one study, respiratory muscle function testing was performed,
SUMMARY We carried out a prospective randomized controHed trial to Investigate the effects of short-term refeeding (16 days) in 10 malnourished Inpatients with chronic obstructive pulmonary dI.._ (COPD). Six patients were randomized to receive sufficient nasoenterlcally admlnlstanld calories to provide a total caloric int8Ice efI'NIl to 1,000 IlCIliabove their ueuallntake. The other four ptItIeftts were sm- ted, receiving only 100 1liCal1IIOf'8......urements of nutritional status, respiratory muscle atrength and endurance, adductor poH\cis function, and pulmonary function were performed initially and at atuGly end. The reted group gained Significantly more _Ight and showed significant i n c _ In maximal expiratory " . . . , . and mean sustained iftapIratory pressure. There were no significant changes In the maximal Inspiratory pressure or In adductor pollicls function. In malnourished Inpatients with COPO, short-term refeedlng leads to Improvement In respiratory muscle enGIurance and In some parameters of respiratory muscle strength In the absence of demonAM REV RESPIR DIS 1890; 142:283-288 strable changes In peripheral muscle function.
showing a slight improvement. We therefore studied the effects of inpatient refeeding on a sample of COPD patients, using a randomized control group that underwent sham feeding. In addition to the previously reported parameters of respiratory muscle strength testing, we measured respiratory muscle endurance. Methods Study Population Ten patients with moderate malnutrition « 85"10 ideal body weight) and clinically stable COPD who had an FEY,/FYC ratio of 35 to 70% of predicted and a response to bronchodilators of less than 15% were selected for the study. Patients who were clinically unstable, or who had congestive heart failure, an active respiratory infection, malabsorption, or diabetes mellitus, were excluded. Patients underwent an initial workup consisting of standard nutritional assessment, including body weights and anthropometry, metabolic measurements, dietetic consultation, pulmonary function tests,arterial blood gases, adductor pollicis muscle function, and respiratory muscle testing. Patients consented to the passage of a nasoenteric feeding tube prior to randomization.
Study Design Before admission to the hospital, each patient completed a 1-wk dietary diary as an outpatient and was encouraged to maintain that level of food intake while in the hospital. Having given informed consent, each patient was
admitted to the hospital, initial evaluations were performed, and a nasoenteric tube was placed in the distal duodenum or jejunum under fluoroscopic visualization. The patients were randomly divided between a fed group and a control group in a blinded fashion by one of the investigators (JDR). Food intake while in the hospital was carefully monitored by a dietitian (PAB) who weighed foodstuffs in order to ensure that an accurate evaluation ·of caloric intake was performed. The group that was to be refed randomly then received supplemental calories in the form of Isocal (Mead-Johnson, Belleville, Ontario, Canada) given nocturnally to equal at least 1,000 kcal above their usual intake, or 1.7 times the resting energy expenditure (REE), whichever was greater. Adjustments ofthe nasoenteric feeds were made by the dietitian. The control group received an equivalent volume of a dilute solution of Isocal containing less than 100 kcal per night. The patients were fed (Received in original form August 8, 1989 and in revised form January 26, 1990) 1 From the Divisions of Gastroenterology and Respirology, Department of Medicine, University of British Columbia, Vancouver, British Columbia, Canada. 1 Supported by grants from the British Columbia Lung Association, the Volunteer Services Association (University Hospital, UBC Site), and Bristol-Myers (Mead-Johnson Division). 3 Correspondence and requests for reprints should be addressed to J. Scott Whittaker, M.D., Department of Medicine, St. Paul's Hospital, 1081 Burrard Street, Vancouver, BC, Canada V6Z IY6.
283
284
WHITTAKER, RYAN, BUCKLEY, AND ROAD
nasoenterically for 16 ± 3 days, following which the initial parameters were repeated. Both the patients and the observers (CFR, JSW) performing the respiratory and adductor pollicis muscle function tests were blinded as to whether the patients were being refed. The dosages of medications such as theophylline-containing compounds, beta,-agonists, or other treatments were not altered during the study. Three patients who were receiving theophylline preparations, two in the refed group and one in the control group, had serum levels within the therapeutic range.
Nutritional Assessment Nutritional assessment included standard anthropometric measurements (body weight, midarm muscle circumference, triceps skinfold thickness) and serum levels of albumin, transferrin, and absolute lymphocyte count. Resting energy expenditure was measured in the postabsorptive state using the Sensormedics MMC Horizon metabolic cart and canopy system (Sensormedics, Anaheim, CAl.
Adductor Pollicis Function The adductor pollicis function was measured as previously described (6). Each patient's right hand was warmed to 37° C using warmed moist towels, and the thumb was placed in a sling connected to a pressure transducer. A thermal chart recorder (Model 3000; Gould Inc., Cleveland, OH) was used to measure the forces generated at 10, 20, 30, 50, and 100 Hz. In addition, the maximal relaxation rate at 30 Hz was measured (expressed as percent force lossl1O ms). Each measurement was repeated. The force generated at 10 Hz was compared with the maximal force generated (FlO/Fmax). The average of the values of FlO/Fmax and maximal relaxation rate was calculated for each patient. These values have been previously shown to be highly reproducible (6).
Pulmonary Function Testing Spirometry, lung volumes, and carbon monoxide diffusing capacity were measured using the Cybermedic Pulmonary Function System (Spinnaker, 1987; Software 3.04). Lung volumes were measured using the helium dilution technique and diffusing capacity was determined by the single breath method. Blood gas samples were taken from the radial artery (Flo,: room air) and analyzed (ABL 300 Blood Gas Analyzer; Radiometry, Copenhagen).
Respiratory Muscle Performance Both strength and endurance of the respiratory muscles were measured. Measurements were performed while the patients were seated and wearing noseclips. Respiratory muscle strength was assessed as maximum static inspiratory (MIP) and expiratory (MEP) pressures. Patients performed the maneuvers through an occluded valve (Model 2700; Hans Rudolph Inc., Kansas City, MO) containing a small air leak (internal diameter 0.6 mm) to minimize artifact resulting from contrac-
tion of facial muscles. Pressures were measured using a differential pressure transducer (±350 cm H 2 0) (Model DP45-16; Validyne Co., Northridge, CAl, recorded on a thermal chart recorder (Model 3000; Gould Inc., Cleveland, OH) and read directly from the analog tracing. The system was calibrated before each test using a water manometer. MIP was measured near residual volume (VR) and MEP near total lung capacity. Pressures were sustained for greater than 1 s. Repeated measurements were obtained until variability was less than 50/0 over three measurements. The highest value achieved was reported. Normal values for MIP and MEP were taken from Black and Hyatt (17). Inspiratory muscle endurance was assessed by measuring peak (PM peak) and mean (PM) inspiratory mouth pressures during an incremental threshold loading test (18). Patients inspired through a twoway valve (Hans Rudolph Inc.) attached via the inspiratory limb to a device consisting of a weighted plunger and an inspiratory port with an orifice of 6.6 cm'. Weights were added externally and increasing weight on the plunger required increased pressure to open the inspiratory port when' the stopper was in place. Once a threshold inspiratory mouth pressure (PM) was generated, flow was largely independent of PM. Starting with no additional weight, weights were added in 50-g increments at 2-min intervals and the PM peak and PM corresponding to the maximum load tolerated for at least 1 min were measured. Measurements were recorded in an identical fashion to that outlined above for MIP and MEP. PM peak was averaged over 5 breaths. PM was obtained on line by passing the PM signal through a second-order, low-pass filter with a time constant of 20 s. No training was given for this test, as the 2-min incremental format was designed to allow learning of appropriate breathing patterns at low weights. This simple test of respiratory muscle endurance has been shown to be reproducible (18, 19).
Statistical Analysis For each variable, the change from baseline
to the end of the study was calculated. The differences between the groups were then tested using Student's unpaired t test. A level of 5% was chosen as being significant. Since all multiple comparison correction procedures assume complete independence of the variables, which was not the case in this study, no correction was made. Results
Baseline Data Baseline data, including anthropometric measurements, are shown in table 1. For all 10 patients, the average percentage of ideal body weight was 78.5 (using Metropolitan Life Tables, 1959), indicating a modest degree of malnutrition. All patients were at least 151l,70 below ideal body weight as required by the entrance criteria. No patient was above the 25th percentile for midarm muscle circumference or triceps skin fold thickness. There were six patients in the refed group (Group A) and four patients in the sham-fed group (Group B). There were no differences between the two groups with respect to baseline data. The blood values drawn (including electrolytes, calcium, magnesium, phosphorous, creatinine, and urea) were normal in both groups. These values did not change significantly over the period of the study. Pulmonary Function Data These data are presented in table 2 for Groups A and B. There were no differences between the groups at entry with respect to the pulmonary function data, and there were no differences over time in either the refed or the control group. Similarly, baseline arterial blood gases were not different between the groups (table 3).
TABLE 1 COMPARISON OF BASELINE DETAILS AND NUTRITIONAL STATUS BETWEEN THE GROUPS OF PATIENTS ENTERED INTO THE STUDY
Number Sex, M:F Age, yr Weight, kg Height, cm Ideal body weight, % MAMC:J:, cm Triceps skinfold, mm Albumin, giL Transferrin, giL Lymphocyte count, x 10·/L Creatinine-height index, % • Group A = refed group. Group B = sham·fed group. MAMC = midarm muscle circumference.
t
*
Group A'
Group Bt
6
4 1:3 64 ± 17 45.6 ± 9.9 157 ± 10
71 50.3 169 76 21.4 7.2 39 2.5 2.099 74
± 7 ± 9.2 ± 8 ±
5
± ± ± ± ± ±
2.3 2.1 3 0.4 1 .246 17
82 ± 3 19.4 6.6 39 3.0 1.750 53
± 3.7 ± 0.8 ± 2 ± 0.3
± 0.532 ± 15
285
EFFECTS OF REFEEOING IN COPO
TABLE 2
3000
COMPARISON OF PULMONARY FUNCTION DATA BETWEEN REFED AND SHAM-FED PATIENTS BEFORE AND AFTER REFEEDING 2000
Group A* Post§
Pre:!:
Percent Predicted
Group Bt Pre
Post 1000
FEV, FVC FEV,/FVC
40 67 56 163 109 151 52
VR TLC VRlTLC DLeoNA
± ± ± ± ± ± ±
27 27 13 88 13 29 28
44 73 56 146 107 143 49
± ± ± ± ± ± ±
27 27 14 28 14 23 29
47 83 56 121 106 124 46
± ± ± ± ± ± ±
27 41 20 63 26 31 12
48 ± 93 ± 56 ± 113 ± 100 ± 119 ± 48 ±
29 54 15 47 35 26 26
* Group A = refed group. = sham-fed group.
t Group B
*
Harne
Hospital
Group A
Home
Hospital
Group B
Fig. 1. Daily caloric intake. The refed group received approximately 1,000 kcal more than the control group while they were in the hospital.
Pre = value before feeding.
§ Post = value after feeding.
Caloric Intake The caloric intakes of the two groups are shown in figure 1. It is evident that there was significantly greater intake in the group that was refed and that the control group did not eat more than their baseline home values, as determined by the dietary diaries. In all cases, the refed patients received 1,000 kcal above their baseline intake, since in every case this was greater than the increment necessary to achieve 1.7 times the resting energy expenditure as determined by indirect calorimetry using the Sensormedics metabolic cart. Refed patients gained weight over the refeeding period (Aweight = 2.4 kg), whereas the control patients failed to gain weight (Aweight = -0.6 kg). Adductor Pollicis Muscle Function The adductor pollicis function is shown in figure 2. When compared with the results of previous studies (6), adductor pollicis function was normal at baseline in this study and neither the FlO/Fmax ratio nor the maximal relaxation rate changed following refeeding. Respiratory Muscle Function The respiratory muscle function data are shown in table 4 and figure 3. Refed patients, when compared with control subjects, showed significant increases in MEP (AMEP refed group 34 ± 10 cm H 2 0, control group 4 ± 6 em H 2 0) and TABLE 3 BASELINE ARTERIAL BLOOD GASES Group A* pH Peo, Po, Heo,
7.40 43 71 26
± 0.04 ± 4
± 13 ± 1
• Group A = reled group. t Group B = sham·led group.
Group Bt 7.40 46 56 26
± 0.03 ± 9
± g ± 5
PM (APM refed group 11.4 ± 2.9 em H 2 0, control group 0.4 ± 0.1 em H 2 0). MIP and PM peak did not change significantly. Discussion
This study shows that stable malnourished COPO patients can gain weight during inpatient enteral feeding and that the improvement in nutritional status is associated with improved respiratory muscle performance. These changes take place over a relatively short refeeding period (16 ± 3 days). The refed patients consumed 2,489 ± 249 kcal/day while in the hospital, representing 1750Je of their home caloric intake or 220OJo of their measured REE. Refeeding was generally well tolerated apart from transient pharyngeal discomfort in three patients and one episode of pharyngitis, which responded promptly to an oral broad spectrum antibiotic without the necessity of removing the enteral tube. Our patients gained weight and had significant improvements in expiratory muscle strength and inspiratory muscle endurance. Furthermore, linear regression analysis showed significant correlations between weight gain and changes inMIPandMEP(r = 0.67andr = 0.73, respectively). Of the recent studies examining refeeding in COPO patients, it is interesting that only in those studies in which weight gain was achieved has an associated improvement in respiratory muscle performance been demonstrated. In a controlled trial by Efthimiou and colleagues (13), 3 months of supplemental oral nutrition produced significant improvements in nutritional status including a weight gain of 4.2 kg and significant increases in respiratory muscle strength, hand grip strength, and decreased sternocleidomastoid fatigability. Wilson and coworkers (16), in an uncontrolled inpatient study, also noted im-
proved respiratory and peripheral muscle function in association with weight gain. Other refeeding studies, where little or no weight gain resulted, have failed to demonstrate significant changes in respiratory muscle function (12, 20). In the latter studies, difficulties in achieving weight gain may have related to poor tolerance for the orally administered nutritional supplement or a tendency for some patients to reduce their normal oral intake when receiving an oral supplement. None of our patients had intolerance for the supplemental feed. We believe that this related to the method of administration of the feed, whereby patients were given a slow overnight infusion through a nasojejunal tube. Although we also noted a small reduction
40
30
~
20
u:
10
Pre
Pas1
Group A
a a:
"'
'"'"
.2
E
10
0
~
8
a: ~
6
)(
0;
Pas1
Group B
12
c:
,g
Pre
.l!! .E
g
:::> ~ 4 :::;:
2
Pre
Pas1
Group A Fig. 2. Adductor pollicis function.
Pre
Post
Group B
286
WHITTAKER, RYAN, BUCKLEY, AND ROAD
TABLE 4 INDIVIDUAL RESPIRATORY MUSCLE FUNCTION DATA IN REFED AND SHAM·FED PATIENTS BEFORE AND AFTER REFEEDING Patient Number Group A:j::j: 1 2 3 4 5 6 Mean ± SE Group B§§ 7 8 9 10 Mean ± SE P Value"
Weight t. (kg)
MEPt
MIP' Pre'
Post"
2.7 1.9 3.0 3.3 -1.0 4.7 2.4 0.8
62 91 93
84
43 72 68 9
107 78 40 113 81 11
0.5 -0.7 -1.2 -1.0 -0.6 0.4 0.02
34 102 27 20 46 19
27 105 27 30 47 19
44
t.tt
62
0 -7 14 34 -3 41 13 8 -7 3 0 10 1.5 3.5 0.31
PM Peak:j:
PM Mean§
Pre
Post
t.
Pre
Post
t.
Pre
Post
t.
98 99 71 107 91 92 5
104 114 134 129 123 150 126 6
6 15 51 58 16 59 34 10
30 48 70 21 13 30 35 9
54 69 91 42 30 36 54 9
24 21 21 21
12.1 16.8 18.1 5.4 4.0 11.4 11.3 2.3
18.8 33.5 45.6 16.1 11.4 15.4 23.5 5.4
6.7 16.7 27.5 10.7 7.4 4.0 11.4 2.9
48 129 54 72 76 16
39 139 52 88 80 23
-9 10 -2 16 4
35 30 19 19 26 4
13.4 12.7 6.7 4.4 9.3 2.2
12.7 16.1 6.0 4.0 9.7 2.9
-0.7 3.4 -0.7 -0.4 0.4 0.1 0.03
83
6
27 62
17 15 30 11
0.05
17 6 18 2 -8 32 -2 -4 5 10 0.14
• MIP: maximal inspiratory pressure. t MEP: maximal expiratory pressure. PM Peak: peak inspiratory pressure during threshold loading test. § PM Mean: mean inspiratory pressure during threshold loading test. , Pre: value before feeding . •• Post: value alter feeding. tt .
