Physiologic evidence for high-frequency chest wall oscillation and positive expiratory pressure breathing in hospitalized subjects with cystic fibrosis.

In people with cystic fibrosis (CF), the most common lethal inherited disease affecting people of Caucasian ethnicity, the effective loosening and removal of airway mucus are crucial to enhanced life expectancy and decreased morbidity. (1) Infected airway secretions contain proteases that destroy lung tissue. (2) Peripheral airways, which are less than 2 mm in diameter, lose their stability secondary to lung tissue destruction and tend to collapse, trapping air and mucus. (3) The collapse of smaller peripheral airways creates areas of nonhomogeneous ventilation distribution. (4) As lung function continues to deteriorate, abnormalities in ventilation distribution worsen, leading to ventilation-perfusion mismatching, hypoxemia, and pulmonary hypertension (5) and eventually respiratory failure and death. (1,2,5) Respiratory failure accounts for greater than 80% of CF-related deaths in the United States. (6) Understandably, airway clearance techniques (ACTs) are critical components of the daily regimen of care performed by people with CF.

Putative goals of ACTs are to decrease airway obstruction and airflow limitation and to improve ventilation distribution through the mobilization and removal of airway mucus. (7) Two independent ACTs, high-frequency chest wall oscillation (HFCWO) and positive expiratory pressure (PEP) breathing, have been shown to be effective at loosening and removing airway mucus in hospitalized people with CF. (8-10) Mucus weight was greater after HFCWO than after traditional airway clearance interventions involving postural drainage and manual percussion and vibration techniques (CPTs) when 29 subjects received each intervention 3 times a day on alternating days for 4 days. (10) In contrast, mucus weights with HFCWO (8,9) and PEP breathing (9) were similar to sputum weights with CPTs during (8,9) and up to 24 hours after (8) treatment. Thus, both HFCWO and PEP breathing are effective at removing airway mucus; however, information on the concomitant effects of these 2 ACTs on ventilation distribution is scant.

To date, there have been few reports examining the effects of ACTs on ventilation distribution or gas mixing, even though it has been assumed that ACTs promote improvements in ventilation distribution. (7,11) Arens et al (8) assessed the phase III alveolar slope during the single-breath nitrogen (N9) washout test to discover that overall ventilation distribution improved equally for HFCWO and CPT treatment groups. Recently, Darbee and colleagues (12) showed that a single PEP breathing treatment improved gas mixing, a measure of the extent to which an inspired volume of gas mixes with gas already present in the lungs, without improving ventilation distribution.

Because of the limited information regarding the effects of HFCWO and PEP breathing on ventilation distribution, it is unclear whether one of these techniques is more efficacious than the other. We decided to compare the physiologic effects of HFCWO and PEP breathing on ventilation distribution and gas mixing for people with CF by using a single-breath inert gas technique. The single-breath inert gas technique has the capacity to provide information about ventilation distribution and gas mixing. We wanted to determine the efficacy of each airway clearance intervention and to determine whether the physiologic effects of each intervention would differ in subjects during an acute phase of pulmonary disease exacerbation (within 48 hours of hospital admission) versus a subacute phase of exacerbation (within 48 hours of hospital discharge). For this study, we investigated a group of subjects who had moderate to severe obstructive disease and who were undergoing treatment for an acute exacerbation of their CF-related lung infections. Therefore, the purpose of this investigation was to examine the physiologic effects of HFCWO and PEP breathing on ventilation distribution, gas mixing, lung volume, expiratory airflow, and arterial blood oxyhemoglobin saturation (Sp[O.sub.2]) for subjects with CF at hospital admission and discharge.

Method

Subjects

From a pool of 196 subjects who had CF and who were monitored at the University of Kentucky Cystic Fibrosis Center, 112 subjects experienced 232 hospitalizations for treatment of pulmonary exacerbations during the 18-month study period. Experimental procedures were explained to 41 subjects who had CF, who were admitted to the University of Kentucky Medical Center Hospital, and who met study inclusion criteria. Fifteen subjects with CF documented by a sweat test (13) agreed to participate and were enrolled in the study.

Subjects were eligible to participate if they were being hospitalized for treatment of an acute exacerbation of their CF-related chronic obstructive lung disease, were able to perform lung function testing according to standard guidelines, (14) were at least 7 years of age, and were regarded as medically stable by their primary CF physician. Subjects who had a history of pneumothorax were excluded for safety reasons related to breathing against resistance. No subjects were on prescribed daytime oxygen use at the time of the study. All subjects performed HFCWO on an outpatient basis 1 to 3 times daily before admission, and no subjects performed daily PEP breathing. All subjects who had CF and who were admitted to the hospital for treatment of pulmonary exacerbations performed HFCWO 3 times daily whether or not they were study participants.

Subject characteristics at study entry are shown in Table 1. Informed consent was obtained from all study volunteers and from parents (for subjects younger than 18 years of age) before participation.

Experimental Protocol

Each subject visited the lead author's pulmonary laboratory located in the College of Health Sciences at the University of Kentucky on 4 separate days. Single-breath inert gas tests and spirometry were performed before and immediately after HFCWO or PEP breathing treatments to assess ventilation distribution, gas mixing, and lung function on 2 separate successive days within 48 hours of hospital admission and again on 2 separate successive days within 48 hours of hospital discharge. Because most subjects are discharged at the end of their intravenous antibiotic treatment, we were able to identify the likely discharge date and used the 2 days before discharge to study the subjects in the subacute phase.

Subjects were assigned to treatment order by numbering them consecutively, 1 through 15, at study entry. On the basis of a coin toss at admission, subject 1 and all odd-numbered subjects were randomly assigned to perform HFCWO on day 1 and PEP breathing on day 2, and even-numbered subjects performed PEP breathing on day 1 and HFCWO on day 2. At discharge, subjects received treatment in the order opposite the treatment order at admission. Three subjects were discharged while continuing to receive intravenous antibiotics. For these 3 subjects, final testing was performed within 48 hours of the time at which intravenous antibiotics were discontinued. Subjects received an average of 10 days (range=7-14) of intravenous antibiotics. The average length of hospital stay was 11 days (range=9-15).

Interventions

For HFCWO, a model 103 Vest airway clearance system* was used while subjects were seated upright in a chair. (15) Subjects were fitted with a nonstretch, vinyl-coated polyester inflatable vest, which was worn over the entire thorax as shown in Figure 1. The vest was closed at the front with 3 buckles and fit snugly when subjects inhaled to total lung capacity (TLC). (15) Two ports, located on the front panels of the vest, were connected to the air-pulse generator via 2 large-bore tubes. The air-pulse generator consisted of an air blower that delivered air pressure to the inflatable vest and a rotary valve that produced alternating positive and zero pressures. (15) During treatment, the vest was inflated so that a background air pressure was created when the setting of 5 was selected from a scale ranging between 1 and 10 (arbitrary units). A middle range for background pressure was selected because, although the volume of inspired air during spontaneous breathing has been shown to be higher during high background pressure than during low background pressure, high background pressure is known to lower end-expiratory lung volume (EELV) more than low background pressure for the same oscillation frequency. (16) In addition, there were no differences in the volumes of expired air between the low and the high background pressure settings during spontaneous breathing for the same oscillation frequency. (16)

Oscillation frequency was set at 10 Hz for the initial 15 minutes and was increased to 15 Hz during the last 15 minutes of treatment. (16) Oscillation frequencies of 10 and 15 Hz were selected because these 2 frequencies were previously identified to generate peak airflow during spontaneous breathing for subjects with CF. (16) We limited the oscillation frequency to 15 Hz because airflow during spontaneous tidal volume breathing, which is the movement of air in and out of the lungs during resting, has been shown to decrease during expiration as oscillation frequency increases beyond 15 Hz. (16)

A hand-foot switch, controlled by the subjects, activated and deactivated oscillations. Every 5 minutes, subjects deactivated the oscillations, inhaled to TLC, activated the oscillations at 10 or 15 Hz, and performed a forced expiratory maneuver to a low lung volume, which resulted in coughing, in order to clear airway secretions. Each subject determined how many forced expiratory maneuvers, followed by coughing, were necessary in order to clear airway secretions at each 5-minute interval. Six cycles consisting of 5 minutes of treatment followed by forced expiratory maneuvers and coughing were completed by all subjects.

Low PEP was generated by breathing through a face mask fitted with a 1-way valve, an expiratory resistor, and a pressure manometer (17) as shown in Figure 2. During low-PEP breathing, a resistor with an internal diameter that provided a steady PEP of 10 to 20 cm [H.sub.2]O during expiration, while the subject was breathing through the PEP mask, was used. (12,17) On the basis of clinical observations made by the lead author, low PEP as opposed to high PEP (>20 cm [H.sub.2]O) was selected for use in the present study because high-PEP breathing is not well tolerated and cannot easily be performed by people experiencing an acute pulmonary exacerbation. The pressure manometer provided visual feedback so that a steady PEP of 10 to 20 cm [H.sub.2]O was maintained during tidal exhalations and exhalations were slightly active. (12,17) Expiratory resistor internal diameters and mean sustained expiratory pressures generated during low-PEP breathing for each subject are shown in Table 2. Subjects breathed against the expiratory resistance for 8 breaths, removed the PEP mask, and were encouraged to perform a forced expiratory maneuver to a low lung volume, which resulted in coughing, in order to clear airway secretions. Each subject determined how many forced expiratory maneuvers, followed by coughing, were necessary in order to clear airway secretions. A total of 8 to 10 cycles consisting of 8 breaths were performed over 30 minutes.

Subjects inhaled an aerosolized solution containing 0.5 mL of albuterol and normal saline by using a PARI Master nebulizer ([dagger]) during the HFCWO and PEP breathing treatment interventions. The nebulizer was interfaced with the inspiratory port of the one-way valve during PEP breathing so that the aerosol was inhaled through the PEP mask. (18,19) Subjects held the nebulizer in their hand during HFCWO treatment. Nebulized albuterol was administered during both ACTs in an effort to duplicate treatments routinely performed by our subjects during hospitalizations and at home. Subjects at our CF center perform bronchodilator therapy during HFCWO treatment.

Measurements

The distribution of ventilation (phase III [N.sub.2] slope data expressed as percentages of predicted values) and gas mixing (dilution index values expressed at an absolute lung volume [D[I.sub.VL]]) were measured by use of a single-breath inert gas test (20) in accordance with standard guidelines. (21) Subjects were prompted to perform a slow inhalation of a test gas mixture containing 5% helium (He), 5% sulfur hexafluoride (S[F.sub.6]), and 90% oxygen gases, from residual volume (RV) to TLC, followed by a slow controlled expiration back to RV. Figure 3 shows the distribution of ventilation depicted by single-breath washout curves of exhaled [N.sub.2] gas concentrations plotted against exhaled lung volume for a subject without CF (top panel) and a subject with CF (bottom panel).

A regression test was performed on the [N.sub.2] data versus the lung volume data between the onset of phase III and the onset of phase IV of the single-breath curve. (20) The slope of the regression line represented the N2/volume slope of phase III, which served as a measure of the uniformity of the distribution of ventilation and which was the result of incomplete gas mixing. (20,22) The closer the phase III slope was to horizontal, the more uniform was the distribution of ventilation. Upward sloping of the line away from horizontal represented an increase in the phase III slope and a poorer distribution of ventilation. Phase III alveolar slope data for expired [N.sub.2] were expressed as percentages of predicted values. (23) The greater the percentage of the predicted value for the [N.sub.2] slope, the poorer the distribution of ventilation.

Expired He, [N.sub.2], and S[F.sub.6] concentrations were expressed as D[I.sub.VL] (12,20,22) to determine whether there were any changes in expired gas concentrations after treatment. Higher D[I.sub.VL] for He, [N.sub.2], and S[F.sub.6] gases after treatment than before treatment would indicate improved mixing of the inspired test gas with gas already present in the lungs.

Vital capacity and expiratory flow measurements were obtained by simple spirometry according to standard methods (14) by use of a MedGraphics PC SpiroCard interfaced with the Office Medic software programs ([double dagger]) and a computer. The measurements were expressed as percentages of predicted values. (24,25)

Noninvasive, continuous-pulse oximetry (Nellcor N-200([section])) was performed with a finger probe attached to the right index finger to estimate Sp[O.sub.2]. Arterial blood oxyhemoglobin saturation was measured before treatment, continuously throughout treatment, and for several minutes after treatment until heart rate and Sp[O.sub.2] stabilized. Arterial blood oxyhemoglobin saturation measurements were obtained during both ACTs because HFCWO-induced decreases in EELV have the potential to decrease Sp[O.sub.2], (16) and low-PEP breathing has the potential to increase Sp[O.sub.2] during and after treatment. (12,17)

Data Analysis

To ensure that this study had adequate power, a power analysis was conducted. The results of the power analysis indicated that for a large effect size at an alpha level of .05, the study required 15 subjects per treatment group.

Group means and standard deviations were calculated for demographic data, for percentages of predicted [N.sub.2]/volume slope data, for D[I.sub.VL] for He, [N.sub.2], and S[F.sub.6], for Sp[O.sub.2], and for percentages of predicted forced vital capacity (FVC), forced expiratory volume in 1 second (FE[V.sub.1]), and forced expiratory flow between 25% and 75% vital capacity (FE[F.sub.25%-75%]) at study enrollment. To determine acute changes during treatment sessions and differences in treatment at hospital admission and discharge, a 3-way repeated-measures analysis of variance was performed (treatment [HFCWO or PEP breathing] x time [before or after treatment] x hospital status [admission or discharge]) with time and hospital status as the repeated variables (Statistica 6.1 software package ([parallel])). Analysis of Sp[O.sub.2] data was performed as for the other variables, with the exception that because Sp[O.sub.2] was measured continuously throughout the airway clearance treatment, the high and low Sp[O.sub.2] values during treatment were identified, thus yielding 4 levels for time (before, high, low, and after) but 2 levels for the other variables. Significant F ratios were followed up with Tukey honestly significant difference post hoc methods to identify specific differences. Significance was set at a P value of <.05.

Results

Effects of Airway Clearance on Pulmonary Function at Hospital Admission and Discharge

Pulmonary function was measured before and after airway clearance treatments at admission to and discharge from the hospital (Tab. 3). At admission to the hospital, pulmonary function values measured before airway clearance treatments were not different on the 2 testing days (Tab. 3).

Airway clearance treatments were associated with changes in FVC and FE[V.sub.1]. Specifically, a significant interaction between time and hospital status was observed for FVC. This interaction demonstrated that both HFCWO and PEP breathing resulted in average improvements in FVC of 13% (P<.0002) during the acute stage of exacerbation (admission) but not during the subacute stage (discharge) (Tab. 3). A significant main effect of time was observed for FE[V.sub.1], indicating that during both the acute stage of exacerbation and the subacute stage, HFCWO and PEP breathing treatments resulted in improvements in FE[V.sub.1] (P<.0002) (Tab. 3).

The airway clearance treatments at admission and discharge had no effect on FE[V.sub.1]/FVC and FE[F.sub.25%-75%]. Additionally, there were no differences in the responses of any of the pulmonary function measures between the HFCWO and the PEP breathing treatments at either admission or discharge (Tab. 3).

Effects of Airway Clearance on Ventilation Distribution and Gas Mixing at Hospital Admission and Discharge

Ventilation distribution and gas mixing were measured before and after airway clearance treatments during both the acute stage (admission to the hospital) of exacerbation and the subacute stage (just before discharge from the hospital) (Tab. 4, Figs. 4 and 5). At admission to and discharge from the hospital, the values for N2 slope, which provides an index of overall ventilation distribution homogeneity, were not different before either airway clearance treatment (Tab. 4, Fig. 4).

[FIGURE 4 OMITTED]

A main effect of time was observed for [N.sub.2] slope, indicating that treatment with either HFCWO or PEP breathing at both admission and discharge was associated with a reduction in [N.sub.2] slope (P=.011) (Tab. 4, Fig. 4). This reduction indicates that ventilation was more uniformly distributed throughout the lungs after either airway clearance treatment during both the acute exacerbation stage and the subacute stage (Tab. 4, Fig. 4).

The D[I.sub.VL] for the 3 test gases (He, [N.sub.2], and S[F.sub.6]) measure how well an inspired test gas mixes with gas already present in the lungs. At admission to and discharge from the hospital, the D[I.sub.VL] for He, [N.sub.2], and S[F.sub.6] were not different before either airway clearance treatment (Tab. 4, Fig. 5).

[FIGURE 5 OMITTED]

There were main effects of time for the D[I.sub.VL] for He, [O.sub.2], and S[F.sub.6] (Tab. 4, Fig. 5). This finding indicates that treatment with either HFCWO or PEP breathing at both admission and discharge was associated with improvements in gas mixing. The D[I.sub.VL] for He, [N.sub.2], and S[F.sub.6] increased, on average, 8% (P<.0004), 9% (P<.0002), and 10% (P<.0003), respectively (Tab. 4, Fig. 5).

Effects of Airway Clearance on Sp[O.sub.2] at Hospital Admission and Discharge Arterial blood oxyhemoglobin saturation was measured before, during, and after airway clearance treatments at both admission to and discharge from the hospital. There were no differences in Sp[O.sub.2] before either treatment at admission to and discharge from the hospital (Fig. 6).

[FIGURE 6 OMITTED]

A significant interaction between time and treatment was observed for Sp[O.sub.2]. This interaction demonstrated that during both the acute stage of exacerbation and the subacute stage, Sp[O.sub.2] responses differed between HFCWO and PEP breathing airway clearance treatments. Specifically, HFCWO resulted in decreases in Sp[O.sub.2] during treatment at both admission and discharge (P<.00004) (Fig. 6). Immediately after treatment, Sp[O.sub.2] returned to pretreatment levels. Conversely, airway clearance with PEP breathing at both admission and discharge was associated with increases in Sp[O.sub.2] during treatment (P<.00004) (Fig. 6). These increases were not sustained after treatment.

Discussion

In this investigation, we evaluated the physiologic responses to 2 airway clearance interventions, HFCWO and low-PEP breathing, in subjects who had moderate to severe CF and who required hospitalization because of an exacerbation of their chronic lung disease. The goal of this investigation was to use the physiologic response data obtained during the application of HFCWO and low-PEP breathing to improve the prescription of airway clearance treatment in this patient population. The most important finding in this investigation was that both the HFCWO and the PEP breathing interventions were similarly efficacious in improving pulmonary function, ventilation distribution, and gas mixing in subjects who had CF and who were experiencing an exacerbation. However, distinct differences in the underlying physiologic mechanisms between the 2 treatments were identified, and these differences may affect clinical decision making during prescription of an airway clearance treatment intervention in this patient population.

Effects on Sp[O.sub.2]

The most striking difference between the HFCWO and the PEP breathing treatments was the observation that the HFCWO treatment was associated with decreases in Sp[O.sub.2] during treatment, whereas the PEP breathing treatment produced modest, but significant, increases in Sp[O.sub.2] during treatment (Fig. 6). When subject Sp[O.sub.2] values are 94% or above, values that correspond to partial pressures for oxygen in the low to middle 70s, as we observed in our study subjects, the decreases in Sp[O.sub.2] with HFCWO are likely to be clinically insignificant. However, the observed decreases in Sp[O.sub.2] may be important to note, because the lower a subject's pretreatment Sp[O.sub.2], the more likely desaturation will occur and the more likely the subject will experience hypoxia during HFCWO treatment.

Among reports of HFCWO use by subjects with CF, (8-10,15) only Arens et Al (8) measured and reported subject Sp[O.sub.2] responses. In contrast to the findings in the present study, improvements in Sp[O.sub.2] occurred during and up to 1 hour after treatment on days 7 and 14 of hospitalization for a pulmonary exacerbation in both the CPT and the HFCWO groups, but there were no differences in Sp[O.sub.2] between the 2 treatment interventions. (8) Interestingly, at enrollment, our subjects had less central and peripheral airway obstruction than subjects in the study of Arens et al, (8) as indicated by the percentages of the predicted FE[V.sub.1] ([bar.X] [+ or -] SD: 55 [+ or -] 21 versus 33.8 [+ or -] 2.4) and FE[F.sub.25%-75%] (37 [+ or -] 29 versus 13.5 [+ or -] 2.0) values, yet the mean Sp[O.sub.2] was slightly lower in our study subjects. In addition, our study subjects exhibited more ventilation distribution inhomogeneity, as indicated by the percent predicted [N.sub.2] slope values ([bar.X] [+ or -] SD: 914 [+ or -] 567) in the present study compared with the percent predicted [N.sub.2] slope values in the study of Arens et al (8) ([bar.X] [+ or -] SD: 601 [+ or -] 75).

Improvements in Sp[O.sub.2] during low-PEP breathing in the present study were similar to improvements in Sp[O.sub.2] reported by other authors. (17) Arterial blood oxyhemoglobin saturation increased after a 20-minute low-PEP breathing treatment, peaked at 35 minutes after treatment, and never dropped below baseline for a group of subjects who had CF and who had considerably more central airway obstruction, as indicated by an FE[V.sub.1] that was 34% of the predicted value, (17) than the subjects in our study, who had an FE[V.sub.1] that was 55% of the predicted value.

Different Mechanisms Leading to Improved Ventilation Distribution and Gas Mixing

Although there were no differences between the effects of the HFCWO and PEP breathing treatments on ventilation distribution and gas mixing in our study, there were differences in the underlying physiologic mechanisms of the 2 treatments that led to the improvements, and these differences warrant further discussion. The decreases in the phase III alveolar slopes after both airway clearance treatments resulted from more complete gas mixing and led to improvements in ventilation distribution. The increases in expired gas concentrations for He, [N.sub.2], and S[F.sub.6] after the treatments reflected the improved efficiency with which an inspired gas mixture containing a test gas mixed with gas already present in the lungs. (26)

HFCWO mechanisms. Although the exact mechanism for improvements in gas mixing and ultimately ventilation distribution during HFCWO is speculative, several theories have been suggested. (27) One postulate is that pendelluft may increase during HFCWO. (28-30) Pendelluft is the movement of air between neighboring lung units (the functional unit of gas exchange, which consists of structures containing alveoli that reside distal to the end of the terminal bronchiole) that have different time constants. (31) The time constant for a lung unit is defined as the time required for a lung unit to empty or fill and is equal to the product of its resistance to airflow and its compliance. (31) Thus, air movement within the lungs is dependent on airway diameter and tissue elasticity. (31) Time constants are slow when lung units have low distensibility and high airway resistance, such as in CF-related lung disease. Parallel lung units, present in the same lung region, normally fill and empty at about the same rates. (31) However, in obstructive lung disease, parallel lung units frequently fill and empty at different rates. (4,12) During HFCWO, pendelluft may increase the recirculation of air, thereby increasing alveolar ventilation for previously closed or underventilated lung units. (27,29) The results are improvements in gas mixing and homogenization of expired gas concentrations from these neighboring lung units. (29,30)

High-frequency chest wall oscillation delivered at 10 and 15 Hz has been shown to decrease functional residual capacity (FRC) and to increase tidal volume and airflow in subjects who were healthy (16,32) and in subjects with obstructive lung disease. (16,32,33) Functional residual capacity, or EELV (the volume of gas remaining in the lungs at the end of expiration), decreased during HFCWO because of the positive pressure applied over the chest wall. (16,28,32,33) End-expiratory lung volume, a dynamic and constantly changing breath-by-breath lung volume, is commonly elevated in subjects with chronic obstructive pulmonary disease, such as CF secondary to air trapping. (3,34) This elevation of EELV provides an important physiologic mechanism by which small-airway closure is minimized and gas exchange is sustained in subjects with chronic obstructive pulmonary disease. (2)

Although decreases in FRC have been shown to accompany increases in tidal volume and alveolar ventilation at low oscillation frequencies, (28) HFCWO delivered at high oscillation frequencies potentially leads to small-airway closure and deterioration in gas exchange for subjects who already have expiratory airflow limitations. (3,16,28,34) During HFCWO intervention, we purposefully selected lower oscillation frequencies, of 10 and 15 Hz, from a range of 5 to 25 Hz, and a midrange background pressure in order to maximize oscillated airflow (16) and oscillated tidal volume (16) and to minimize reductions in EELV. (16,33) In this regard, Jones et al (16) reported reductions in EELV to 90% of baseline pre-HFCWO values. Although reductions in EELV were observed, no deterioration in Sp[O.sub.2] occurred during or after HFCWO treatment because the subjects were breathing 50% supplemental oxygen. (16) The authors speculated, however, that hypoxia could occur for subjects with CF during a 30-minute HFCWO treatment while breathing room air because of the reductions in EELV associated with HFCWO. (16) Therefore, the observed decreases in Sp[O.sub.2] during HFCWO treatment in the present study may be important because the lower the subject's pretreatment Sp[O.sub.2], the more likely desaturation is to occur.

In light of the small decreases in Sp[O.sub.2] during HFCWO, it is possible that some small-airway closure occurred and may have reduced the contribution made by small-airway inhomogeneities to the phase III alveolar slope during single-breath testing and caused the phase III slope to move downward toward horizontal. (28) However, the effects of any decreases in lung volume likely were offset by the effects of improved gas mixing during HFCWO, which preserved gas exchange in our subjects in both the admission and the discharge HFCWO treatment sessions.

PEP breathing mechanisms. Low-PEP breathing also was associated with marked improvements in ventilation distribution (Tab. 4, Fig. 4) and gas mixing (Tab. 4, Fig. 5), but through physiologic mechanisms different from those of HFCWO. Improvements in gas mixing after low-PEP breathing in the present study were similar to findings reported previously from our laboratory. (12) Our data suggest that the improvements in gas mixing and FVC likely were attributable to a decrease in the partial or complete obstruction of smaller, peripheral airways. The low expired He, [N.sub.2], and S[F.sub.6] concentrations (Tab. 4, Fig. 5) and the [N.sub.2] phase III alveolar slope values (Tab. 4, Fig. 4) measured before the interventions reflected heterogeneity of time constants within the peripheral airways of our subjects with CF, confirming the presence of fast and slow time constants. During 30 minutes of resistance breathing during PEP treatment, peripheral airways were dilated, facilitating the ongoing exhalation of RV gas. (12,35) The ongoing exhalation of RV gas generated airflow through the smaller airways (12,35) and facilitated faster filling and emptying times for lung units, but particularly for slow lung units. (12,35) Gas mixing improved because time constants for lung units, which are dependent on airway diameter and tissue elasticity, became faster and additional gas volume could be exhaled during the prolonged expiration. (12,35) This additional exhaled gas volume led to the increased FVC. We attribute the improvements in Sp[O.sub.2] after PEP breathing to improvements in gas mixing, in which a larger volume of an inspired gas mixture containing a test gas mixed with a smaller volume of gas already present in the lungs. (12,22)

[FIGURES 4-5 OMITTED]

Decreases in the phase III alveolar slopes for [N.sub.2] gas indicated that there were improvements in ventilation distribution after PEP breathing treatment. This finding was different from findings reported previously from our laboratory. (12) Previously, phase III alveolar slopes for He gas were not found to be different after 20 minutes of PEP breathing. (12) We attribute this finding to the high diffusivity of He, which permitted the gas to diffuse into poorly ventilated regions before PEP breathing treatment and therefore minimized our ability to measure any changes in ventilation distribution that might have occurred as a result of the PEP breathing treatment. In the present study, we used heavier, resident [N.sub.2] gas which, in contrast to He gas, had unequal concentrations throughout the lungs due to poorly ventilated regions before treatment. During PEP breathing treatment, airways were opened and [N.sub.2] gas likely diffused into previously closed regions as a result of breathing against positive pressure. (12,35) The improvements in the homogeneity of ventilation distribution were reflected by the decreased [N.sub.2] phase III alveolar slopes after PEP breathing treatment.

Role of Sputum Collection in Studies of Airway Clearance Treatments

Sputum collection was not performed in this study. Based on our previous findings, (12) we do not believe that sputum collection, within the context of repeated pulmonary function testing and airway clearance treatment interventions, reflects only the efficacy of an ACT. We believe that deep breathing, coughing, and the prolonged, forceful exhalations performed during pulmonary function testing facilitated the loosening and removal of sputum, which led to the cumulative increases in sputum amounts observed during control, low-PEP, and high-PEP conditions in our previous study. (12) We believe that, in the evaluation of an airway clearance treatment, it is important to consider other indexes reflecting airway clearance, such as Sp[O.sub.2], ventilation distribution, and gas mixing, which are physiologically meaningful and have important implications for physical therapist clinical practice.

Limitations

In the present study, the subjects used bronchodilator therapy during both airway clearance treatments. Because PEP breathing transmits a back pressure to the airways, it is conceivable that the administration of the bronchodilator will be altered during this treatment and thus may affect results. However, a previous study that examined the effect of PEP breathing on the delivery of metered-dose inhaled albuterol showed that there were no differences in total drug dosages with PEP breathing and without PEP breathing. (18) Additionally, in another study, (19) an inhaled bronchodilator was administered with PEP breathing and without PEP breathing; the authors found that peak expiratory flow increased similarly with the administration of the bronchodilator with PEP breathing and without PEP breathing. (19) These findings suggest that the administration of albuterol with PEP breathing or without PEP breathing is unlikely to be a major confounding variable.

Conclusion

The results of this investigation indicated that both HFCWO and low-PEP breathing airway clearance interventions were similarly efficacious in improving pulmonary function, ventilation distribution, and gas mixing for subjects who had CF and who were experiencing an acute exacerbation of their pulmonary disease. These improvements in ventilation distribution and gas mixing were associated with small, yet significant, increases in Sp[O.sub.2] during PEP breathing. Alternatively, the use of HFCWO was associated with improvements in ventilation distribution and gas mixing and small decreases in Sp[O.sub.2]. These findings led to our recommendation that, for people who have moderate to severe CF-related lung disease and who are experiencing an acute exacerbation requiring hospital admission, it is essential to monitor Sp[O.sub.2] during airway clearance treatments. People who have low pretreatment Sp[O.sub.2] levels could desaturate to unacceptable levels during HFCWO therapy. These individuals may benefit from the use of low-PEP breathing therapy during an acute exacerbation of their lung disease. Both HFCWO and low-PEP breathing interventions were associated with improvements in ventilation distribution and gas mixing. Therefore, we recommend either HFCWO or low-PEP breathing for people stable CF-related lung disease because, once the acute lung infection has been treated and lung function improves, it may be possible to maintain Sp[O.sub.2] at acceptable levels with either treatment and still experience the benefits of improved gas mixing. For people in whom HFCWO results in desaturation and who find that low-PEP breathing is too fatiguing to perform during the early stages of treatment for an acute exacerbation, CPT may be indicated because it has been shown that CPT and low-PEP breathing are equally effective at secretion removal in subjects with CF. (9) For people with stable lung function, both HFCWO and low-PEP breathing are associated with similar improvements in pulmonary function, ventilation distribution, and gas mixing; thus, patient preference should be considered in the prescription of a specific airway clearance treatment.

References

(1) Staab D. Cystic fibrosis: therapeutic challenge in cystic fibrosis children. Eur J Endocrinol. 2004;151(suppl 1):S77-S80.

(2) Iannuzzi MC. Cystic fibrosis: genetics. In: Davis PB, ed. Cystic Fibrosis. New York, NY: Marcel Dekker Inc; 1993:1-27.

(3) Desmond KJ, Coates AL, Martin JG, Beaudry PH. Trapped gas and airflow limitation in children with cystic fibrosis and asthma. Pediatr Pulmonol. 1986;2:128-134.

(4) Brown JS, Gerrity TR, Bennett WD. Effect of ventilation distribution on aerosol bolus dispersion and recovery. J Appl Physiol. 1998;85: 2112-2117.

(5) Dantzker DR, Patten GA, Bower JS. Gas exchange at rest and during exercise in adults with cystic fibrosis. Am Rev Respir Dis. 1982;125: 400-405.

(6) Cystic Fibrosis Foundation Patient Registry: 2003 Annual Data Report to the Center Directors. Bethesda, Md: Cystic Fibrosis Foundation; 2004.

(7) Tecklin JS. The patient with airway dysfunction: preferred practice pattern 6C. In: Irwin S, Tecklin JS, eds. Cardiopulmonary Physical Therapy: A Guide to Practice. St Louis, Mo: Mosby Inc; 2004:285-329.

(8) Arens R, Gozal D, Omlin KJ, et al. Comparison of high-frequency chest compression and conventional chest physiotherapy in hospitalized patients with cystic fibrosis. Am J Respir Crit Care Med. 1994;150:1154-1157.

(9) Braggion C, Cappelletti LM, Cornacchia M, et al. Short-term effects of three chest physiotherapy regimens in patients hospitalized for pulmonary exacerbations of cystic fibrosis: a crossover randomized study. Pediatr Pulmonol. 1995;19:16-22.

(10) Kluft J, Beker L, Castagnino M, et al. A comparison of bronchial drainage treatments in cystic fibrosis. Pediatr Pulmonol. 1996;22: 271-274.

(11) Sciaky A, Stockford J, Nixon E. Treatment of acute cardiopulmonary conditions. In: Hillegass EA, Sadowsky HS, eds. Essentials of Cardiopulmonary Physical Therapy. Philadelphia, Pa: WB Saunders Co; 2001:647-675.

(12) Darbee JC, Ohtake PJ, Grant BJB, Cerny FJ. Physiological evidence for the efficacy of positive expiratory pressure as an airway clearance technique in patients with cystic fibrosis. Phys Ther. 2004;84:524-537.

(13) Gibson LE, Cooke RE. A test for concentration of electrolytes in sweat in cystic fibrosis of the pancrease utilizing pilocarpine by iontophoresis. Pediatrics. 1959;23:545-549.

(14) American Thoracic Society. Standardization of spirometry: 1994 update. Am J Respir Crit Care Med. 1995;152:1107-1136.

(15) Warwick WJ, Hansen LG. The long-term effect of high-frequency chest compression therapy on pulmonary complications of cystic fibrosis. Pediatr Pulmonol. 1991;11:265-271.

(16) Jones RL, Lester RT, Brown NE. Effects of high-frequency chest compression on respiratory system mechanics in normal subjects and cystic fibrosis patients. Can Respir J. 1995;2:40-46.

(17) Falk M, Kelstrup M, Andersen JB, et al. Improving the ketchup bottle method with positive expiratory pressure, PEP, in cystic fibrosis. Eur J Respir Dis. 1984;65:423-432.

(18) Rau JL, Torniainen M. Combining a positive expiratory pressure device with a metered-dose inhaler reservoir system using chlorofluorocarbon albuterol and hydrofluoroalkane albuterol: effect on dose and particle size distributions. Respir Care. 2000;45:320-326.

(19) Christensen EF, Norregaard O, Dahl R. Nebulized terbutaline and positive expiratory pressure in chronic obstructive pulmonary disease. Pneumologie. 1991;45:105-109.

(20) Fowler WS. Lung function studies, III: uneven pulmonary ventilation in normal subjects and patients with pulmonary disease. J Appl Physiol. 1949;2:283-299.

(21) Anthonisen NR, Danson J, Robertson PC, Ross WRD. Airway closure as a function of age. Respir Physiol. 1969;8:58-65.

(22) Van Liew HD, Mahajan RK. Display of the alveolar plateau of single-breath tests in "dilution index" format. J Appl Physiol. 1989;67: 1699-1703.

(23) Buist AS, Ross BB. Quantitative analysis of the alveolar plateau in the diagnosis of early airway obstruction. Am Rev Respir Dis. 1973; 108: 1078-1087.

(24) Knudson RJ, Slatin RC, Lebowitz MD, Burrows B. The maximal expiratory flow-volume curve: normal standards, variability, and the effects of age. Am Rev Respir Dis. 1976;113:587-600.

(25) Knudson RJ, Lebowitz MD, Holberg CJ, Burrows B. Changes in the normal expiratory flow-volume curve with growth and aging. Am Rev Respir Dis. 1983;127:725-734.

(26) Olgilvie CM, Forster RE, Blackemore WS, Morton JW. A standardized breath holding technique for the clinical measurement of diffusion capacity of the lung for carbon monoxide. J Clin Invest. 1957;36: 1-17.

(27) Chang HK. Mechanisms of gas transport during ventilation by high-frequency oscillation. J Appl Physiol. 1984;56:553-563.

(28) Harf A, Zidulka A, Chang HK. Nitrogen washout during tidal breathing with superimposed high-frequency chest wall oscillation. Am Rev Respir Dis. 1985;132:350-353.

(29) Lehr J, Drazen JM, Westerman PA, Zatz SL. Regional expansion of excised dog lungs during high frequency ventilation [abstract]. Fed Proc. 1982;41:1747.

(30) Isabey D, Harf A, Chang HK. Alveolar ventilation during high-frequency oscillation: core dead space concept. J Appl Physiol. 1984;56: 700 -707.

(31) Otis AB, McKerrow CB, Bartlett RA, et al. Mechanical factors in the distribution of pulmonary ventilation. J Appl Physiol. 1956;8:427-443.

(32) Perry RJ, Man GCW, Jones RL. Effects of positive end-expiratory pressure on oscillated flow rate during high-frequency chest compression. Chest. 1998;113:1028-1033.

(33) Dosman CF, Zuberbuhler PC, Tabak JI, Jones RL. Effects of positive end-expiratory pressure on oscillated volume during high-frequency chest compression in children with cystic fibrosis. Can Respir J. 2003; 10:94-98.

(34) Featherby EA, Weng TR, Crozier DN. Dynamic and static lung volumes, blood gas tensions, and diffusing capacity in patients with cystic fibrosis. Am Rev Respir Dis. 1970;102:737-749.

(35) Oberwaldner B, Evans J, Zach M. Forced expirations against a variable resistance: a new chest physiotherapy method in cystic fibrosis. Pediatr Pulmonol. 1986;2:358-367.

([dagger]) PARI Respiratory Equipment Inc, 7493 Whitepine Rd, Richmond, VA 23237.

([double dagger]) Medical Graphics Corp, 350 Oak Grove Pkwy, St Paul, MN 55127.

([section]) Mallinckrodt Inc, a division of Nellcor-Puritan Bennet Co, 675 McDonnel Blvd, Hazelwood, MO 63042.

([parallel]) StatSoft Inc, 2300 E 14th St, Tulsa, OK 74104

JC Darbee, PT, PhD, is Assistant Professor, Department of Rehabilitation Sciences, Division of Physical Therapy, College of Health Sciences, University of Kentucky, 900 S Limestone St, Lexington, KY 40536 (USA) ([e-mail]). Address all correspondence to Dr Darbee.

JF Kanga, MD, is Professor of Pediatrics and Chief, Division of Pediatric Pulmonology, Department of Pediatrics, School of Medicine, University of Kentucky.

PJ Ohtake, PT, PhD, is Associate Professor, Department of Rehabilitation Sciences, University at Buffalo, The State University of New York, Buffalo, NY. Dr Ohtake is supported by grants from the American Lung Association and the Interdisciplinary Research and Creative Activities Fund-University at Buffalo (IRCAF).

Dr Darbee provided concept/idea/research design, data collection, project management, fund procurement, and facilities/equipment. Dr Darbee and Dr Ohtake provided writing and data analysis. Dr Kanga provided subjects. Dr Darbee and Dr Kanga provided institutional liaisons. The authors thank Hill-Rom, of St Paul, Minn, for supplying The Vest airway clearance system for this study.

Approval for this investigation was granted by the University of Kentucky Medical Institutional Review Board.

This study was funded by a grant from the Medical Center Research Fund at the University of Kentucky.

This research was presented as a platform presentation at the Combined Sections Meeting of the American Physical Therapy Association; February 23-27, 2005; New Orleans, La.

This article was received August 23, 2004, and was accepted May 24, 2005.

Table 1.
Demographics of Participants (8 Male, 7 Female)

Characteristic                    [bar.X]    SD

Age (y)                            17.5      4.2
Height (cm)                       159        9
Weight (kg)                        46.6      6.9
Body mass index (kg/[m.sup.2])     18.3      2.0

Table 2.

Expiratory Resistor Internal Diameters (IDs) and Mean Sustained
Expiratory Pressures (SEPs) Generated During Low Positive Expiratory
Pressure (PEP) Breathing

          Admission                     Discharge

Subject   Resistor ID   SEP             Resistor ID   SEP
No.       (mm)          ([H.sub.2]O])   (mm)          ([H.sub.2]O])

 1        2.5           20              2.0           20
 2        2.0           18-20           2.0           20
 3        2.0           15-18           2.0           20
 4        2.0           18-20           2.0           18-20
 5        2.0           18-20           1.5           18-20
 6        2.0           20              2.0           20
 7        1.5           20              1.5           20
 8        1.5           20              1.5           20
 9        1.5           15              1.5           18
10        1.5           18              1.5           18-20
11        2.0           20              1.5           20
12        1.5           10              1.5           15
13        2.0           15              2.0           20
14        1.5           20              1.5           20
15        1.5           18              1.5           20

Table 3.

Lung Function at Admission and Discharge (a)

                                  HFCWO

                                  Admission

                                  Pre               Post      Pre

Parameter                         [bar.X]   SD      [bar.X]   SD

FVC (% predicted)                 70        18      77 (b)    21
FE[V.sub.1] (% predicted)         54        20      60 (b)    22
FE[V.sub.1]/FVC                    0.68      0.15    0.67      0.13
FE[F.sub.25,-75%] (% predicted)   38        32      40        35

                                  HFCWO

                                  Discharge

                                  Pre

Parameter                         [bar.X]   SD      [bar.X]   SD

FVC (% predicted)                 79        20      83        19
FE[V.sub.1] (% predicted)         66        26      69 (b)    24
FE[V.sub.1]/FVC                    0.70      0.14    0.71      0.12
FE[F.sub.25,-75%] (% predicted)   47        37      47        37

                                  PEP Breatjing

                                  Admission

                                  Pre               Post

Parameter                         [bar.X]   SD      [bar.X]   SD

FVC (% predicted)                 66        18      77 (b)    21
FE[V.sub.1] (% predicted)         56        21      60 (b)    23
FE[V.sub.1]/FVC                   0.67       0.13    0.67      0.13
FE[F.sub.25,-75%] (% predicted)   37        27      37        29

                                  PEP Breatjing

                                  Discharge

                                  Pre               Post

Parameter                         [bar.X]   SD      [bar.X]   SD

FVC (% predicted)                 79        20      80        19
FE[V.sub.1] (% predicted)         64        24      66 (b)    25
FE[V.sub.1]/FVC                    0.69      0.14    0.70      0.15
FE[F.sub.25,-75%] (% predicted)   45        36      48        38

(a) HFCWO=high-frequency chest mall oscillation. PEP=low positive
expiratory pressure, Pre=before treatment, Post=after treatment,
FVC=forced vital capacity, FE[V.sub.1]=forced expirator volume in 1
second, FE[F.sub.25%-7%]=forced expi[ator,v flmc between 25% and 75%
vital capacity.

(b) Significant change between pretreatment and posttreament values.

Table 4. Single-Breath Data (a)

                      HFCWO

                      Admission           Discharge

                      Pre                 Post

Parameter             [bar.X]    SD       [bar.X]      SD

[N.sub.2] slope       922        631      843 (b)      564
  (% predicted)
D[I.sub.VL] for He      3.6        1.3      3.8 (b)      1.2
D[I.sub.VL] for
  [N.sub.2]             3.2        1.4      3.5 (b)      1.1
D[I.sub.VL] for
  S[F.sub.6]            3.6        1.5      3.9 (b)      1.3

                      PEP Breathing

                      Admission           Discharge

                      Pre                 Post

Parameter             [bar.X]    SD       [bar.X]      SD

[N.sub.2] slope       759        660      680 (b)      571
  (% predicted)
D[I.sub.VL] for He      3.7        1.5      4.0 (b)      1.4
D[I.sub.VL] for
  [N.sub.2]             3.4        1.4      3.6 (b)      1.3
D[I.sub.VL] for
  S[F.sub.6]            3.8        1.6      4.3 (b)      1.7

                      PEP Breathing

                      Admission           Discharge

                      Pre                 Post

Parameter             [bar.X]    SD       [bar.X]      SD

[N.sub.2] slope       90 (b)     522      822 (b)      500
  (% predicted)
D[I.sub.VL] for He     3.4         1.0      3.8 (b)      1.4
D[I.sub.VL] for
  [N.sub.2]            3.1         1.0      3.5 (b)      1.3
D[I.sub.VL] for
  S[F.sub.6]           3.4         1.0      3.9 (b)      1.5

                      PEP Breathing

                      Admission           Discharge

                      Pre                 Post

Parameter             [bar.X]    SD       [bar.X]      SD

[N.sub.2] slope       755        605      705 (b)      554
  (% predicted)
D[I.sub.VL] for He      3.7        1.5      3.9 (b)      1.5
D[I.sub.VL] for
  [N.sub.2]             3.4        1.4      3.7 (b)      1.4
D[I.sub.VL] for
  S[F.sub.6]            3.9        1.8      4.1 (b)      1.6

(a) HFCWO=high-frequency chest wall oscillation, PEP=low positive
expiratory pressure, Pre=before treatment, Post-after treatment,
[N.sub.2]=nitrogen, D[I.sub.VL]=dilution index salues expressed at an
absolute lung volume, He=helium, S[F.sub.6]=snllut hexaflouride.

(b) Significant change benceen preticattnent and post treatment
Values.