CORE Metadata, citation and similar papers at core.ac.uk Provided by ElsevierVol. - Publisher97 (2003) Connector 647^653

Orthopnea and inspiratory e¡ort in chronic heart failure patients

w z S. NAVA*,M.TERESA LAROVERE ,F.FANFULLA ,P.NAVALESI*,M.DELMASTRO* w AND A. MORTARA

w *Respiratory Intensive Care Unit, Fondazione S.Maugeri, Istituto Scienti¢co di Pavia, Italy Cardiology Department, Fondazione S.Maugeri Istituto Scienti¢co di Montescano, Italy, zRespiratoryDepartmentFondazioneS.Maugeri Istituto Scienti¢co di Montescano, Italy

Abstract Study objective: Orthopnea is a typical feature of patients with chronic heart failure (CHF), the factors con- tributing to it are not completely understood.We investigated changes in dyspnea and other respiratory variables, in- duced by altering posture (from sitting to supine) in 11 CHF patients (NYHA classes II^IV) and 10 control subjects. Methods and Results: We measured dyspnea (Borg scale), the diaphragm pressure time product per minute (PTPdi/m, index of metabolic consumption), and mechanical properties ofthe lung (lung compliance (C,L) and resistances (R,L).CHF patients also underwent a trial of non-invasive mechanicalventilation (NIMV) inthe supine positionin order to ascertain whether unloading theinspiratorymuscles could somehowrelieve dyspnea.While sitting the PTPdi/minwas significantly higher in CHF patients than in controls (181754 cm H2O s/min vs.96732; Po0.05). Assuming a supine position caused no major changes in controls, whereas CHF patients showed a significant worsening in dyspnea, a rise in PTPdi/ min (243797 po0.01) and R,L (4.7 71.2 cm H2O/L ssittingvs.7.972.5 supine; Po0.01) and a decrease in C,L (0.0870.02 L/cm H2Osittingvs. 0.07+0.01supine; Po0.05). Applying NIMV to supine CHF patients significantly reduced the PTPdi/min to 81742 (Po0.001).Changes in dyspnea, produced by varying position or applying NIMV, were signifi- cantly correlated with PTPdi/min (r=0.80, Po0.005 and r=0.58, Po0.01, respectively). Conclusions: CHF patients had a higher PTPdi/min than controls when sitting, and assuming a supine position induced severe dyspnea, a large rise in R,L,andareductioninC,L so that PTPdi/min increased further.Orthopnea was strongly correlated with the increased diaphragmatic effort. r 2003 Elsevier Science Ltd. Allrightsreserved. doi:10.1053/rmed.2003.1495, available online at http://www.sciencedirect.com

Keywords CHF; dyspnea; diaphragm; inspiratory muscle; non-invasive ventilation.

INTRODUCTION return might also reduce lung compliance by increasing pulmonary hydrostatic pressure and causing interstitial Orthopnea is a common feature in patients with chronic edema (3^5). Both increased air£ow resistance and de- heart failure (CHF). Surprisingly, however, little is known creased lung compliance may lead to greater energy ex- about the mechanisms underlying this symptom. A re- penditure by the diaphragm and other respiratory cent study has shown that air£ow resistance, as assessed muscles, an event which has been shown to be linked to by the oscillation technique, increases signi¢cantly in the genesis of dyspnea (6). Indeed, regardless of the spe- CHF patients, but not in normal subjects, when they as- ci¢c cause, a sudden increase in respiratory impedance sume a supine position (1). In another study, performed provoked by lying down should produce, together with on patients with acute heart failure, Duguet et al.(2)de- an abrupt rise in inspiratory e¡ort, the onset of orthop- monstrated a relationship between expiratory £ow lim- nea (2). In order to evaluate this hypothesis we measured itation and orthopnea, exaggerated by changing posture lung mechanics, inspiratory muscle e¡ort, and sensation from sitting to supine.Whether or not this mechanism is of dyspnea in the sitting and supine position in 11stable also responsible for orthopnea in stable CHF is not CHF patients and compared these results with those ob- known. Indeed, in CHF patients, increased blood venous tained in a control group of normal subjects. Further- more, we applied inspiratory pressure support plus Recieved10 June 2002, accepted on revised form11November 2002 continuous positive airway pressure to the supine CHF Correspondence should be addressed to: Stefano Nava, MD, patients to verify whether unloading the inspiratory Respiratory Intensive Care Unit, Fondazione S.Maugeri, Istituto Scienti¢co di Pavia,Via Ferrata 8, 27100 Pavia, Italy. muscles was somehow associated with a relief of dys- Fax: 0382 592 075; E-mail: [email protected] pnea. 648 RESPIRATORY MEDICINE

MATERIAL AND METHODS sured using the balloon-catheter technique (7). Pressure at the airway opening (Paw) was measured via a side Subjects port. Eleven in-patients with stable CHF with no history of Respiratory mechanics were assessed using Mead and neurological or chronic respiratory disease were re- Wittenberger’s technique (8). Inspiratory pulmonary re- cruited.Two patients were in New York Heart Associa- sistance (R,L)andelastance(E,L) were calculated by ¢tting tion functional class II, seven in class III and two in class IV. the equation of motion of a single-compartment model All of them were receiving diuretics and ACE-inhibitor using multilinear regression. drugs,whichwerecontinuedonthedayofthestudy. Dynamic PEEPi (PEEPi,dyn) was measured according They were in a phase of clinical stability and were ad- to Appendini et al.(9).The pressure time integrals of the mitted to the hospital for a medical control and even- diaphragm and the other inspiratory muscles were calcu- tually for a short period of rehabilitation. Patients with lated per breath (PTPdi/b and PTPes/b, respectively) and X-ray evidence of pulmonary edema or congestion were per minute (PTPdi/min and PTPes/min) (10). excluded from the study.Data from a control group of10 Relative changes in lung volumes were recorded by matched subjects were used for comparison. Spiro- means of inductive plethysmography. A rib cage band metric and hemodynamic data on patients and controls was placed at the level of the axilla and an abdominal are shown inTable1.Written consent was obtained from one was positioned just above the umbilicus. Calibration all subjects, and the protocol was approved by the Fon- was performed with the isovolume maneuver (11). dazione Maugeri Ethics Committee. Subjective ratings of dyspnea were made during the various trials using Borg’s scale. Arterial blood gases (BG) were measured in samples ta- Measurements ken from the radial artery (ABL 550 Radiometer,Copen- hagen, Denmark). Static and dynamic lung volumes were assessed The above-mentioned variables, excluding arterial bybody plethysmography (MasterLab-Jaeger,Hochberg, blood gases and body plethysmography, were recorded Germany). in the CHF patients and control subjects both in sitting Flow was measured with a pneumotachograph and supine positions after15min of spontaneous breath- (Screenmate Box 0586, Jaeger GmbH, Hochberg, Ger- ing, unless there was the onset of severe (Borg scale46) many), connected to a rigid mouthpiece. dyspnea, so that the measurements were recorded once Airway pressure was measured through tubing in- reached that dyspnea value. At the end of this part of the serted in the mouthpiece and connected to a di¡erential study, the CHF patients were administered non-invasive pressure transducer (Honeywell7300 cm H2O, Free- mechanical ventilation (NIMV) through a nose mask for port, IL, U.S.A.). VT was obtained by integration of the 20 min while they were in a supine position. The follow- £ow signal.The inspiratory (TI), expiratory (TE)andtotal ing pressures were used: 10cm H2O of inspiratory aid breathing cycle (TTOT ) duration, respiratory frequency and 5 cm H2O of expiratory-positive airway pressure. (RR), and duty cycle (TI/TTOT ) were calculated as average ECG, SaO2 and end-tidal CO2 (ETCO2)weremonitored values of10 consecutive breaths, after 5 min of breathing. continuously. Baseline frompeakchangesin esophageal (Pes), gastric All signals were collected using a personal computer (Pga), and transdiaphragmatic (Pdi) pressures were mea- equipped with an A/D board, and stored at a sampling rate of100 Hz.The mean value of each physiological vari- able during the 5 min of recording was used for analyses. Results are presented as mean+standard deviation (SD). TABLE 1. Patients’characteristicsinthe two groups Di¡erences between physiological parameters recorded Variables CHF patients Control subjects in sitting and supine positions were analysed by paired Student’s t-test, both for CHF patients and normal sub- Age (year) 57.078.4 51.6711.5 jects. Di¡erences between patients and controls were 7 7 BMI 22.1 3.5 24.9 6.4 analysed using unpaired Student’s t-test. Pearson’s coe⁄- 7 7 FEV1(% pred) 84.1 16.8 98.6 2.5 cient was calculated to assess correlation between vari- FVC (% pred) 89.7716.6 99.472.9 ables. A P value 0.05 was chosen as the threshold of FEV1/FVC 76.679.17 9. 5 73.8 o TLC (% pred) 92.5712.3 101.575.2 statistical signi¢cance. FRC (% pred) 89.277.0 9 7.7 77. 3 LVEF (%) 22.679.4 N A Wedge pressure (mm Hg) 21.376.6 NA RESULTS BMI, body mass index; LVEF,left ventricular ejection fraction. The participants completed the protocols without di⁄culty except one patient who was able to tolerate ORTHOPNEA AND INSPIRATORY EFFORT 649 the supine position for only 2 min. Data from that crease. NIMV, applied only to the CHF patients, particular trial were, therefore, discarded from analysis. signi¢cantly reduced the energy expenditure of the dia- phragm and the other inspiratory muscles. Figure1 illus- trates the percentage changes from sitting position in Ventilatory pattern the PTPdi and PTPes per minute in the CHF and control Table 2 illustrates the ventilatory patterns of the CHFpa- subjects during unassisted breathing. tients and controls in both the sitting and supine posi- tions and during NIMV administration. In the sitting positionVTwas smaller and breathing frequency higher, Lung mechanics albeit not statistically signi¢cantly so, in the CHF pa- Figure 2 upper part is a box-whisker plot comparing re- tients, so that minute ventilation was similar in the two sistancesin the two groups.In the sitting position control groups.The change to a supine position resulted in a de- subjects and CHF patients had similar resistances, crease in VT in both groups, but this decrease achieved although they were slightly higher in the latter group. statistical signi¢cance only in the CHF patients (Po0.05). Changing to a supine position signi¢cantly increased the During NIMV, minute ventilation was statistically higher. resistances only in CHF patients. Lung compliance was This was due to a signi¢cant increase inVTwhile respira- signi¢cantly lower in supine CHF patients than in sitting tory rate decreased; at the same time SaO2 increased ones. PEEPi,dyn was similar in both groups in the sitting and ETCO2 decreased. position (1.06+1.22 cm H2OforCHFsitvs.0.5+0.1for controls sit; P=n.s.), but signi¢cantly worsened with a change in posture only in the CHF patients Inspiratory muscle strength (1.6 4+1.35 cm H2OCHFsupvs.CHFsit;Po0.05). In the sitting position diaphragm strength, measured by Posture-induced changes in end-expiratory lung vo- sni¡ Pdi, was statistically signi¢cantly lower than in nor- lumes, as assessed by inductive plethysmography, were F mal subjects (92.15714.2 cm H2OCHFsitvs.120.3716.1 138ml in patients with CHF vs. 583 ml in control Controls sit; Po0.01). Assuming a supine position pro- subjects. duced a signi¢cant decrease, but only in CHF patients (81.6713.7 cm H 2OCHFsup[CHFsitvs.CHFsup; Po0.01]). Dyspnea The passage from a sitting to supine position induced a Inspiratory muscle e¡ort signi¢cant deterioration in Borg’s score of dyspnea in CHF patients (from 1.371to4.872.1; Po0.01), while Changes induced by posture were negligible for controls, the application of NIMV to the already supine patients but statistically di¡erent in the CHF patients. Table 3 was associated with a signi¢cant improvement in the shows the tidal Pdi (Pdi T), the PTPdi/b, the PTPdi/min, score.Of interest the increasing in Borg score recorded the PTPes/b and PTPes/min values in the two groups re- from the sitting to supine position was signi¢cantly corded in both postures. All the parameters of muscle (Po0.005) correlated to the percent increase in PTPdi e¡ort in the sitting position were signi¢cantly higher in (Fig. 3 upper part). Similarly, the decrease in Borg score CHF patients than in the control subjects, while assum- observed during NIMV was signi¢cantly (Po0.01) corre- ing a supine position induced a further signi¢cant in- lated to the percentdecrease in PTPdi (Fig. 3 lower part).

TABLE 2. Ventilatory pattern during the two postures and during non-invasive mechanical ventilation (NIMV) in the two groups.

Variables CHF sitting CTR sitting CHF supine CTR supine CHF NIMV VT (ml) 4997146 591791451 7108 5557102 7587138* w Respiratoryrate (breath/min) 18.175.3 14.772.8 18.975.0 15.072.6 14.674.5 Ve (l/min) 8.63 71.45 8.5370.94 8.0971.29 8.3272.05 11.173.2* Ti/TOT (%) 37.475.6 36.373.5 39.877.4 3 5. 5 72.9 34.673.3 z ETCO2 (mm Hg) 34.5 75.0 NA 35.174.7 NA 31.774.4 z SaO2 %94.671.2 92.573.4 97.271.3 *Po0.01CHF supine vs.CHFNIMV. w Po0.05 CHF supine vs.CHFNIMV. zPo0.01CHF sitting and supine vs.CHFNIMV. 650 RESPIRATORY MEDICINE

TABLE 3. Diaphragm and inspiratory muscle function during the two postures and during non-invasive mechanical ventilation (NIMV) inthe two groups.

Variables CHF sitting CTR sitting CHF supine CTR supine CHF NIMV

w,z PdiT (cm H 2O) 7.2372.72 4.4771.25* 8.9673.12 4.8971.59 3.6872.45 w,z PesT (cm H 2O) 6.3672.313.85 71.07* 7. 3 8 72.97 3.9171.4 4 2.9772.52 w,} PgaT (cm H 2O) 1.4870.68 0.7770.48* 1.3270.37 0.9370.510.73 70.32 z w,z PTPdi/b (cm H2O s) 10.36 74.96 6.5471.75* 13.1576.26 6.8571.60 5.372.5 z w,z PTPes/b (cm H2O s) 9.33 73.67 4.8971.42* 10.9175.54 5.0671.22 4.272.1 z w,z PTPdi/min (cm H2O s/min) 181.2754.6 96.6732.8* 243.2797.0 103.5744.2 81.8742.4 z w,z PTPes/min (cm H2O s/min) 166.6 + 48.3 83.5 + 27.3* 226.9+84.3 89.5+42.0 71.7733.7 *Po 0.01CHF sitting vs CTR sitting. z Po0.01CHF sitting vs.CHF supine. w Po0.001CHF supine vs.CHFNIMV. }Po0.01CHF supine vs.CHFNIMV. zPo0.01CHF sitting vs CHF NIMV.

a CHF patient lies down, due to an increase of the elastic (i.e. decrease in lung compliance) and resistive (i.e. in- crease in lung resistance) loads, and that this rise is very 40 strictly correlated with orthopnea. Conversely, orthop- * 35 * nea was alleviated by the application of mechanical venti- 30 lation which, by reducing inspiratory muscle e¡ort, 25 decreased the sense of breathlessness. 20 The sensation of dyspnea has been linked to several 15 factors.One of the most important of these is the bur- 10 den that the inspiratory muscles have to face (6,16).While 5 this may be not the case in patients a¡ected by a 0 ‘‘chronic’’ load, such as COPD patients, any ‘‘acute PTPdi/min PTPes/min PTPdi/min PTPes/min change’’ in this load may induce a profound variation in the sense of breathlessness (2). In CHF patients orthop- CHF CONTROL nea often occurs very quickly, and in this study it was ac- companied by a simultaneous increase in energy % CHANGES from SITTING POSITION expenditure by the diaphragm and other inspiratory FIG.1. Percentage changes from sitting to supine position in muscles. On the other hand NIMV, able to unload the in- the pressure^ time product per minute of the diaphragm (PTPdi/min) and of the other inspiratory muscle (PTPes/min) in spiratory muscle by about three-fold, had a salutary ef- the two groups, during spontaneous breathing. fect on dyspnea which was signi¢cantly reduced. *=Po0.01. O’Donnell et al.(17) showed in these patients that during the early stage of cycle ergometer exercise, inspiratory aid alone did not appreciably a¡ect dyspnea, while CPAP DISCUSSION alone had a modest e¡ect, despite the endurance time improving in both trials. In the present study we have ap- Dyspnea and orthopnea in particular are common clini- plied a combination of the two ventilatory aids, likely on cal features in patients a¡ected by chronic heart failure the one hand to reduce the elastic load and that due to (12). It has been claimed that dyspnea is in£uenced by PEPPi, dym (using CPAP) and on the other the resistive the central nervous system’s perception of inspiratory loads (using Pressure Support) (18), so that the net e¡ect motor output (13), that may increase with a reduction in on dyspnea may have been enhanced.Respiratory muscle respiratory muscle strength (14,15). As a matter of fact, weakness has been demonstratedin stable CHF patients, in patients a¡ected by chronic obstructive pulmonary (14,15,19), so it has also been claimed that reduced diseases (COPD), dyspnea seems to be well correlated strength of these muscle may play a role in the develop- with the portion of the maximal inspiratory pressure ment of exertional and resting dyspnea. Most studies ex- that each patient has to face during inspiration rather amining this aspect were, however, performed using than the maximal output that can be generated (16). In volitional maneuvers and particularly static maximal the present study we have ¢rst shown that energy ex- mouth pressure, during which it may be di⁄cult to ex- penditure of the diaphragm increases signi¢cantly when clude that the subject’s muscle activation is submaximal. ORTHOPNEA AND INSPIRATORY EFFORT 651

R,L (cmH20/L x s) r = .79623 p<0.005 14 6 + 12 5

4 10

3 8 2 6 DELTA BORG 1 4 0 0 102030405060 2 % DELTA PTPdi

0 r= 0.5776 p<0.01 CHF sitting CTR sitting CHF supine CTR supine 6 + CHF Vs CHF P<0.01 sit sup 5 COMPLIANCE,L (cmH20/L/s) 4 0.13

3 0.11 2 * 1 0.09 DELTA BORG decrease 0 0 102030405060 0.07 % DELTA PTPdi decrease

FIG.3. Linearregression analysisbetween absolute changesin 0.05 dyspnea (Borg) and percentage increase in PTPdi when passing from the sitting to supine position in CHF patients (upper part) 0.03 CHF sitting CTR sitting CHF supine CTR supine and between absolute changes in dyspnea (Borg) and percen-

+ CHF Vs CHF P<0.05 tage decrease in PTPdi when NIMV was applied in supine CHF sit sup (lower part). FIG.2. Box^Whisker plot comparing resistances (upper part) andlung compliance (lower part)during the two posturesinthe two groups. PTPdi includes the threshold due to PEEPi, the compo- Smallsquare=mean nent due to lung elastance and ¢nally that due to the re- Large square=+1St D. spiratoryresistances (21).PEEPi, dym did not signi¢cantly Ver tical line = +1.96St D. +=Po 0.01CHF sitting vs.CHFsupine. change in our patients.This is apparently in contrast with *=Po0.05 CHF sitting vs.CHF supine. the results of a study by Duguet et al.(2)whoreported the onset or exacerbation of £ow limitation in patients with acute left heart failure. Flow limitation leads to dy- Usingmeasurements that are less dependent on patients’ namic hyperin£ation, through incomplete lung emptying, aptitude, such as maximal sni¡ transdiaphragmatic pres- with the persistence of an alveolar pressure above atmo- sure and cervical magnetic nerve stimulation, Hughes spheric pressure at the end of expiration not dependent and coworkers (14) concluded that global respiratory on an increase in breathing frequency which was rela- muscle strength is well preserved in CHF patients and tively constant despite the change in posture. In our that only relatively mild weakness occurs in the dia- study we found only a small, not signi¢cant increase in phragm, but that this is unlikely to be clinically impor- PEEPi,dyn on changing posture, probably related to the tant. Indeed, the same authors in a further study mild increase in breathing frequency.On the other hand, showed that slowing of inspiratory muscle relaxation oc- in CHF patients we have found only small (138 ml) curs with walking to severe breathlessness (20), directly changes in end-expiratory lung volume, as assessed by suggesting that severe loading of the inspiratory muscles inductive plethysmography, when they were lying supine, is a feature of dyspnea. suggesting that £ow limitation probably did not occur. Lying for a few minutes resulted in a signi¢cant in- This is in keeping with the ¢ndings of Yap et al. (1) The crease in the patients a¡ected by CHF. The PTPdi per present study relies much on the recording of PTPdi to breath depends on the time of inspiration and the tidal assess the of O2 cost of breathing.This has beenvalidated Pdi generated during each inspiration (10). Since the two and/or assessed in several studies both in normals and postures did not in£uence the respiratory timing in patients a¡ected by di¡erent pathologies, even though either group, the increase in the metabolic consumption we are aware that this measure is prone to criticisms was due to the use of a higher portion of Pdi, with re- especially when lung volumes and mean inspiratory £ow spect to the maximal Pdi generated. Partitioning of vary.As a matter of fact changes in lung volumes due to 652 RESPIRATORY MEDICINE changes in body position were rather small as well as In normal subjects, the supine posture induces a ce- those in mean inspiratory £ow. Moreover, we are also phalic shift of the diaphragm and anincreasein intrathor- con¢dent that PTPdiis a rather good estimate of the me- acic blood volume that reduces resting lung volume; this tabolic cost of breathing since it increased with a conco- may account for the rise in resistances. It has, therefore, mitant increase in resistance and decrease in compliance been postulated that to interpret changes in pulmonary with the supine position. mechanics induced by lying supine, changes in lung vo- LungresistanceswerefairlysimilarinsittingCHFpa- lume need to be monitored (1). Yap et al.(1)monitored tients and controls but increased signi¢cantly (approxi- absolute lung volumes during impedance measurements mately doubling) when the CHF patients assumed a by integrating the £ow signals, estimating FRC during supine position. This phenomenon was recorded a long the supine position, assuming that TLC did not change time ago using a balloon-catheter technique (3) and has from sitting to supine position. We directly assessed been very recently con¢rmed using the so-called forced end-expiratory lung volume (EELV), the main determi- oscillation technique applied at the mouth during tidal nant to interpret changes in pulmonary mechanics, by breathing (1). In fact, Yap et al.(1) demonstrated a very si- using inductive plethysmography. As a matter of fact, milar risein totalrespiratoryresistance to thatobserved our results for both normal subjects and CHF patients inthepresentstudy.Thepresentstudywasnotaimedat were very similar to those obtained by Yap et al.(1),so, investigating the possible mechanisms inducing a rise in since EELV changed only very slightly in CHF, it is unlikely lung resistances, but it is unlikely that this rise can be at- that supine reduction in static lung volumes accounted tributed to a reduction in FRC when supine that, if pre- for changes in pulmonary mechanics. sent, could cause an increased, gravity-determined In conclusion, dyspnea worsened considerably in pa- closure of small airways. On the other hand, it has been tients with chronic heart failure on adopting a supine shown that this increase in resistance is not related posture. The present study showed that breathlessness either to re£ex bronchoconstriction, since it was only was strictly related to a rapid increase in energy expen- slightly attenuated by treatment with an inhaled mus- diture of the diaphragm, due in particular to a rise in the carinic antagonist, or to the development of peri-airway elastic and resistive loads against which the inspiratory edema and increased airway closure (1), so further stu- muscle must work.The application of NIMV to these pa- dies are required to investigate the mechanisms of these tients when they were lying supine signi¢cantly reduced changes. the inspiratory e¡ort and simultaneously the sense of Dynamic lung compliance in the sitting position was dyspnea. slightly, but not signi¢cantly, lower in CHF patients than in the controls subjects, but changing position resulted in a signi¢cant decrease in the former group. Evans and REFERENCES coworkers (12) also found a mild reduction in static lung 1. Yap JC, Moore DM, Cleland JG, et al. Effect of supine posture on compliance in CHF patients at rest, but this did not cor- respiratory mechanics in chronic left ventricular failure. Am JRespir relate with either treadmill exercise capacity or dyspnea. Crit Care Med 2000; 162(4 Pt 1): 1285–1291. The recording of static lung compliance has the advan- 2. Duguet A, Tantucci C, Lozinguez O, et al. 2000; Expiratory flow tage, with respect to recording the dynamic one, of re- limitation as a determinant of orthopnea in acute left heart failure. £ecting only the role of lung elasticity, and not of lung J Am Coll Cardiol 1; 35: 690–700. 3. Bondurant S, Hickman JB, Isley JK. Pulmonary and circulatory resistance as well. Unfortunately, the measurement of effects of acute pulmonary vascular engorgement in normal static compliance has the objective disadvantage of subjects. J Clin Invest 1957; 36: 59–66. having a poor coe⁄cient of repeatability (12), and of 4. Bondurant S, Mead J, Cook CD. A re-evaluation of the effects of being quite di⁄cult to perform in patients, especially acute central congestion on pulmonary compliance in normal when orthopnea develops, so that after a few attempts subjects. J Appl Physiol 1960; 15: 875–877. 5. Stock MC, Davis DW, Manning JW, et al. Lung mechanics and to measure it in the ¢rst 3 patients, we decided to oxygen consumption during spontaneous ventilation and severe record only the dynamic component.It has been claimed heart failure. Chest 1992; 102(1): 279–283. that the reduction in compliance is a result of a decrease 6. Killian KJ, Bucens DD, Campbell EJM. The effect of brething in functioning alveoli, since the close relationship patterns on the perceived magnitude of added loads to breathing. J between absolute static lung compliance and FRC, Appl Physiol 1982; 52: 578–584. 7. Baydur A, Behrakis PK, Zin WA, et al. A simple method for and reduced regional ventilation in CHF patients, are assessing the validity of the esophageal balloon technique. Am Rev both consistent with the above mentioned hypothesis Respir Dis 1982; 126: 788–791. (22). 8. Mead J, Whittenberger JL. Physical properties of human lungs The present study has some inherent limitations. measured during spontaneous respiration. JApplPhysiol 1953; 5: We excluded all patients known to have COPD or to 770–796. 9. Appendini L, Patessio A, Zanaboni S, et al. Physiologic effects of be at risk of it. As shown inTable1, spirometric evaluation positive end-expiratory pressure and mask pressure support ruled out signi¢cant obstruction despite about half the during exacerbations of chronic obstructive pulmonary disease. patients being smokers or ex-smokers. Am J Respir Crit Care Med 1994; 149: 1069–1076. ORTHOPNEA AND INSPIRATORY EFFORT 653

10. Sassoon CSH, Lodia R, Rheeman CH, et al. Inspiratory muscle 17. O’Donnell, D’Arsigny C, Raj S,et al. Ventilatory assistance work of breathing during flow by, demand flow, and continuous improves exercise endurance in stable congestive heart failure. flow systems in patients with chronic obstructive pulmonary Am J Respir Crit Care Med 1999; 160: 1804–1811. disease. Am Rev Respir Dis 1992; 145: 1219–1222. 18. Nava S, Ambrosino N, Rubini F, et al. Effect of nasal 11. Stradling JR, Chadwick GA, Quirk C, et al. Respiratory inductance pressure support ventilation and external PEEP on diaphragmatic plethysmography: calibration, techniques, their validation and function in severe stable COPD patients. Chest 1993; 103: the effects of posture. Bull Eur Physiopathol Respir 1985; 21: 143–150 317–324 19. McParland C, Resch EF, Krishnan B, et al. Inspiratory muscle 12. Evans SA, Watson L, Cowley AJ, et al. Static lung compliance in weakness in chronic heart failure: role of nutrition and electrolyte chronic heart failure: relation with dyspnoea and exercise capacity. status and systemic myopathy. Am J Respir Crit Care Med Apr; 151: Thorax 1995; 50: 245–248. 1101–1107. 13. Elayara L, Becklake MG, Volta C, et al. Relationship between 20. Hughes PD, Hart N, Hamnegard CH, et al. Inspiratory muscle chronic dyspnea and expiratory flow limitation in COPD patients. relaxation rate slows during exhaustive treadmill walking in Am J Respir Crit Care Med 1996; 154: 1726–1734 patients with chronic heart failure. Am J Respir Crit Care Med 14. Hughes PD, Polkey MI, Harrus ML, et al. Diaphragm strength 2001; 163: 1400–1403. in chronic heart failure. Am J Respir Crit Care Med 1999; 160: 21. Appendini L, Purro A, Patessio A, et al. Partitioning of 529–534. inspiratory muscle work-load and pressure assistance in ventila- 15. Ambrosino N, Opasich C, Crotti P, et al. Breathing pattern, tor-dependent COPD patients. Am J Respir Crit Care Med 1996; ventilatory drive and respiratory muscle strength in patients with 154: 1301–1309. chronic heart failure. EurRespirJ 1994; 7: 17–22. 22. Gibson GJ, Pride NB. Lung distensibility: the static pressure- 16. Killian K, Jones N. Respiratory muscles and dyspnea. Clin Chest volume curve of the lungs and its use in clinical assessment. Br J Dis Med 1988; 9: 237–248. Chest 1976; 70: 143–148.