Respiratory system impedance in patients with acute left ventricular failure : pathophysiology and clinical interest FLORENCE

To investigate the relationship between alterations in lung mechanics and acute pulmonary vascular congestion, repeated measurements of the respiratory system impedance (Zrs) were performed in 11 patients with and in seven without acute left ventricular failure. Indexes of Zrs were obtained by calculating the average and slope of the resistance and reactance in low (10 to 20 Hz) and high (20 to 50 Hz) frequency intervals. Zrs indexes in patients with ventricular failure differ significantly from those in patients without failure. Pulmonary vascular congestion is regularly associated with an abnormal frequency dependence of resistance at low frequencies and with an increased resonant frequency. Discriminant analysis of Zrs indexes allows 92% correct classification of pulmonary capillary wedge pressures lower than and those equal to or higher than 18 mm Hg. Zrs differences between patients with and without left ventricular failure are consistent with the presence of a small airways obstruction even in patients with mild left ventricular failure. Furthermore, use of Zrs indexes permits moderate and severe pulmonary vascular congestion to be distinguished from one another and this is probably due to a significant narrowing of the large airways during severe left ventricular failure. Circulation 73, No. 3, 386-395, 1986. ALTERATIONS in lung function have been shown to be related to hemodynamic impairment.' Pulmonary vascular distension due to increased pulmonary blood volume appears as one of the early processes involved in the perturbations of lung mechanics.4 1 A specific increase in the small airways resistance, which may be due to "competition for space between the vessel and airway in the bronchovascular sheath," has been observed in the dog with an increase in left atrial pressure from 0 to 15 mm Hg.4 Similarly, a reversible decrease in lung compliance with increasing left atrial pressure in the phase of rapid rise of pulmonary blood volume has been described in the isolated rabbit lung.5 In man, the rapid increase in left end-diastolic pressure observed during anginal syndrome has been shown to be accompanied by a significant fall in lung compliance and in specific conductance.2 From the Institut de Physiopathologie Clinique, Centre Hospitalier Universitaire Vaudois (CHUV), and Institut de Physique Appliquee, Ecole Polytechnique Fed6rale de Lausanne (EPFL), Lausanne, Switzerland. Supported by a special grant from the Ecole Polytechnique Federale de Lausanne, Switzerland, and by the Swiss National Science Foundation grant 3.849-0.83. Address for correspondence: Florence B. Depeursinge, M.D., Institut de Physiopathologie Clinique, BH 19.632, CHUV, 1011 Lausanne, Switzerland. Received July 30, 1985; revision accepted Nov. 7, 1985. The existence of small airways dysfunction in pulmonary vascular congestion is suggested by an increased closing volume` and the development of an abnormal frequency dependence of the total pulmonary resistance (RT). 1 Indeed, the frequency dependence of RT has been shown to be closely related to that of lung compliance,9" which is considered as one of the most sensitive indicators of small airways obstruction. 12 Using forced sinusoidal oscillations, Dubois et al."3 first described the frequency dependence of the RTat frequencies above the spontaneous one. With the same technique, Interiano et al.' observed a frequency dependence of RTin the 3 to 9 Hz range in patients with acute myocardial infarction that was reversible within the following 2 to 3 weeks. RT at 3 Hz was significantly correlated with the pulmonary capillary wedge pressure (Ppcw). In a further study, Gray et al. 1 did not find such a close relationship between RT and pulmonary vascular pressures. They did show, however, that the frequency dependence of RT was probably related to a decrease in functional residual capacity (FRC), which was significantly correlated with pulmonary vascular pressures. Quantification and interpretation of the frequencydependent mechanical response of the lung underwent CIRCULATION 386 by gest on A ril 5, 2017 http://ciajournals.org/ D ow nladed from PATHOPHYSIOLOGY AND NATURAL HISTORY-LEFT VENTRICULAR FAILURE considerable development with the advent of the forced random excitation technique. The measurement of the respiratory system impedance (Zr) by the forced random noise technique was introduced by Michaelson et al. 14 Different lung models have since been studied and several algorithms have been developed to partition RT into central and peripheral components.'5-17 Little information is available on the resistance and reactance frequency dependence above 10 Hz in cardiac patients. Moreover, the two studies of the oscillatory resistance in patients with acute myocardial infarction included patients with only mild pulmonary vascular congestion. 1 3 New results are reported in this article, which describes the influence on Zrs Of pulmonary vascular congestion induced by acute left ventricular failure. Zrs indexes are compared with P and their relationship is discussed. Patients and methods Patients. Eighteen patients were investigated during hospitalization in the coronary care unit of the Centre Hospitalier Universitaire Vaudois in Lausanne. The group without ventricular failure (group 1) includes seven patients with acute uncomplicated myocardial infarction. The absence of left ventricular failure was assessed by the following criteria: absence of an S3 or summation gallop, absence of adventitious pulmonary sounds, absence of cardiomegaly (cardiothoracic index less than 0.5), and chest x-ray labeled class 0 according to McHugh's classification.18 The ventricular failure group (group 2) includes 1 1 patients with acute ventricular failure, induced by acute myocardial infarction (nine patients), a hypertensive crisis (one), or severe arrhythmia (one). In spite of treatment, moderate-to-severe clinical signs of pulmonary vascular congestion were present in all patients with ventricular failure. Hemodynamic measurements were therefore made in all patients in this group. Clinical data and initial hemodynamic values for patients in group 2 are listed in table 1. During the initial study, four to nine Zrs measurements were made in each patient within a period of 1 to 5 days. The first Zrs measurement was taken within 24 hr after the onset of myocardial infarction in all but one group 1 patient (53 hr). In group 2, the first Zrs measurement was taken within 4 hr after insertion of the Swan-Ganz catheter in all but two patients (23 and 60 hr). Sixty-four Zrs measurements were obtained simultaneously with hemodynamic measurements in the patients in group 2. The follow-up study (3 to 13 weeks after the first investigation) included standard lung function tests, obtained in all group 1 and in nine group 2 patients, and Zrs measurements in six patients of each group. The results of the lung function tests are given in table 2. No patient had clinical signs of pulmonary vascular congestion. Three patients in group 2 had chronic obstructive pulmonary disease (COPD), based on a typical history of chronic bronchitis, residual volume/total lung capacity greater than 0.4, and forced expiratory volume in 1 sec/forced vital capacity lower than 0.6. Hemodynamic measurements. Right heart catheterization was performed by the flow-directed balloon-tipped catheter technique.'9 The pressures were measured with a Statham P-23 Db transducer, positioned 5 cm below the sternal angle. Cardiac output was determined by the thermodilution technique with a KMA 3500 computer. Follow-up lung function tests. Lung volume and airways resistance were measured with a pressure-corrected body plethysmograph Pulmorex SAB (Fenyves & Gut)20 and respiratory flow was determined with a Fleisch No. 3 pneumotachograph. Results of lung function tests were compared with the predicted values published by Bates et al.2' Zrs measurements. Zrs measurements were obtained at the bedside with the forced excitation technique. 14 The corresponding apparatus was developed at the Ecole Polytechnique Feddrale de Lausanne. We shall first give a general description of the principle of the forced oscillations technique (figure 1). Small amplitude acoustical pressure oscillations (a forcing noise) in the frequency band 1 to 50 Hz are generated at the patient's mouth by a loudspeaker. The induced flow oscillations superim-