Improved technique for estimating pleural pressure from esophageal balloons.

Transcription

1 Improved technique for estimating pleural pressure from esophageal balloons J. MILIC-EMILI, J. MEAD, J. M. TURNER, AND E. M. GLAUSER2 Department of Physiology, Haruard School of Public Health, Boston, Massachusetts MILIC-EMILI, J., J. MEAD, J. M. TURNER, AND E. M. GLAUSER. Improved technique for estimating pleural pressure from esophageal balloons. J. Appl. Physiol. I g(2) : 27-2 I I. I g64.-esophageal pressure has been measured in eight healthy men during breath holding (glottis open) at various fixed lung volumes with a rubber balloon (length: IO cm; perimeter: 3. cm; wall thickness: ca..6 mm) containing various volumes of air. The tip of the balloon was placed at a constant distance of 4 cm from the nares. Esophageal pressure was found to increase with balloon volume at all lung volumes but not uniformly, the effect of balloon volume being greatest at both extremes of the vital capacity. As a result, lung volume-pressure curves obtained from balloons containing volumes of gas such as are commonly used are distorted. One can avoid these distortions by repeating measurements at different balloon volumes and extrapolating esophageal pressure to zero balloon volume, or by making measurements at very small balloon volumes. A close approximation of the extrapolated pressures was obtained with the present balloon containing.2 ml air. The esophageal pressures at or near zero balloon volume probably reflect closely local absolute pleural pressures over the full vital capacity range in some subjects and above 2% of the vital capacity in most subjects. lung volume-pressure curves M EASUREMENTS OF VOLUME-PRESSURE relationships of human lungs are usually based on indirect determination of pleural pressure obtained from the esophagus. The most widely used method for measuring esophageal pressure employs air-containing latex balloons sealed over catheters which transmit balloon pressure to manometers (I, 4, g, I 3). The pressure within the balloon placed in the esophagus will be the same as local pleural pressure only if the pressure difference across all intervening structures (balloon wall, esophageal wall, and various mediastinal structures) is zero. Over a certain Received for publication I 3 August I This study was supported in part by National Institutes of Health Grants HTS-379, E-I I 7, and H-339, and Institute of General Medical Sciences. Grant GM Present address: Woman s Medical College of Pennsylvania, Philadelphia, Pa. range of balloon volumes there is only a negligible pressure drop across the wall of the balloon itself. Within this range, however, when the balloon is placed in the esophagus balloon pressure increases with balloon volume, presumably as a result of distension and displacement of the esophageal wall and surrounding structures (9, I 3). As a result, the pressure in the esophageal balloon tends to be more positive than pleural pressure, a fact that has been generally recognized. Since the magnitude of this difference was unknown, little use has been made of esophageal pressure as a measure of absolute pleural pressure. On the other hand, it has been thought that variations in pleural pressure would be reliably recorded from such balloons, the assumption being that the influence of balloon volume per se on pressure does not change with lung volume. The present study shows that this is not the case. An effect of balloon volume on the recorded pressure was found which increased at both extremes of the vital capacity. As a result, volumepressure curves for lungs based on esophageal pressures are distorted at these lung volumes. These distortions can be avoided if one repeats measurements at various balloon volumes and extrapolates esophageal pressures to zero balloon volume or, more practically, can be rendered negligible by making measurements at smaller balloon volumes than have commonly been used in the past. In addition, the results suggest that esophageal pressures at or near zero balloon volume probably reflect closely local absolute pleural pressures. MATERIALS AND METHODS Esophageal pressure was measured in eight healthy men during breath holding (glottis open) at various fixed lung volumes with a balloon containing various volumes of air. The physical characteristics and vital capacities of the subjects are given in Table I. All measurements were made with the subjects in the upright posture. Lung volumes were measured with a g-liter Collins spirometer. The esophageal balloon was made of rubber and had the following dimensions: length, IO cm; perim-

2 28 MILIC-EMIL& MEAD, TURNER, AND GLAUSER eter, 3. cm; wall thickness, ca..6 mm. The balloon was sealed over the end of a polyethylene catheter (i.d., o. 14 cm; o.d., o. r g cm; length, IOO cm), which had a number of spirally arranged holes in the part covered by the ba.lloon. The volume-pressure characteristics of the balloon used are given in Fig. 2. The catheter was connected to a pressure transducer (Sanborn model 268B). Pressure was recorded on a direct-writing oscillograph (Sanborn Poly-Viso). The volume displacement coefficient of the catheter-manometer unit was about.2 ml/cm H2. As a result of the small volume. + IO I I. pes. cm H 9!. TABLE I. Physical characteristics and vital cahacities (IX> of subjects -2 Subject JT FS DL JS ME NF JM RK Am, yr Weight, kg Height, cm WE? I I = I I -1 SEC- I FIG. I. Esophageal pressure during lung deflation in subject ME. Lung volumes, expressed as per cent of the vital capacity, are indicated on the tracing. Balloon volume: 2 ml. displacement coefficient of the catheter-manometer unit used, changes in balloon volume resulting from pressure changes were small. The balloon was introduced through the nose into the esophagus and placed at a distance of 4 cm from the balloon tip to the nares. Prior to each measurement the balloon was emptied by having the subject perform a Valsalva maneuver while the catheter was open to the atmosphere. The balloon was then connected to a syringe and 8 ml air were introduced in order to distend the walls of the balloon evenly. Following this, air was withdrawn into the syringe until the desired amount (.2- ml) remained in the balloon, and the catheter was connected to the manometer. All measurements were made at balloon volumes where there was a negligible pressure drop across the balloon wall. Changes in balloon volume, resulting from pressure variations in the closed balloon-catheter-manometer system, were taken into account. To obtain consistent results it was found necessary to standardize the procedure as follows. Before each measurement the subject performed three full inspirations. At the end of the third full inspiration, FIG. 2. Broken line: pressure-volume relationship of balloon in air. Solid lines : esophageal pressure-balloon volume curves obtained at different lung volumes, expressed as per cent of the vital capacity. As a result of the small volume displacement coefficient of the catheter-manometer system used, changes in balloon volume due to compression or expansion of gas were negligible. Dotted line : esophageal pressure-balloon volume relationship for a system with a volume displacement coefficient of. ml/cm HZO, the balloon containing 2 ml of gas at atmospheric pressure. The pressures at the various lung volumes that would be obtained with such a system are indicated by the intercepts of the dotted line with the iso-lung volume curves. he exhaled stepwise to fixed lung volumes until residual volume was attained. Each lung volume was maintained (glottis open) long enough to measure the static esophageal pressure (in general for 2-4 set). During the period of breath holding esophageal pressure did not change appreciably, except at total lung capacity where it decreased markedly with time (). At this lung volume only the most negative values of esophageal pressure were taken into consideration. All data were obtained during lung deflation and were repeated at least two times. L4 typical record is shown in Fig. I. RESULTS Figure 2 illustrates the relationship between esopha.geal pressure (Pes) and balloon volume (Ves) obtained at various lung volumes in subject ME. At a constant lung volume esophageal pressure increased linearly with balloon volume. The change in esophageal pressure resulting from a unit change in balloon volume, i.e., APes/AVes, varied at different lung volumes, being greatest at the extremes of the vital capacity. Figure 3 shows corresponding lung volume-esophageal pressure curves. In five of the other subjects studied, the relationship between Pes and Ves at any given lung volume was also found to be approximately linear. Their values of Apes/

3 ESOPHAGEAL 8 PRESSURE I v 1. I. I. % vc Ves, ml 4 (2) 2 FIG. 3. Solid lines: lung volume-esophageal pressure curves at different balloon volumes from data in Fig. 2. Broken line: lung volume-pressure curve based on esophageal pressure extrapolated to zero balloon volume, i.e., the intercepts on the ordinate of the esophageal pressure-balloon volume curves in Fig. 2. Dotted line: lung volume-esophageal pressure curve for a catheter-manometer system with a volume displacement coefficient of. ml/cm Hz with 2 ml of air introduced into the balloon at atmospheric pressure APes/AVes, cmh&l/mi FIG. 4. Change in esophageal pressure per unit change in balloon volume (A Pes/A Ves) at various fixed lung volumes for the six subjects in whom the relationship between esophageal pressure and balloon volume was linear. DISCUSS1 ON Our results confirm the previous finding that esophageal pressure increases with balloon volume (9, I 3). In addition we have found that the effect of balloon volume on esophageal pressure is greatest at both extremes of the vital capacity, particularly so at the high lung volumes. As a result, with increasing balloon volume, lung volume-pressure curves based on balloon pressures are displaced in the direction of more positive pressures, but not uniformly, being displaced more at high and low than at intermediate lung volumes. Curves based on pressures obtained by extrapolation to zero balloon volume probablv reflect more closely the actual volume-pressure behavior of lungs. A close approximation of extrapolated curves was obtained with the present balloon containing a volume of.2 ml. The results of the present investigation raise the question whether the pressures at zero balloon volume correspond to absolute pleural pressures. The results obtained at low lung volumes are interesting in this connection. In four of our subjects esophageal pressures at zero balloon volume were still slightly subatmospheric at residual volume. Since at residual volume the lungs are not completely collapsed, negativity of pleural pressure +I t 1 I I 1 1 t pes, cm Hz JM IO AVes obtained at different lung volumes are shown in Fig. 4 together with the results obtained on the subject in Fig. 2. In all these subjects APes/AVes increased as total lung capacity or residual volumes were approached. In the other two subjects studied (JM and RK), the relationship between balloon volume and esophageal pressure was approximately linear only at balloon volumes greater than about 2 ml (Fig. ). In both subjects the changes in esophageal pressure resulting from changes in balloon volume tended again to be greatest at the extremes of the vital capacity. The lung volume-pressure curves based on esoph.ageal pressures extrapolated to zero balloon volume for the eight subjects studied are given. in Fig. 6. FIG.. fixed lung linear. Esophageal volumes I in pressure-balloon volume curves at various a subject in whom this relationship was not at this lung volume may be assumed, and the pressures observed in these individuals are reasonable. In the other four subjects, at lung volumes lower than about 2 % of the vital capacity, esophageal pressures at zero balloon volumes increased disproportionately and became supra-atmospheric. It seems likely that in these instances esophageal pressures deviated from pleural pressures as a result of compression of the esophagus and

4 21 MILIC-EMILI, MEAD, TURNER, AND G-LAUSER balloon by mediastinal structures. It should be noted that at lung volumes higher than about 2 % of the vital capacity the behavior of esophageal pressure at zero balloon volume was, in general, similar for all subjects studied. We have tentatively concluded that esophageal pressure at or near zero balloon volume probably approximates closely local absolute pleural pressure in all of our subjects when lung volume is above about 2oY of the vital capacity. In some subjects it probably approximates local pleural pressure at all lung volumes. In the past, esophageal pressure has commonly been measured at relatively high balloon volumes. For the most part measurements have been made in the middle range of lung volumes where the effect of balloon volume is comparatively small. The pressures recorded, however, were presumably more positive than pleural pressures. In a few studies esophageal pressures were also measured at the extremes of the vital capacity, where balloon volume may produce considerable distortions. Accordingly, it is not surprising that the pressures obtained in these studies at total lung capacity and at residual volume were, in general, appreciably more positive than the pressures obtained in the present investigation by extrapolation to zero balloon volume, particularly so when relatively large balloon volumes were used. Published results are surnmarized in Table 2. In the previous studies balloons of various dimensions IO 8 2 t1 -IO ;7 FIG. 6. Lung volume-pressure curves based on esophageal pressure extrapolated to zero balloon volume for the eight subjects. have been used; the shorter and narrower the balloon, the larger are the changes in pressure produced by a given balloon volume and, hence, the larger the distortions (I 3). An additional source of error is introduced by catheter-manometer systems with large volume displacement coefficients (I 2, I 3), as illustrated in Figs. 2 and 3. Finally, errors may also be introduced if the balloon extends into the upper third of the esophagus (I I )* The dimensions of the balloon in the present study represent an attempt to minimize the effect of balloon volume while avoiding the upper part of the esophagus. Accordingly, the balloon had a larger perimeter than in most previous studies. Its length was, however, limited to I o cm, since with longer balloons it is difficult to avoid the upper third of the esophagus unless part of the bal I FIG. 7. Esophageal pressure-balloon volume curves at 4 and 8oY of the vital capacity. Solid circles: experimental results from Fig.. Solid lines- esophageal pressure-balloon volume curves calculated, using values for constants indicated in the figure, according - to the following equation: -. Pes = Peso < d2 X Esp )( Grad X Ves where Peso is the esophageal pressure at zero balloon volume at the uppermost point of the IO-cm long balloon, Esp is the specific elastance of the esophagus (i.e., the ratio of pressure change to balloon volume for a unit length of the esophagus), and Grad is the pressure gradient along the esophagus (i.e., the change in pressure at zero balloon volume per unit variation in level along the esophagus). Peso and Grad were obtained as described elsewhere (I I ) ; Esp was obtained from pressure-volume measurements made with a short balloon (length, 2 cm; perimeter, 3. cm; wall thickness,.6 mm) as previously described (I 3). TABLE 2. Comparison of esophageal pressures obtained at residual volume (RV) and total lung capacity (TLC) by p revious workers with present results Investigators No. of Subj. Age, yr Balloon Vol., ml Pes at RV, cm H2D Pes at TLC, cm H2 Mead et al. (IO) Slagter and Heemstra (14) Frank et al. (3) Knowles et al. (7) Car et al. (2) Hyatt (6) Macklem and Becklake (8) Macklem and Becklake (unpubl. results) Present results I2 II 4 1 I2 8 (32-7 ).-1 (19-22 > 2 (+18 to +36) 28 (22-47) I 33 (28-4 ) 2 +8 (19-4 > I 49 o- 3 (26-36 > I-2 3 (24-3 >. (-2 to -33) -3 (-21 to -4) (-1 to -27) -3o* -3 (-3 to -34) -33 (-23 to -4) 34 (27-43 ) +.8 (-2 to +g)* -38 (-29 to -6) Values are means, with ranges in parentheses. * Consistent values of Pes at RV were obtained only in six subjects.

5 ESOPHAGEAL PRESSURE 211 loon extends into the stomach. In addition, the reduced wall thickness of the present balloon together with the small volume displacement coefficient of the cathetermanometer unit made it possible to obtain accurate measurements of esophageal pressure at smaller balloon volumes than in the past. In agreement with the results of previous studies (9, 13) the relationship between esophageal pressure and balloon volume was, in general, approximately linear. In two of our subjects, however, this was not true at low balloon volumes. We felt that this might be explained by nonuniformity of esophageal pressure over the segment of esophagus encompassing the balloon. If a balloon containing a small volume of gas is placed in a region of the esophagus where pressure is not uniform from point to point, the gas will be displaced to the more negative points, and the remainder of the balloon will be collapsed. With increasing balloon volume the pressure inside the balloon increases as a result of distention and displacement of the esophageal wall and surrounding structures. Accordingly, the segment of balloon containing gas must increase progressively until it comprises the whole length of the balloon. Since the longer the length of the air-containing segment the smaller the strain on the esophageal wall and its surrounding structures induced by a given change in balloon volume, the slope APes/AVes will decrease with increasing balloon volume until the balloon is distended along its entire length. Changes in esophageal pressure resulting from any further increase in balloon volume will reflect the volume-pressure relationship of the segment of esopha.gus REFERENCES I. BUYTENDIJK, H. J. Oesophagusdruck en longelasticiteit. Groningen : Oppenheim, I gag. 2. CARO, C. G., j. BUTLER, AND A. B. DUBOIS. J. C&z. Invest. 39: 73, *9* 3. FRANK, N. R., J. MEAD, AND B. G. FERRIS. J. Clin. Invest. 36: 168, FRY, D. L., W. W. STEAD, R. V. EBERT, R. I. LUBIN, AND H. S. WELLS. J. Lab. Clin. Med. 4 : 664, I 92.. HUGHES, R., A. J. MAY, AND J. G. WIDDICOMBE. J. Physiol., London 146: 8, HYATT, R. E. Am. Rev. Resfirat. Diseases 83 : 676, KNOWLES, J. H., S. K. HONG, AND H. RAHN. J. AfipZ. Physiol. 14: 2, 99 encompassing the entire balloon, which appears to be linear within our experimental limits. Since the topography of esophageal pressure had been measured in three of our subjects (I I), it was possible to test this hypothesis. In line with our expectations, in subjects A4E and NF, who exhibited a linear relationship between esophageal pressure and balloon volume, the esophageal pressures at zero balloon volume were found to be relatively uniform in the segment of the esophagus encompassing the present balloon. This was not the case for subject JM, one of the two subjects with nonlinear Pes versus Ves curves. Since in this subject in the region of the esophagus occupied by the present balloon a) esophageal pressure at zero balloon volume increased approximately linearly with descent along the esophagus (I I) (Fig. 4), and b) the mechanical properties of the esophagus, as expressed by the ratio of pressure change to balloon volume for a unit length of the esophagus or specific elastance of the esophagus (13), have been shown to be uniform in the segment of esophagus concerned, the relationship between Pes and Ves can be predicted. Figure 7 shows the predicted relationship between Pes and Ves at 4 and 8% of the vital capacity together with the experimental data for subject JA4. It can be seen that the experimental results are in good agreement with the predicted values, lending further support to the hypothesis that nonuniformity of esophageal pressure accounted for the instances of nonlinearity in the relationship between esophageal pressure and balloon volume IO. II MACKLEN, P. T., AND M. R. BECKLAKE. Am. Rev. Resfirat. Diseases 87 : 47, MEAD, J., M. B. MCILROY, N. J. SELVERSTONE, AND B. C. KRIETE. J. Appl. Physiol. 7 : 491, I 9. MEAD, J., I. LINDGREN, AND E. A. GAENSLER. J. Clin. Invest. 34: 1, 19. MILIC-EMILI, J., J. MEAD, AND J. M. TURNER. J. A@Z. Physiol. I g : 2 I 2, I 964. MILIC-EMILI, J., AND J. M. PETIT. J. ApfZ. Physiol. 14 : 3,rgsg. PETIT, J. M., AND J. MILIC-EMILI. J. A$@. Physiol. 13 : 481, 198. SLAGTER, B., AND H. HEEMSTRA. Acta Physiol. Pharmacol. iveerz. 4: =