Measuring PA Wedge Pressure: Nurses vs. Doctors

Abstract & Commentary

The goal of this study was to determine whether there were inter- and intraobserver differences in measures of pulmonary artery occlusion (wedge) pressure (Ppao) made by physicians vs. critical care nurses (CCN) and, if differences were found, to determine what factors influenced this variability. Over a one-year period, CCN were asked to obtain a paper tracing twice a day at the same time that they recorded Ppao values using the bedside monitor. The tracings were labeled by a single individual, who indicated the mode of ventilation and also assessed each tracing for the presence of ventricular waves, defined as phasic increases in Ppao of 4 mmHg or more that corresponded to the shock therapy (ST) interval of the electrocardiogram (ECG), and also measured the respiratory phasic variation (RPV) of all Ppao tracings, defined as the maximum-minimum value on each tracing. The two physicians (chief of critical care [CCMD] and chief of cardiology [CARD]) independently interpreted each tracing on two separate occasions, in blinded fashion, and these values were compared with CCN values.

During the study interval, 147 measurements of Ppao were performed on 40 patients with a mean age (+ SE) of 62.5 ± 2.2 years and a mean APACHE II score of 21.5 ± 0.8. Thirty-four readings were excluded from analysis because the tracings did not contain an appropriate Ppao tracing (n = 28) or were mislabeled (n = 6). Of the tracings (n = 113) interpreted for the study, 96 were obtained during positive pressure ventilation and 17 were obtained during spontaneous ventilation. Agreement between the two physicians, determined by correlation coefficients, was 0.91 for the CCMD and 0.87 for the CARD. Agreement between different observers was 0.83 for CCMD-CARD, 0.66 for CARD-CCN, and 0.67 for CCMD-CCN. Differences between the physicians and nurses for 95% of the readings (limits of agreement) ranged from -5.3 to 7.3 mmHg (CCMD-CARD) to -9.2 to 7.6 mmHg (CARD-CCN).

Neither the use of positive pressure ventilation nor the presence of ventricular waves in the Ppao tracing predicted physician-nurse or physician-physician disagreement in measurement. However, differences for tracings with RPV > 8 mmHg were significantly greater than for tracings with variations < 8 mmHg (P < 0.05, except for CCMD-CCN, P = 0.10). As a final step, three tracings (1 with no RPV, 2 with large RPV) were presented to a convenience sample of 23 CCN and 18 physicians (8 board-certified cardiologists, 4 board-certified critical care physicians, and 6 cardiology or pulmonary fellows). All but one nurse and one fellow identified the Ppao within 4 mmHg of the mean value for the tracing with minimal RPV. For the two tracings with a larger RPV, variability was twice that observed for the tracing with minimal RPV. (Al-Kharrat T, et al. Am J Respir Crit Care Med 1999;160:415-420.)


This study demonstrates the presence of significant differences in the measurement of Ppao tracings among physicians and between physicians and CCN when tested in a clinical practice setting. As such, findings confirm those obtained from prior studies that have described significant observer variability in Ppao measurement when tested using a questionnaire format. In this study, values recorded by 50 nurses with varied experience in critical care were compared to those reported by two physicians with extensive experience in critical care. Despite these differences in sample size, nurse-physician differences were not substantially greater than physician-physician differences.

The major explanation appeared to be related to two factors: different methods of identifying end-exhalation and large RPV. The CARD identified end-exhalation on the basis of ventilator status. For patients on mechanical ventilation, he chose end-exhalation as the trough values of Ppao. When patients were breathing spontaneously, he chose peak values of Ppao. In contrast, the CCMD determined end-exhalation by combining knowledge of ventilator status and examining RPV, assuming that exhalation is usually longer than inhalation. For ventilated patients with large RPVs, he chose the peaks for Ppao interpretation if the peaks were of longer duration than the troughs.

Thus, some of the variability appeared to be related to different methods of identifying end-exhalation. Criteria used by the CCNs were not identified in the article, although Al-Kharrat and colleagues indicated that the nursing department provided extensive training, with annual skill evaluation. Standard practice used by the CCN to determine Ppao involved adjusting the on-screen horizontal axis (cursor) tangential to the point on the Ppao tracing believed to be end-exhalation. A digital Ppao reading was then displayed and recorded on the patient’s chart.

Because nurses who participated in this study were recruited from the medical and cardiac intensive care units (ICUs), it is likely that nursing staff of the two units used different criteria for identifying end-exhalation, similar to the CARD and CCMD. In addition, there was substantial variability when physicians and nurses interpreted tracings with large (> 8 mmHg) RPV, most likely due to difficulty in identifying end-exhalation.

These findings are important because they identify two variables that can affect Ppao readings, both of which can potentially be influenced through education. Critical care practitioners can be taught strategies to increase accuracy when there are large RPV, such as examination of tracings for respiratory phasic cycle lengths, and using this analysis to more accurately identify end-exhalation. If large RPV are present, this factor should be noted when reporting Ppao values, since the likelihood of error appears to be greater. Such information is important because inaccurate interpretation of Ppao can lead to incorrect management decisions, including failure to treat or inappropriate treatment.

The factor that caused the most variability in pulmonary artery pressure occlusion (Ppao) readings was:

a. large respiratory phasic variations.

b. level of practitioner training.

c. physician specialty.

d. years of ICU experience.

e. primary disease process of patient.