Detecting Intrinsic PEEP by Clinical Examination
Abstract & Commentary
Synopsis: Intrinsic PEEP can be detected at the bedside by simple inspection, palpation, and auscutation in a large proportion of instances, even by clinicians without extensive training or experience.
Source: Kress JP, et al. Am J Respir Crit Care Med 1999;159:290-294.
This study sought to determine whether clinicians of varying training and experience could detect the presence of intrinsic positive end-expiratory pressure (PEEPi; "auto-PEEP") in mechanically ventilated patients by physical examination alone. Kress and associates at the University of Chicago asked seven attending critical care physicians, 18 residents in internal medicine, and 15 fourth-year medical students to examine a total of 71 patients who were being ventilated for a variety of conditions. Examiners were instructed to observe and palpate each patient’s chest and to perform auscultation, and then to state whether they thought PEEPi was present. If signs of PEEPi were detected by either inspection/palpation or auscultation, PEEPi was recorded as being present. Patients were each examined daily by the clinicians up to three times within a four-day period. Kress et al simultaneously determined the presence or absence of PEEPi in each instance using the waveform display of the Puritan Bennett 7200 ventilator or a Bicore CP-100 pulmonary monitor connected to a Siemens Servo 900C ventilator.
Altogether, 503 observations were made on the 71 patients, of whom 28% had acute hypoxemic respiratory failure, 23% had ventilatory failure, 15% had postoperative respiratory failure, 14% had sepsis or shock, and 20% had been intubated for airway protection or because of weakness. PEEPi was detected by expiratory waveform analysis in 351 of the 503 observations (70%). The incidence of PEEPi in the different diagnostic categories ranged from 84% among patients with ventilatory failure to 44% in those intubated for airway protection or weakness. The majority of the study patients were making active respiratory efforts at the time of study, and thus PEEPi was quantifiable in only 86 of the 503 observations. The magnitude of PEEPi was more than 5 cm H2O in 14 of these 86 measurements.
The clinical examination, taken as a whole, was better at correctly predicting the presence of PEEPi than was any individual component. For all examiners, physical examination had a sensitivity of 72%, a specificity of 91%, and a positive predictive value of 95% for detecting the presence of PEEPi. The likelihood ratio for detecting PEEPi by clinical examination was 8.35, which translated to a 95% post-test probability of PEEPi in this population, whose pretest probability (true prevalence) was 70%. The negative predictive value (for correctly predicting the absence of PEEPi) was only 58%, with a likelihood ratio of 0.31.
Attending physicians were better at detecting PEEPi than residents and students in all categories, but the specificity and positive predictive values for PEEPi based on physical examination were high at all levels of training and experience. The specificity and positive predictive values for PEEPi were each 1.0 for the attending intensivists; these corresponded to 0.82 and 0.89 for the residents and to 0.86 and 0.93 for the students, respectively. The sensitivity for PEEPi was 0.75 for the attending intensivists, 0.68 for the residents, and 0.69 for the medical students.
Kress et al conclude that the clinical examination is good for detecting PEEPi at all experience levels. On the other hand, they also conclude that the clinical examination is only modestly useful for ruling out PEEPi and recommend the use of other techniques for looking for PEEPi if it is not detected by clinical examination.
COMMENT BY DAVID J. PIERSON, MD, FACP, FCCP
Clinical methods for quantitating or estimating PEEPi in intubated patients include the end-expiratory airway occlusion technique, the simultaneous recording of pressure and flow at the airway opening, esophageal pressure manometry, and the stepwise addition of external PEEP. In addition, the presence of PEEPi can be detected by observation of the flow-vs.-time expiratory waveform available on some ventilators or ancillary monitoring systems, although it cannot be quantitated by this technique. Each of these methods requires that the clinician suspect and specifically look for PEEPi, and most of them require special maneuvers or apparatuses not always available. Because PEEPi is both common and physiologically important, any means for simplifying its detection and making clinicians more widely aware of its presence would be most welcome.
In this study, Kress et al have demonstrated that PEEPi can be detected by routine physical examination in a high percentage of instances if clinicians know how to look for it. Further, they show that the detection of PEEPi does not require specialist training or years of clinical experience. The simple technique used in this study is as follows:
| Table-How to Detect PEEPi by Physical Examination |
| Look at and palpate patient’s chest during exhalation |
| • Is there continuous inward movement of the chest wall that persists up to the moment the next breath occurs? |
| Auscultate the chest |
| • Do breath sounds persist throughout exhalation up until the moment the next breath occurs? |
Even though PEEPi during mechanical ventilation was first brought to clinical attention 17 years ago (Pepe PE, Marini JJ. Am Rev Respir Dis 1982;126:166-170), many clinicians remain unaware of its importance and do not routinely look for it in managing patients. Wider recognition of the incidence and physiologic importance of PEEPi has been hampered by the fact that it cannot be detected unless one looks for it. However, Kress et al have shown that it is easy to look for, using skills possessed by everyone who works in the ICU. The key is to think of PEEPi and to use those skills on a routine basis in every ventilated patient.
There are several potential problems with this study, although these do not change its main message. There are different clinical varieties of PEEPi, including that caused by obstructive lung disease, that seen in patients with acute lung injury and high minute ventilation requirements, and that produced by active expiratory effort, and these were all lumped together in this study. In addition, any clinical interpretation of the varying frequencies of PEEPi in the different modes of mechanical ventilation (which ranged from 64% with intermittent mandatory ventilation to 84% with continuous positive airway pressure) is hampered by a lack of information about which patients were placed on which modes and why.
The incidence of PEEPi (70%) in this report is high, and no doubt in many instances the pressure itself was not clinically important. Although quantititative measurement of PEEPi was possible only 17% of the time, when this was done the PEEPi exceeded 5 cm H2O in only 13% of the measurements. If this proportion held for the whole population and for all the encounters, clinically important PEEPi would have been present in about one examination in seven. Although far less than the 70% detected by examination of the waveforms, this figure would still constitute a clinically important incidence for a phenomenon that can seriously impair cardiac function and lead to life-threatening barotrauma.
Techniques for detecting and/or quantitating PEEPi include:
a. end-inspiratory occlusion technique.
b. stepwise addition of external PEEP.
c. transduction of intragastric pressure.
d. All of the above
e. None of the above
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content