Editorials |

Perfluorocarbon Fluid as a Mediator of Pulmonary Barotrauma : A Potential Hazard of Liquid Ventilation FREE TO VIEW

Bob Demers, BS, RRT
Author and Funding Information

Carmel, CA

Correspondence to: Bob Demers, BS, RRT, Demers Consulting Services, 225 Crossroads Blvd, Suite 415, Carmel, CA 93923

Chest. 2000;117(1):8-10. doi:10.1378/chest.117.1.8
Text Size: A A A
Published online

In this issue of CHEST (see page 191), Ferreyra et al report their findings using partial liquid ventilation (PLV) in an animal model. This is a cutting-edge technology, and the article makes for fascinating reading. The perfluorocarbon fluid used in this technique is hyperdense, having a specific gravity of 1.92. This attribute poses a potential hazard for human patients, owing to the elevated hydrostatic pressure that this extremely dense liquid is able to exert.

Consider a case wherein PLV might be employed for an adult patient in the erect posture. In order to illustrate the pattern of distribution of the instillate, we are going to employ the modified roentgenogram shown in Figure 1. This posteroanterior chest radiograph was obtained in preparation for a CT scan, for which horizontal lines are automatically scribed onto the radiograph at 1-cm intervals. As perfluorocarbon fluid is instilled into the airway, it will percolate into the most dependent regions of the lung. Ferreyra et al continued to instill perfluorocarbon into the endotracheal tube until it reached the level of the teeth of the (animal) subject. If this methodology were duplicated in an erect human subject, the pattern of perfluorocarbon distribution would resemble that shown in Figure 2. We have overdrawn a line at each of the centimeter markers, beginning at the lateral costophrenic sulcus, upward to the level of the main stem bronchi. This vertical distance is observed, in this case, to be 8 cm. The posterior sulcus lies below this level, but we have chosen the lateral sulcus as our zero point because it is easier to visualize on the chest radiograph. Perfluorocarbon would rise to the 8-cm level within the lungs, at which point additional fluid would rise within the endotracheal tube. This additional instillate would be prevented from entering the portion of each lung overlying the 8-cm scribe line because the newly loculated gas lying above each hilum would have no route of egress.

Owing to the high density of the perfluorocarbon instillate, the hydrostatic pressure gradient would be far greater than that seen if an equal volume of water had been instilled. Hence, the alveoli at the level of the sulcus would be subjected to a pressure of approximately 15 cm H2O. We have specified the resultant pressure gradient on an axis drawn along the right-hand margin of Figure 2.

Let us assume that a positive end-expiratory pressure (PEEP) of 10 cm H2O were imposed upon the lungs at this point, in a manner analogous to the study of Ferreyra et al. This would result in an augmentation of pressure by that margin throughout the lung. Thus, the alveoli at the lung bases would be exposed to a pressure of approximately 25 cm H2O at end-expiration, not the 10-cm level that the unwary clinician might assume would prevail if he or she failed to take the additional perfluorocarbon-related gradient into account. If a cyclical pressure were now to be imposed upon the lungs in order to undertake PLV, the gradients throughout the lung would escalate accordingly. Pressure fluctuations are propagated at the speed of sound through intrapulmonary gas, and at a marginally higher velocity through liquid. This translates to a speed of about one foot per millisecond. Because pressures are transmitted undiminished throughout the body of any liquid, we can anticipate that any variation in pressure occurring at the central airways will be reflected throughout the lungs almost instantaneously.

Finally, we will assume that a peak inspiratory pressure (PIP) of 30 cm H2O were to be selected. This would be in keeping with current and widespread clinical practice, aimed at avoiding the excessive alveolar pressures that have been shown to be barotraumatic. In this situation, an additional increment of 20 cm H2O would be superimposed cyclically upon the end-expiratory pressure as positive-pressure inflation supervened. In accordance with our previous analysis, a PIP of this magnitude would result in a peak alveolar pressure at the lung bases of approximately 45 cm H2O. Obviously, the failure to account for the augmented alveolar pressures mediated by the presence of perfluorocarbon results in substantial escalations in pressure in dependent lung zones. To be sure, this factor is of such considerable magnitude that it is likely to completely nullify clinicians’ strategies to limit PIP to levels designed to protect against barotrauma, and they would not even be aware that such efforts were being frustrated.

What might be done to avoid incurring excessive pressures during PLV secondary to the mechanism described here? To the extent that the prevailing pathology is confined to a specific segment of the lungs, clinicians would be well advised to orient that zone lowermost. Subsequent introduction of perfluorocarbon will thus tend to selectively fill the impaired regions of the lung. In the best of situations, a unilateral disease process would be present, which would allow for the patient to be placed in the lateral decubitus position. This would be fortuitous, because the lateral dimension of the lung is considerably smaller than its vertical dimension. This would, in turn, result in an attenuated perfluorocarbon-related pressure head in comparison to that which would be obtained in the erect posture. Irrespective of the orientation of the patient, the clinical team would also do well to instill a volume of perfluorocarbon that will rise to a level that falls short of the entry point(s) of the airway(s) of the lung(s) that are subject to instillation. Determination of this point would, of course, be facilitated by the availability of a fluoroscope. Ensuring that the fluid level does not extend into the affected bronchus (or bronchi) would obviate the need to subsequently impose obligatory PEEP in order to enhance the distribution of tidal ventilation. Finally, the clinical team should consider limiting peak inspiratory pressure to a level that prevents excessive distention of the most dependent lung zones. If the fluoroscopically confirmed vertical dimension of the perfluorocarbon bolus residing in the lung(s) were, for example, 6 cm, the prevailing end-expiratory pressure at the base of this fluid column would be (6 × 1.9 =) 11 cm H2O in the absence of PEEP. This would prompt the clinical team to restrict inspiratory pressures to ≤ 19 cm H2O, if 30 cm H2O were the selected target for maximum inflation pressure. The narrowed range of pressures thus imposed might obligate the clinical team to employ a technique that was developed for the express purpose of preventing barotrauma-permissive hypercapnia.

Readers might consider this litany of precautions to be quite elaborate, insofar as they necessitate the procurement of a fluoroscope, not to mention the requirement for additional bedside calculations. On the other hand, a decision to undertake PLV itself obliges us to employ technologic tools that are intrinsically elaborate. Implementation of a few additional precautions in such a high-tech environment does not appear to be too much to ask. In the longer term, it would be useful for pharmacophysiologists to identify an oxygen-bearing fluid with a lower specific gravity, ideally approaching that of water, than that exhibited by perfluorocarbon.

Almost 2 decades ago, Dr. Alfred Fishman1coined a catchy phrase that might be considered the pulmonologist’s call to arms:“ Down with the good lung!” The physiology that undergirds this axiom relates to the fact that (lesion-free) pulmonary parenchyma in dependent zones is considerably better ventilated, and vastly better perfused, than is parenchyma in nondependent zones. Exceptions to the“ Fishman Rule” are rarely encountered,2 but liquid ventilation constitutes one such exception. Earlier, we suggested that the lesion-containing portion of the lung be oriented downward in preparation for instilling perfluorocarbon simply because, in that portion of the lung(s) wherein instillate resides, ventilation (although not oxygenation) will be zero. And, because the fluid will“ put the squeeze” on alveolar capillaries, perfusion will be marginal to absent. Perfluorocarbon-filled alveoli will thus resemble a so-called “silent unit,” with the notable difference being that oxygenation will persist at a presumably brisk rate, despite the absence of the ventilation associated with bulk gas flow. Because they are almost silent, perhaps we should refer to these units as“ reticent.”

Figure Jump LinkFigure 1. Figure 1. A posteroanterior chest radiograph taken in preparation for a CT scan.Grahic Jump Location

Figure Jump LinkFigure 2. Figure 2. Pressure gradient ascribable to perfluorocarbon liquid after instillation.Grahic Jump Location


Fishman, A (1981) Down with the good lung [editorial].N Engl J Med304,537-538. [PubMed] [CrossRef]
Demers, RR Down with the good lung (usually) [editorial].Respir Care1987;32,849-850


Figure Jump LinkFigure 1. Figure 1. A posteroanterior chest radiograph taken in preparation for a CT scan.Grahic Jump Location
Figure Jump LinkFigure 2. Figure 2. Pressure gradient ascribable to perfluorocarbon liquid after instillation.Grahic Jump Location



Fishman, A (1981) Down with the good lung [editorial].N Engl J Med304,537-538. [PubMed] [CrossRef]
Demers, RR Down with the good lung (usually) [editorial].Respir Care1987;32,849-850
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543