The Science Journal of the American Association for Respiratory Care

2002 OPEN FORUM Abstracts

A Ventilator Protocol for Fulminating Non-Cardiogenic Pulmonary Edema

John H. Boynton Jr. RRT, Kenneth Hawkins RRT, Dean Holland, RRT, Jason Higgins, RRT, Grant O'Keefe MD MPH, Parkland Health and Hospital System and UTSWMC, Dallas, Texas

Background: For the purpose of our protocol, the setting and stabilization of ventilating end pressure during the initial seventy-two hours post injury is fundamental. The system end pressure is set and maintained at 2cmH2O above the lower inflection point as identified on a static pressure-volume curve. Our protocol is intended to decrease mortality in this high-risk patient population. Methods: The Oxygenation Index (OI)1 of ten ventilated surgical/trauma ICU patients (chosen in succession) that present with fulminating non-cardiogenic pulmonary edema will be compared with OI value (47) which has been associated with 100% mortality in ARDS.1 The patients evaluated are as follows: two patients each with 95% total body surface area (TBSA) burns, one gastric bypass, one hip replacement, and six with pulmonary contusions. All patients were in the initial phase of fluid resuscitation or were immediately post-op. Cardiac origin of the edema was ruled out in each case via ECG's, Swan-Ganz readings, and clinical presentation. Results: All patients had OIs > 47. Patient OI's at protocol initiation ranged from 50-93. Of the ten patients studied, eight patients survived, one patient (pulmonary contusion) died, and one patient was not included in the final evaluation secondary to a breach in protocol. Experience: In the process of refining our approach, we discovered three aspects of management, which were especially important. First, patients must not develop respiratory efforts for the initial 24-72 hours post onset. Our team used paralytic agents to prevent patient respiratory effort. Second, frothy secretions must be removed without compromising system end pressure. We used short bursts (10-40sec.) of O2 in the range of 25-liters/minute through an in-line suction catheter advanced to tracheal level to displace frothy secretions. Third, end pressure should not be compromised during ventilator disconnects. We successfully maintained end pressure during disconnects by cross clamping the ETT with padded hemostats.

We used OI as a marker in our evaluation because no other commonly used index for oxygenation incorporates the importance of ventilating pressures. Fort et al1 associated OIs greater than 47 with 100% mortality. Although Fort's work was done in a broader population, we believe that OI provides an appropriate index for severity of illness in our population. Our initial experience with this protocol suggests a decrease in mortality. Our findings during this evaluation are consistent with concepts presented in Amato's2 work in electrical impedance tomography. We recognize that this is a limited look at a complicated problem and much more work remains. We will continue to gather data as it becomes available, however, the small number of patients that are included each year make a single center study difficult. A multi-center study of prospective patients may be the best approach.


1. High-frequency oscillatory ventilation for adult respiratory distress syndrome-A pilot study Crit Care Med 1997; 25: 937-947

2. Regional pressure volume curves by electrical impedance tomography in a model of acute lung injury. Crit Care Med. 2000 Jan; 28(1): 178-83.