Acute cor pulmonale in acute respiratory distress syndrome submitted to protective ventilation: Incidence, clinical implications, and prognosis Antoine Vieillard-Baron, MD; Jean-Marie Schmitt, MD; Roch Augarde, MD; J. L. Fellahi, MD; Sebastien Prin, MD; Bernard Page, MD; Alain Beauchet, MD; François Jardin, MD B ased on an echocardiographic definition (1), massive pulmo- nary embolism and acute re- spiratory distress syndrome (ARDS) are the two main causes of acute cor pulmonale (ACP) in adults. ACP com- plicating massive pulmonary embolism has been found to be reversible in many studies, and this reversibility, leading to a high recovery rate with an adequate treatment, has been documented by two- dimensional echocardiography (2). Con- versely, ACP complicating ARDS, which was also documented by two-dimensional echocardiography (3), was associated in the past with high mortality (3– 6). This was consistent with previous anatomical studies showing that ARDS irreversibly damages the pulmonary microvascula- ture (7). In recent years, protective ventilation, with airway pressure limitation, has led to major changes in respiratory support of ARDS patients. A recent clinical study performed in our unit demonstrated marked improvement in ARDS survival rate with this respiratory strategy (8). Thus, in the present prospective study, transesophageal echocardiography (TEE) was used to update the exact prevalence, clinical implications, and prognosis of ACP in an ARDS population submitted to protective ventilation. PATIENTS AND METHODS Patients. During a 5-yr period, from Janu- ary 1996 to February 2001, among 934 pa- tients free from prior cardiopulmonary disease and requiring emergency mechanical ventila- tion for an acute episode of respiratory failure, 75 met the ARDS criteria (9) on the first day of respiratory support and were enrolled in a longitudinal TEE study. Monitoring and Severity Indexes. Hemody- namic monitoring included heart rate by an electrocardiographic lead, systemic arterial pres- sure by an indwelling radial artery catheter, cen- tral venous pressure, and continuous pulse oxymetry. Daily transthoracic echocardiography (TTE) was used to evaluate qualitatively both right and left ventricular function and to mea- sure cardiac output by the Doppler technique (10). Additionally, the first TTE study performed at admission, which was also the first day of mechanical ventilation for each patient (day 1), permitted exclusion of any previously undiag- nosed chronic cor pulmonale. On the third day (day 3) of mechanical ventilation, we calculated the simplified acute physiology score (SAPS II) (11), logistic organ dysfunction score (LODS) (12), and lung in- jury severity score (LISS) (13). Circulatory failure, defined as hypotension (systolic arterial pressure 90 mm Hg by in- vasive monitoring) despite apparently ade- quate fluid resuscitation (central venous pres- sure 12 mm Hg) and requiring continuous infusion of a vasoactive agent, was present in 59 cases at day 3. Vasoactive agents were used at the lowest dosage permitting a systolic ar- From the Medical Intensive Care Unit (AVB, JMS, RA, JLF, SP, BP, FJ) and the Department of Biostatis- tics (AB), University Hospital Ambroise Paré, Assis- tance Publique Hôpitaux de Paris, Boulogne Cedex, France. Copyright © 2001 by Lippincott Williams & Wilkins Context: The incidence of acute cor pulmonale (ACP), a fre- quent and usually lethal complication of acute respiratory distress syndrome (ARDS) during traditional respiratory support, has never been re-evaluated since protective ventilation gained acceptance. Objective: We performed a longitudinal transesophageal echo- cardiographic (TEE) study to determine whether this incidence, and its severe implications for prognosis, might have changed in our unit as we altered respiratory strategy. Design: Prospective open clinical study. Setting: Medical intensive care unit of a university hospital. Patients: Seventy-five consecutive ARDS patients given respi- ratory support with airway pressure limitation (plateau pressure <30 cm H 2 O). Interventions: ACP was defined as a ratio of right ventricular end-diastolic area to left ventricular end-diastolic area in the long axis >0.6 associated with septal dyskinesia in the short axis during TEE examination. Results: Normal right ventricular function was present in 56 patients, whereas right ventricular dysfunction was observed in 19 patients after 2 days of respiratory support. ACP was associ- ated with pulmonary artery hypertension, increased heart rate, and decreased stroke index. Significant impairment of left ven- tricular diastolic function was also seen. All echo-Doppler abnor- malities were reversible in patients who recovered, and the mor- tality rate was the same in both groups (32%). However, ACP patients who recovered required a longer period of respiratory support. A multivariate analysis individualized PaCO2 level as the sole factor independently associated with ACP, suggesting that ACP development in ARDS is influenced by the severity of lung damage and/or the respiratory strategy. Conclusion: Evaluation of right ventricular function by TEE in a group of 75 ARDS patients submitted to protective ventilation re- vealed the persistence of a 25% incidence of ACP, resulting in detrimental hemodynamic consequences associated with tachycar- dia. However, ACP was reversible in patients who recovered and did not increase mortality. (Crit Care Med 2001; 0:●●●–●●●) KEY WORDS: acute cor pulmonale; acute respiratory distress syndrome; circulatory failure; transesophageal echocardiography; protective ventilation 1551 Crit Care Med 2001 Vol. 29, No. 8