Severe sepsis and septic shock in patients with pre-existing non-cardiac pulmonary hypertension: contemporary management and outcomes.

Objective: To review treatment and outcomes of septic shock in patients with pulmonary hypertension (PH) managed at a tertiary care institution.

Design, Setting And Patients: We identified consecutive patients with non-cardiac PH (non-Group 2 in the World Health Organization classification) who were treated for septic shock in four intensive care units at a tertiary care institution between July 2004 and July 2007. Patients with a left ventricular ejection fraction < 50%, diastolic dysfunction, pericardial effusion or significant valve disease were excluded. Descriptive statistics were used to analyse the data.

Main Outcome Measures: Hospital mortality, duration of vasopressor and ventilatory support, length of hospital and ICU stay.

Results: The final group for analysis comprised 82 patients. The major causes of PH were chronic obstructive pulmonary disease, interstitial lung disease and portopulmonary hypertension. PH was mild in 46 patients (56%), moderate in 21 (26%) and severe in 15 (18%). Vasopressor treatment was initiated in 69 patients (84%) within the first 48 hours: noradrenaline was most commonly used (53 patients, 65%), and 51 patients (62%) were treated with more than one agent. Sixty-seven patients (82%) were mechanically ventilated, and 33 (40%) required renal replacement therapy. Fortythree patients (52%) survived to hospital discharge; 23 (28%) remained alive at 1 year. Hospital mortality increased with severity of PH: 28% in mild, 67% in moderate and 80% in severe PH. Nonsurvivors were more likely to have plateau pressures beyond 30 cm H(2)O while mechanically ventilated within the first 48 hours in the ICU (56% v 29%, P = 0.03), to develop atrial fibrillation (AF) (46% v 12%, P < 0.001), and to require longer vasopressor support (mean, 5.3 v 2.6 days, P = 0.003). In a multivariate logistic regression analysis, severity of PH (odds ratio [OR], 1.55; 95% CI, 1.04-2.46; P = 0.04), new-onset AF (OR, 6.51; 95% CI, 2.24-22.07; P < 0.001) and longer duration of vasopressor support (OR, 1.15; 95% CI, 1.03-1.34; P = 0.04) were associated with increased hospital mortality.

Conclusions: The severity of PH, new-onset AF, and longer vasopressor support were associated with poor outcomes in patients with PH who developed severe sepsis and septic shock.