Although extrapulmonary ALI/ARDS is a common clinical entity, most animal models used to study this disease are induced by direct lung injuries. Our intention was therefore to investigate whether a condition resembling ALI/ARDS develops during the course of a fecal peritonitis in pigs; in that case experimental peritonitis would also prove as a clinically relevant ARDS model.

In 10 anesthetized, mechanically ventilated, and instrumented pigs fecal peritonitis was induced by inoculating autologue feces pellets suspended in saline. Mechanical ventilation was set with VT = 8 ml/kg, FiO₂ to reach a SaO₂ target of >90%, PEEP = 10 cmH₂O if PaO₂/FiO₂ > 300 and 12 cmH₂O if PaO₂/FiO₂ < 300, and respiratory rate to obtain a PaCo₂ of 35–45 mmHg. Before as well as 12 and 24 hours after peritonitis induction we measured the PaO₂/FiO₂ ratio, the total compliance of the respiratory system (C), calculated as VT/(P_plateau – PEEP) and inspiratory airway resistance (R_i) calculated as (P_max – P_plateau)/mean inspiratory flow. Data are mean [range].

For data see Table 1. During the course of the 24-hour study period, six of 10 animals developed gas exchange deteriorations consistent with the ARDS definition; two further animals fulfilled the gas exchange referred to as ALI. Impairment in lung mechanics over time is reflected by the decreasing C values.

We conclude that an ALI/ARDS-like state is developed by most pigs during fecal peritonitis and that this peritonitis model may therefore serve as an extrapulmonary ARDS model. However, this condition develops after a prolonged period of approximately 12–18 hours, and the severity of the condition in single animals may be less predictable when compared with ARDS models induced by direct lung injury. Furthermore, it should be emphasized that pulmonary function in pigs is markedly different from humans in as much as no collateral ventilation exists in this species 1, and that pulmonary blood flow regulation is far more susceptible to hypoxia in pigs when compared with other species including humans 2. Lung function data derived from pig models should therefore always be interpreted cautiously if clinically relevant conclusions have to be drawn.