Introduction: Administration of bone marrow-derived cells produces beneficial effects in experimental extrapulmonary acute respiratory distress syndrome (ARDS). However, there are controversies regarding the effects of timing of cell administration and initial insult severity on inflammatory response. We evaluated the effects of bone marrow-derived mononuclear cells (BMDMC) in two models of extrapulmonary ARDS once lung morphofunctional changes had already been installed. Methods: BALB/c mice received lipopolysaccharide (LPS) intraperitoneally (5 mg/kg in 0.5 ml saline) or underwent cecal ligation and puncture (CLP). Control mice received saline intraperitoneally (0.5 ml) or underwent sham surgery. At 24 hours, groups were further randomized to receive saline or BMDMC (2 × 106) intravenously. Lung mechanics, histology, and humoral and cellular parameters of lung inflammation and remodeling were analyzed 1, 3 and 7 days after ARDS induction. Results: BMDMC therapy led to improved survival in the CLP group, reduced lung elastance, alveolar collapse, tissue and bronchoalveolar lavage fluid cellularity, collagen fiber content, and interleukin-1β and increased chemokine (keratinocyte-derived chemokine and monocyte chemotactic protein-1) expression in lung tissue regardless of the experimental ARDS model. Intercellular adhesion molecule-1 and vascular cell adhesion molecule-1 expression in lung tissue increased after cell therapy depending on the insult (LPS or CLP). Conclusions: BMDMC therapy at day 1 successfully reduced lung inflammation and remodeling, thus contributing to improvement of lung mechanics in both extrapulmonary ARDS models. Nevertheless, the different inflammatory responses induced by LPS and CLP resulted in distinct effects of BMDMC therapy. These data may be useful in the clinical setting, as they suggest that the type of initial insult plays a key role in the outcome of treatment.