BPD is the most common morbidity associated with RDS. Treatment with steroids can limit disease severity; it is thus of utmost importance to identify patients at risk of developing BPD. Aim of the study: To determine whether lung function test parameters obtained within 48 hours of birth can predict BPD. Subjects: 43 non-consecutive premature neonates ventilated for RDS in a tertiary care Pediatric ICU. All were treated with a single 200 mg/kg dose of Curosurf*. Interventions: Disease severity (a/AO2 ratio: PaO2 mmHg/FIO2-PaO2mmHg), Rrs (cmH2O/ml/s) and compliance, Crs, (ml/cmH2O/kg) (by passive expiration, Sensor Medics 2100) were assessed at baseline (HO, before surfactant administration) and at H6, H18, H24 and H48. Statistical analysis: BPD was defined as oxygen requirement at 28 days post-natal age and 36 weeks of gestation, with compatible radiographic changes. The risk of DBP was determined by the odds ratio (95% CI). Results: Eight patients had BPD (BW 1180±250g, GA 29±1.6weeks) (mean±SD), 31 had a favorable outcome (1900±700g; 32.2±3weeks) and 4 died. Risk for BPD (OR, 95% CI) were as follows: GA < 29 weeks (6.75; 1.19 - 38.4), BW < 1300 g(8.67;1.55 - 48.5), HO a/AO2 < 0.12 (6.25; 1-39), HO Crs < 0.4 (0.46; 0.08-2), HO Rrs > 0.15 (3; 0.52-17.1), H24 a/A 02 < 0.35 (6.36; 0.6-62.7), H24 Crs < 0.6 (1.33; 0.12-14.9), H24 Rrs > 0.16 (11.2; 1-125.6). Conclusion: The risk of subsequent BPD was higher in the most preterm neonates with severe initial disease, and possibly H24 Rrs. We speculate that the last parameter could be related to smaller airways and/or early airway inflammatory changes. This has to be confirmed by further studies.