Rationale: A U-shaped relationship should exist between lung volume and pulmonary vascular resistance (PVR), with minimal PVR at FRC. Thus, positive end-expiratory pressure (PEEP) in patients with acute respiratory distress syndrome (ARDS) should increase PVR if it induces significant lung distension compared with recruitment. However, this has never been proved in patients. Objectives: To study the effects of PEEP on PVR according to lung recruitability, evaluated by the recruitment-to-inflation (R/I) ratio. Methods: In patients with ARDS, we measured hemodynamic (pulmonary artery catheter), echocardiographic, and ventilatory variables (including esophageal pressure) at both low PEEP and higher PEEP by 10 cm H2O. Preload responsiveness was assessed by the passive leg-raising test at high PEEP. Measurements and Main Results: We enrolled 23 patients, including 10 low recruiters (R/I <0.5) and 13 high recruiters (R/I ⩾0.5). Raising PEEP from 4 (2-5) to 14 (12-15) cm H2O increased PVR in low recruiters (from 160 [120-297] to 243 [166-380] dyn·s/cm5; P < 0.01), whereas PVR was unchanged in high recruiters (from 224 [185-289] to 235 [168-300] dyn·s/cm5; P = 0.55). Right-to-left ventricular end-diastolic area ratio simultaneously increased in low recruiters (from 0.54 [0.50-0.59] to 0.64 [0.56-0.70]; P < 0.01) while remaining stable in high recruiters (from 0.70 [0.65-0.79] to 0.68 [0.58-0.80]; P = 0.48). Raising PEEP decreased cardiac index only in preload responsive patients. Conclusions: PEEP increases PVR only when it induces significant lung distension compared with recruitment according to the R/I ratio. Tailoring PEEP on this recruitability index should mitigate its hemodynamic effects.