Background: Diagnosis of co/superinfection in patients with Acute Respiratory Distress Syndrome (ARDS) is challenging. The FilmArray Pneumonia plus Panel (bioMérieux, France), a new rapid multiplex Polymerase Chain Reaction (mPCR), has never been assessed on a blinded protected telescope catheter (PTC) samples, a very common diagnostic tool in patients under mechanical ventilation. We evaluated the performance of mPCR on PTC samples compared with conventional culture and its impact on antibiotic stewardship. Methods: Observational study in two intensive care units, conducted between March and July 2020, during the first wave of the COVID-19 pandemic in France. Results: We performed 125 mPCR on blinded PTC samples of 95 ARDS patients, including 73 (77%) SARS-CoV-2 cases and 28 (29%) requiring extracorporeal membrane oxygenation. Respiratory samples were drawn from mechanically ventilated patients either just after intubation (n = 48; 38%) or later for suspected ventilator-associated pneumonia (VAP) (n = 77; 62%). The sensitivity, specificity, positive, and negative predictive values of mPCR were 93% (95% CI 84–100), 99% (95% CI 99–100), 68% (95% CI 54–83), and 100% (95% CI 100–100), respectively. The overall coefficient of agreement between mPCR and standard culture was 0.80 (95% CI 0.68–0.89). Intensivists changed empirical antimicrobial therapy in only 14% (18/125) of cases. No new antibiotic was initiated in more than half of the CAP/HAP pneumonia-suspected cases (n = 29; 60%) and in more than one-third of those suspected to have VAP without affecting or delaying their antimicrobial therapy. Conclusions: Rapid mPCR was feasible on blinded PTC with good sensitivity and specificity. New antibiotics were not initiated in more than half of patients and more than one-third of VAP-suspected cases. Further studies are needed to assess mPCR potential in improving antibiotic stewardship.