Randomization was performed at a 1:1 ratio using a parallel group design. Randomization lists were generated at the biostatistical center using the software nQuery, release 4, and based on these lists, numbered envelopes were provided and used for randomization (stratified for center and using blocks of random length). For each center, a separate spate list was generated, and closed envelopes were supplied to the participating centers. Envelopes were opened only by the treating physician. The randomization number and treatment were recorded in the ID screening and enrollment list, dated and signed. The signed sheet was then stored at the participating center. Random treatment allocation was used to protect against selection bias. Concealment bias was not present, because the person who was recruiting patients was informed after recruitment about the assigned study arm. The primary and secondary endpoints were objectively measurable, which excluded information bias. Intervention was inhalation with Iloprost (20 µg/3times per day in 10 ml NaCl 0.9% for 5 days) or inhalation of NaCl 0.9% (10 ml) as Placebo [11 (link)].
Free full text: Click here
