Idiopathic interstitial pneumonias are often complicated by pulmonary hypertension, increasing morbidity and mortality. There are no approved treatments for pulmonary hypertension associated with idiopathic interstitial pneumonia (PH-IIP). We aimed to evaluate the efficacy and safety of riociguat in patients with PH-IIP. RISE-IIP was a double-blind, randomised, placebo-controlled study done at 65 pulmonary hypertension and interstitial lung disease centres in 19 countries to evaluate the efficacy and safety of riociguat in patients with PH-IIP. Eligible patients were adults (aged 18-80 years) diagnosed with idiopathic interstitial pneumonia (as per American Thoracic Society/European Respiratory Society/Japanese Respiratory Society/Latin American Thoracic Association guidelines), forced vital capacity (FVC) of at least 45%, 6MWD of 150-450 m, WHO functional classes II-IV, precapillary pulmonary hypertension confirmed by right heart catheterisation, systolic blood pressure of at least 95 mm Hg, and no signs or symptoms of hypotension. Patients were randomly allocated (1:1) using an interactive voice and web response system to riociguat (0·5-2·5 mg three times daily) or placebo for 26 weeks (main study), after which they could enter an open-label extension in which all patients received riociguat. The primary endpoint was change in 6-min walking distance (6MWD) in the intention-to-treat population. Prespecified safety variables included adverse events and serious adverse events, laboratory parameters, and adverse events of special interest (haemoptysis and symptomatic hypotension), assessed in the intention-to-treat population. This trial is registered with ClinicalTrials.gov, number NCT02138825. Between June 4, 2014, and May 5, 2016, we enrolled 229 participants. After the exclusion of 82 participants, 147 were randomly allocated to treatment (73 to riociguat, 74 to placebo). The study was terminated early (median treatment duration 157 days [range 6-203]) at the request of the data monitoring committee owing to increased serious adverse events (main study: 27 [37%] of 73 participants in the riociguat group vs 17 [23%] of 74 in the placebo group) and mortality in patients receiving riociguat, and the absence of efficacy signals in the riociguat group. 11 patients died in the main study (eight in the riociguat group, three in the placebo group), and nine died in the extension phase (one in the riociguat group, eight in the former placebo group; all received riociguat). In the main study, the most common adverse events were peripheral oedema (16 [22%] of 73 in the riociguat group vs seven [9%] of 74 in the placebo group) and diarrhoea (11 [15%] vs seven [9%]). The most common serious adverse events were worsening of interstitial lung disease (main study: six [8%] of 73 in the riociguat group vs five [7%] of 74 in the placebo group) and pneumonia (four [5%] vs one [1%]). Riociguat did not improve 6MWD versus placebo at 26 weeks (least-squares mean difference 21 m; 95% CI -9 to 52). In patients with PH-IIP, riociguat was associated with increased serious adverse events and mortality, and an unfavourable risk-benefit profile. Riociguat should not be used in patients with PH-IIP. Bayer AG and Merck & Co.