INTRODUCTION: In mainland China, there are no special care centers (long-term acute care, weaning, chronic care facilities) for patients requiring prolonged mechanical ventilation (PMV). Our goal was to characterize the prevalence and outcome of patients undergoing PMV in Chinese intensive care units (ICUs). METHODS: A prospective 1-d prevalence study was performed at 55 ICUs, with 28-d follow-up. RESULTS: On the observation day, 622 adult patients occupied ICU beds. Enrollment criteria were met by 302 subjects receiving invasive mechanical ventilation, of which 109 (36.1%) had received ventilation for more than 21 d (median 51, 21-3,419), which was defined as PMV. During the following 28 d, another 45 subjects were classified as receiving PMV, but only 5% (3/58) of the subjects who were newly admitted to the ICU on the study day received PMV. Thirty-six (22.9%) of the 157 subjects receiving PMV were weaned, and 81 (51.6%) continued ventilation in the ICU. In the logistic regression analysis, age > 74 y (odds ratio = 2.78, 95% CI 1.05-7.40, P = .041) and chronic congestive heart failure (odds ratio = 12.23, 95% CI 1.48-101.05, P = .020) were associated with failure to wean in 28 d, while acute respiratory distress syndrome (ARDS) as the reason for mechanical ventilation (odds ratio = 0.14, 95% CI 0.04-0.52, P = .003) was associated with successful weaning. CONCLUSION: The number of subjects receiving PMV was surprisingly high in this cross-section of Chinese ICUs. In the following 28 ICU days, only a small proportion of these subjects were weaned. Age and chronic heart dysfunction were high risk factors for weaning failure.