This prospective randomized multicenter study is intended to investigate tolerance and effectiveness of thoracic drainage conducted by Seldinger technique with small drains, or by a surgical-like technique with large armed drains, in intensive care units patients.

Drainage of pleural effusion and pneumothorax is a common feature in Resuscitation, Intensive Care Units (ICU) and Continuing Care Units (CCU). Although they are associated with a low incidence of complications (ranging from 0 to 8%), some of these can become fatal if they are associated with a visceral puncture (liver, spleen, lung parenchyma or heart by instance). It has been reported in the literature that complications were greater in case of drainage with large diameter drains set up by so-called "surgical-like" technique.

The choice of the type of chest tube is usually guided by the indication of drainage or the habits and / or experience of the practitioner. In the case of liquid pleural effusions, it may be preferable to use small diameter drains, whereas in the case of suspicious thick effusions such as empyema or blood, it may be preferable to use drainage drains of a larger diameter. However, results of retrospective analyzes seem to suggest the versatile and effective use of small-bore chest tubes in any of these indications without increasing complications' rates such as clogging.

However, no prospective randomized controlled trial (RCT) has studied this issue to date. Therefore, the investigators propose to perform a multicenter RCT in ICU and CCU patients requiring pleural drainage for any indication or underlying disease.

This prospective RCT is intended to investigate tolerance and effectiveness of thoracic drainage conducted by Seldinger technique with small drains, or by a surgical-like technique with large armed drains. Furthermore, they want to estimate the respective costs, identify the difficulties related to both strategies, recognize associated practices (ultrasound-guidance, implantation site, operator's competence), and finally point out the secondary determinants of tolerance and effectiveness.

Inclusion Criteria:

• Of-age patient (>18years)
• Patient admitted in ICU or CCU
• Patient requiring a pleural drainage, semi-urgent or planned
• Patient with a social security insurance

Exclusion Criteria:

• Patient under guardianship
• Severe or uncompensated bleeding disorders
• Thoracic trauma at the acute phase (<6 hours)
• Compressive pneumothorax requiring immediate and urgent needle exsufflation
• No thoracic drainage (whatever the technique used) performed previously during the same stay in ICU or CCU.

Since the study is based on two different techniques, the masking of participants, care providers and local investigator is impossible.