How Long Should You Wait Before Removing a Chest Tube?

Abstract & Commentary

By Leslie A. Hoffman, RN, PhD, Department of Acute/Tertiary Care, School of Nursing, University of Pittsburgh, is Associate Editor for Critical Care Alert.

Dr. Hoffman reports no financial relationship to this field of study.

Synopsis: A normal chest radiograph 3 hours after placing a chest tube on water seal drainage effectively excluded development of a clinically significant pneumothorax.

Source: Schulman CI, et al. How long should you wait for a chest radiograph after placing a chest tube on water seal? A prospective study. J Trauma. 2005;59:92-95.

Standard teaching advocates placing a chest tube on water seal in patients with a prior hemothorax or pneumothorax and obtaining a chest radiograph (CXR) the following morning. This study was conducted to determine if this time interval could be shortened, thereby permitting earlier discharge and reducing costs. The sample included 119 patients admitted to a trauma service with a unilateral chest tube, no pneumothorax on prior CXR, tube drainage < 300 mL for 24 hours, no evidence of air leak on continuous suction, No significant residual hemothorax or empyema and, if on mechanical ventilation, ≤ 10 cm H2O positive end-expiratory pressure (PEEP). The chest tube was inserted as a consequence of blunt trauma (64%), penetrating trauma (28%), or a general surgery emergency (8%). Of those enrolled, 38% required mechanical ventilation. Once patients met entry criteria, they were placed on water seal drainage without suction. A CXR was obtained in 3 hours (mean 3.1 ± 2.1 hours later) and also the following morning (mean 17.6 ± 8.0 hours later). Each CXR was reviewed by an attending radiologist.

After chest tube removal, 31 (26%) patients had a pneumothorax, of which 22 were early and 9 late. Three patients in the early group had a clinically significant pneumothorax (n = 2) or an increase in size of pneumothorax on the follow-up CXR (n = 1). None of the patients in the late group had a clinically significant pneumothorax or required further attention. There were no significant differences between those who did and did not develop a pneumothorax in regard to previous drainage, injury type, or requirement for mechanical ventilation. Those with a late pneumothorax had had their chest tubes for a shorter time before being placed on water seal (2.3 days vs 3.8 days; P < .05).


There is no well-established standard of care regarding the time patients should be observed on water seal drainage to exclude the potential of prematurely removing a chest tube. Textbooks provide varying advice, and formal education on chest tube management is not common. Few studies have examined the question of when a CXR should be obtained to exclude the potential of a clinically significant pneumothorax. A common practice is to obtain a CXR immediately after placing the chest tube on underwater seal drainage and again the next morning. The risk of a clinically significant pneumothorax appears to be small. However, there is concern that premature removal will result in the need for chest tube reinsertion and prolong recovery. Consequently, a short trial of water seal drainage appears warranted.

Based on findings from this study, the authors advocate an algorithm designed to maximize patient safety, minimize time before chest tube removal, and lead to earlier discharge (see Figure).

With this protocol, the patient begins the day with a water seal trial without suction, followed by a 3-hour confirmatory CXR. If the CXR is negative, the chest tube is removed and a second CXR is obtained 3 hours later. If the second CXR is negative, the patient can be discharged home the same day, if their condition otherwise permits.

The limitations of the study include a relatively small sample size, a one-institution setting, and a small number of general surgery patients, making generalization to the latter population less well defined. Of note, there was a relatively high incidence of early (18%) and late pneumothorax (8%) which the authors attribute to special attention from the radiologist. Over half of these pneumothoraces (19/31) were small apical air collections that resolved by the next morning without any intervention. Results of this study suggest the potential to move more quickly to discharge using the proposed protocol, thereby saving resources and speeding patient recovery.