Preliminary report of a prospective, randomized trial of underwater seal for spontaneous and iatrogenic pneumothorax

J Am Coll Surg. 2007 Jan;204(1):84-90. doi: 10.1016/j.jamcollsurg.2006.09.012. Epub 2006 Nov 28.


Background: Management of pneumothorax has traditionally been tube thoracostomy and -20 cm H2O suction. The purpose of our study was to determine if underwater seal in iatrogenic and spontaneous pneumothoraces is safe and efficacious and if small-caliber chest tubes are appropriate for routine use in pneumothorax.

Study design: From April 2001 through October 2003 patients with iatrogenic or spontaneous pneumothorax were enrolled in this prospective, randomized trial. Small-bore catheters were inserted. Initial management was 1 hour -20 cm H2O suction, chest radiography, and randomization into -20 cm H2O suction, -10 cm H2O suction, or underwater seal. Tubes were discontinued at 48 hours if there were no pneumothoraces and no air leaks. Those with air leaks and recurrent pneumothoraces persisting 5 days underwent pleurodesis. The primary end point was successful chest tube removal at 48 hours. The secondary end point was need for pleurodesis.

Results: Twenty-nine patients were analyzed. Seven were randomized to -20 cm H2O suction, 11 to -10 cm H2O suction, and 11 to underwater seal. Most (59%, 17 of 29) chest tubes were successfully removed 48 hours after placement: 57% (4 of 7) after -20 cm H2O suction, 73% (8 of 11) after -10 cm H2O suction, and 45% (5 of 11) after underwater seal (p = 0.48). Seven (24%) required pleurodesis: 29% (2 of 7) after -20 cm H2O suction, 27% (3 of 11) after -10 cm H2O suction, and 18% (2 of 11) after underwater seal (p = 0.70).

Conclusions: Early underwater seal appears to be safe for treating iatrogenic and spontaneous pneumothoraces. It can achieve comparable frequencies of early chest tube removal and avoidance of operation compared with traditional management. A larger, multi-institutional study should be performed to demonstrate that pneumothorax treatment can effectively incorporate small-caliber tubes and underwater seal.

Publication types

  • Randomized Controlled Trial

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Aged, 80 and over
  • Chest Tubes
  • Female
  • Follow-Up Studies
  • Humans
  • Iatrogenic Disease
  • Immersion*
  • Male
  • Middle Aged
  • Pneumothorax / etiology*
  • Pneumothorax / therapy*
  • Prospective Studies
  • Suction / instrumentation
  • Treatment Outcome
  • Water*


  • Water