Pneumothorax is present in about 20% of blunt major trauma cases. Insertion of an intercostal tube drainage is one effective treatment, however it is unclear whether the thoracostomy has more advantages if placed in the ventral (2.-3. intercostal space) or lateral (4.-6. intercostal space) approach. The aim of this study was to determine, whether there are any differences between the two approaches in respect of malposition and complications. The data from 851 consecutive patients, admitted to our trauma centre from January 2000 to June 2004, was collected and analysed prospectively. The inclusion criteria were: ISS > or = 16, insertion of an intercostal tube and subsequent thoracic computed tomography. Epidemiological and physiological data were analysed together with the location of the tube (ventral or lateral). The attending physician was free to choose the location of insertion. Chest tubes placed both on-scene and in-hospital chest tubes were investigated. Malpositions, defined as extrathoracic, abdominal, parenchymal or interlobal positions, were analysed by reviewing the computed tomography of the thorax (CT). Complications, like injuries to vessels or organs, infection or empyaema were analysed using our standardised prospective trauma protocol. Furthermore, the rate of clinically relevant malfunctions due to malposition was investigated as well as the number of chest tubes that had to be replaced. One hundred and one chest tubes were inserted in 68 patients with multiple trauma (mean age 40.7, ISS=38.1, AIS thorax=3.9). In 21 cases a ventral approach was chosen (20.8%) and in 80 a lateral approach (79.2%). CT revealed malposition in two of the ventrally placed tubes (9.5%) and in 20 of the laterally placed tubes (25%) (p=0.15, Fisher's exact test). One tube was identified in a subcutaneous location 17 chest tubes, after ventral approach all of them as a result of lateral approaches, were placed in the interlobe. No interlobal positions were observed in the ventral group. The interlobal position was found to be significantly higher in the lateral approach (p=0.013, Fisher's exact test). Clinically relevant malfunction was diagnosed in 6 of the 22 malpositioned chest tubes (5.9%). These tubes had to be repositioned, one was placed ventrally, the other five were placed laterally. In our setting physicians preferred the lateral approach on-scene as well as in-hospital. In every fifth patient malpositioning of the tube was observed, mostly interlobal after lateral chest tube, however only few were associated with relevant clinical malfunctions. The probability of interlobal malpositioning is significantly higher when using the lateral approach as opposed to the ventral approach. Correction of malpositioned and ineffective chest tubes was necessary in every 17th case. No statistically significant difference between the two approaches for functional malposition was observed. Hence both approaches for emergency chest tube insertion seem to be equally justified.