Thoracolumbar bursT fracTures, shorT x long fixaTion: a meTa-analysis Fraturas toracolombares do tipo explosão, Fixação curta x longa: uma metanálise Fracturas toracolumbares tipo explosión, Fijación corta x larga: un meta-análisis ABSTRACT The number of fixed segments in the surgical treatment of thoracolumbar burst fractures remains controversial. This study aims to compare the results of short and long fixation in thoracolumbar burst fractures through a meta-analysis of studies published recently. MEDLINE and Cochrane databases were used. Randomized controlled trials and non-randomized comparative studies (prospective and retrospective) were selected. Data were analyzed with the software Review Manager. There was no statistically significant difference in the Cobb angle of preoperative kyphosis. Long fixation showed lower average measurements postoperatively (MD = 1.41; CI = 0.73-2.08; p<0.0001) and in the last follow-up (MD = 3.98; CI = 3.22-4.75; p<0.00001). The short fixation showed the highest failure rates (RD = 4.03; CI = 1.33-12.16; p=0.01) and increased loss of height of the vertebral body (MD = 1.24; CI = 0.49-1.98; p=0.001), with shorter operative time (MD = -24.54; CI = -30.16 --18.91; p<0.00001). There was no significant difference in blood loss and clinical outcomes. The high rates of kyphosis correction loss with short fixation and the lower correction rate in the immediate postoperative period were validated. There was no significant difference in the blood loss rates because arthrodesis was performed in a short segment in the analyzed studies. The short fixation was performed in a shorter operative time, as expected.No study has shown superior clinical outcomes. The short fixation had worse rates of kyphosis correction in the immediate postoperative period, and increased loss of correction in long-term follow-up, making the long fixation an effective option in the management of this type of fracture.