This paper presents clinical experience with the use of total parenteral nutrition (TPN) in patients undergoing major spinal surgery, as well as the identification of possible risk factors involved in the extended duration of TPN. Although TPN is an expensive therapy, it may even represent a cost saving in patients with a high risk of developing malnutrition. Therefore, it is necessary to identify those patients with the greatest likelihood of presenting nutritional deterioration in order to select those who would benefit from post-operative TPN. During the period under analysis (1995-1998), 137 patients were operated on and could be analyzed retrospectively. The variable considered were: age, sex, weight, body mass index, height, diagnosis, type and time of surgery involved, bleeding, duration of TPN and number of days admitted. The statistical analysis was carried out using the SPSS 6.1 for Windows software package. It comprised a descriptive univariate study, a bivariate comparative means analysis using Student's t test and multi-variant analysis using multiple regression. In the bivariate analysis, a greater duration of TPN was found in those patients subjected to circumferential arthrodesis (p = 0.0001), as well as in patients diagnosed as having kyphoscoliosis following neuromuscular disease (p = 0.0041). The multi-variant analysis gave a correlation statistically significant duration of TPN and surgery time (p = 0.00001), intra-operative bleeding (p = 0.01) and weight (p = 0.0002). No link was found between the duration of TPN and age, nor with the number of days spent in hospital.