Background: The indications for surgical feeding tube (SFT) placement in trauma patients are poorly defined. Patient selection is critical as complications from SFTs have been reported in up to 70% of patients. A previous analysis by our group determined that 25% of the SFTs we placed were unnecessary and that older patients, patients with head and spinal cord injuries, and patients who needed a tracheostomy were more likely to require long-term SFTs. Following this study, we modified our institutional guidelines for SFT placement. We hypothesized that a more selective placement strategy would result in fewer unnecessary SFTs.
Methods: A retrospective review of all adult patients from 2012 to 2016 with an intensive care unit length of stay longer than 4 days and an SFT placed during admission was conducted. This group was compared to data collected prior to our change in practice (2007-2010). Data from 2011 were excluded as a washout period. "Necessary" SFT use was defined per established guidelines as either daily use of the SFT through discharge or for 28 days or longer and "unnecessary" SFT use as all others.
Results: Two hundred fifty-seven SFTs were placed from 2007 to 2010 and 244 from 2012 to 2016. Following implementation of our selective SFT placement strategy, unnecessary SFT placement decreased from 25% in 2007 to 2010 to 8% in 2012 to 2016 (p < 0.0001). Significant predictors of necessary SFT placement by univariate regression were as follows: increasing age (odds ratio [OR] 1.03/year; 95% confidence interval [CI], 1.01-1.04), head injury (OR, 2.80; 95% CI, 1.71-4.60), cervical spinal cord injury (OR, 4.42; 95% CI, 1.34-14.50), and need for tracheostomy (OR, 1.41; 95% CI, 2.21-7.67). The overall complication rate was 11% (9% in the selective group vs. 13% in the preselective group, p = 0.2574) and was highest following open SFT placement (22%).
Conclusion: A selective placement strategy for SFTs in our trauma population resulted in fewer unnecessary SFTs and a trend toward fewer complications. Surgical feeding tubes should be placed through a percutaneous approach whenever possible.
Level of evidence: Therapeutic study, level IV.