Background: It is difficult to predict which patients with a postsurgical short bowel will require long-term parenteral nutrition.
Methods: We performed a retrospective prognostic study for the time to home parenteral nutrition or death from malnutrition (nonautonomy), on the basis of 103 patients with a residual short bowel of 17 to 150 cm. The influence of anatomic variables was summarized through the use of Cox regression model.
Results: Of the 103 patients included, 24 lost nutritional autonomy. Three anatomic variables were identified as having independent predictive information; remaining small bowel length (measured on small bowel x-rays; p = .0001), and jejunoileal anastomosis (p = .01) promoted autonomy, whereas end jejunostomy (p = .002) increased the risk of losing nutritional autonomy.
Conclusions: On the basis of these results and on the relative weight of these variables, high-risk patients for loss of nutritional autonomy were defined as those with jejunoileal anastomosis and a remaining small bowel length < 35 cm, patients with jejunocolic anastomosis and remaining small bowel length < 60 cm, and patients with an end jejunostomy and remaining small bowel length < 115 cm. This classification was thereafter validated on a prospective series of 32 patients.