Survival of home parenteral nutrition-treated patients: 20 years of experience at the Mayo Clinic.

OBJECTIVE: To present the largest single institutional review of demographics, associated primary diseases, and survival of patients receiving home parenteral nutrition (HPN).
MATERIAL AND METHODS: We conducted a retrospective review of medical records of all Mayo Clinic patients treated with HPN between 1975 and 1995. The probability of survival was calculated by using Kaplan-Meier analysis.
RESULTS: In the 225 study patients requiring HPN (median age, 51 years), the main underlying primary diseases were as follows: inflammatory bowel disease (IBD) (N = 50), nonterminal active cancer (N = 39), ischemic bowel (N = 35), radiation enteritis (N = 32), motility disorder (chronic pseudo-obstruction) (N = 26), and adhesive intestinal obstruction (N = 18). Other conditions included intestinal and pancreatic fistula, refractory sprue, dumping syndrome, and protein-losing enteropathy. The overall probability of 5-year survival during HPN was 60%. The probability of survival at 5 years based on the primary disease was 92% for IBD, 60% for ischemic bowel, 54% for radiation enteritis, 48% for motility disorder, and 38% for cancer. The probability of 5-year survival stratified by age at initiation of HPN was as follows: younger than 40 years, 80%; 40 through 60 years, 62%; and older than 60 years, 30%. Most deaths during therapy with HPN were attributable to the primary disease. Among the 20 patients who died of an HPN-related cause, 11 deaths were from catheter sepsis, 4 from liver failure, 2 from venous thrombosis, and 2 from metabolic abnormalities.
CONCLUSION: Survival of HPN-treated patients is best predicted on the basis of the primary disease and the age at initiation of HPN. Patients with IBD and age younger than 40 years have a better 5-year survival in comparison with other groups. Most deaths during treatment with HPN are a result of the primary disease; HPN-related deaths are uncommon.