Home parenteral support in patients with incurable cancer. Patient characteristics of importance for catheter related complications and overall survival.

Purpose: It is uncertain if home parenteral support (HPS) is of advantage in patients with incurable cancer and intestinal failure, functional obstruction or severe malabsorption. From a single centre cohort we present characteristics of patients with incurable cancer treated with HPS.

Methods: Over a ten year period (2005-2015) data were retrospectively collected on patients with incurable cancer discharged on HPS from a Danish tertiary referral centre. Data on socio-demographics, catheters and parenteral nutrition, catheter related complications, re-admissions and mortality were analysed. The inflammation based score; modified Glasgow prognostic score (mGPS) was investigated as a prognostic score by Cox proportional hazard regression analyses adjusted for sex, age, diagnosis, and pathophysiological conditions.

Results: Eighty patients with incurable cancer, aged 25.1-83.6 (median 63.8) were identified. Patients with gynaecologic cancer accounted for 25% of the cohort, thus women predominated. Short bowel syndrome was more prevalent in the patients with gynaecologic or lower gastrointestinal cancer compared to the upper gastrointestinal cancer. Catheter related complications occurred in a minority of patients (31%); most frequent was catheter related bloodstream infection (CRBSI). CRBSI rate was overall 0.97 pr 1000 catheter days, depending on diagnosis. Eleven percent had several infections, and 75% did not have any. Patients self-administering the catheter were younger, less frail and had fewer CRBSI events. Re-admissions were prevalent, and only one fifth of the patients had no re-admissions after initiation of HPS. Patients with mGPS 0 or 1 survived significantly longer, median 372 (CI 39-2006) days versus patients scoring 2 in mGPS, median 43 (CI 6-578) (p < 0.01). In patients with mGPS 0 or 1 survival at six months was 75% and in patients with mGPS 2, 20%. In multivariate cox regression analyses mGPS 2 was a significant predictor of mortality (HR 4.66, 95% CI 2.65-8.20, p < 0.01).

Conclusions: It is feasible to offer HPS to patients with incurable cancer. Frequency of catheter related infections is acceptable but most patients will be re-readmitted after initiation of HPS. Predictors of survival in patients with incurable cancer on HPS may include mGPS. However, our study does not give a clear answer; when to prescribe HPS and who might possible benefit from the treatment in patients with incurable cancer.