Banko D, Rosenthal N, Chung J, Lomax C, Washesky PF. Comparing the risk of bloodstream infections by type of parenteral nutrition preparation method: A large retrospective, observational study.
Clin Nutr ESPEN 2019;
30:100-106. [PMID:
30904208 DOI:
10.1016/j.clnesp.2019.01.011]
[Citation(s) in RCA: 8] [Impact Index Per Article: 1.6] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/20/2018] [Revised: 01/16/2019] [Accepted: 01/24/2019] [Indexed: 11/26/2022]
Abstract
BACKGROUND AND AIMS
Complications such as blood stream infections (BSI) have been observed with the administration of parenteral nutrition (PN). Prior published studies reported the incidence of BSI for inpatient hospitalizations by comparing patients treated with custom compounded parenteral nutrition to those treated with premixed multichamber bag (MCB) formulations. Previous publications grouped patients treated with MCBs into a single category and no distinction was made between patients receiving only a MCB and those receiving a MCB supplemented with manual additions. This Study aims to assess differences in risk of blood stream infection, cost, and clinical outcomes among patients receiving multichamber bag parenteral nutrition products only (MCB-only), MCB with additions (MCB-addition), and compounded (COM) PN products using seven years of Premier Healthcare Data from 688 hospitals in the United States of America.
METHODS
Adult inpatients who were discharged between 01/01/2008 and 12/31/2014, had a hospital length of stay ≥3 days and received PN during the index hospitalization were analyzed. PN preparation method was determined by billing charge descriptions. BSI was defined as having primary or secondary ICD-9 diagnosis codes of 038.x (septicemia), 995.91 (sepsis), 995.92 (severe sepsis), and 790.7 (bacteremia). Multivariable regression models were used to assess effect of PN preparation on patient outcomes, adjusting for confounders.
RESULTS
84,564 patients were analyzed (MCB-only: 6.3%; MCB-addition: 14.8%; COM: 78.9%). Multivariable analysis indicated that compared to COM group, MCB-addition group had similar risk of BSI (7.0% vs. 6.8%, P > 0.05) and a 2.7% lower average total hospitalization cost ($28,072 vs. $28,861, P < 0.05) but had a higher PN treatment cost ($1135 vs. $1,031, P < 0.05) and a higher percentage of being discharged to rehabilitation or other acute care facilities (39.4% vs. 31.1%, P < 0.05). MCB-only group had lower risk of BSI and hospitalization cost.
CONCLUSIONS
In the U.S., compounded PN is the most commonly used in clinical practice followed by MCB with additions. MCB-addition group had similar BSI risk with COM. The slightly lower overall cost in MCB-addition group may be offset by higher post-hospitalization care cost to providers and payers under bundled payment methods in the U.S.
Collapse