Safety and Ecacy of Multi-chamber Bag Parenteral Nutrition

Background There is a signicant degree of debate regarding the use of standardized parenteral solutions. Multi-chamber bag parenteral nutrition (MCB-PN) showed advantages over compounded PN in previous literature. Meanwhile, some literature has shown the limitations in use of MCB-PN. This study was conducted to evaluate the nutritional ecacy and safety of commercially available MCB-PN. Methods All adult hospitalized patients who have been on MCB-PN for at least seven consecutive days at King Faisal Specialist Hospital & Research Center in Riyadh from January 2015 until December 2019 were included. Laboratory parameters were evaluated before PN started, which was used as a baseline, and every seven days while on PN. Primary outcomes were the percentage of patients who achieved calculated target calories and protein and the percentage of patients who developed electrolyte abnormalities. The secondary outcome was a percentage of adverse drug reactions during the treatment period. treatment


Introduction
Parenteral nutrition preparations are considered a high-alert medication according to ISMP Medication Safety Self-Assessment® for High-Alert Medications. The chances for errors, contamination, and complications related to PN therapy are high. Standardization is a suggested strategy to ensure patient safety associated with PN therapy. As mentioned in the 2007 American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.), position statement on PN standardization, a standardized, commercial PN product available from a manufacturer requires limited compounding. 1− 2 Ready-made PN formulations are prepared in single-container or multi-chamber bags, usually cited as "premixed" despite the fact that they need mixing in the pharmacy before administration. Ready-made formulations have been marketed as safer and more e cient delivery systems for nutrients compared with traditional formulations compounded manually or by automated compounder machines. 3 The marketed two-chamber bag contains amino acids in one chamber and dextrose in the other chamber.
Bags with and without added electrolytes in nal volumes of 1 L and 2 L have been available in markets over past years. 1 The lipid can be given through a Y-connector inserted in the IV catheter. While threechamber bags contain the same elements of the two-chamber bags, the third chamber contains lipids. 1 Multichambered PN with compounded formulations are an available formulation choice for hospitalized patients to best meet an institution's patient needs. 3 The bene ts of using the ready-made parenteral solutions compared to the customized solutions include reduced medication error throughout the whole preparation process, reduced incidents of bloodstream infections (BSIs), and possible lower cost. [4][5] To accommodate particular patient demands for electrolytes, vitamins, and minerals, these positive impacts must be weighed against the need for individual adaptation. 6 The ready-made parenteral solutions are made with a xed number of calories and electrolytes designed to t the majority of the population. Thus, there are major limitations, including that ready-made solutions cannot be used in populations requiring higher protein requirements (e.g., obese, underweight, and CRRT patients), a possible inability to maintain electrolyte balances in patients who require higher or lower requirements (e.g., HF patients, CKD or AKI, and post bone marrow transplant), and di culty achieving target calories/day while maintaining the patient's hydration status. 7 In conditions of impaired lipid metabolism, which may occur in patients with renal failure, diabetes mellitus, pancreatitis, impaired liver function, hypothyroidism, and sepsis, SmofKabiven should be prescribed with caution. 8 SmofKabiven® is a premixed PN available on the market. Approved by the FDA in August 2014 to be used in patients 2 years and older, SmofKabiven® is the rst premixed PN added to King Faisal Specialist Hospital's formulary in 2015. SmofKabiven is a three-chamber container available in different pack sizes: 960 ml, 1080 ml, 1200 ml, 1320 ml, 1440 ml, 1560 ml, 1680 ml, 1800 ml, and 1920 ml. Each bag contains amino acids (3.2 g/100 mL), dextrose anhydrous (7.1 g/100 mL), and lipids (2.8 g/100 mL) with electrolytes.
Each component is located in a separate chamber. These components must be mixed before administration and should be administered via a central line, as the maximum allowed peripheral osmolarity is 900 mOsmol/L for adults. 7− 8 Contraindications for use, as de ned by the manufacturer, are hypersensitivity to sh-, egg-, soya-, or peanut protein or corn (maize) and corn products or to any of the active substances or excipients; severe hyperlipidemia; severe liver insu ciency; severe blood coagulation disorders; congenital errors of amino acid metabolism; severe renal insu ciency without access to hemo ltration or dialysis; unstable conditions (such as extreme post-traumatic conditions, uncompensated diabetes mellitus, acute myocardial infarction, stroke, embolism, metabolic acidosis, severe sepsis, hypotonic dehydration and hyperosmolar coma); uncontrolled hyperglycemia; and elevated serum levels of any of the included electrolytes. In addition, the manufacturer includes uncontrolled hyperglycaemia contraindications to infusion therapy, such as acute pulmonary oedema, hyperhydration, and decompensated cardiac insu ciency. Previous literature compared the use of compounded monobags with industrially manufactured three-chamber bags. MCB-PN had proven bene ts compared to customized PN in terms of reduced cost and reduced risk of BSIs. This study was conducted to evaluate the nutritional e cacy and safety of MCB-PN. In terms of e cacy, the percentage of patients who achieved calculated target calories and protein and the percentage of patients who developed electrolyte imbalances were evaluated. In terms of safety, the incidence of BSI and MCB-associated complications developed during the treatment period were evaluated.

Methods
The study was a retrospective chart review, approved by the local institutional review board (IRB). We included all adult hospitalized patients who had been on MCB-PN consecutively for at least 7 days at King Faisal Specialist Hospital & Research Center in Riyadh (KFSHRC-R) from January 2015 until December 2019. We excluded patients who received MCB-PN for less than 7 days. The following was collected for this study. In terms of e cacy, we evaluated the percentage of patients who achieved calculated target calories and protein and the percentage of patients who developed electrolyte abnormalities. In terms of safety, we evaluated the incidence of central line-associated bloodstream infections (CLABSIs) and MCB-associated complications developed during the treatment period. The data collection sheet was formulated by using the REDCAP system for entering and analyzing valuables (please refer to appendix A). Total calories, total macronutrients g/day (proteins, dextrose, and fat), electrolyte contents, and the volume provided from SmofKabiven® are available in the product information reference (please refer to appendix B). Laboratory parameters were determined and used as a baseline before the PN started and at day 7 after starting PN, including hematological parameters (hemoglobin, red blood cell count, white blood cell count, platelet count, percentage of granulocyte, and hematocrit), parameters of blood coagulation function (prothrombin time and activated partial thromboplastin time), liver function parameters (alanine aminotransferase, γ-glutamyl transpeptidase, alkaline phosphatase, total bilirubin, and serum albumin), blood chemistry and electrolytes (triglycerides, total cholesterol, blood glucose, serum urea nitrogen, creatinine, sodium, potassium, chloride, and phosphate), and C-reactive protein (CRP) and prealbumin (as a short-term marker for nutrition status). The changes in prealbumin levels between POD 1 (i.e., before administration of PN) and POD 7 (i.e., after administration of PN for 7 consecutive days) were reviewed using the hospital electronic healthcare record (EMR) system (ICIS). The institutional review board (IRB) of hospital institution approved the study, before data collection started. The study was performed in compliance with all ethical principles of King Faisal Specialist Hospital & Research Center.

Statistical Analysis
Baseline and demographic characteristics were summarized using descriptive statistics (means, standard deviations) for continuous variables, such as age, and percentages for categorical variables, such as race and ethnicity. Data were entered in the REDCAP system. Data were presented as means ± SD. Statistical signi cance was set at p < 0.05. The results are presented in percentages and tables by the REDCAP system. A chi-square test was done for post hoc analysis.

Study Population
Results are stated descriptively as means (95% con dence intervals) and proportions. Of the 660 hospitalized patients on MCB-PN from January 2015 until December 2019, 231 patients met the inclusion criteria. Among the 231 patients, 51.5% were males, and 48.5% were females. Furthermore, 21.6% were underweight (BMI less than 18.5); 44.6% were normal weight (BMI 18.5 to 24.9); 18.6% were overweight (BMI 25 to 29.9); and 15.1% were obese (BMI 30 or more) ( Table 1-3). Eligible patients weighed between 18 and 125 kg and had nutrition requirements between 1500-2000 kcal/d. Patients who received MCB-PN for less than 7 days were excluded. Of the patients, 77.9% were admitted electively for surgery; 45.5% were for hyperthermic intraperitoneal chemotherapy (HIPEC); 4.44% were for partial or total gastrectomy; 6.66% were for bowel resection; 2.8% were for a Whipple procedure; and 40.6% were for other procedures. Furthermore, 35.9% were admitted with a chief complaint of GI symptoms; 2.2% were admitted for nutritional support; and 14.3% were admitted for other reasons.

Maintenance electrolytes infusion
During the treatment period, 43.3% were on potassium chloride infusion; 2.2% were on phosphate infusion; 14.7% were on magnesium sulfate infusion; and 0.4% were on calcium gluconate, with MCB-PN and close monitoring of electrolyte serum levels daily.
Secondary outcomes There was no documented incidence of CLABSI. However, 7.8% of study subjects developed hyperglycemia; 2.2% developed hypoglycemia; 10.4% developed liver injury; 1.3% developed hypertriglyceridemia; 3.0% developed uid overload; and 1.3% developed refeeding syndrome. After stopping MCB-PN, 74.5% of the population started an oral diet; 1.7% shifted to enteral feeding; 0.9% died; 22.5% shifted to customized PN; and 1.3% developed SmofKabiven®-associated complications (i.e., liver injury, hypertriglyceridemia, and hyperkalemia). During the treatment period, 22.5% of the population shifted from MCB-PN to customized PN for the following reasons: 71.1% due to electrolyte imbalances, 23.1% due to not achieving their nutritional targets in term of calories and protein, 3.8% due to AKI, 9.6% due to uid overload, and 3.8% due to other reasons.

Discussion
Standardization is suggested to ensure patient safety associated with PN therapy. [1][2] Published literature has demonstrated the advantages of standardized formulations in hospitalized individuals who received parenteral therapy. 1-2 Literature comparing the standardized and customized PN is limited, not commonly of good quality, and comes from Europe. 1,13−19 Almost all of these studies have shown that MCB-PN is less pricey than customized nutrition is. However, MCB-PN has been shown to have higher costs or equivalent costs. [14][15][16][17][18][19][20] By analyzing the data, there are several confounding factors and variables to consider, including compounding hours, the organization's policies, compounding machine costs, type of premix (whether three-chambered or two-chambered bags with additional intravenous lipid emulsion), administration tube costs, nurse time, staff salaries for pharmacy and nursing, and outside added electrolyte boluses when using premixed PN. 1,21 A prospective open-label randomized control trial of 100 patients comparing MCB-PN to customized PN was published in Advances in Pharmacology and Pharmacy. In this study, MCB-PN was not superior to customized PN formulas in terms of safety and e cacy, but it was a less expensive formula. 2,7 Fewer adverse drug reactions and metabolic disturbances were shown in patients who received premixed PN than those in patients receiving tailored PN. 2,7 In an EPICOS study, an international, multicenter, prospective, open-label, controlled trial studied the impact of PN delivery systems on the incidence of BSIs in critically ill patients. They suggested that the use of MCB-PN could be considered a nutritional support option with fewer incidences of BSIs compared to a customized formula. 8 A multicenter, randomized study done in ve Chinese hospitals compared MCB-PN to customized PN formulations. Among 240 patients, prealbumin levels rose dramatically in the study population by 2.70 ± 5.69 mg/dL (P < .001). The preparation time decreased dramatically by almost 8 minutes in the study population (P < .001), and no signi cant differences in safety parameters were de ned in their study in terms of adverse reactions and 30-day mortality. 10 In our study, we aimed to study standardized PN formulation regarding the e cacy of nutrition support and safety in hospitalized patients. Among patients receiving MCB-PN, only 44% achieved target calories, and 29.6% achieved the target protein dose. Overweight and obese patients may be at higher risk of not achieving target protein requirements while on MCB-PN. A smaller percentage of patients experienced adverse drug reactions and metabolic abnormalities during the treatment period. Individualized nutritional assessment and evaluation before starting MCB-PN are warranted to ensure appropriate use of PN formulas. Compared to previous published studies, our study included more patients in different hospital settings, a longer PN intervention ( 7 days), and an ad hoc analysis to examine the primary outcomes in all patients with different BMIs.

Limitations
In evaluating the results of this analysis, some limitations should be noted. First, the unavailability of an indirect calorimetry device and the use of estimated equations (e.g., Harris-Benedict equations and Mi in) may not give a patient's accurate nutritional requirements and targets. Second, e cacy was determined by achieving the target calories and protein speci ed by an assigned clinical pharmacist every day. Although these factors are signi cant, they are not the only indicators of effectiveness in terms of support for nutrition. Third, confounding variables were not controlled because the nature of our study was retrospective. Finally, this study is a single arm; we did not compare the ndings of the study with standard therapies.

Conclusion
In conclusion, among patients receiving MCB-PN, patients with BMI more than 30 kg/m 2 are less likely to achieve the recommended target protein requirements (p-value < 0.05). Less than 60% of all included patients with different BMIs achieved target caloric requirements. Electrolyte abnormalities remained high in all included patients while on MCB-PN.

Declarations
Funding (information that explains whether and by whom the research was supported) no funding support was provided. The primary author, Muna Islami, declares the following on behalf on me and co-authors: Neither the paper nor portions of it have been previously published (except as an abstract has been presented as poster in ASPEN21), The manuscript has not been proposed for publication in another journal and it will not be published anywhere until the editorial process at SN COMPREHENSIVE CLINICAL MEDICINE is completed Availability of data and material (data transparency) All data and records generated throughout the course of the study will be kept con dential in alignment with King Faisal Specialist Hospital and Research center (KFSHRC) Policies and the Primary investigators and co-investigators will be the only ones to have access to the study data and records for the purposes of conducting the study. All data will be stored and built using a study data management systems, (RedCap,), the data stored in REDCap's back end database, the software application employs multiple methods to protect against users who may try to identify and exploit any security vulnerabilities in the system. Data will be only accessed and used by investigators during study period.
Code availability (software application or custom code) All data will be stored and built using a study data management systems, (RedCap,).  Primary Outcomes n = 231