Collin Anderson

Frequency and Severity of Parenteral Nutrition Medication Errors at a Large Children’s Hospital After Implementation of Electronic Ordering and Compounding

INTRODUCTION The Institute for Safe Medication Practices has stated that parenteral nutrition (PN) is considered a high-risk medication and has the potential of causing harm. Three organizations--American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.), American Society of Health-System Pharmacists, and National Advisory Group--have published guidelines for ordering, transcribing, compounding and administering PN. These national organizations have published data on compliance to the guidelines and the risk of errors. The purpose of this article is to compare total compliance with ordering, transcription, compounding, administration, and error rate with a large pediatric institution. METHOD A computerized prescriber order entry (CPOE) program was developed that incorporates dosing with soft and hard stop recommendations and simultaneously eliminating the need for paper transcription. A CPOE team prioritized and identified issues, then developed solutions and integrated innovative CPOE and automated compounding device (ACD) technologies and practice changes to minimize opportunities for medication errors in PN prescription, transcription, preparation, and administration. Thirty developmental processes were identified and integrated in the CPOE program, resulting in practices that were compliant with A.S.P.E.N. safety consensus recommendations. Data from 7 years of development and implementation were analyzed and compared with published literature comparing error, harm rates, and cost reductions to determine if our process showed lower error rates compared with national outcomes. RESULTS The CPOE program developed was in total compliance with the A.S.P.E.N. guidelines for PN. The frequency of PN medication errors at our hospital over the 7 years was 230 errors/84,503 PN prescriptions, or 0.27% compared with national data that determined that 74 of 4730 (1.6%) of prescriptions over 1.5 years were associated with a medication error. Errors were categorized by steps in the PN process: prescribing, transcription, preparation, and administration. There were no transcription errors, and most (95%) errors occurred during administration. CONCLUSION We conclude that PN practices that conferred a meaningful cost reduction and a lower error rate (2.7/1000 PN) than reported in the literature (15.6/1000 PN) were ascribed to the development and implementation of practices that conform to national PN guidelines and recommendations. Electronic ordering and compounding programs eliminated all transcription and related opportunities for errors.

Physical Compatibility of Sodium Glycerophosphate and Calcium Gluconate in Pediatric Parenteral Nutrition Solutions

BACKGROUND The solubility of inorganic calcium and phosphate in parenteral solutions can be complicated in pediatrics due to the dosing of calcium and phosphorus at the saturation point. The purpose of this study was to test the solubility of sodium glycerophosphate (NaGP) with calcium gluconate in pediatric parenteral nutrition (PN) solutions. METHODS Five PN solutions were compounded by adding calcium gluconate at 10, 20, 30, 40, and 50 mEq/L and corresponding concentrations of NaGP at 10, 20, 30, 40, and 50 mmol/L. Each of the 5 solutions was compounded using 1.5% and 4% amino acids, cysteines, and lipids. Compatibility was evaluated by visual inspection (precipitation, haze, and color change). Solutions were evaluated microscopically for any microcrystals and measured by a turbidimeter for changes in turbidity. Solutions were further analyzed using United States Pharmacopeia 788 standards. Six hundred seventy-one PN solutions were compounded at various concentrations and evaluated for visual stability. RESULTS Compatibility testing showed no changes in the PN solution in any of the concentrations tested. Microscopically, no microcrystals were detected. The turbidimeter measurements had changes of ≤ 0.14 nephelometric turbidity units for all test solutions. There were no visual changes in any of the 671 PN solutions. CONCLUSION It is recommended that NaGP replace sodium phosphate in PN solutions. This would eliminate the concern of calcium and phosphorus precipitation and the need of any saturation curves.

Physical Compatibility of Calcium Chloride and Sodium Glycerophosphate in Pediatric Parenteral Nutrition Solutions

BACKGROUND Calcium and phosphate precipitation is an ongoing concern when compounding pediatric parenteral nutrition (PN) solutions. Considerable effort has been expended in producing graphs, tables, and equations to guide the practitioner in prescribing PN that will remain stable. Calcium gluconate is preferred over calcium chloride when compounding PN because of its superior compatibility with inorganic phosphates. PN solutions containing calcium gluconate carry a higher aluminum load than equivalent solutions compounded with calcium chloride, leading to increased potential for aluminum toxicity. This study tested the solubility of calcium chloride in PN solutions compounded with an organic phosphate component, sodium glycerophosphate (NaGP), in place of sodium phosphate. METHODS Five PN solutions were compounded by adding calcium chloride at 10, 20, 30, 40, and 50 mEq/L and corresponding concentrations of NaGP at 10, 20, 30, 40, and 50 mmol/L. Each of the 5 solutions was compounded using 1.5% and 4% amino acids, cysteine, and lipids. The physical stability was evaluated by visual inspection (precipitation, haze, and color change). Solutions were evaluated microscopically for any microcrystals using U.S. Pharmacopeia <788> standards. RESULTS Compatibility testing showed no changes in the PN solution in any of the concentrations tested. Calcium chloride was found to be physically compatible with NaGP in PN at the tested concentrations. CONCLUSION Utilization of NaGP in PN solutions would eliminate the need for precipitation curves and allow for the use of calcium chloride. Compounding with NaGP and calcium chloride allows the practitioner a mechanism for reducing the aluminum load in PN.

Correlation of weight-based i.v. immune globulin doses with changes in serum immunoglobulin G levels

PURPOSE Correlations between the i.v. immune globulin (IVIG) dose and the change in serum immunoglobulin G (IgG) concentration with three methods of weight-based dosing were investigated. METHODS A retrospective medical records review was conducted to identify patients in a multicenter healthcare system who received IVIG over a 10-year period and had serum IgG measurements within two days before and two days after the date of IVIG administration. For each of the 11 adults and 7 adolescents identified, the relationship between the weight-based dose of IVIG (determined using actual body weight [BW], adjusted BW, or ideal BW [IBW]) and the resulting change in serum IgG was evaluated. Correlation coefficients and corresponding p values for the three dosing methods were calculated and compared. RESULTS Among adult patients, the correlation of IVIG dose and postdose change in serum IgG was strongest with dosing by IBW (correlation coefficient [r], 0.83 [p < 0.05] versus r values of 0.73 and 0.70 for dosing by adjusted BW or actual BW, respectively [p = 0.05 for both]); the corresponding r values in adolescent patients were 0.99, 0.99, and 0.95, respectively (p < 0.005 for all). There were no statistically significant differences between the r values calculated for the three weight-based dosing methods in either adults or adolescents. CONCLUSION In adult patients, the correlation between the dose of IVIG and the change in IgG level was strongest when doses were calculated using IBW; comparable degrees of correlation were observed with the three evaluated methods of weight-based dosing in the adolescent population.

Physical and Chemical Compatibility of Injectable Acetaminophen During Simulated Y-Site Administration

Purpose The physical and chemical compatibility of intravenous acetaminophen with commonly administered injectable medications was evaluated. Methods Simulated Y-site evaluation was accomplished by mixing 2 mL of acetaminophen (10 mg/mL) with 2 mL of an alternative intravenous medication and subsequently storing the mixture in a polypropylene syringe for 4 hours. The aliquot solutions were visually inspected and evaluated for crystal content at 4 hours by infusing 4 mL of the medication mixture through a 0.45-μm nitrocellulose filter disc. Medication mixtures that were selected for chemical stability testing were analyzed by high-performance liquid chromatography at 0, 1, and 4 hours using a Zorbax Eclipse Plus C18, 4.6 × 100 mm, 3.5-μm column for separation of analytes with subsequent diode-array detection. Medications were considered chemically compatible if the concentrations of all components were >90% of the original concentrations during the 4 hour simulated Y-site compatibility test. Results U.S. Pharmacopeial Convention (USP) standards for physical particle counts were met for acetaminophen injection (10 mg/mL) when combined with cefoxitin, ceftriaxone, clindamycin, dexamethasone, diphenhydramine, dolasetron, fentanyl, granisetron, hydrocortisone, hydromorphone, ketorolac, meperidine, methylprednisolone, midazolam, morphine, nalbuphine, ondansetron, piperacillin/tazobactam, ranitidine, and vancomycin. Injectable acetaminophen is incompatible with acyclovir and diazepam and therefore should not be administered concomitantly with either of these products. Further testing confirmed the chemical compatibility of acetaminophen with ceftriaxone, diphenhydramine, granisetron, ketorolac, nalbuphine, ondansetron, piperacillin/tazobactam, and vancomycin. Conclusion All medications tested with acetaminophen were physically compatible except for acyclovir and diazepam. All 8 medications tested for chemical compatibility with acetaminophen were stable over the 4 hour simulated Y-site administration study.

Assessing Selenium, Manganese, and Iodine Status in Pediatric Patients Receiving Parenteral Nutrition

Background: Pediatric patients who are receiving parenteral nutrition (PN) unsupplemented with trace minerals can become deficient. Due to shortages in trace mineral products and the 2004 American Society for Parenteral and Enteral Nutrition report stating that individualized trace element supplementation may be warranted, a review was conducted concerning the trace minerals selenium (Se), manganese (Mn), and iodine (I). Method: A retrospective review of pediatric patients receiving PN that contained Se and Mn was conducted to determine if a difference existed between them and patients receiving PN without Se and Mn. Statistical analysis was done to assess a difference between trace mineral levels and the time to deficiency between supplemented and unsupplemented patients. Unsupplemented I patients had urine I levels assessed to determine deficiencies in patients receiving PN. Results: Plasma Se levels were measured at a mean of 20 days for supplemented patients (n = 131) and 19 days for nonsupplemented patients (n = 57) with no difference between groups (P = .2973). Plasma Mn levels were measured at a mean of 28 days, showing no statistical difference (P = .721). Of the 177 nonsupplemented I patients, 74% demonstrated I deficiencies without supplementation. Conclusions: Time to the development of a Se, Mn, or I deficiency is important to guide supplementation of exclusive PN in children when trace mineral products are short in supply. Our retrospective experience supports assessment of the trace minerals Se at 21 days and Mn at 30 days. It also suggests that some pediatric patients receiving PN are deficient in I.

Stability of Fentanyl Citrate, Hydromorphone Hydrochloride, Ketamine Hydrochloride, Midazolam, Morphine Sulfate, and Pentobarbital Sodium in Polypropylene Syringes

Purpose: Determine the stability of fentanyl 10 mcg/mL in 0.9% sodium chloride, fentanyl 10 mcg/mL in 5% dextrose, fentanyl 50 mcg/mL, hydromorphone 100 mcg/mL in 0.9% sodium chloride, ketamine 10 mg/mL, midazolam 0.4 mg/mL in 5% dextrose, midazolam 5 mg/mL, morphine 1 mg/mL in 0.9% sodium chloride, morphine 1 mg/mL in 5% dextrose, and pentobarbital 50 mg/mL when stored as single drug entities at room temperature in polypropylene syringes. Methods: Four 5 mL samples of each drug and concentration were prepared in 10 mL polypropylene syringes. The samples were stored at ambient room temperature in a locked cabinet. Triplicate determinations of drug concentration for each sample were performed initially, on day 50 or 51, and on day 100 using high-performance liquid chromatography with diode-array detection. Results: With the exception of the hydromorphone 100 mcg/mL dilution, all compounds were found to contain greater than 95% of their initial concentration remaining at 100 days. Each sample remained clear and colorless when visually inspected.

Stability of dexmedetomidine 4 μg/mL in polypropylene syringes

PURPOSE The results of a study to determine the long-term (up to 14 days) stability of diluted dexmedetomidine kept in polypropylene syringes under typical pharmacy storage conditions are presented. METHODS Four samples of dexmedetomidine injection diluted to 4 μg/mL in 0.9% sodium chloride were prepared and divided into 25-mL portions for storage in syringes at ambient room temperature (20-25 °C) with light exposure or under refrigeration (5 °C) in darkness. At 24 and 48 hours, the percentage of the initial dexmedetomidine concentration remaining in all samples was assessed via high-performance liquid chromatography with diode-array detection; further stability testing of the refrigerated samples was performed on days 7 and 14. At each time point, the test samples were visually inspected for color, clarity, and signs of formation of particulate matter. RESULTS As determined by chromatographic analyses, the samples of diluted dexmedetomidine stored in syringes at room temperature exhibited a loss of drug concentration of <10% over 48 hours; the refrigerated samples exhibited a loss of drug concentration of <5% over 14 days. All of the syringe-stored samples remained clear and colorless on visual inspection for the duration of the study. CONCLUSION Dexmedetomidine diluted to 4 μg/mL in 0.9% sodium chloride injection was stable for at least 48 hours at 20-25 °C and 14 days at 5 °C when stored in polypropylene syringes.