Evaluation of a Process Monitoring Method for Compounding Parenteral Nutrition with the Baxter EM2400 in a Hospital Pharmacy Department
-
Chelsey Collins
Chelsey Collins studied Pharmacy at Philipps-University in Marburg, Germany. She completed her Diploma of Pharmacy at the Martin-Luther University in Halle, Germany. She is currently preparing her PhD thesis at the Pharmacy Department of the University Medical Center, Johannes Gutenberg-University, Mainz, Germany on “Analysis of the critical steps in the preparation of PN and evaluation of possible optimization methods”. Her research interests include the aseptic preparation of parenteral nutrition with an automated compounding device and methods to assure the quality of the compounded products.
and
Irene Krämer
Irene Krämer is currently Director of the Pharmacy Department, University Medical Center, Johannes Gutenberg-University Hospital, Mainz and is also a Professor for clinical pharmacy at the Pharmacy School of Johannes Gutenberg-University. She completed her postdoctoral thesis in Pharmaceutical Technology entitled: Development, quality assurance, and optimization of ready-to-use parenteral solutions in the integrated cancer care concept. Her special interests include oncology pharmacy, infectious diseases and aseptic drug preparation. She is doing research projects in the field of physicochemical and microbiological stability of cytotoxic drugs, compatibility of admixtures of nebulizer solutions and monitoring of medication compliance.
Abstract
Background
A method for process monitoring of the Baxter ACD, Exactamix EM2400, by chemically analyzing the concentrations of the ingredients in parenteral nutrition test preparations was evaluated.
Methods
In the study, three different test preparations were developed consisting of four ingredients varying in volume and concentration, which simulated actual PN products. Each test preparation was produced in triplicate by a trained pharmacy technician using the Exactamix EM2400 from Baxter (Baxter International Inc.). The process was repeated on three consecutive days using the same conditions. The amount of each ingredient in the test preparation was measured in an external contract laboratory using European Pharmacopoeia methods. Based on USP monographies and the Guidelines on the Safe Use of Automated Compounding Devices for the Preparation of Parenteral Nutrition from the ASHP the device was tested to be accurate to deliver within 5 % of the amount programmed.
Results
The study showed that in most cases the ingredient delivery of the automated compounder deviated less than 3 % from the expected concentrations. For certain ingredients out of specification results were detected and analyzed. By resetting the flow factor, it was possible to optimize the performance of the ACD.
Conclusion
The study emphasizes the need for process monitoring of the Exactamix EM2400 during the initial installation and on a regular basis for each ingredient to ensure the accurate delivery of ingredients. Further methods need to be analyzed to determine the most feasible method to regularly conduct process monitoring tests on an ACD in a hospital pharmacy setting.
Introduction
Parenteral nutrition (PN) plays a vital role in critically ill patients. In general, the multicomponent PN mixtures for pediatric patients are aseptically prepared in pharmacy-based admixture services, where automated compounding devices (ACD) are widely used for mixing nutritive ingredients, electrolytes and vitamins. Reliable performance of the compounding device is essential for the safety of the patients. It is well accepted that the use of an ACD can increase the quality and productivity in comparison to manual preparation [1, 2, 3]. However, the ACD must be consistently capable of operating properly and within the acceptable tolerance limits. Routine maintenance, calibration and checking of the proper function are obligatory. Quality assurance measures are to be implemented in the pharmacy department in order to ensure the accurate delivery of the PN additives [1]. As a matter of routine an ACD specific method for process monitoring must be developed and conducted according to a predefined monitoring plan. This study describes the experimental trial of a method for testing the accuracy of the Exactamix EM2400 compounder as an essential part of monitoring the pumping process.
There are no specific regulations for the implementation and utilization of ACDs to be found in the PIC/S PE 010-4 Guide or in the European Pharmacopeia [4, 5]. Of note, regulations can be found in the United States Pharmacopoeia and in US-Guidelines [1, 6]. According to the USP monograph<797>the accuracy of an ACD can be determined in different ways. The accuracy can be tested by measuring the volume delivered to an appropriate volumetric container and by weighing the pumped volume on a precision scale. Gravimetric controls can also be used as process controls. In addition, accuracy can be tested by determining the content of certain ingredients in PN admixtures using quantitative analytical methods. The intermediate precision of the ACD can be determined on the basis of the day-to-day variations in performance of the accuracy measures [6]. Ongoing i. e. daily, process monitoring measures should be established for assessing the performance of the compounder. Results should be reported and regularly reviewed in the assessment of trends and other long-term measures of performance [1]. There is only few literature available reporting accuracy checks as quality assurance measures of ACDs [7, 8, 9]. To our best knowledge there are no studies published regarding routine assessments of accuracy in the delivery of correct amounts of nutrients by the Baxter EM2400 ACD. The aim of our study was to evaluate the delivery of ingredients by this ACD marketed by Baxter (Baxter International Inc.). The method chosen for process monitoring was the chemical analysis of the concentration of each delivered ingredient in the final PN product compounded with this ACD. According to the USP<797>, the amount of each ingredient should be within the limits set in the monograph of each individual ingredient, e. g. not less than 95.0 % and not more than 105.0 % for dextrose monohydrate in Dextrose injection [6]. In the ASHP Guidelines on the Safe Use of Automated Compounding Devices for the Preparation of Parenteral Nutrition it is stated that the ACD should be accurate within 5 % of the amount programmed with verification of the amount pumped versus the programmed amount for each ingredient [1]. We set the same accuracy limits.
Materials and methods
Exactamix EM2400 compounder
The study was performed in the aseptic preparation unit of the Pharmacy Department of the University Medical Center in Mainz, Germany, using the Exactamix EM2400 (EM2400) compounder (serial number: 452138) manufactured by Baxter International Inc. The EM2400 compounder has 24 ports for compounding admixtures up to 24 ingredients (Figure 1). This ACD is a micro and macro compounder for volume delivery starting at 0.2 mL into TPN bags up to a maximum of 5000 mL. A peristaltic pump system delivers the quantities of components using volumetric measurements. An integrated scale gravimetrically verifies the delivered volume by weighing the final product. The individual ingredients are only weighed during the pumping process if they have volumes above 100 mL.
Baxter Exactamix EM2400 Compounder (Figure provided by Baxter International Inc).
Flow factor settings
The flow factor is a control parameter which is determined based on parameters like the density, type and volume of a source product. The flow factor is set for each ingredient and adjusts the pumping speed and thereby the correctness of the pumped volume at the predefined port. For the micro ingredients (in most cases volumes<5 mL pumped) one flow factor is set for slower pumping speeds for volumes delivered up to 10 mL and a second flow factor is set for higher pumping speeds for volumes delivered above 10 mL. For the macro ingredients (in most cases volumes>10 mL pumped) a single flow factor is set applicable to all pumping speeds.
To determine the flow factors, the actual delivered volume of each individual ingredient is measured. For the macro ingredients, volumes of 100 mL are pumped 3 times and the average actual pumped volume is calculated based on the weight measured by the device internal scale and the density of the ingredient. Based on the actual delivered volume compared to the expected volume, a flow factor is determined with the work sheet provided by Baxter. The flow factor is set at a value at which the actual volume delivered is approximately within ±1 % of the desired delivery.
For micro ingredients, a flow factor for each pumping speed is determined. These flow factors are determined by comparing the actual delivered volume to the expected volume of 5 mL and 20 mL for the slower and higher pumping speed, respectively. An external scale must be used to determine these flow factors. The source container of each ingredient is removed, weighed and reconnected to the EM2400. Then, the EM2400 is programmed to pump 5 mL of each ingredient and 30 mL water (necessary to flush the lines). Afterward, the individual source containers are removed from the EM2400 and weighed again. The difference in mass and the actual delivered volume is calculated. Using the Baxter work sheet, the flow factor is calculated and set to allow the actual delivery of ingredients to be within approximately ±1 % of the desired delivery. For the flow factor for the higher pumping speed setting, the measurement is conducted by pumping a volume of 20 mL for each ingredient and 30 mL of water.
Once a flow factor is determined, the correctness is verified by pumping each ingredient three more times. If the accuracy limit is exceeded the flow factor has to be re-adjusted and checked again.
The flow factors were adjusted according to the training received during the initial installation using the Baxter handbook and provided worksheet. The worksheet is an excel spread sheet developed by Baxter with the necessary formulas for the calculations. For the weighing of the micro ingredients, a calibrated Sartorius scale LC4801POCEMS (Serial number: 50105921) was used. The flow factors were initially set with the Baxter technicians and were scheduled to be re-evaluated every six months based on the recommendation given by the ACD manufacturer.
Design of the test preparations
Three types of PN preparations were planned. Test preparation A was prepared with ingredients at amounts appropriate for a 1000 g neonate based on the ESPGHAN guidelines [10]. Two further test preparations were prepared containing higher (test preparation B) and lower (test preparation C) amounts of each ingredient. Each preparation contained glucose, an amino acid solution, sodium chloride, magnesium aspartate and sterile water for injection (s. Table 1).
Characteristics of the ingredients and source products used for the test preparations. The batch-specific concentrations of the source products were measured in the external lab and used to calculate the theoretical concentrations of the test preparations prepared with the ACD.
| Ingredient (Abbreviation) | Source product | Manufacturer | Container | Density [g/cm3] | Measured concentration/ percentage rate of nominal concentration |
|---|---|---|---|---|---|
| Glucose (Glucose) | Glucose 70 % (Lot # 144238062) | B.Braun | 500 mL glass bottle | 1.26 | 703.7 g/L 100.5 % |
| Amino acid solution (AA) | Aminoven infant 10 % (Lot # 16ID1089) | Fresenius Kabi | 1000 mL glass bottle | 1.03 | 5.26 g/L* 101.6 % |
| Sodium chloride (NaCl) | Natrium chlorid 5.85 % (Lot # 224340) | Serag Wiessner | 250 mL glass bottle | 1.04 | 60.63 g/L 103.7 % |
| Magnesium aspartate (MgAsp) | Magnesium aspartat 0.5 M (Lot # 20150326-01) | Manufactured in the pharmacy department | 100 mL plastic bottle | 1.083 | 500.3 mmol/L 100.1 % |
| Sterile Water for Injection (water) | Aqua ad inj. (Lot # 152118151) | B.Braun | 1000 mL bag | 1 |
N-Acetyl-Tyrosine used as index component for the amino acid content.
The composition of test preparations A, B, C is described in Table 2 and the volume of each ingredient for each test preparation in Table 3. Test preparation A was compounded with a final volume of 100 mL in 125 ml Baxter TPN BAGs (EVA 125 mL LL Fill Port, Lot number D028CA). Test preparation B was compounded with a final volume of 300 mL in 500 ml Baxter TPN BAGs (EVA 500 mL LL Fill Port, Lot number E058CC) and contained three times the amounts of A at the same concentrations. Test preparation C was compounded with a final volume of 60 mL in 60 mL BD syringes (Becton Dickinson, distributor: Loccioni Humancare, Italy, Lot number 150205). These syringes were connected with a Syringe Extension Line (Baxter, Lot number: B053ZC). For test preparation C, the amounts of NaCl, AA and MgAsp were 50 % of the amounts used in test preparation A. The amount of glucose was 25 % of the amount in test preparation A to challenge the ACD with a very low volume of highly viscous glucose 70 %. This also allowed for the amount of water to be>30 mL which was necessary because the EM2400 requires at least 30 mL of one ingredient to flush the line when using the syringe adapter. This was achieved by adjusting the final volume to 60 mL.
Composition of test preparation A, B, C.
| Ingredient | Test preparation A (100 mL) | Test preparation B (300 mL) | Test preparation C (60 mL) | |||
|---|---|---|---|---|---|---|
| Amount | Concentration | Amount | Concentration | Amount | Concentration | |
| AA | 3 g | 30 g/L | 9 g | 30 g/L | 1.5 g | 25 g/L |
| N-Acetyl-Tyrosine* | 0.155 g | 1.55 g/L | 0.466 g | 1.55 g/L | 0.078 g | 1.29 g/L |
| MgAsp | 0.5 mmol | 5 mmol/L | 1.5 mmol | 5 mmol/L | 0.25 mmol | 4.17 mmol/L |
| NaCl | 3 mmol | 30 mmol/L | 9 mmol | 30 mmol/L | 1.5 mmol | 25 mmol/L |
| Glucose | 12 g | 120 g/L | 36 g | 120 g/L | 3 g | 50 g/L |
| Water | 48.9 mL | 146.6 mL | 38.7 mL | |||
N-Acetyl-Tyrosine used as index component for the amino acid content.
Volumes of ingredients for test preparation A, B, C.
| Ingredients | Test preparation A (100 mL) | Test preparation B (300 mL) | Test preparation C (60 mL) |
|---|---|---|---|
| Volume [mL] | Volume [mL] | Volume [mL] | |
| Glucose | 17.10 | 51.40 | 4.30 |
| NaCl | 3.00 | 9.00 | 1.50 |
| MgAsp | 1.00 | 3.00 | 0.50 |
| AA | 30.00 | 90.00 | 15.00 |
| Water | 48.90 | 146.60 | 38.70 |
Compounding of test preparations
Special care was taken to control the conditions of the test environment. The production of the samples occurred at room temperature which was 24 °C on average (checked and documented daily). The product preparation was conducted by an experienced pharmacy technician in an aseptic environment. The pharmacy technicians were trained by Baxter representatives. All of the ingredients and materials were approved for use in Germany. During each test preparation period, the ACD was equipped with material from the same batches. The material used for each study day were one 24 Port Valve Assembly (Baxter, Lot number: 1018180), two 174 Lines (Baxter, Lot number: 801773), three 175 Lines (Baxter, Lot number: 801818) and one Syringe Extension Line (Baxter, Lot number: B053ZC). The compounding occurred with the EM2400. Test preparations were accepted within the range of ±3 % deviation of the total nominal weight as indicated by the internal scale of the EM2400.
Test preparations A, B, and C were prepared in triplicate on three consecutive days, leading to 9 products per test preparation and 27 products in total.
Storage and transport of test preparations
The products were prepared in the afternoon of the study day and sent overnight to an external contract laboratory (Supramol Laboratory, Friedberg, Germany) for end product testing. The samples were sent in insulated shipping boxes (EEP multiple use shipper, Storopack) which are validated to remain between 2 °C and 8 °C inside of the container for approximately 40 hours when used with the validated cooling packs. The products arrived at the laboratory the next morning. One sample was withdrawn from each product to determine the concentrations of each ingredient in triplicate. Analysis was performed within 16–18 hours after product preparation.
Analysis of test preparations
Supramol Laboratory was chosen due to its experience with the analysis of nutrition formulations and GMP certification by the German health authorities. Prior to the study, the source ingredients were sent to the Supramol Laboratory and the batch-specific concentrations of the source ingredients were measured. The actual concentrations (see Table 1) were used to calculate the concentrations of the ingredients in the test preparations. The test preparations consisted of ingredients which are routinely used in the pharmacy and could be tested with the validated analytical methods in the contract laboratory. The methods were validated in the contract lab to be appropriate for parenteral nutrition products based on the current Ph. Eur. and GMP standards. Prior to the measurements the analytical methods were validated by the contract lab.
The concentrations of the individual ingredients were determined by a validated HPLC assay (N-Acetyl-L-tyrosine, glucose) and titration (titration with silver nitrate for sodium chloride and titration with EDTA for magnesium aspartate). The average concentration of each ingredient and pumped volume was calculated based on the 9 test preparations A, B and C each. The ratio of the delivered concentrations to the expected concentrations were calculated as percentage. Accuracy of pumping was determined as the deviation of the mean of the delivered concentration from the expected concentration for each ingredient in each test preparation in percent (according to [6]). Precision and intermediate precision was determined with the relative standard deviations with the intra-series variance and the sum of the intra- and the inter-series variances, respectively.
Additional tests for Out of specification results
In the case an ingredient was not delivered within the accuracy range of 5 % the questionable sample was planned to be re-tested in a second external laboratory using flame atomic absorption spectroscopy for quantitative measurement of the metal ions. If the out of specification results were confirmed, the settings of the ACD such as the flow factors pertaining to the ingredient out of specification had to be reevaluated. If a flow factor had to be reset (see paragraph Flow Factor Settings above), additional test preparations had to be compounded. This was the case for the NaCl solution as source ingredient. Additional test preparations were scheduled. Test preparation D contained 3 mL of NaCl 5.85 % solution and 97 mL of water and test preparation E consisted of 12 mL NaCl 5.85 % solution and 388 mL water. To obtain comparable results, the same concentration as in test preparation A and B were chosen.
Results
Test preparations were accepted within the range of ±3 % deviation of the total nominal weight. All the test preparations fulfilled this requirement according to the internal gravimetric verification of the EM2400.
The ratio of the delivered concentrations to the expected concentrations as percentage for each ingredient in each test preparation A-C is shown in Figure 2. The delivered concentration of glucose succeeded the accuracy limit of 105 % in test preparation C. The delivered concentration of NaCl went considerably below the lower limit of 95 % in each type of test preparation.
Percentage of the delivered concentration compared to the expected concentration of each ingredient. (mean of n=9 for each test preparation).
Results of the process monitoring for the 3 test preparations A-C prepared in triplicate on 3 consecutive days (n = 9 for each test preparation) with the EM2400.
| Test Preparation | A (100 mL) | B (300 mL) | C (60 mL) | A (100 mL) | B (300 mL) | C (60 mL) | A (100 mL) | B (300 mL) | C (60 mL) | A (100 mL) | B (300 mL) | C (60 mL) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Ingredient | Glucose | Sodium Chloride | Magnesium Aspartate | N-Acetyl-Tyrosine | ||||||||
| Expected Volume [mL] | 17.1 | 51.4 | 4.3 | 3 | 9 | 1.5 | 1 | 3 | 0.5 | 30 | 90 | 15 |
| Expected Concentration | 120.33 g/L | 120.57 g/L | 50.40 g/L | 31.13 mmol/L | 31.13 mmol/L | 25.94 mmol/L | 5.00 mmol/L | 5.00 mmol/L | 4.17 mmol/L | 1.58 g/L | 1.58 g/L | 1.32 g/L |
| Accuracy [%] | 1.29 | ‒1.03 | 9.76 | ‒12.85 | ‒10.31 | ‒14.35 | ‒1.29 | ‒1.46 | ‒2.69 | ‒1.96 | ‒1.66 | ‒3.04 |
| Precision [%] | 0.50 | 0.19 | 0.44 | 2.09 | 1.79 | 34.32 | 0.61 | 0.22 | 0.76 | 0.17 | 0.25 | 0.23 |
| Intermediate Precision [%] | 1.16 | 0.72 | 0.77 | 4.58 | 2.34 | 33.76 | 0.85 | 0.47 | 1.97 | 1.48 | 1.50 | 1.15 |
Note: Red numbers are out of specifications.
The accuracy, precision and intermediate precision for the concentrations of the four ingredients pumped with the EM2400 are shown in Table 4 for test preparation A, B, and C. The three parameters were in most cases below (±)3 % independent from the type of ingredient and volume.
The accuracy for the glucose concentrations was about ±1 % in test preparation A and B, but was out of specification in test preparation C with an accuracy of 9.76 %. However, the intermediate precision was 0.77 % in preparation C.
The accuracy of the NaCl concentration varied between ‒10.31 % (test preparation B) and ‒14.35 % (test preparation C) compared to the expected concentration and was thereby far out of specification. The intermediate precision amounted to 2.34 % for test preparation B, 4.58 % for test preparation A and 33.76 % for test preparation C. The out of specification sodium concentrations were confirmed using flame atomic spectroscopy (Biochem GmbH, Karlsruhe, Germany). Therefore, we assumed a technical problem specific to the port in the valve assembly of the EM2400 which delivers NaCl. After resetting the flow factor for the NaCl solution, the delivery revealed acceptable accuracy. Because different flow factors are set for ingredients delivered at volumes ≤ 10 mL and>10 mL, two types of further test preparations were prepared (n=9) and the chloride concentrations were titrated. With an accuracy of ‒4.04 % for test preparation D and 1.69 % for test preparation E the adjustment of the flow factors led to NaCl concentrations in the PN products delivered within the predefined accuracy limits. The intermediate precision for test preparation D amounted to 0.40 % (n=9 on 3 consecutive days). Intermediate precision for test preparation E amounted to 2.03 % (n=9 on 3 consecutive days).
Discussion
PN mixtures are often aseptically prepared by hospital pharmacies using automated compounding devices like the Baxter EM2400. ACDs like the Baxter EM2400 are qualified by the manufacturer by conducting a design qualification (DQ) which documents that the design and purpose of the device are suitable. When the device is installed in the pharmacy department installation qualification (IQ) and operational qualification (OQ) are conducted. Thereby it is documented that the device was installed based on the manufacturer’s recommendations and operates within the established specifications [11]. Ongoing quality assurance measures for assessing the performance of the compounder in daily practice are to be specified and implemented in the pharmacy department. Process monitoring for proper function, accuracy and precision are compulsory to ensure the correct delivery of ingredients when compounding PN with an ACD [6]. We designed three different test PN mixtures to be prepared on three consecutive days and planned a quantitative analysis of the ingredients as accuracy monitoring for the Baxter EM2400. Test preparations were designed to simulate typical PN preparations while also covering upper and lower operation limits of the Baxter EM2400.
The target ingredients were similar in density except for the Glucose 70 % solution which is also highly viscous. Although Glucose 70 % is not widely used in hospital pharmacies when preparing PN with an ACD it might be appropriate in specific settings, e. g. in pharmacies with the primary focus on individual PN for neonates. Using a glucose solution with a lower concentration as source ingredient would lead to volumes which would be too high for the needs of some of the extremely premature neonates. Glucose 70 % has a density of 1.26 g/cm3. At very low volumes the accuracy of delivery of such viscous solutions can be affected. Therefore, Baxter does not recommend the use of Glucose 70 %. The decreased accuracy in test preparation C can be due to the high viscosity of the ingredient. After discussing the issue with Baxter, it was decided to lower the speed of the Glucose 70 % delivery to improve the accuracy. Weight-based checks confirmed the improved accuracy. In general, increased care should be taken when ingredients with high viscosities are used for compounding. This information about delivery speed adjustments for other ingredients with high viscosities should be made available in the documents provided by the manufacturer and taken into consideration during installation.
Nevertheless, the ingredients, except NaCl 5.85 % solution, were delivered accurate within the ±5 % limit from the expected concentration, which shows an appropriate delivery from the Baxter EM2400. We were not able to explain the distinct deviation of the NaCl concentration in the test preparations A, B and C or the extremely high intermediate precision of the NaCl concentration in test preparation C. The variation in NaCl for test preparation C does not correlate with the delivery of small volumes taking the accuracy and intermediate precision of the other ingredients such as MgAsp into consideration. Therefore, an issue which specifically affected the NaCl delivery was expected. The study was scheduled to be conducted with the initial flow factors settings before the next re-evaluation took place. However, since other possible errors were ruled out it was decided to re-evaluate the flow factors prior. Only the adjustment of the flow factors allowed for NaCl 5.85 % solution to accurately be delivered. This shows the need for ongoing process monitoring and possibly flow factor adjustments when necessary. In this case only the NaCl port was affected and required adjustments. Adjustments could, however, be necessary for other ingredients in the future. Therefore, it is necessary to test each ingredient and port used. The ingredients in this study were limited to the validated methods of the external lab. The observed fluctuations prove, however, that it is necessary to test every ingredient especially critical ingredients such as potassium chloride [12]. It is known, that variations between source product batches and changes over time can lead to differences in the delivery of the ingredients. Consequently, Baxter recommends re-evaluating the flow factors at least every six months even when the ingredients remain the same.
Moreover, based on the study results it is highly recommended to conduct ongoing process monitoring for the delivery of each ingredient or at least the critical ones like potassium regularly [12]. This is especially true when preparation is done on a patient individual basis and not batch-wise. In the latter case the amounts of the ingredients are usually analyzed during end-product testing before releasing the batch. In the case of patient-individual PN compounding a dummy product containing a defined amount of each ingredient could be prepared. However, chemical analysis of each ingredient is not feasible in daily practice as process monitoring for the ACD, especially when the tests have to be assigned to an external lab. This agrees with a past accuracy study of older ACD models which also discussed the limitations of this method [8]. There is a need for a more efficient method for process monitoring to be implemented and conducted on a preassigned regular basis. Planning of the test preparations in this study was originally intended to simulate typical PN products for neonates and to deliver typical amounts of ingredients with the ACD. However, it has now been concluded that it would be more useful to focus on the operating limits of the device taking the different flow factors into consideration.
Further evaluation will focus on the measurement of volumes delivered by the device for each ingredient. The performance qualification of two ACDs conducted by Hennache et al. analyzed the delivered volume by weighing the delivered product in a volumetric flask [9]. With the Baxter EM2400 it is not possible to measure the volumes or weights of each delivered ingredient because of technical reasons. One way to measure the delivered volume is measuring the remaining volume in the source product containers. This can be done by weighing the source product containers before and after pumping of a defined volume and calculating the removed volume based on the density. This type of process monitoring is less time-consuming and more cost-efficient than chemical analysis. A future study will be conducted to test the feasibility of this method and to make a proposal for ongoing process monitoring of the Baxter EM2400 in a hospital pharmacy setting.
Finally, continuous support of the compounder should be a part of the contractual agreement with the ACD manufacturer [1]. The performance requirements and the responsibilities of the ACD manufacturer and the pharmacy department should also cover quality assurance and be conducted cooperatively. Today, the EM2400 user is dependent on the Baxter work sheet when checking and setting the flow factors, which should be made more transparent to the user.
Conclusion
The study showed that the Baxter EM2400 can accurately deliver different amounts of ingredients while adjustments for individual ingredients may be necessary to obtain proper performance. The inaccurate delivery of one ingredient highlights the importance of ongoing quality assurance measures. Results must be used for corrective actions. Deviations caused by specific parameters such as highly viscous source products like Glucose 70 % can be detected and problem solving adjustments elaborated with the ACD manufacturer. Solutions to such issues obtained during quality performance measurements should be made available by the ACD manufacturer to all customers in the official documents of the ACD. Close cooperation of the pharmacy department with the ACD manufacturer regarding continuous monitoring of the proper performance of the ACD should be contractually agreed in order to ensure product quality and the safety of the patient at all times.
Pharmacy departments must be aware of the responsibility for initial and ongoing measures of performance of the ACD and should implement a systematic monitoring plan. Simulating the preparation of actual PN products and quantitative analysis of the ingredients proved to be suboptimal. For ongoing process monitoring the method is too complex and too costly. More feasible methods for process monitoring of compounding parenteral nutrition with the EM2400 are to be evaluated.
About the authors
Chelsey Collins studied Pharmacy at Philipps-University in Marburg, Germany. She completed her Diploma of Pharmacy at the Martin-Luther University in Halle, Germany. She is currently preparing her PhD thesis at the Pharmacy Department of the University Medical Center, Johannes Gutenberg-University, Mainz, Germany on “Analysis of the critical steps in the preparation of PN and evaluation of possible optimization methods”. Her research interests include the aseptic preparation of parenteral nutrition with an automated compounding device and methods to assure the quality of the compounded products.
Irene Krämer is currently Director of the Pharmacy Department, University Medical Center, Johannes Gutenberg-University Hospital, Mainz and is also a Professor for clinical pharmacy at the Pharmacy School of Johannes Gutenberg-University. She completed her postdoctoral thesis in Pharmaceutical Technology entitled: Development, quality assurance, and optimization of ready-to-use parenteral solutions in the integrated cancer care concept. Her special interests include oncology pharmacy, infectious diseases and aseptic drug preparation. She is doing research projects in the field of physicochemical and microbiological stability of cytotoxic drugs, compatibility of admixtures of nebulizer solutions and monitoring of medication compliance.
Conflict of interest statement: Chelsey Collins has received samples of products from B.Braun, Baxter International, Fresenius Kabi, Serag Wiesner and Codan to conduct workshops on the preparation of parenteral nutrition. Her research is financed through research funds from the University Medical Clinic in Mainz. Part of the research funding was provided by Fresenius Kabi as a grant to support research in the field of parenteral nutrition. She received financial support through B.Braun in the form of a scholarship to attend the Congress of the European Association of Hospital Pharmacists. The author also received compensation from ADKA and LAK Rhineland-Palatinate for presenting as a speaker and from Baxter International for participating in advisory boards. The author confirms that the content and conclusion of this study was not affected by the support of any company. The journal’s publication ethics and publication malpractice statement available at the journal’s website was read and it is hereby confirmed that the authors comply with all its parts applicable to the present scientific work.
References
1. ASHP Council on Professional Affairs. ASHP guidelines on the safe use of automated compounding devices for the preparation of parenteral nutrition admixtures. Am J Health Syst Pharm 2000;57:1343–1348.10.1093/ajhp/57.14.1343Search in Google Scholar PubMed
2. Dickson LB, Somani SM, Herrmann G, Abramowitz PW. Automated compounder for adding ingredients to parenteral nutrient base solutions. Am J Hosp Pharm 1993;50:678–682.10.1093/ajhp/50.4.678Search in Google Scholar PubMed
3. Bonnabry P, Cingria L, Sadeghipour F, Ing H, Fonzo-Christe C, Pfister RE. Use of a systematic risk analysis method to improve safety in the production of paediatric parenteral nutrition solutions. Qual Saf Health Care 2005;14:93–98.10.1136/qshc.2003.007914Search in Google Scholar PubMed
4. Pharmaceutical Inspection Co-operation Scheme. Document PE 010–4, Guide to good practices for the preparation of medicinal products in healthcare establishments. Geneva (CH): PIC/S, 2014:1–56.Search in Google Scholar
5. Council of Europe. European Pharmacopoeia, 9th ed. Strasbourg (FR): Council of Europe, 2016.Search in Google Scholar
6. United States Pharmacopeial Convention. <797>Pharmaceutical compounding – sterile preparations. Rockville (US): The United States Pharmacopeia and National Formulary (USP 37 – NF 32), 2014:567–610.Search in Google Scholar
7. Johnson R, Coles BJ, Tribble DA. Accuracy of three automated compounding systems determined by end-product laboratory testing and comparison with manual preparation. Am J Health Syst Pharm 1998;55:1503–1507.10.1093/ajhp/55.14.1503Search in Google Scholar PubMed
8. Crill CM, Hak EB, Helms RA Accuracy of parenteral nutrition solutions compounded with automated systems and by hand. Am J Health Syst Pharm 2005;15:2345–2348.10.2146/ajhp050322Search in Google Scholar
9. Hennache T, Clabaut G, Mustapha L, J-M D, Marçon F Performance qualification of two processor-controlled dosing pumps for the production of mixed infusion solutions in a closed system. Pharm Technol Hosp Pharm 2016;1:29–35.10.1515/pthp-2015-0008Search in Google Scholar
10. Koletzko B, Goulet O, Hunt J, Krohn K, Shamir R Group PNGW. 1. Guidelines on paediatric parenteral nutrition of the european society of paediatric gastroenterology, hepatology and nutrition (ESPGHAN) and the european society for clinical nutrition and metabolism (ESPEN), supported by the european society of paediatric research (ESPR). J Pediatr Gastroenterol Nutr 2005;41:S1–S4.Search in Google Scholar PubMed
11. European Commission Directorate-General for Health and Food Safety. EU Guidelines for good manufacturing practice for medicinal products for human and veterinary use annex 15: Qualification and validation. Brussels: (BE): Eudralex, vol. 4, 2015:16.Search in Google Scholar
12. Driscoll DF. Delivery of nutritional therapy: Quality assurance of automated compounding devices. Nutrition 1996;12:651–652.10.1016/S0899-9007(96)00185-2Search in Google Scholar PubMed
© 2017 Walter de Gruyter GmbH, Berlin/Boston
Articles in the same Issue
- Frontmatter
- Graphical Abstract
- Editorial
- Quality Assurance in Hospital Pharmacy Compounding Units is a Multi Player Game
- Research Articles
- Pilot Study Evaluating the Efficacy of Four Cleaning Solutions and Two Types of Mops in Delimited Areas of a Floor Contaminated with Cyclophosphamide
- Evaluation of a Process Monitoring Method for Compounding Parenteral Nutrition with the Baxter EM2400 in a Hospital Pharmacy Department
- Analytical Control of Pediatric Chemotherapy Preparations with a UV-Raman Automaton: Results After 18 Months of Implementation and Development of A Suitable Method for Low Volume Preparations
- Short Communication
- Differences in Disinfection Protocols of Raw Material Vials Intended for the Production of Parenteral Nutrition Mixtures
- Opinion Paper
- Resolution CM/Res(2016)2 and Centralised Intra Venous Additive Services (CIVAS): Challenges and Opportunities
Articles in the same Issue
- Frontmatter
- Graphical Abstract
- Editorial
- Quality Assurance in Hospital Pharmacy Compounding Units is a Multi Player Game
- Research Articles
- Pilot Study Evaluating the Efficacy of Four Cleaning Solutions and Two Types of Mops in Delimited Areas of a Floor Contaminated with Cyclophosphamide
- Evaluation of a Process Monitoring Method for Compounding Parenteral Nutrition with the Baxter EM2400 in a Hospital Pharmacy Department
- Analytical Control of Pediatric Chemotherapy Preparations with a UV-Raman Automaton: Results After 18 Months of Implementation and Development of A Suitable Method for Low Volume Preparations
- Short Communication
- Differences in Disinfection Protocols of Raw Material Vials Intended for the Production of Parenteral Nutrition Mixtures
- Opinion Paper
- Resolution CM/Res(2016)2 and Centralised Intra Venous Additive Services (CIVAS): Challenges and Opportunities
