Drug incompatibility is a problem especially when managing patients in intensive care units (ICU). It can provoke physical (physical incompatibilities) and/or chemical (chemical incompatibilities) reactions. Physical incompatibilities result in visible (precipitate, color change, gas production) and invisible (subvisible particles, variations in pH) reactions. Chemical incompatibilities can lead to a decrease in drug bioavailability, drug degradation, and/or production of toxic products. Physicochemical incompatibilities have been reported in several observational studies in ICUs.1–5 Methodological differences in these studies lead to highly variable results in the frequency of incompatibilities ranging from 0.2 to 25% of drug combinations used.
Preventing incompatibility is important for the safe administration of injectable drugs. Although several handbooks and databases deal with the subject6 and 2 recent works have assessed these incompatibilities,7,8 there were discrepancies among available references and data are often missing or incomplete.9–11 The use of separate catheter lumens can prevent contact between incompatible drugs, but there are some limitations to this solution: a slight increase in the risk of infection with multilumen central venous catheters (CVC) versus single-lumen catheters and an increase in resistance to flow.12–15 Moreover, because the number of catheter lumens is typically lower than the number of drugs infused, it is necessary to use multiport manifolds. The recent new multilumen infusion access devices may prevent incompatibility and are worth assessing.
The aim of this study was to evaluate the impact of multilumen infusion access devices connected to a single-lumen CVC on the occurrence of known drug incompatibility through a controlled in vitro study comparing a standard set with 2-port manifold and 1-m extension set and 2 multilumen infusion access devices: a 3-lumen extension set and a 9-lumen extension set (Edelvaiss-Multiline; Doran International, Toussieu, France).
This laboratory in vitro study did not need ethical approval. A well-documented incompatible combination of 2 drugs was used to perform the study.6,7,9,11 Midazolam-furosemide incompatibility results from an acid-base reaction. Mixing a furosemide solution (alkaline) with a midazolam solution (acid) decreases the pH in the mixture sufficiently to result in immediate furosemide precipitation with the formation of a visible milky-white precipitate. Furosemide (10 mg/mL, lot number 102551; Renaudin, Itxassou, France), midazolam (5 mg/mL, lot number F3031; Mylan, Saint-Priest, France), and saline (Freeflex 500 mL, lot number 13CIL081; Fresenius Kabi, Sèvres, France) were infused simultaneously through 3 infusion sets, which differed in concentration and dead volume (Fig. 1):
1. A standard single-lumen set with 2-port manifold and 1-m extension set (RPB2310; Cair LGL, Civrieux d’Azergues, France) with dead volume equal to 7.1 mL;
2. A 3-lumen infusion access device (VSET+M; Doran International) that consists of a central tube with an antireflux valve and 2 flexible, low-dead-volume tubes (dead volume = 0.046 mL) reserved for furosemide and midazolam infusions (Fig. 2).
3. Another multilumen infusion access device (Edelvaiss-Multiline; Doran International) consisting of an extension set with 8 accesses connected to 9 separate lumens in a single tube (outside diameter = 4 mm, length = 150 cm). Seven accesses are for drug infusion and each is connected to a peripheral lumen (dead volume = 0.9 mL). The 8th access with high-flow-rate capacity is intended for the carrier fluid. It is connected to 2 lumens (peripheral and central) for a total dead volume of 2.9 mL. The fluids administered through the 8 ports mix at the tube outlet. Precipitate formation may differ according to the distribution of accesses; therefore, 3 different access combinations were used in relation to the distance between midazolam and furosemide, with 1 port solely dedicated to saline: (1) closest to the saline port: furosemide on access 7 and midazolam on access 1 (F7/M1); (2) at an intermediate distance: furosemide on access 6 and midazolam on access 2 (F6/M2); and (3) furthest away from the saline port and closest together: furosemide on access 4 and midazolam on access 3 (F4/M3; Fig. 1).
Three 50-mL syringes were prepared just before each experiment: 1 filled with 10, 5, or 2.5 mg/mL furosemide diluted in saline, 1 filled with 1 mg/mL midazolam diluted in saline, and 1 with pure saline. Simultaneous infusions were performed using previously purged syringe pumps and extension lines (diameter = 1.5 mm, length = 150 cm) to connect the syringes to the infusion device. A transparent extension line (diameter = 1 mm, length = 25 cm) simulating a single-lumen CVC was added at the distal end of the infusion set. All tests were made at room temperature between 18°C and 22°C.
Physical compatibility assessment was performed by 2 methods: (1) visual inspection of precipitate in the extension line against a black background, and (2) light obscuration subvisible particle count test according to the European Pharmacopeia.16 The visual inspections were performed by a trained pharmacist. Measurements of pH were made on drug solutions in the syringes and at the egress of the infusion device using a pH meter (PHM201 MeterLab; Radiometer Analytical, Villeurbanne, France). The infusion passed the visual inspection test if it revealed no precipitate resembling white smoke and no particle deposition in any part of the transparent terminal extension line simulating the CVC. If no precipitate and no particle deposition were observed, the light obscuration subvisible particle count test was performed on a 25-mL solution sample collected at the end of the terminal extension line simulating the CVC. Particle counts were taken using a particle counter (APSS-2000; PMT, Dourdan, France). The test was performed under the conditions described in chapter 2.9.19 of the 7.5th European Pharmacopeia urging caution about equipment, air bubble contamination, and validation of the work environment.16 Sampling consisted of 4 successive 5-mL aliquots from the solution sample. The result obtained for the first portion was disregarded. The mean number of particles for the sample to be examined was calculated from the results of the 3 remaining portions. The infusion condition complied with the subvisible particle count test as long as the average number of particles present in the sample tested did not exceed 25/mL for particle sizes ≥10 µm and 3/mL for particle sizes ≥25 µm. The solution sample did not satisfy the test if the threshold was exceeded.
Midazolam was infused at the rate of 2 mL/h to obtain drug delivery of 2 mg/h. The furosemide infusion rate was established in relation to drug concentration to obtain drug delivery of 20 mg/h: 2, 4, and 8 mL/h with 10, 5, and 2.5 mg/mL concentrations, respectively. The saline infusion rate was initially set at 100 mL/h. Preliminary internal work had shown that the sequence of drug administration had no impact on this incompatibility. The infusion of furosemide and saline started first, followed by midazolam 3 minutes later. The infusion was first subjected to visual inspection for 15 minutes and the solution sample for the particle count test was collected after 5 minutes of midazolam infusion. Each infusion condition was repeated 3 times by the same person with a new infusion set. One visual inspection was performed per trial. If no precipitate was observed for all 3 trials, 3 particle counts per trial were performed.
If no precipitate and no particle deposition was observed during the visual inspection for all 3 trials and if the average number of subvisible particles was <25/mL for particle sizes ≥10 µm and <3/mL for particle sizes ≥25 µm for all 3 trials, new experiments with new devices were performed using a saline infusion rate decreased by 10 mL/h until physical incompatibility was detected (Fig. 2).
The lowest saline infusion rate value to satisfy the 2 tests for all 3 trials is reported for each infusion set.
The results are presented in Table 1. Initial pH values for drug solutions in syringes were 3.47 for midazolam at 1 mg/mL and 8.36, 8.62, and 8.77 for furosemide at 2.5, 5, and 10 mg/mL, respectively. The pH value of saline was 5.62. For each furosemide concentration, the lowest saline infusion rate value related to the infusion condition to reveal no physical incompatibility differed according to the infusion devices. The pH values of mixed solution at the egress of the infusion device related to the lowest saline infusion rate value were similar whatever the infusion device and access combination. The standard set with 2-port manifold and 1-m extension set revealed visible precipitate even at the highest saline flow rate (100 mL/h) and whatever the furosemide concentration (10, 5, or 2.5 mg/mL). The VSET+M device revealed no visible precipitate, no particle deposition, and an average number of particles <25/mL for particle sizes ≥10 µm and <3/mL for particle sizes ≥25 µm but only at 5 and 2.5 mg/mL furosemide concentrations and with a saline infusion rate of 100 and 50 mL/h, respectively. When using F7/M1 and F6/M2 accesses (closest to the saline port), the Edelvaiss-Multiline device passed the 2 tests regardless of the furosemide concentration. The use of F7/M1 and F6/M2 accesses allowed for the largest reduction in saline flow rate (down to 50, 30, and 20 mL/h for 10, 5, and 2.5 mg/mL furosemide concentrations, respectively). However, results for the F4/M3 access (furthest away from the saline port and closest together) were not satisfactory (precipitate formation) despite saline flow rates of 100 mL/h with the 10 and 5 mg/mL concentrations of furosemide. This access also required 100 mL/h saline for the 2.5-mg furosemide concentration. For each multilumen access device, the saline flow rate satisfying the tests differed according to furosemide concentration. In all cases, when the infusion condition passed the visual inspection (i.e., no precipitate and no particle deposition), the subvisible particle count revealed an average number of particles <25/mL for particle sizes ≥10 µm and <3/mL for particle sizes ≥25 µm.16
The aim of our study was to test the ability of multilumen access devices to prevent physical incompatibility between 2 drugs. Three factors were found to affect physical compatibility: drug concentration, carrier fluid flow rate, and the infusion device itself.
For the standard set, precipitation occurred whatever the furosemide concentration. In specific infusion conditions, the use of multilumen infusion access devices revealed no physical incompatibility between the 2 drugs and is capable of preventing it: the 3-lumen device only for the lowest furosemide concentration and the Edelvaiss-Multiline 9-lumen device regardless of the furosemide concentration, although this result was obtained for only 2 access combinations (F7/M1 and F6/M2). The other access combination (F4/M3) was unable to prevent physical incompatibility.
Our methodology of visible inspection made it possible to determine the presence or absence of visible particles according to infusion conditions. The furosemide/midazolam incompatibility is clearly detectable by a nontrained pharmacist, anesthesia resident, technician, or nurse. Drug concentrations throughout the study are those typically used in clinical practice in continuous IV infusion. The concentration of furosemide varied from 10 to 2.5 mg/mL, although the concentration of midazolam was kept constant at 1 mg/mL. If no precipitate was visually observed, the subvisible particle count test made it possible to detect invisible particles. All solution samples with no visible particles conformed to the subvisible particle count test as defined by the European Pharmacopeia.16 Under our conditions, the absence of visible particles appears to be predictive of low particulate contamination of the solution sample at the egress of the CVC.
Because midazolam-furosemide incompatibility is pH-dependent, the impact of the furosemide concentration was predictable. The dilution of furosemide in saline produces an alkaline solution. Mixing a furosemide solution with an acid solution (e.g., midazolam) decreases the pH in the mixture sufficiently to result in furosemide precipitation.
Our main hypothesis was that fluid dynamics differ according to infusion devices and accesses, which modify contact time between the 2 drugs and saline. In the case of the 2-port manifold and 1-m extension set, furosemide and midazolam are immediately in contact and therefore precipitate because they have insufficient time to be diluted with the saline. In the case of the multilumen access device, the drugs are separated until they reach the catheter inlet. The lumen arrangement of this set may induce fluid dynamics at the infusion device egress that result in dilution before contact between the 2 drugs. In the case of the Edelvaiss-Multiline device, the lumen arrangement, that is, 2 carrier-fluid lumens, 1 central and 1 peripheral situated between the lumens of drugs to be infused, may prevent contact between drug-concentrated solutions until their dilution at the outlet. Therefore, the conditions under which drugs are mixed in the CVC must be taken into consideration because their impact seems to be greater than that of the pH levels of mixed solutions at the end of the extension line, which were similar whatever the infusion device and access combination. The above result is obtained only for the accesses on either side of the peripheral carrier-fluid access and not for the other combinations where the carrier fluid plays a minor role. Moreover, it is obtained with an acceptable saline flow rate not exceeding 1200 mL per day.
There are several limitations to this study. Our assessment was limited to a 2-drug combination inducing pH-dependent incompatibility. The results may vary with other drug combinations. The midazolam concentration was unchanged during the study to maintain high acidity. Our results should be confirmed by testing other drug combinations. The number of tests per condition was in accordance with the European Pharmacopoeia.16 This number is not sufficient for a statistical analysis of results, particularly when every replication has the same outcome. The visual inspections could be more valid if performed by a person blinded to the devices, saline flow rates, and furosemide concentrations being used. Moreover, we studied only 3 combinations of infusion access for the 9-lumen infusion device. The fluid dynamics hypothesis needs to be confirmed by testing more access combinations and more flow rates.
The direct clinical applicability of this study is limited, because it simulated infusion via a single-lumen catheter whereas most patients in the ICU requiring continuous infusion of potentially incompatible solutions will probably have a multilumen catheter. Studies of the infusion of incompatible drugs via multilumen catheters have demonstrated that the staggered orifices of the multilumen catheter reduce the phenomena of drug incompatibility.17–19 Because the number of catheter lumens is usually less than the number of drugs infused, there is still a risk of drug incompatibility, and multilumen catheters apparently increase the risk of infection.12–15 Multilumen infusion access devices therefore offer a new way of preventing drug incompatibility. The Edelvaiss-Multiline device prevents physical furosemide-midazolam incompatibility with results depending on the access combination. These first in vitro results obtained under specified conditions must be confirmed by further studies replicating common situations in clinical practice (changeover of drug infusion pump, interruption and resumption of drug flow, changes in drug flow rate) and by assessing other drug incompatibilities to validate their use.
The infusion condition revealed no physical incompatibility if no precipitate and no particle deposition were observed during the visual inspection for all 3 trials and if the average number of subvisible particles was <25/mL for particle sizes ≥10 µm and <3/mL for particle sizes ≥25 µm for all 3 trials.
Name: Aurélie Foinard, MSc.
Contribution: This author helped collect and analyze the data and write the manuscript.
Attestation: Aurélie Foinard has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Conflicts of Interest: The author has no conflicts of interest to declare.
Name: Bertrand Décaudin, PhD.
Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.
Attestation: Bertrand Décaudin has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.
Conflicts of Interest: Bertrand Décaudin reports receiving reimbursements of travel expenses related to medical congresses from Doran International.
Name: Christine Barthélémy, PhD.
Contribution: This author helped write the manuscript.
Attestation: Christine Barthélémy has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Conflicts of Interest: The author has no conflicts of interest to declare.
Name: Bertrand Debaene, PhD.
Contribution: This author helped design the study, analyze the data, and write the manuscript.
Attestation: Bertrand Debaene has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Conflicts of Interest: Bertrand Debaene reports receiving reimbursements of travel expenses related to medical congresses from Doran International.
Name: Pascal Odou, PhD.
Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.
Attestation: Pascal Odou has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Conflicts of Interest: Pascal Odou reports receiving reimbursements of travel expenses related to medical congresses from Doran International.
This manuscript was handled by: Dwayne R. Westenskow, PhD.
The authors thank Prof. Olivier Mimoz (Surgical Intensive Care Unit, Centre Hospitalier Universitaire de Poitiers, Poitiers, France) for reviewing the manuscript. The authors also thank the manufacturer Doran International (Toussieu, France) for providing technological expertise on this work.
1. Gikic M, Di Paolo ER, Pannatier A, Cotting J. Evaluation of physicochemical incompatibilities during parenteral drug administration in a paediatric intensive care unit. Pharm World Sci. 2000;22:88–91
2. Fahimi F, Ariapanah P, Faizi M, Shafaghi B, Namdar R, Ardakani MT. Errors in preparation and administration of intravenous medications in the intensive care unit of a teaching hospital: an observational study. Aust Crit Care. 2008;21:110–6
3. Taxis K, Barber N. Incidence and severity of intravenous drug errors in a German hospital. Eur J Clin Pharmacol. 2004;59:815–7
4. Tissot E, Cornette C, Demoly P, Jacquet M, Barale F, Capellier G. Medication errors at the administration stage in an intensive care unit. Intensive Care Med. 1999;25:353–9
5. Wirtz V, Taxis K, Barber ND. An observational study of intravenous medication errors in the United Kingdom and in Germany. Pharm World Sci. 2003;25:104–11
6. Trissel LA. Handbook on Injectable Drugs. 15th ed. 2011 Bethesda, MD American Society of Health-System Pharmacists
7. De Giorgi I, Guignard B, Fonzo-Christe C, Bonnabry P. Evaluation of tools to prevent drug incompatibilities in paediatric and neonatal intensive care units. Pharm World Sci. 2010;32:520–9
8. Smith WD, Karpinski JP, Timpe EM, Hatton RC. Evaluation of seven i.v. drug compatibility references by using requests from a drug information center. Am J Health Syst Pharm. 2009;66:1369–75
9. Kanji S, Lam J, Johanson C, Singh A, Goddard R, Fairbairn J, Lloyd T, Monsour D, Kakal J. Systematic review of physical and chemical compatibility of commonly used medications administered by continuous infusion in intensive care units. Crit Care Med. 2010;38:1890–8
10. Kalikstad B, Skjerdal A, Hansen TW. Compatibility of drug infusions in the NICU. Arch Dis Child. 2010;95:745–8
11. Chiu MF, Schwartz ML. Visual compatibility of injectable drugs used in the intensive care unit. Am J Health Syst Pharm. 1997;54:64–5
12. Bouza E, Guembe M, Muñoz P. Selection of the vascular catheter: can it minimise the risk of infection? Int J Antimicrob Agents. 2010;36 Suppl 2:S22–5
13. Zürcher M, Tramèr MR, Walder B. Colonization and bloodstream infection with single- versus multi-lumen central venous catheters: a quantitative systematic review. Anesth Analg. 2004;99:177–82
14. Dezfulian C, Lavelle J, Nallamothu BK, Kaufman SR, Saint S. Rates of infection for single-lumen versus multilumen central venous catheters: a meta-analysis. Crit Care Med. 2003;31:2385–90
15. Templeton A, Schlegel M, Fleisch F, Rettenmund G, Schöbi B, Henz S, Eich G. Multilumen central venous catheters increase risk for catheter-related bloodstream infection: prospective surveillance study. Infection. 2008;36:322–7
16. European Pharmacopoeia Commission. . Particulate contamination: sub-visible particles. In: European Pharmacopeia. 20127th ed:3559–62 Author
17. Collins JL, Lutz RJ. In vitro study of simultaneous infusion of incompatible drugs in multilumen catheters. Heart Lung. 1991;20:271–7
18. Reyes G, Mander GS, Husayni TS, Sulayman RF, Jaimovich DG. In-vivo evaluation of simultaneous administration of incompatible drugs in a central venous catheter with a decreased port to port distance. Crit Care. 1999;3:51–3
19. Jaimovich DG, Rose WW. In vivo evaluation of simultaneous administration of incompatible drugs via a double-lumen peripheral catheter. Crit Care Med. 1990;18:1164–6