See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/269283147

Incidence of intravenous drug incompatibilities in intensive care units

Article in Biomedical Papers · November 2014

DOI: 10.5507/bp.2014.057 · Source: PubMed

CITATIONS READS

23 2,238

4 authors:

Ondrej Machotka Jan Manak

Charles University in Prague University of Hradec Králové
4 PUBLICATIONS 23 CITATIONS 60 PUBLICATIONS 2,022 CITATIONS

SEE PROFILE SEE PROFILE

Kuběna Aleš Antonín Jiri Vlcek

The Czech Academy of Sciences Charles University in Prague
112 PUBLICATIONS 2,279 CITATIONS 116 PUBLICATIONS 1,365 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Ondrej Machotka on 13 September 2017.

The user has requested enhancement of the downloaded file.

 Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2015 Dec; 159(4):652-656.

Incidence of intravenous drug incompatibilities in intensive care units

Ondrej Machotkaa, Jan Manakb, Ales Kubenac, Jiri Vlceka

Aims. Drug incompatibilities are relatively common in inpatients and this may result in increased morbidity/mortal-
ity as well as add to costs. The aim of this 12 month study was to identify real incidences of drug incompatibilities in
intravenous lines in critically ill patients in two intensive care units (ICUs).
Methods. A prospective cross sectional study of 82 patients in 2 ICUs, one medical and one surgical in a 1500-bed uni-
versity hospital. One monitor carried out observations during busy hours with frequent drug administration. Patients
included in both ICUs were those receiving at least two different intravenous drugs.
Results. 6.82% and 2.16% of drug pairs were found to be incompatible in the two ICUs respectively. Among the most
frequent incompatible drugs found were insulin, ranitidine and furosemide.
Conclusions: The study showed that a significant number of drug incompatibilities occur in both medical and surgi-
cal ICUs. It follows that the incidence of incompatibilities could be diminished by adhering to a few simple rules for
medication administration, following by recommendations for multiple lumen catheter use. Future prospective studies
should demonstrate the effect of applying these policies in practice.

Key words: medical error, medical, medication safety, drug administration, drug incompatibilities, intensive care
units

Received: April 12, 2014; Accepted with revision: October 16, 2014; Available online: November 6, 2014
http://dx.doi.org/10.5507/bp.2014.057
a
Department of Social and Clinical Pharmacy, Faculty of Pharmacy in Hradec Kralove, Charles University in Prague, Hradec Kralove, Czech
Republic
b
Department of Gerontology and Metabolism, University Hospital Hradec Kralove
c
The Institute of Information Theory and Automation, Academy of Sciences of the Czech Republic, Prague
Corresponding author: Ondrej Machotka, e-mail: [email protected]

INTRODUCTION and infections5. Several studies have specifically focused

on intravenous medication administration errors and in-
During the treatment of inpatients, medication errors vestigated the incidence6-10.
and unintended side effects are two relatively common Physicochemical incompatibility is a typical medi-
drug-related problems which may result in increasing cation error in the administration of parenteral drugs.
morbidity / mortality as well as add to the cost of the Incompatibility is defined as the reaction of intravenous
therapy1,2. Medication errors have been identified as the drugs resulting in solutions that are no longer optimal for
main factor limiting the effectiveness and safety of phar- the patient after they are mixed - the stability or structure
macotherapy3. The types and frequency of drug related of the drugs is altered by physical or chemical reactions.
problems, especially medication errors in preparation and Changes in stability can cause changes in drug effec-
administration, have dramatic implications for the overall tiveness; the increased size of microparticles has been
quality of nursing care4. frequently linked to consequences such as therapeutic fail-
Currently only a limited number of physicians and ure, catheter occlusion, or, in the worst cases, embolism11.
pharmacists have concerned themselves with monitoring A number of studies confirm that this is a significant
these problems in the Czech Republic, with even fewer and increasing problem. Taxis and Barber7 have shown
doing so outside the hospital environment. The authors of that about half of all ward-based intravenous drug prepara-
early studies and analyses were concentrated on prescrip- tions and administrations are incorrect. These incompat-
tion and dispensing errors as well as mistakes caused by ibilities produce about 20% of all medication errors and
patient non-compliance. In contrast, our study was con- almost 89% of the overall errors in the administration of
ducted in the hospital environment of an ICU. drugs8,12.
Especially in the ICU, rapid response to an emergency The aim of this study was to survey the incidence and
is often crucial. For this reason, drugs are usually admin- extent of the incompatibilities in a hospital in the Czech
istered parenterally. Other reasons for this are that drugs Republic and in Eastern Europe with the ultimate goal
are often poorly absorbed via the oral route or patients are of making recommendations for reducing the risk of in-
unable to receive drugs in other ways. Poorly prepared or compatibilities.
wrongly administered parenteral therapy can cause inter
alia thrombus formation, severe hypersensitivity reactions

652
 Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2015 Dec; 159(4):652-656.

METHODS infusions which are incompatible when infused through

the same IV line15. There were no standard operating pro-
The study was carried out as a comparison of two cedures regarding the incompatibilities in the University
ICUs at the 1500-bed University Hospital Hradec Kralove. Hospital Hradec Kralove during the data collection.
The first was a 20-bed medical ICU specializing in geron- For analysis of the data in the prospective studies, we
tology and metabolism, the second a 12-bed surgical ICU used graph coloring theory16. The averages of lumens per
treating adult patients after major surgeries and accidents. patient in both ICUs were compared using t-tests.
The prospective study led to the identification of real The study was approved by the hospital Ethics
situations of drug incompatibilities. Data collection was Committee.
conducted gradually from January 2011 to December
2011. 50 patients in the medical ICU and 32 patients in
the surgical ICU were assessed prospectively. Only pa- RESULTS
tients who were administered at least two different intra-
venous drugs were included. The data were collected by The results are shown in Table 1.
the direct observation of nurses administering intravenous The first prospective study (n=50; mean ± S.D. age;
medication. For this study we focused on drug incom- 59.3 ± 15.0 years; 54% female) was performed in the medi-
patibilities excluding both TPN and drugs not found in cal ICU. The patients were maintained on a total of 318
the database13. The observation was performed by one IV drugs (mean ± S.D., 6.3 ± 3.0) featuring 64 different
monitor (an independent graduate student in the clini- brands of medications on day two of their ICU stay. Out
cal pharmacy doctoral program) during busy hours with of a total of 318 IV drugs, 88.7% (n=282) were found in
frequent drug administration (6-10 a.m.). The observation the database of drug incompatibilities. Out of 220 drug
was undisguised: nurses and doctors who had previously pairs given to the patients through one intravenous line
been informed about the goals of the study, the first of its which were found in the database, 6.82% (n=15) of the
kind in the Czech Republic. Nurses were clearly informed pairs were incompatible, the most frequent compounds
they would not be identified in to ensure anonymity. being insulin, ranitidine, furosemide and ciprofloxacin
As the source of incompatibilities, Trissel’s Handbook (see Table 2).
on Injectable Drugs13 was used. This database contains The second prospective study (n=32; mean ± S.D. age;
incompatibility information on 359 different intravenous 58.3 ± 17.2 years; 28% female) was performed in the surgi-
active ingredients and is often mentioned as the gold cal ICU. The patients were maintained on a total of 207
standard in IV drug incompatibilities14. We focused only IV drugs (mean ± S.D., 6.5 ± 2.4) featuring 48 different
on Y-site incompatibilities, indicating two separate drug brands of medications on day two of their ICU stay. Out

Table 1. Results of our study.

Medical ICU Surgical ICU

No. patients 50 32
No. total administered IV drugs 318 207
Mean of administered drug/patient 6.3 6.5
No. administered drugs found in the database (%) 282 (88.7) 177 (85.5)
No. brands of medications 64 48
No. drug pairs 220 139
Mean (median) of lumens of catheters per patient 1.88 (1.50) 2.41 (3.00)
Incompatible pairs (%) 15 (6.82) 3 (2.16)

Table 2. Incompatible drug pairs coinfused through a common IV line.

Drug combinations (brand names) No. incompatibilities

Drug A Drug B Medical ICU Surgical ICU
Insulin (Humulin) Ranitidine (Ranital) 5 0
Vancomycin (Edicin) Omeprazole (Helicid) 2 0
Furosemide (Furosemid Biotika) Ciprofloxacin (Ciprofloxacin Kabi) 2 1
Furosemide (Furosemid Biotika) Fluconazole (Mycomax) 2 0
Insulin (Humulin) Noradrenalin (Noradrenalin Leciva) 2 0
Ciprofloxacin (Ciprofloxacin Kabi) Hydrocortisone (Hydrocortison Valeant) 1 0
Potassium chloride (Kalium chloratum Leciva) Methylprednisolone (Solu-Medrol) 1 0
Magnesium sulfate (Magnesium sulfuricum Biotika) Amiodarone (Cordarone) 0 1
Midazolam (Midazolam B. Braun) Omeprazole (Helicid) 0 1

653
 Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2015 Dec; 159(4):652-656.

of a total of 207 IV drugs, 82.9% (n=145) of them were catheter lumens used for the administration of IV drugs
found in the database of drug incompatibilities. Out of in the two different ICUs. The average lumens per patient
139 drug pairs found in the database and given to the used in the gerontology and metabolism ICU (1.88) was
patients through one intravenous line, 2.16% (n=3) of the significantly (P=0.008) lower than the average in the surgi-
pairs were incompatible, the most frequent compounds cal ICU (2.41).
being ciprofloxacin, furosemide, midazolam, omeprazole, The differences in the types of IV drugs administered
amiodarone and magnesium sulfate (see Table 2). in the ICUs could be another reason for the dissimilar
results. Although the total amount of administered drugs
per patient was higher in both studies conducted in the
DISCUSSION surgical ICU, the number of brands of used medication
was always lower in the surgical ICU in comparison with
Errors in the administration of intravenous drugs have the medical ICU.
been analyzed in a number of studies7-10,17,18, with intrave- Other studies7,9,10,15,19 focused on intravenous drug
nous drug incompatibilities comprising one subgroup of incompatibilities show an error rate of incompatibilities
these problems. Especially in intensive care, during which similar to the results observed in our study (3%, 18.6%,
parenteral drug administration is often complicated by 0.53%, 5.8%, 3.4%). On the other hand, for reasons such
the fact that the number of the concurrently administered as differences in sources of incompatibilities, study sam-
drugs exceeds the number of available infusion lines, in- ples, types of incompatibilities, methods of collecting data
travenous drug incompatibilities represent a significant as well as interpretation methods, our results cannot be
problem. compared with those in the literature (see Table 3).
Our study was designed to identify the real state of Despite differences between studies, the incidence of
intravenous drug incompatibilities at one University hos- incompatibilities may still be seen as comparable with
pital in the Czech Republic. The incompatibilities dis- other studies made in this field.
covered could suggest a lack of interest in the described Based on a comparison with Bertsche et al.15 it is
problem, knowledge deficiencies and/or the absence of a possible to show relatively clear strategies towards im-
clear strategy or effective tools for reducing the frequency provement; the establishment of standard operational
of incompatibilities. procedures, the institution of compatibility charts and
Although the frequency of the incompatibilities does the education of ICU nurses in this area have been dem-
not seem particularly high, the substances found most fre- onstrated to show promising results.
quently were often vital drugs like insulin, antibiotics, an- On the other hand, the establishment of new SOPs
tiarrhythmics and catecholamines. Even a small decrease can lead to other types of complications, e.g. the shown
in the efficiency of these drugs can cause a significant link between switching to multi-lumen catheters and a
impact on patients in ICUs. subsequent rise in catheter infections20, the increase in
Comparing the two ICUs involved in our study, the the volume of drug solvent influencing fluid balance in
differences in the results are quite interesting. The discrep- critically ill patients, as well as the escalating economic
ancies may simply be caused by the differing number of burden and increasing demands on ICU staff. These phe-

Table 3. Comparison of similar studies focused on intravenous incompatibilities.

Our study Bertsche et al.15 Taxis, Barber7 Gikic, Paolo19 Tissot, Cornette9 Westbrook et al.10
No. patients 82 25 106 19 26 Undocumented
involved
No. total 525 160 430 Undocumented Undocumented 568
administered IV
drugs
Mean of 6.4 6.4 4.1 6.5 Undocumented Undocumented
administered drug/
patient
No. drug pairs 359 516 Undocumented 175 102 Undocumented
Database of Trissel’s KIK database Manufacturer’s More database Data Data systematically
incompatibilities Handbook13 instructions sources systematically researched
researched
Incompatible pairs 18 (5.01) 30 (5.8) 12 (3.0) 6 (3.4) 19 (18.6) 3 (0.53)
(%)
Method of data Undisguised Undisguised Disguised Undisguised Undisguised Undisguised bedside
collecting bedside bedside bedside bedside bedside observation
observation observation observation observation observation

654
 Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2015 Dec; 159(4):652-656.

Table 4. Possible distribution of IV drugs in different lumens of a catheter.

Medical ICU Surgical ICU

Lumen 1 Lumen 2 Lumen 1 Lumen 2
Ciprofloxacin Phenytoin Ciprofloxacin Phenytoin
Clindamycin Furosemide Amiodarone Furosemide
Metoclopramide Omeprazole Metoclopramide Insulin
Vancomycin Hydrocortisone Dobutamine Midazolam
Insulin Potassium phosphate Omeprazole Teicoplanin
Potassium chloride Magnesium sulfate Potassium chloride Piperacillin/tazobactam
Clarithromycin Noradrenalin Noradrenaline
Midazolam Ranitidine Propofol
Fluconazole Ceftriaxone Ranitidine
Methylprednisolone Sufentanil
Fluconazole

nomena will be monitored in our next study, which will be Another limitation of our study is the fact that we did
focused on the assessment of the new SOPs in practice. not observe the clinical implications for patients. This
In our study we have partially addressed the prob- would have been very difficult in a routine clinical en-
lem of switching to multi-lumen catheters. Taking into vironment due to the difficult recognition and differen-
consideration all of the medications administered in the tiation of symptoms of incompatibilities. For example a
prospective studies in both ICUs in terms of graph color- decrease in the activity of one particular drug is difficult
ing theory11, it seems that only the two-lumen catheter to establish in the complexity of the treatments used in
remains a viable solution for preventing the problem of the ICU.
incompatibilities, thus switching to multi-lumen catheters The recurrent simplification of medicines to active
should be eliminated as an option. Table 4 shows an exam- substances was a drawback as well. Thus it was also nec-
ple of the potential distribution of IV drugs in our study in essary to limit the medicines studied to those listed in
terms of excluding the possibility of incompatibilities. The the selected database. Despite the fact that our database
other IV drugs can be added arbitrarily to any one of the - Trissel’s Handbook13 is primarily designated for the
groups without presenting problems. The only exceptions market in the United States, it is often taken as the gold
are with patients on total parenteral nutrition. For the standard among incompatibility databases14. The reasons
sake of simplicity for nursing staff, one guideline should for exclusion of TPN in the prospective studies were the
be that a single catheter with one lumen designated only variability in nutrients composition, the dependence of
for parenteral nutrition is recommended. the incompatibility of individual components on their
Our study has some limitations. First, the prospective concentration, and the lack of data on incompatibility of
observational studies provided real data regarding e.g. spe- parenteral nutrition solutions.
cific routes of administration. Nevertheless, some disad- Working toward the detection and elimination of drug
vantages of prospective observational studies are the long incompatibilities and other drug related problems has
time needed for data collection, additional demands on been acknowledged as a key goal for the clinical pharma-
the data collectors and ICU staff, as well as difficulty of cist21. Employing pharmacists with this specialization has
ensuring that studies are blind. In choosing to use this still not become sufficiently widespread in Czech hospi-
type of study we kept in mind these advantages and dis- tals, but it has been on the rise in recent years.
advantages. The use of prospective observational studies This study has demonstrated that although a signifi-
is in our opinion necessary to identify the real state of cant number of drug incompatibilities occur in both medi-
incompatibilities. cal and surgical ICUs, in fact only a limited number of
On the other hand, this could present the problem of medications are involved in these interactions. Besides
the observer interfering with the administering nurse. In the amount and type of medications, the most important
our opinion, attempting a totally blind collection of these determinant of the incidence of medication incompatibili-
data is unrealistic. Disguised observation seems to be the ties is the usage of single- or multiple lumen intravenous
optimal method but we chose undisguised observation catheters.
with the hope of the possibility of only limited interfer-
ence with the duties of the nurses who cooperated with
our study. It is significant that the nurses in our study were CONCLUSION
in fact unaware of how to solve problems with incompat-
ibilities, most likely due to difficulties in obtaining data Based on our results, the incidence of drug incompat-
about the compatibility and incompatibility of adminis- ibilities could be theoretically diminished by adhering to a
tered IV drugs (lack of information in SPC). few simple rules for medication administration and follow-
ing a set of recommendations for multiple lumen catheter

655
 Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2015 Dec; 159(4):652-656.

use. Future prospective studies should demonstrate the 7. Taxis K, Barber N. Ethnographic study of incidence and severity of
effect of applying these policies in practice. intravenous drug errors. BMJ 2003;326:684-7.
8. Taxis K, Barber N. Incidence and severity of intravenous drug errors
in a German hospital. Eur J Clin Pharmacol 2004;59:815-7.
Acknowledgement: The authors would like to thank all 9. Tissot E, Cornette C, Demoly P, Jacquet M, Barale F, Capellier G.
participating nurses for the successful collaboration. Medication errors at the administration stage in an intensive care
This work was supported by grant No 53410-C-2010 unit. Intensive Care Med 1999;25:353-9.
10. Westbrook JI, Rob MI, Woods A, Parry D. Errors in the administra-
of the Charles University Grant Agency and supported tion of intravenous medications in hospital and the role of correct
in part by grant SVV 267 005 administered by Charles procedures and nurse experience. BMJ Qual Saf 2011;20:1027-34.
University in Prague. 11. Nemec K, Kopelent-Frank H, Greif R. Standardization of infusion solu-
Author contributions: OM: study conception and design, tions to reduce the risk of incompatibility. Am J Health-Syst Pharm
2008;65:1648-54.
data collection, analysis and interpretation, drafting and 12. Taxis K, Barber N. Causes of intravenous medication errors: an eth-
manuscript revision; JM: study design, data interpretation, nographic study. Qual Saf Health Care 2003;12:343-8.
manuscript revision; AK: data analysis and interpretation; 13. Trissel LA. Handbook on Injectable Drugs, 15th edition: American
JV: study conception, data interpretation, manuscript revi- Society of Health-System Pharmacists 2008.
14. De Giorgi I, Guignard B, Fonzo-Christe C, Bonnabry P. Evaluation of
sion; all authors: final approval. tools to prevent drug incompatibilities in paediatric and neonatal
Conflict of interest statement: None declared. intensive care units. Pharm World Sci 2010;32:520-9.
15. Bertsche T, Mayer Y, Stahl R, Hoppe-Tichy T, Encke J, Haefeli WE.
Prevention of intravenous drug incompatibilities in an intensive
care unit. Am J Health-Syst Pharm 2008;65:1834-40.
REFERENCES 16. Harris JM, Hirst JL, Mossinghoff MJ. Combinatorics and Graph
Theory: Springer 2008.
1. Van den Bemt PM, Egberts TCG, de Jong-van den Berg LT, Brouwers 17. Tissot E, Cornette C, Limat S, Mourand JL, Becker M, Etievent JP,
JR. Drug-related problems in hospitalized patients. Drug Safety Dupond JL, Jacquet M, Woronoff-Lemsi MC. Observational study
2000;22:321-33. of potential risk factors of medication administration errors. Pharm
2. Bates DW, Cullen DJ, Laird N, Petersen LA, Small SD, Servi D, Laffel G, World Sci 2003;25:264-8.
Sweitzer BJ, Shea BF, Hallisey R, Vandervliet M, Nemeskal R, Leape LL. 18. Han PY, Coombes ID, Green B. Factors predictive of intravenous
Incidence of adverse drug events and potential adverse drug events fluid administration errors in Australian surgical care wards. Qual
- implications for prevention. JAMA 1995;274:29-34. Saf Health Care 2005;14:179-84.
3. Lesar TS, Briceland L, Stein DS. Factors related to errors in medication 19. Gikic M, Di Paolo ER, Pannatier A, Cotting J. Evaluation of physico-
prescribing. JAMA 1997;277:312-3. chemical incompatibilities during parenteral drug administration in
4. Barker KN, Allan EL. Research on drug-use-system errors. Am J a paediatric intensive care unit. Pharm World Sci 2000;22(3):88-91.
Health-Syst Pharm 1995;52:400-3. 20. Reed CR, Sessler CN, Glauser FL, Phelan BA. Central venous cath-
5. Bruce J, Wong I. Parenteral Drug Administration Errors by Nursing eter infections: concepts and controversies. Intensive Care Medicine
Staff on an Acute Medical Admissions Ward During Day Duty. Drug 1995;21(2):177-83.
Safety 2001;24:855-62. 21. Wiffen P, Mitchell M, Snelling M, Stoner N. Oxford handbook of clini-
6. Schneider MP, Cotting J, Pannatier A. Evaluation of nurses’ errors cal pharmacy. Oxford, UK: Oxford University Press 2012.
associated in the preparation and administration of medication in
a pediatric intensive care unit. Pharm World Sci 1998;20:178-82.

656

View publication stats