Original research

Implementation of pharmacists’ monitoring for

intravenous drug compatibility
Young Ju Cheon,1 Kyong nam Ye,1 Kyong ha Park,2 Jung Bo Kim,1 Ji Eun Yoon,1
Min Kyung Lee,1 Jung Tae Kim,1 Sook Hee An ‍ ‍3
1
Department of Pharmacy, ABSTRACT
Kyung Hee University Hospital Background Injectable medication errors primarily WHAT IS ALREADY KNOWN ON THIS TOPIC
at Gangdong, Seoul, South ⇒ Injectable medication errors primarily occur
Korea occur during preparation and administration. Currently,
2
ONESglobal Inc, Seoul, South South Korea is experiencing chronic pharmacist during preparation and administration, and
Korea shortages. Moreover, pharmacists have not routinely inquiries regarding these errors are directed to
3
College of Pharmacy, conducted prescription monitoring for intravenous the pharmacy.
Wonkwang University, Iksan, ⇒ As hospital pharmacists in Korea are
compatibility. In the present study, we analysed the
Jeollabuk-­do, South Korea
implementation of a pre-­issue monitoring program experiencing chronic personnel shortages,
using recently released cloud-­based software to provide injectable orders in all wards have not been
Correspondence to
Professor Sook Hee An, College information on intravenous compatibility in the pharmacy monitored.
of Pharmacy, Wonkwang at a general hospital in South Korea. ⇒ Several solutions for the physicochemical safety
University, Iksan, Jeollabuk-­do, Objectives The aims of this study were to determine information of parenteral medication have been
South Korea; ​shan7@​wku.​ac.​kr implemented, but integration with hospital
whether adding an intravenous drug prescription review
Received 9 December 2021 to pharmacists’ actual work scope could promote patient information systems has been minimal.
Accepted 9 August 2022 safety, and to assess the impact of this new task on
WHAT THIS STUDY ADDS
Published Online First pharmacists’ workload.
1 September 2022 Methods Data on intravenous drugs prescribed in the ⇒ The new task (intravenous compatibility
EAHP Statement 4: Clinical intensive care unit and haematology-­oncology ward review using a computerised program) allowed
Pharmacy Services. were prospectively collected during January 2020. pharmacists to identify quickly intravenous drug
Four quantitative items were evaluated: the run-­time, pairs with incompatibility risk and prevent their
intervention ratio, acceptance ratio, and the information administration.
completeness ratio with regard to the compatibility of ⇒ Although a digital tool helped monitor
intravenous drugs. intravenous compatibility, the tool’s range
Results The mean run-­time of two pharmacists was of compatibility information was incomplete.
18.1 min in the intensive care unit and 8.7 min in the Therefore, research to find more experimental
haematology-­oncology ward (p<0.001). Significant studies reporting physicochemical evidence
differences were also found between the intensive should continue.
care unit and the haematology-­oncology wards in HOW THIS STUDY MIGHT AFFECT RESEARCH,
terms of the mean intervention ratio (25.3% vs 5.3%, PRACTICE OR POLICY
respectively; p<0.001) and the information completeness
⇒ The intravenous compatibility check procedure
ratio (38.3% vs 34.0%, respectively; p=0.007).
However, the mean acceptance ratio was comparable can be embedded into the workflow of the
(90.4% in the intensive care unit and 100% in the hospital pharmacist, helping to reduce potential
haematology-­oncology ward; p=0.239). The intravenous medication errors.
⇒ This study provides methodological information
pairs that most frequently triggered interventions were
tazobactam/piperacillin and famotidine in the intensive for hospitals that have not yet implemented
care unit, and vincristine and sodium bicarbonate in the intravenous compatibility monitoring programs.
haematology-­oncology ward.
Conclusion This study suggests that despite a shortage
of pharmacists, intravenous compatibility can be
pharmacists review the drug name, dosage, dosing
monitored before issuing injectable products in all wards.
method, and incompatibilities of injectable medica-
Because the prescribing pattern of injections varies
tions, such as anticancer drugs, special total paren-
across wards, pharmacists’ tasks should be established
teral nutrition, and neonatal intensive care unit
accordingly. To improve the completeness of information,
(NICU) injections.
efforts to generate more evidence should continue.
A pharmacist prepares ready-­to-­administer inject-
able syringes or infusion bags for some wards, such
© European Association of as haematology-­ oncology, the intensive care unit
Hospital Pharmacists 2023. No
commercial re-­use. See rights INTRODUCTION (ICU), and the NICU. However, injections in other
and permissions. Published In the USA, parenteral medications must be prepared wards are issued in the finished product form (eg,
by BMJ. by pharmacists; however, in South Korea (hereafter, a vial or ampoule), prepared, and administered by
To cite: Cheon YJ, Ye Knam, Korea), these medications are prepared by nurses, nurses. Problems may occur due to the septic envi-
Park Kha, et al. which may lead to an increased risk of medication ronment where preparation is performed and there
Eur J Hosp Pharm errors, unsafe disposal, and the re-­use of contam- is insufficient information available on the physico-
2023;30:101–106. inated syringes.1 At most medical institutions, chemical safety of injections.2
Cheon YJ, et al. Eur J Hosp Pharm 2023;30:101–106. doi:10.1136/ejhpharm-2021-003187    101
 Original research
Medication errors in injections prepared by nurses primarily
occur during preparation (33.6%) and administration (66.4%).3
The four representative types of errors (wrong mixture, wrong
volume, wrong rate, or incompatibility) account for 91.7% of
errors made by nurses.4 Therefore, the administration informa-
tion for intravenous (IV) injections must be verified to ensure
patient safety. It is particularly important to monitor IV incom-
patibility, an error that becomes evident in the administration
process as the drug may precipitate and cause catheter obstruc-
tion, venous irritation, and possible pulmonary or renal embo-
lism.5 There have been cases of death after the administration of
lactated Ringer’s solution mixed with ceftriaxone, a well-­known
incompatible mixture, as well as fatal cardiac arrest after a fluid
injection of antibiotics mixed with gastrointestinal drugs.6 7 Figure 1 The process of pharmacists’ intervention in the new task.
A study conducted in Korea found that 88.1% of healthcare
providers who required assistance from the pharmacy depart-
ment were nurses. Out of 1749 inquiries that the pharmacy conducted for injection prescriptions in the surgery, internal
received in the hospital, the most frequent questions related to medicine, and ICU wards. The workflow of the new pharmacist
injection medication stability and compatibility (n=500).8 This task was also designed. Then, the new task was formally intro-
finding suggests that nurses in Korea have limited access to infor- duced to the pharmacy, and all pharmacists were instructed to
mation on IV injections. In Korea, Lexi-­comp Inc, Micromedex use the program.
IV Compatibility, the website Stabilis version 4.0, and the King This study aimed to confirm whether adding this new task
Guide to Parenteral Admixtures are used to obtain safety infor- for IV compatibility monitoring to the pharmacists’ actual work
mation regarding IV injections. These online services or text- scope promoted patient safety, as well as to investigate the
books provide physicochemical safety information on various impact of this program on pharmacists’ workload.
parenteral medications, but have expensive subscription fees
and limited information on the injection medications produced METHODS
in Korea. Korean companies, such as KIMS and BIT Druginfo, The TF team introduced the new service workflow and trained
only provide the labelling information for products and do not all pharmacists at KUHG on the task. The study focused on the
provide compatibility information. Thus, medical professionals ICU and the haematology-­oncology ward. The ICU has many
who need compatibility information on injections frequently parenteral prescriptions for inpatients, and many high-­risk anti-
inquire at the pharmacy, resulting in an excessive workload for cancer drugs are used in the haematology-­oncology ward. Two
pharmacists. pharmacists with more than 5 years of hospital practice expe-
In Korea, approximately 10% of active licensed pharmacists rience conducted this new task of reviewing IV compatibility.
work in hospitals. Due to legislation stipulating one or more The pharmacists searched for incompatible IV pairs, confirmed
pharmacists per 30–80 inpatients per annual average of over 300 the information, and then provided the content to nurses by
beds (Table 5-­2, Enforcement Rules of Medical Service Act; The phone or an EMR message before the drugs were administered
Ordinance of the Ministry of Health and Welfare No. 851, 31 (figure 1). They prospectively collected task-­related data during
Dec 2021, Partial Amendment), hospital pharmacists are always their actual workdays from 1 January 2020 to 31 January 2020.
understaffed. As a result, only a few tertiary hospitals regularly
monitor parenteral prescriptions, although this is an important Definition of terms
part of the pharmacist’s job.9 New task: The IV prescription review through the linked
Kyunghee University Hospital at Gangdong (hereafter, ConnectDI program in the ICU and haematology-­oncology ward
KUHG) in Seoul, Korea is a general hospital with about 650 beds before issuing a drug.
and 12 pharmacists on the dispensing team. Before this study, Compatible: Physicochemical compatibility among IV drug
the dispensing team monitored non-­parenteral drugs by using an pairs, which the literature describes as no change in colour or
inspection program to review each patient, intervene as needed, gas formation and no precipitation during preparation.
and record those actions. However, this system did not involve Incompatible: Physicochemical incompatibility among IV
monitoring for inappropriate injections. drug pairs based on the literature or approval information, and
In 2019, ConnectDI (ONESglobal Co, Ltd, Seoul, Korea) was IV pairs showing uncertain or contradictory results.
released in Korea. This cloud-­ based program provides infor- Intervention: The act of delivering information on IV compat-
mation about injections produced in Korea in addition to drug ibility risk to nurses via phone or an EMR note. When the
information based on Trissel’s guide. Through this program, program showed incompatible results (incompatible or uncer-
the KUHG pharmacy implemented parenteral prescription tain in any route), a pharmacist confirmed the content, then
reviews to improve patient safety. First, a task force (TF) team delivered the pharmacist’s comment to nurses via phone or a
was composed of the Drug Regulatory Affairs Team Leader, nursing note in the EMR. A pharmacist confirmed adherence to
Dispensing Team Leader, QI Leader, IT Specialist at KUHG, and the intervention within 24 hours (figure 1).
the CEO of ONESglobal. Run-­time: A pharmacist’s working time to review IV compati-
The TF team linked ConnectDI to KUHG hospital informa- bility using the ConnectDI linked program on EMR.
tion system (HIS) from July to August 2019, then developed the Intervention frequency: The frequency of interventions by a
new electronic medical record (EMR) view that reviewed the pharmacist about IV compatibility.
reconstitution, dilution, and compatibility information for IV Acceptance frequency: The frequency of nurses’ acceptance
medications. Before the review program was officially launched of the pharmacist’s intervention (ie, that two IV drugs were not
on the pharmacy EMR, in October 2019, two trials were simultaneously administered through the same infusion line).
102 Cheon YJ, et al. Eur J Hosp Pharm 2023;30:101–106. doi:10.1136/ejhpharm-2021-003187
 Original research

Table 1 Definitions of the four outcomes

Indicator Definition Calculation
Run-­time The required time taken by pharmacists to conduct the The required time from the start of the task to the end of the task
new task (recorded in minutes)
Intervention ratio (%) The frequency of interventions by the pharmacist for (The sum of interventions)/(the sum of inpatients prescribed IV drugs) ×
compatibility results that have been checked 100
Acceptance ratio (%) The frequency of acceptance of the intervention (The sum of accepted interventions)/(the sum of interventions) × 100
Information completeness ratio (%) The number of drug pairs with compatibility information (The sum of drug pairs with compatibility information)/(the sum of
among the prescribed IV drug pairs theoretical drug pairs) × 100
IV, intravenous.

Data collection Intervention and acceptance ratio

The following data were collected to evaluate whether the new The ICU pharmacist conducted 54 interventions, averaging
task had a significant impact on patient safety: the frequency of 3.4±1.7 (range 1–6) per day. Therefore, the intervention
interventions, acceptance, IV pairs (compatible, incompatible), ratio for IV drugs was 25.3%. The ICU nurses accepted 49 of
and the IV compatibility information shown on the program 54 interventions (averaging 3.1±1.7 per day), corresponding
for the prescribed IV pairs. The following data were collected to an acceptance ratio of 90.4% (table 3). The five interven-
to assess pharmacists’ workload: the number of inpatients, the tions that were not accepted were as follows; pantoprazole and
number of IV prescriptions, and the working day and required vancomycin (one case), pantoprazole and midazolam (one case),
time of pharmacists. piperacillin/tazobactam and famotidine (two cases), and furo-
semide and vasopressin (one case). The physician’s order was
Outcomes discontinued in two of these cases, and the medication pair was
To quantitatively assess the value of the new task, the following administered through the same IV line in the other three cases.
items were defined: (1) run-­time, (2) ratio of interventions to In one case, although a nurse read the intervention message, she
patients, (3) acceptance ratio of interventions completed by did not hand it over, and then a pharmacist educated the nurses
nurses, and (4) completeness of the compatibility information about the incompatibility risk. However, in two cases, the inter-
(table 1). vention messages were not delivered properly. The haematology-­
To check the completeness of the information on theoretical oncology ward pharmacist provided 17 interventions, averaging
compatibility, the percentage of compatibility results provided 0.9±0.7 (range 0–2) per day, with an intervention ratio of 5.3%.
by ConnectDI on the number of drug-­drug pairs was investi- The haematology-­oncology ward nurses accepted all interven-
gated in each ward. We compared both wards to identify differ- tions, with an acceptance ratio of 100.0% (table 3).
ences in pharmacists’ workload with the new task. The intervention ratio was significantly different between
The study protocol was approved by the institutional review both wards (p<0.001). Converting these results to a per-­patient
board of KUHG (IRB no. KHNMC 2020-­02-­027). basis, IV incompatibility risk occurred once for every four
patients in ICU and once per 19 patients in the haematology-­
oncology ward. Both acceptance ratios were high (90.4% and
Statistical analysis
100%, respectively), without a significant difference between
Frequency analysis was conducted for the data collected for each
them (p=0.239) (table 3).
endpoint. To compare the tasks of the ICU and haematology-­
oncology ward, the independent t-­test or the Mann-­Whitney U
test was applied for data with or without a normal distribution, Information completeness
respectively. The significance level was p<0.05. PASW Statistics The average number of theoretical IV pairs in the ICU was
version 18.0 (SPSS Inc, Chicago, IL, USA) was used for the statis- 498.6±127.4 per day. The daily mean number of IV pairs with
tical analysis. compatibility information was 189.5±46.8. Therefore, the
information completeness ratio in the ICU was 38.3±3.5%.
The daily mean number of theoretical IV pairs in the
RESULTS
haematology-­oncology ward was 294.1±109.4, while the daily
Pharmacists’ run-time mean number of IV pairs with compatibility information was
The ICU pharmacist worked for 16 days in January 2020 and
100.7±43.1. Therefore, the information completeness ratio in
monitored a total of 213 patients, averaging 13.3±2.1 per day.
the haematology-­oncology ward was 34.0±5.2% (table 3). The
In total, 3663 IV drugs were prescribed, averaging 227.1±43.6
information completeness ratio of two wards was significantly
per day and 17.1±9.0 per patient. The pharmacist provided a
total of 54 interventions, averaging 3.4 per day, and her run-­time
was 18.1±5.4 min per day.
The haematology-­oncology ward pharmacist worked for 20 Table 2 Pharmacists’ workload in conducting the new task
days over the same period and monitored a total of 338 patients, ICU ONC
averaging 16.9 per day. The total number of IV drugs prescribed Variables Mean±SD Mean±SD P value
was 3511, averaging 175.6±42.2 per day and 10.4±6.5 per
Number of patients per day 13.3±2.1 16.9±3.6 0.001
patient. The haematology-­oncology ward pharmacist provided
Number of IV drugs per day 227.1±43.6 175.6±42.2 0.001
a total of 17 interventions, averaging 0.9 per day, and her run-­
time was 8.7±3.2 min per day. The pharmacists’ run-­time for Number of IV drugs per patient 17.1±9.0 10.4±6.5 <0.001
conducting the new task showed a statistically significant differ- Pharmacist’s run-­time (minutes) 18.1±5.4 8.7±3.2 <0.001
ence (p<0.001) (table 2). ICU, intensive care unit; IV, intravenous; ONC, haematology-­oncology ward.

Cheon YJ, et al. Eur J Hosp Pharm 2023;30:101–106. doi:10.1136/ejhpharm-2021-003187 103

Original research
different (p=0.007). For the IV drugs with compatibility infor-
Table 3 Frequency of intravenous compatibility-­related parameters
mation, the compatibility percentage was 68.3±4.1% in the ICU
in the ICU and ONC ward
and 68.2±5.8% in the haematology-­oncology ward; this small
ICU ONC difference was not significant (p=0.990).
Variables Mean±SD Mean±SD P value
Intervention ratio
Frequent incompatible pairs
 Number of interventions per day 3.4±1.7 0.9±0.7 <0.001
The most frequent incompatible pairs in the ICU were tazo-
 Intervention ratio (%) 25.3±11.7 5.3±5.3 <0.001
bactam/piperacillin and famotidine (15 total interventions).
Acceptance ratio
Furosemide was the most common incompatible drug in the ICU
 Number of accepted interventions 3.1±1.7 0.9±0.1 <0.001*
when paired with clarithromycin, levofloxacin, metoclopramide,
per day
midazolam, labetalol, nicardipine, vasopressin, and vecuronium,
 Acceptance ratio (%) 90.4±17.8 100.0±0.0 0.239*
resulting in 18 interventions by a pharmacist. The most common
Information completeness
incompatibility in the haematology-­oncology ward was vincris-
 Number of theoretical drug pairs 498.6±127.4 294.1±109.4 <0.001
tine and sodium bicarbonate (seven times), followed by pantopra-
per day
zole and vancomycin (three times). The number of intervention
  Number without compatibility 309.1±84.1 193.5±69.0 <0.001
information per day pairs was 19 in the ICU and five in the haematology-­oncology
  Number with compatibility 189.5±46.8 100.7±43.1 <0.001
ward. The high-­frequency intervention pairs were substantially
information per day different between the two wards (table 4).
   Number of compatible drug 129.4±32.7 70.0±33.6 <0.001
pairs per day
   Number of incompatible drug 60.1±16.6 30.7±11.1 <0.001
DISCUSSION
pairs per day The KUHG pharmacy established a new task for pharmacists to
 Information completeness (%) 38.3±3.5 34.0 ± 5.2 0.007 review IV compatibility before administering drugs to patients,
Intervention ratio = (the sum of interventions)/(the sum of inpatients prescribed intervene in cases of incompatibility, and then record the contents
injections) × 100; acceptance ratio = (the sum of accepted interventions)/(the sum and results of the intervention. The KUHG pharmacy has long
of interventions) × 100; information completeness = (the sum of IV drug pairs with desired to carry out this task, but doing so was not possible
compatibility information)/(the sum of prescribed IV drug pairs) × 100. previously due to a lack of human resources and the absence of
p<0.05 indicates statistical significance. appropriate computer programs. However, a computer program
*Mann-­Whitney U test.
for this task was released recently in Korea, and the TF team was
ICU, intensive care unit; IV, intravenous; ONC, haematology-­oncology ward.
assembled to develop and implement this new task. This study
shows significant findings in relation to new task utilisation.

Table 4 Pharmacists’ interventions for incompatible intravenous drug pairs

Drug A Drug B Route Effect/result Frequency
Intensive care unit 54
 Tazobactam/piperacillin Famotidine Y-­site Particles form immediately 15
 Pantoprazole Vancomycin Y-­site Colour change after 10 hours 7
 Pantoprazole Midazolam Y-­site Precipitate forms immediately 5
 Furosemide Levofloxacin Y-­site Cloudy precipitate forms 4
 Furosemide Midazolam Y-­site White precipitate forms immediately 4
 Furosemide Nicardipine Y-­site Precipitate forms immediately 4
 Furosemide Vecuronium Y-­site Precipitate forms immediately 2
 Propofol Methylprednisolone Y-­site White precipitate forms immediately 2
 Acetylcysteine Cefepime Y-­site Over 10% cefepime loss in 1 hour 1
 Acetylcysteine Ceftazidime Y-­site Over 20% ceftazidime loss in 1 hour 1
 Hydrocortisone Ciprofloxacin Y-­site White cloudiness forms immediately 1
 Hydrocortisone Midazolam Y-­site White precipitate forms immediately 1
 Furosemide Clarithromycin Y-­site White cloudiness forms immediately 1
 Furosemide Metoclopramide Y-­site Precipitate forms immediately 1
 Furosemide Labetalol Y-­site White precipitate forms immediately 1
 Furosemide Vasopressin Y-­site White precipitate forms in 5 to 15 min 1
 Pantoprazole Cisatracurium Y-­site Turbidity increases 1
 Sulfamethoxazole/trimethoprim Linezolid Y-­site Physically incompatible 1
 Tazobactam/piperacillin Amiodarone Y-­site White hazy form in 24 hours 1
Hematology-­oncology ward 17
 Vincristine Sodium bicarbonate Y-­site White precipitate forms in 30 min 7
 Tazobactam/piperacillin Famotidine Y-­site Particles form immediately 4
 Pantoprazole Vancomycin Y-­site Colour change after 10 hours 3
 Nicardipine Micafungin Y-­site Precipitate forms immediately 2
 Pantoprazole Mannitol Y-­site Particles form 1

104 Cheon YJ, et al. Eur J Hosp Pharm 2023;30:101–106. doi:10.1136/ejhpharm-2021-003187

 Original research
We had two questions related to the research goal: (1) Will and 5.3% in the haematology-­oncology ward. In our study, three
the new task help patient safety? (2) How long will it take to interventions were not accepted because the process of checking
conduct the new task? To address the first question, three param- compatibility was not known or omitted in the handover. In order
eters (pharmacists’ interventions, nurses’ acceptance, and infor- to establish a new task, the pharmacy department must imple-
mation completeness) were analysed. We assessed the daily mean ment more specific training and work closely with the nursing
numbers of IV compatibility errors, interventions, and accepted department. It is also necessary to establish guidelines within the
cases, and evaluated the degree to which the information made hospital on how to act after identifying IV incompatibility.
it possible to detect incompatibility. Our second question was how long it would take for pharma-
Many factors contribute to errors in injections, such as recon- cists to conduct this new task. The run-­time of the ICU pharmacist
stitution, concentration, the administration route, filtering, was 18 min, corresponding to around 3.75% of the pharmacist’s
storage, dilution concentration, administration rate, and compat- time. Negaard et al reported the time spent by a clinical pharmacist
ibility. Information can be found for many of these factors, except on chart review (7%) and counselling on medications (2%).23 In
compatibility, in drugs’ approval paperwork. However, compati- KUHG, the time for prescription review, including non-­parenteral
bility information is more difficult to find and requires a careful drugs, was around 7%. Although previous studies have investigated
analysis of package inserts and cross-­checking of tables, data- pharmacists’ time utilisation, the labels of the tasks and the tasks
bases, and the literature. Although databases, such as Lexicomp themselves were different, making direct comparisons difficult.
or Micromedex, were used by the hospital through subscription Nevertheless, our results suggest that pharmacists could perform
contracts, simultaneous use by multiple hospital workers was this task for IV compatibility (or injection prescription review) using
not feasible, as only a limited number of IP addresses could this program without spending a lot of time.
use the service. Stabilis, a website-­based information service, is The run-­time was also different between the two wards (ICU and
not well-­known in Korea. For these reasons, nurses in Korean haematology-­oncology). Before applying the new task, it is neces-
hospitals often ask pharmacists about compatibility.8 Further- sary to allocate pharmacist resources with reference to the injections
more, compatibility errors accounted for 0.5–18.6% of errors used by the ward. More time should be allocated to pharmacists in
that could occur during the administration of injections, and in special wards such as the ICU. The pharmacists’ time in this study
some cases these may lead to death.10–12 Therefore, we decided will serve as useful reference data. In addition, these results may be
to monitor IV compatibility in this new task. helpful for hospitals that have not yet implemented a task of this
Several studies have investigated monitoring for IV compat- type. This study also has methodological implications for hospitals
ibility. Vijayakumar et al identified IV therapy-­ associated that have not yet launched injection review programs.
drug-­ related problems; among the major problems, actual
incompatibility accounted for 29%.12 Machotka et al reported CONCLUSION
that 18 (5%) of 359 IV pairs in the ICU were incompatible,13 The new task for prescription review of injections could be
while Marsilio et al reported a proportion of 14.6%.11 Our performed in a relatively short time, enabling the identification
theoretical (documented) incompatibility rate was around of IV pairs with incompatibility risk and preventing the risk of
12% (60.1/498.6) (table 3). The difference between our study simultaneous administration. However, there were still many
results and others may be because we included ‘uncertain’ cases IV pairs with unknown compatibility, and additional research
with contradictory compatibility results. Due to differences in and attention are needed to improve administration safety. This
research methods, the reported range of IV incompatibility is result will motivate hospitals that have not yet implemented
relatively broad (5.0–29.0%). Pharmaceutical interventions monitoring for IV compatibility and provide methodological
should be applied to address this potentially serious issue. Our insights regarding how to establish this new task.
new task enabled preventing IV incompatibility risk .
In this study, the information completeness ratio was 34.0– Acknowledgements The authors would like to extend their thanks to all
38.3%. Leopoldino et al researched potential drug incompati- pharmacists at Kyunghee University Hospital in Gangdong (Seoul, South Korea) for
bilities in the neonatal ICU using the Trissel’s 2 IV Compatibility using the intravenous compatibility screening solution and sharing their opinions on
Tool, and known compatibility information was found in 68.8% the solution.
of pairs.14 Kanji et al reported a proportion of 54%,15 Marsilio Contributors YJC, KNY, KHB, JBK, JTK and SHA participated in the research design.
et al described a proportion of 80.0%,11 and D’Huart et al YJC, KNY, KHB, JBK, JEY and MKL performed the experiments and collected the data.
published a rate of 37.3%.16 In other words, several previous YJC and SHA performed the data analyses and are the guarantors of the paper.
studies reported higher compatibility information ratios than Funding This study was funded by a grant from the Korea Hospital Pharmaceutical
ours. In our study, the most frequent IV pairs without compat- Education and Research Center Fund in 2019.
ibility information were acetylcysteine/tranexamic acid, cefaze- Competing interests None declared.
done/tramadol, and propacetamol/metoclopramide. Cefazedone Patient consent for publication Not applicable.
is not included in the Trissel’s 2 database. Therefore, the primary
Ethics approval Not applicable.
reason for a lack of compatibility information is the inclusion of
insufficient information on domestic injections in Korea in the Provenance and peer review Not commissioned; externally peer reviewed.
Trissel’s database. However, a few Korean studies have investi- Data availability statement Data are available upon reasonable request. All
gated compatibility between gabexate and total parenteral nutri- datasets analysed in the current study are available from the corresponding author
upon reasonable request.
tion and published a cross-­check table for the neonatal ICU or for
total parenteral nutrition in the surgical ICU.17–22 Nonetheless, ORCID iD
the compatibility information of injections is still insufficient. It Sook Hee An http://orcid.org/0000-0002-7552-0233
is essential to supplement missing data to improve patient safety.
Marsilo et al reported that 68 pharmaceutical interventions REFERENCES
1 Ministry of Health and Welfare. Comprehensive countermesurement for preventive
were conducted on 100 patients, and all interventions (except management of healthcare associated infections, 2018. Available: http://www.mohw.​
eight cases deemed ‘not applicable’) were applied.11 Our results go.kr/react/al/sal0301vw.jsp?PAR_MENU_ID=04&MENU_ID=0403&page=1&​
for the per-­patient intervention ratio were 25.3% in the ICU CONT_SEQ=345237 [Accessed 1 Dec 2021].

Cheon YJ, et al. Eur J Hosp Pharm 2023;30:101–106. doi:10.1136/ejhpharm-2021-003187 105

 Original research
2 Bertsche T, Niemann D, Mayer Y, et al. Prioritising the prevention of medication 14 Leopoldino RW, Costa HT, Costa TX, et al. Potential drug incompatibilities in the
handling errors. Pharm World Sci 2008;30:907–15. neonatal intensive care unit: a network analysis approach. BMC Pharmacol Toxicol
3 Tissot E, Cornette C, Demoly P, et al. Medication errors at the administration stage in 2018;19:83.
an intensive care unit. Intensive Care Med 1999;25:353–9. 15 Kanji S, Lam J, Johanson C, et al. Systematic review of physical and chemical
4 Westbrook JI, Rob MI, Woods A, et al. Errors in the administration of intravenous compatibility of commonly used medications administered by continuous infusion in
medications in hospital and the role of correct procedures and nurse experience. BMJ intensive care units. Crit Care Med 2010;38:1890–8.
Qual Saf 2011;20:1027–34. 16 D’Huart E, Vigneron J, Demoré B. Physical compatibility of intravenous drugs
5 Guignard B, Gschwind L, Fonzo-­Christe C. Les incompatibilités médicamenteuses en commonly used in intensive care units: an observational study and physical
2015: encore une mission du pharmacien d’établissement de santé ? Pharmactuel compatibility laboratory tests on anti-­infective drugs. Pharm Technol Hosp Pharm
2015;48:132–4 https://pharmactuel.com/index.php/pharmactuel/article/download/​ 2019;4:29–40.
1081/811?inline=1 17 Lee S. Development of a two-­dimensional compatibility table for intravenous Y-­site
6 Gin A, Walker S. Notice to hospitals regarding ceftriaxone-­calcium incompatibility: drug administration in a neonatal intensive care unit]. J Kor Soc Health-­syst Pharm
what’s a clinician to do? Can J Hosp Pharm 2009;62:157–8. 2020;37:64–73.
7 Grissinger M. Analysis of reported drug interactions: a recipe for harm to patients. Pa 18 Lee KH, Heo EJ, Kim YH. Interaction and compatibility between parenteral
administration of drugs and nutrient in the surgical intensive care unit: effects of
Patient Saf Advis 2016;13:137–48 http://patientsafety.pa.gov/ADVISORIES/Pages/​
clinical pharmacist interventions. J Kor Soc Health-­syst Pharm 2021;38:41–51.
201612_137.aspx
19 Choi JY, Lee HY, Jung EJ. Physical compatibility of a gabexate with selected
8 Woo JY, HR G, Jeong JH. Analysis of Q & A cases in drug information center of a
medications. J Kor Soc Health-­syst Pharm 2009;26:362–70 https://www.kshp.or.kr/​
Korean tertiary hospital. J Kor Soc Health-­syst Pharm 2019;36:486–97.
board/bulletin_viw.asp?nbidx=2399&SearchYear=2009&SearchNum=04&Page=1&​
9 Suh SY, Kim SH, Kim SH. Analysis of work activities and proportion of time performed
Mode=nlist
by hospital pharmacists. J Kor Soc Health-­syst Pharm 2020;37:126–50. 20 Kim SM, Lee EK, Min MS. The compatibility of fluorouracil and peripheral
10 De Giorgi I, Guignard B, Fonzo-­Christe C, et al. Evaluation of tools to prevent drug hyperalimentation admixtures. J Kor Soc Health-­syst Pharm 2006;23:291–303
incompatibilities in paediatric and neonatal intensive care units. Pharm World Sci https://www.kshp.or.kr/board/bulletin_viw.asp?nbidx=2272&SearchYear=2006&​
2010;32:520–9. SearchNum=04&Page=1&Mode=nlist
11 Marsilio NR, Silva Dda, Bueno D. Drug incompatibilities in the adult intensive care unit 21 Kim H. Interactions between medications and nutrients in patients receiving
of a university hospital. Rev Bras Ter Intensiva 2016;28:147–53. parenteral and enteral nutrition. J Clin Nutr 2013;5:57–61.
12 Vijayakumar A, Sharon EV, Teena J, et al. A clinical study on drug-­related 22 Bang ES, Sung YW, Kim SL. The physical compatibility of diltiazem and vasopressin
problems associated with intravenous drug administration. J Basic Clin Pharm with usage of 3-­in-­1 parenteral nutrition admixtures. Journal of Korean Society of
2014;5:49–53. Health System Pharmacists 2018;35:196–208 https://www.kshp.or.kr/board/bulletin_​
13 Machotka O, Manak J, Kubena A, et al. Incidence of intravenous drug incompatibilities viw.asp?nbidx=2757&SearchYear=2018&SearchNum=02&Page=1&Mode=nlist
in intensive care units. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub 23 Negaard BJ, Lyons KP, Nichol CL, et al. What does a pharmacist do? A time and
2015;159:652–6. motion study. Res Social Adm Pharm 2020;16:1314–7.

106 Cheon YJ, et al. Eur J Hosp Pharm 2023;30:101–106. doi:10.1136/ejhpharm-2021-003187