Pergamon
International Journal for Quality in Health Cart, Vol. 6, No. 1, pp. 41-45, 1994
Copyright © 1994 Ehevier Science Ltd
Printed in Great Britain. AH rights reserved
1353-4505/94 S7.00+0.00
Blood Usage Review in a Belgian University Hospital
ous savings in the health care sector during the
past decade are one of the main reasons that
Academisch Ziekenhuis Vrije Universiteit Brussels, justify this concern [1]. In the Academisch Ziekenhuis van de Vrije Universiteit Brussels (AZBrussels, Belgium
VUB)—a university hospital of 680 beds—a
*Blood Transfusion Centre Jette, Brussels, Belgium medical audit committee was established in
1991. The committee is charged with the promotion of quality assurance of medical care in
The appropriate use of whole blood, packed the hospital and comprises five permanent
red cells (PRC), fresh-frozen plasma (FFP) and members: a pathologist, a microbiologist, a
platelets was evaluated. The indications, effec- urologist, a nephrologist and the medical directiveness and complications of transfusion were tor. Members of other disciplines can be called
determined on the basis of existing criteria de- upon because of their expertise in specific
scribed in the literature and prevailing practices domains.
in the hospital. A retrospective review of medical
Blood usage review is identified as one of six
records demonstrated that the administration of important medical staff quality assurance funcPRC was unnecessary in 15% of the cases and tions and is considered a top priority by the
that the use of FFP was not indicated hi 67% of prominent American Joint Commission on
the patients. The result of PRC transfusion was Accreditation of Health Care Organizations
inadequate in 4% of the cases. Haemolysis and [2,3]. Hazards associated with transfusions
circulatory overload occurred hi 2% and 4% of including the transmission of diseases such
the patients, respectively. Major hazards such as as Acquired Immunodeficiency Syndrome
Acquired Immunodeficiency Syndrome (AIDS) (AIDS) and hepatitis should be minimized.
and hepatitis were not observed. In order to Recent studies showed, however, that blood
improve the appropriate use of blood com- and blood components are often used in an
ponents, transfusion guidelines were proposed inappropriate way [4-8]. The medical audit
to all medical departments.
committee agreed that assessing the appropriateness of transfusion practice should be an
integral
part of the quality assurance program in
Key words: Blood transfusion, medical audit.
every hospital and decided to initiate a study on
this scope in the AZ-VUB.
INTRODUCTION
J. SCHOTS and L. STEENSSENS*
MATERIALS AND METHODS
The method of medical audit was strictly
applied. The different steps of the audit process
Submitted 15 September 1993; accepted 2 December 1993.
Presented at the 10th ISQA Conference on Quality Assurance in Health Care, Maastricht, The Netherlands, 20-23 June
1993.
Correspondence: Dr J. Schots, Academisch Ziekenhuis Vrije Universiteit Brussels, Laarbeeklaan 101, B-1090 Brussels,
Belgium.
41
Downloaded from by guest on November 14, 2014
The importance of quality assurance has
increased over the last years and has become
apparent in hospital management as well as in
the interest of the policy-makers. The continu-
42
J. Schots and L. Steenssens
Step 1 : Selection of a topic for study
TABLE 1. Number of reviewed medical records per
medical department and per speciality of the requesting physician
Department
Step 2 : Definition of criteria
N
Internal medicine
Gastro-enterology
Haematology/oncology
Nephrology
Geriatry
Neurology
Cardiology
7
7
4
2
1
1
Surgery
Vascular surgery
Abdominal surgery
Plastic surgery
2
2
2
Step 5 : Corrective actions
Gynaecology/urology
Gynaecology/obstetrics
Urology
4
1
Step 6 : Re-evaluation
Critical care
Intensive care/emergency
Anaesthesiology
9
7
Step 3 : Analysis of medical records
I
Step 4 : Analysis of results
Recommendations
I
Imaging/physical sciences
Radiotherapy
Total
are shown in Figure 1 [9]. The purpose of the
study was defined as follows: the evaluation of
the appropriateness of the use of whole blood,
packed red cells (PRC), fresh-frozen plasma
(FFP) and platelets in the hospital. On the basis
of a review of the literature and taking into
account the prevailing medical practices in the
hospital, criteria were established focusing on
the indications of transfusion according to the
product transfused and on the occurrence of
complications. For the use of PRC, criteria
were also defined in order to evaluate the effectiveness of transfusion [10-14]. Each criterium
consists of an element and a standard of 0% or
100% (see Appendix). Transfusion experts
from the Blood Transfusion Centre, from the
Laboratory of Haematology and Haemostasis,
and from the Medical Service of Haematology
were closely involved in determining the evaluation criteria. Infants under 16 years were
excluded from the study as the established criteria only apply to adults. A random sample of
65 transfusion requests received at the Transfusion Centre for adult inpatients over a 3-week
period was taken. The medical records of these
patients were retrospectively reviewed by an
independent physician who was not involved in
1
50
any decision-making regarding patient care. In
the case of repeated transfusions during the
same period of hospitalization only the first
transfusion was included in the study. The
medical record reviewing process consisted of
searching for the elements of the preset criteria
depending on the product transfused.
Transfusion was considered appropriate if
the element was present (standard 100%) or
absent (standard 0%) in the medical record
depending on the value of the attributed standard. Poorly documented medical records lacking essential data for the evaluation of transfusion were excluded. A total of 50 medical
records were analysed. Table 1 shows the number of reviewed records per medical department
and per speciality of the requesting physician.
All transfusions considered inappropriate by
the reviewer for one or more criteria were
finally submitted to the medical audit committee and individually discussed together with
an expert from the Transfusion Centre.
RESULTS
Forty-eight patients had a transfusion of
PRC. Six of them also received FFP. PRC, FFP
Downloaded from by guest on November 14, 2014
FIGURE 1. Steps of the medical audit. Step 1:
selection of a topic for study. Step 2: definition of
criteria. Step 3: analysis of medical records. Step 4:
analysis of results; recommendations. Step 5: corrective actions. Step 6: re-evaluation.
43
Blood Usage Review
TABLE 2. Percentage of inappropriate transfusions
according to the preset criteria*
Transfusions
Total
%
number inappropriate
Indications
Packed red cells (PRC)
Platelets
Fresh-frozen plasma
48
2
6
Effectiveness of PRC transfusion
48
Complications
Haemolytic reaction
Circulatory overload
Other (allergic or febrile
reaction, hepatitis,
HTV seroconversion,...)
50
50
15
0
67
2
4
50
HTV, human immunodeficiency virus.
*See Appendix.
Indications
The use of PRC was unnecessary in seven out
of 48 patients. Four received only one unit of
PRC. These seven patients all suffered from
well-supported chronic anaemia or limited
acute blood loss and had haemoglobin (Hgb) of
more than 8 g/100 ml before transfusion.
Transfusion of platelets was always appropriate. The use of FFP was not indicated according
to criterium No. 4 (see Appendix) in four out of
six cases. Apparently, FFP is often used in cases
of major intra-operative or gastrointestinal
bleeding albeit coagulation tests (prothrombin
time and activated partial thromboplastin time)
are normal. The one patient receiving PRC,
FFP and platelets at the same time suffered
from an extreme hyper-acute blood loss because of a ruptured aneurysm of the aorta. As
whole blood was not immediately available (see
Appendix, criterium No. 2), the simultaneous
transfusion of PRC, FFP and platelets was considered appropriate by the audit committee.
Effectiveness of PRC transfusion
Transfusion of PRC proved inadequate in at
least two cases. One patient received only one
Complications
None of the transfused patients presented
an allergic or febrile reaction. Haemolysis
occurred in one case. In two other cases a
circulatory overload was observed as a result of
an unnecessary or too rapidly administered
transfusion. Finally no mention of hepatitis or
human immunodeficiency virus seroconversion
was found in the medical records within 6
months of the date of transfusion.
DISCUSSION
The analysis of the medical records turned
out to be very troublesome and took up a lot of
time. The essential data for evaluation were
often difficult tofindas the patient records were
sometimes voluminous and often poorly organized. Finally, 23% of the reviewed records were
excluded from the study as a result of missing
vital information. Improving the quality of the
medical record was recognized as a priority by
the audit committee. A new audit, focusing on
this topic, should be considered in the near
future.
Although only a relatively small number of
records, were analysed, this evaluation led to
some interesting findings with regard to transfusion practices in the hospital. Despite the
evident risks of transfusion, blood components
are overused: transfusion of PRC was revealed
as unnecessary in 15% of the cases and FFP
transfusion was judged not indicated in 67% of
cases. Comparisons with similar studies are difficult to make because of differences in local
criteria. Variations in existing practices can
partly be explained by the lack of a broad
consensus with regard to the indications of the
use of blood components.
The audit committee concluded that ongoing
efforts should be made to increase the attention
of the requesting physicians to the risks and
benefits of transfusion, especially as the AZVUB is a university hospital and plays a major
Downloaded from by guest on November 14, 2014
and platelets were simultaneously administered
to one patient. Two patients received platelets
only. Whole blood or FFP alone was never
used. The percentages of inappropriate transfusions according to the predetermined criteria
are shown in Table 2.
unit of PRC, resulting in a Hgb level of less than
10 g/100 ml post-transfusion. In another patient
Hgb was below 10 g/100 ml within 48 h before
discharge. In one case the result of PRC transfusion could not be evaluated as the patient died
within 24 h of transfusion.
44
role in the training of medical students and
residents. Unnecessary transfusions should be
prevented by a higher awareness of the risks as
well as by better decision-making with regard to
the use of blood and blood components based
on written guidelines. The appropriateness of
transfusions should be assessed periodically.
In order to fulfil these recommendations the
following actions were taken by the committee:
the results of the transfusion audit were fed
back to the heads of the medical departments
and the established criteria were proposed as
guidelines for the correct use of blood and blood
components in the hospital. Moreover, a new
audit of transfusion practices was announced
within 6 to 12 months in order to assess the
effectiveness of these actions.
In conclusion, the method of medical audit,
although troublesome andtimeconsuming, was
revealed to be a successful means of evaluating
medical practice in the hospital.
J. Schots and L. Steenssens
3.
4.
5.
6.
7.
8.
9.
13.
REFERENCES
1. Casparie A F, Quality in healthcare (in Dutch).
Medisch Contact A: 4T7,1989.
2. Joint Commission on Accreditation of Healthcare Organizations, Medical Staff Monitoring
14.
Downloaded from by guest on November 14, 2014
Acknowledgements: The authors wish to thank the
members of the audit committee: Prof. L. Tielemans, 10.
Prof. G. K16ppel, Prof. S. Lauwers and Prof. D.
Verbeelen for their support and contribution to this
study. We would also like to thank Prof. J.-P. Gassee, medical director of the Brugmann Hospital, 11.
Brussels, for his comments on an earlier draft of this
paper.
12.
and Evaluation. Blood usage review, 2nd Edn,
pp. 7,25. Joint Commission, Chicago, 1990.
Van Schoonhoven P, Quality patient care and
blood usage review. Transfusion 29:471,1989.
Mozes B, Epstein M, Ben-Bassat I, Modan B
and Hallcin H, Evaluation of the appropriateness
of blood and blood product transfusion using
preset criteria. Transfixion 29: 473,1989.
Kuriyan M, Kim Du and Haverston R, Blood
component transfusion audit: a comprehensive
microcomputer program. Qual Rev Bull 15: 347,
1989.
Coffin C, Matz K and Rich E, Algorithms for
evaluating the appropriateness of blood transfusion. Transfusion 29: 298,1989.
Thompson A, Contreras M and Knowles S,
Blood component treatment: a retrospective
audit in five major London hospitals. / Clin
Pathol 44: 734,1991.
Barnette R E, Fisch D J and Eisenstaedt R S,
Modification of fresh-frozen plasma transfusion
practices through educational intervention.
Transfusion 30: 253,1990.
Gassee J P and Smets P, Medical audit: a method
to evaluate the quality of care in the hospital.
Practical guide (in French), pp. 15-21. Belgische
Vereniging der Ziekenhuizen, Brussels, 1992.
Gassee J P and Smets P, Medical audit: a method
to evaluate the quality of care in the hospital.
Practical guide (in French), pp. 55-57. Belgische
Vereniging der Ziekenhuizen, Brussels, 1992.
National Institutes of Health, Consensus conference on perioperative red blood cell transfusion.
JAMA 260: 2700,1988.
National Institutes of Health, Consensus conference on fresh-frozen plasma. Indications and
risks. JAMA 253: 551,1985.
National Institutes of Health, Consensus conference on platelet transfusion therapy. JAMA 257:
1777,1987.
Kuriyan M, Kim Du and Haverston R, Regional
blood usage review: a quality assessment study
from New Jersey Hospitals. Qual Rev Bull 15:
369,1989.
45
Blood Usage Review
APPENDIX. Indications for blood transfusion in AZ-VUB
Criterion
No.
Indications
1.
2.
3.
4.
Standard
Elements
Packed red cells (PRC)
—acute blood loss (trauma, surgery, digestive,...) >15% of total blood volume
—Hgb s 8 g/100 ml
—symptoms of hypoxia in patients with anaemia
100*
Whole blood
—acute blood loss >25% of total blood volume
—acute bleeding and 4 units of PRC already administered
—exchange transfusion
—extracorporeal circulation
Platelets
—platelet count <20,000/^l
—platelet count <10Q,OOOJft\ and active bleeding or surgical procedure within 12
hours of transfusion
—on advice of haematologjst
100
Effectiveness of PRC-transfusion
5.
—Hgb a 10 g/100 ml post-transfusion
—Hgb a 10 g/100 ml within 72 h before discharge
Complications
6.
—allergic reactions (cutaneous rash; urticaria)
7.
—febrile reaction
8.
—haemolytic reaction (shock, oliguria, disseminated intra vascular coagulation)
9.
—circulatory overload
10.
—hepatitis within 6 months of transfusion
11.
—HIV seroconversion within 6 months of transfusion
100
100
0
0
0
0
0
0
Hgb, haemoglobin; HTV, human immunodeficiency virus; PTT, prothrombin time; APTT, activated
partial thromboplastin time.
•For criterium 1 (PRC) exceptions are: coronary disease, chronic lung disease, cerebral ischaemia,
palliative treatment in terminal patients, preoperative status (indication: Hgb <10 g/100 ml).
Downloaded from by guest on November 14, 2014
Fresh-frozen plasma (FFP)
—active bleeding (trauma, surgery, digestive,...) and PTT <60% or APTT
>55-60 sec (not due to heparin or circulating anticoagulants)
—active bleeding or invasive procedure in patients with congenital deficiencies of
Factors n , V, VH, EX, X, XI and XD
—active bleeding or surgical procedure required in less than 6 h in patients treated
with coumarin
—on advice of haematologist
100