Blood transfusion practice: a nationwide survey in Italy

Abstract

Background

Blood transfusion is a complex activity. Some of the components of this activity are implementation of standard procedures, evaluation of appropriateness of the blood use, methods for bedside identification of patients and the release of blood in emergencies, during out-of-routine hours and from hospitals lacking a Blood Centre. An overview about how these issues are managed in Italy could be of interest.

Materials and methods

A survey dealing with some issues regarding blood release was performed, using a questionnaire sent to 278 Italian Blood Centres.

Results

Out of the 278 Centres, 179 (64%) returned the questionnaire. The geographic distribution of the Centres that responded (80 in the north, 46 in the centre and 53 in the south of Italy) offers a good picture of the Italian reality. Globally there seems to be a relatively uniform application of procedures and guidelines, of methods to identify patients, and of evaluating transfusion appropriateness. The systems used to deal with emergency blood release and blood release in non-routine conditions are more variable. The use of technological resources seems to be disappointingly low.

Discussion

Although many aspects of the blood transfusion process should be improved, the picture that emerged from the survey seems to show, on the whole, a fair quality of blood transfusion practice in Italian Blood Centres.

Introduction

Blood transfusion is a high-volume activity in healthcare. Safe, effective and appropriate transfusion practice is mandatory for a good quality of care to patients who need transfusion as part of their treatment. In Italy, Blood Centres are widely spread throughout the country and are hospital-based. Many of them collect blood directly from donors. Furthermore, one of the main reasons for such a capillary presence is the legally imposed task^1,2 of assessing transfusion appropriateness locally. However, taking into account that the validity of many aspects of blood transfusion has not been rigorously proven by clinical trials, such a high number of often small-sized structures could lead to a risk of excessive variation in local practices. A widespread use of evidence-based standard operating procedures, a starting point in transfusion practice, could be a first indicator of good quality activities.

The evaluation of the appropriateness of blood transfusion has gained a pivotal role in modern Transfusion Medicine.

In 2007, the Italian Society of Transfusion Medicine and Immunohaematology (SIMTI) edited the Standards of Transfusion Medicine³: in chapter D the approval of shared guidelines for optimal use of blood components is strongly recommended. Different approaches are described to verify that the “right” blood component is given for the “right” indication⁴. The knowledge of how Italian Blood Centres deal with this issue is important, given that it is one of the main reason for the widespread presence of these Centres.

For many years, the transfusion of incorrect blood components, and especially the administration to a patient of a blood component intended for a different one, has been the main hazard of transfusion therapy⁵. Proper identification of patients could prevent the great majority of these adverse events; nowadays, technology can provide systems to reduce the risk of mistakes in the identification of patients and ultimately of transfusion errors^6–8. It is, therefore, important to know which instruments are used in Italy to ensure that blood components are given to the “right” patient, also taking into account that these systems seem to require good staff training to avoid technical and operative blocks⁸. Such systems could also help to reduce mistakes in blood delivery in urgent cases and in out-of-hour transfusions, which are more frequently associated with transfusion errors⁹; in these situations, however, good practice planning is of paramount importance.

Technology could also be of help in solving some problems related to blood release in hospitals lacking a Blood Centre: in this condition full control of the transfusion process can create some difficulties, which it may be possible to overcome by the use of remote-controlled refrigerators^10,11.

In order to obtain data about these issues, SIMTI carried out a survey in Italian Blood Centres. The results of this survey were presented in part at the National Conference of Blood Services, held in Terrasini in May 2009.

Materials and methods

A questionnaire was sent to every Italian Blood Transfusion Service with the aim of collecting data about five broad subjects:

1. the spread of standard operating procedures.

2. The actions and instruments used to evaluate transfusion appropriateness.

3. The blood recipient's identification and the rate of use of technology to reduce transfusion errors

4. The ways to approach the problem of blood release to patients in hospitals lacking a Blood Centre.

5. The management of urgent cases and of out-of-hours transfusions.

The questionnaire consisted of 23 multiple choice questions, some of which had subquestions. Furthermore, information on the number of blood units collected by each responder structure was collected, as a indirect measure of the Centre's dimension.

Statistical analysis of data was performed by SPSS Statistics 17,0: analysis of variances (ANOVA) and Pearson's chi-square test were used for group comparisons for continuous and categorical variables, respectively.

Results

Out of the 278 Blood Centres which received the questionnaire, 179 (64.4%) returned it: 80 of these centres were in the North of Italy, 46 in the Centre and 53 in the South of Italy. Two of the Centres did not release blood products and their data were excluded from the analysis.

Altogether these Services collect about 1,900,000 whole blood units/year, that is about 70% of the number of units collected annually in Italy. The mean number of units collected at each Centre, an indirect measure of the structure's dimension, was 10,984 (range, 476–58,604), with differences between geographic areas (mean number of units in the North 14,502; in the Centre 9,342 and in the South 6,826: p=0.000).

Existence of standard operating procedures

The results concerning standard operating procedures are summarised in Table I. Nearly all Centres reported having documented procedures regarding activities related to blood release. Procedures for evaluating transfusion appropriateness were less commonly reported, being used in 77% of Centres.

Table I.

Existence of standard operating procedures.

Question	Yes (n.-%)
In your Centre is there a working documented procedure regarding:
Patient identification	164-93%
Evaluation of request forms for appropriateness	165-93%
Blood release	171-97%
Transfusions in emergency	174-98%
Evaluation of appropriateness in blood component transfusion	137-77%
Evaluation of appropriateness in plasma derivatives administration	108-61%

Still less common were procedures for evaluating the appropriateness of administration of plasma derivatives, which is an understandable result given that plasma derivatives were distributed directly by only 25% of the Blood Centres, although with significant differences between geographic areas (South 13.5%, Centre 17.4%, North 38%: p=0.002): Centres that managed plasma derivatives directly more commonly had a procedure available than did Centres that did not manage plasma derivatives directly (75.6% versus 56.1%, p=0.021).

Evaluation of appropriateness

Table II shows a summary of the results concerning the appropriateness of use of blood components. The large majority of Centres followed local guidelines on the appropriate use of blood components (90%), which were approved by hospital transfusion committees in 80% of cases; regional guidelines were used less commonly (36%). Seventy-seven percent followed guidelines on the appropriate use of plasma derivatives, notably albumin: despite this, guidelines were used to evaluate the appropriateness of requests in only 49% of Centres, some of them (13%) only for unstated indications. Requests for blood components were evaluated prospectively in 67% of cases and retrospectively in 38% of cases, with 17% of Centres performing both; a prospective evaluation of requests for plasma derivatives was performed in 46% of Centres.

Table II.

Evaluation of appropriateness.

Question	Yes (n.-%)
Do you use guidelines for appropriate use of blood component?
Local	159-90%
Hospital Transfusion Committee-approved	141-80%
Regional	63-36%
Do you use guidelines for appropriate use of plasma derivatives?	136-77%
Albumin	83-47%
Intravenous immunoglobulins	36-20%
Antithrombin	11-6%
Prothrombin complex concentrates	9-5%
All the above	50-28%
Do you directly evaluate the appropriateness of requests for plasma derivatives?	64-36%
Only for unstated indications	23-13%
Is the evaluation of the appropriateness of blood components performed?
Retrospectively	67-38%
Prospectively	118-67%
Is the evaluation of the appropriateness of requests for plasma derivatives performed?
Retrospectively	56-32%
Prospectively	82-46%
With regards to your request form:
Is it the same for all blood components?	88-50%
Is it distinct for each blood component?	88-50%
Does it contain a list of stated indications?	70-39%
Does it allow reporting unstated indications?	91-51%
Do you ask for patient's parameters to be reported on the request form? Which ones?	169-96%
Patient's blood group	157-89%
Haemoglobin	150-85%
Platelets	152-86%
International Normalised Ratio	161-91%
Activated partial thromboplastin time	130-73%
Body weight	94-53%
Is fresh frozen plasma expected to be requested by volume?	83-47%
Do you perform periodic statistical analyses of appropriateness of blood use?	90-51%
Quarterly	22-12%
Half-yearly	28-16%
Yearly	49-28%
Are the results of the analysis discussed in the Hospital Transfusion Committee?	115-65%

Half of the Centres utilised a single request form for all blood components, whereas the other half used distinct forms for each type of blood component; the use of distinct forms was more frequent in the North and Centre of Italy (57% and 59%, respectively) than in the South (31%, p=0.005). In more detail, distinct forms seemed to be used more frequently in big centres, given that the mean number of units collected in centres utilising multiple request sheets was higher than that in centres in which a single form was used (12,459 versus 9,509, p=0.038). Indications were reported more frequently when distinct forms were used rather than single forms (50% versus 30%, p=0.012). Fifty-one percent of centres allowed requests for unstated indications, with no difference between whether distinct or single forms were used (54% versus 48%, p=0.407). Substantially all centres asked for the patient's clinical and laboratory parameters to be indicated on the request form; in about half of them fresh-frozen plasma was expected to be requested by volume, rather than by number of units. The results of the appropriateness evaluation are periodically (most often annually) analysed in about half of centres, with geographical variations (North 53%, Centre 63%, South 37%, p=0.028); the results of the analysis are frequently (65%) debated in the hospital transfusion committees, with focused retrospective audits (51%).

Identification of patients

The results concerning the identification of patients are summarised in Table III. Nearly all centres use active identification of patients; technological instruments for bedside checking (wristbands, radio-frequency or biometric devices) are still not used widely.

Table III.

Patient identification.

Question	Yes (n. - %)
Is patient identification performed by:
Active identification	176-99%
Bar-coded wristbands	23-13%
Radio-frequency technology	5-3%
Biometric identification systems	2-1%

Blood transfusion in external hospitals

As shown in Table IV, procedures for blood release in hospitals located far from a Blood Centre did not differ from those used for in-hospital wards. In nearly half of these cases, hospitals are provided with blood refrigerators managed by the personnel, opportunely trained, of other services; automatic blood refrigerators were very rarely used.

Table IV.

Blood transfusion in hospitals lacking a Blood Centre.

Question	Yes (n. - %)
Do procedures regarding blood release differ from those used in in-hospital wards?	22-12%
Are there blood refrigerators managed by personnel of other services (e.g. laboratory staff)?	78-44%
Are there automatic blood refrigerators managed by remote control by a Blood Service?	8-4%

Blood transfusion out of routine hours or in emergency

As shown in Table V, out-of-hours release of blood was more frequently managed by a contactable technician and physician (46%); in 12% of cases only physician was involved. Electronic remote control by physician was used in 11% of the Centres. Red cells for emergency were provided directly close to the wards in 77% of cases, and were almost always group O - RhD negative. In emergency O - RhD negative red cells were used, particularly in patients with an unknown blood group.

Table V.

Transfusion out of routine hours or in an emergency.

Question	Yes (n. - %)
During the night and holidays, is blood released by:
an on-call technician and contactable physician	22-12%
a contactable technician and physician	81-46%
an on-call technician and physician	41-23%
a contactable physician	22-12%
an on-call or contactable technician and electronic data transmission to a physician	19-11%
Do you provide red cells for emergency in blood refrigerators located close to the wards?	136-77%
O - RhD negative	134-76%
O - RhD positive	43-24%
In which emergency cases do you release exclusively O - RhD red cells?
Always	61-34%
For women of childbearing age	8-4%
For patients with unknown blood group	110-62%

Discussion and conclusions

The first significant point to mention is the wide participation of Italian Blood Services in this survey, since analogous initiatives in the recent past did not meet with the same co-operation¹².

With regards to the findings of the survey itself, more than 90% of Blood Services use formally approved procedures for the identification of patients, evaluation of blood request forms and blood release, this representing a good quality indicator.

Reporting indications for transfusion on the request form could make appropriateness evaluation easier and could act as a reminder for clinicians¹³. There was a relevant variability in request forms between Blood Centres: some used forms containing a list of stated indications, allowing requests for unstated conditions or not, some used a single form for all blood components, while others used different forms for different components. The low use of technological tools for bedside checks of patients was quite disappointing and probably due to difficulties in application and to the costs of available instruments.

In accordance with SIMTI statements, shared guidelines were widely spread and usually locally approved; regional guidelines were less common. Plasma derivatives were managed directly in only 25% of the Centres: despite this, guidelines and evaluation of appropriateness for plasma derivatives were frequently used.

Audit approval, that is the prospective evaluation of each transfusion request, should in theory be the most effective method to monitor transfusion appropriateness, as the intervention takes place before the transfusion occurs: however it is very labour-intensive and, in previous studies, did not appear to be more effective than other interventions at reducing inappropriate transfusions^13,14. Anyway, prospective evaluation of transfusion requests was the intervention most frequently used in appropriateness monitoring. Retrospective audits were also used, with discussion of the results in the Hospital Transfusion Committee in nearly two-thirds of the Centres. A further evaluation regarding the use of an electronic blood request (transfusion order-entry), which was proven to be useful in reducing inappropriate transfusions¹⁵, could be of interest.

Many of the Blood Centres (166 over 179 responding structures, 93%) released blood for patients treated in hospitals far from the Centre. Recently there has been much interest in the use of remote-controlled, “self-service” blood refrigerators. On the basis of the data of the survey, in the great majority of cases the procedures adopted for blood release in hospitals far from the Centre were not differ from those utilised for in-hospital wards. Blood was often located in refrigerators at the external hospital and was managed by personnel from other services, trained by Blood Centre staff. “Self-service” blood refrigerators were used very rarely: again, this could be due to the not always easily manageable available devices and to their cost.

The rapid provision of red cell units to patients needing blood urgently is a major issue in Transfusion Medicine. Good planning of the practices to adopt for blood release in emergencies or in out-of-hours is important, as transfusions given outside core hours are more likely to be associated with clinical errors¹⁶. It is common practice to keep stored red cells for an emergency close to the wards; almost always these units are O RhD-negative units and, less frequently, O RhD positive. The survey showed that O RhD negative red cells were used for every patient in one-third of Blood Centres, while in two-thirds they were intended solely for patients with an unknown blood group. Reserving O RhD negative units for women of childbearing age was an uncommon practice.

Out-of-hours blood requests were managed in a variable manner, with a complete range of “on-call” or contactable technicians or physicians. An interesting and innovative practice is the remote control of blood release performed by the physician by means of electronic data transmission, which was used in a minority of cases, but could represent a cost-effective solution. In more than 10% of cases, out-of-hours requests were managed by the physician alone, an apparently opposite trend to the supervisory or consultative role of the blood bank physician¹⁷.

In conclusion, the picture that emerged from the survey seems to show the quality of blood transfusion practice in Italy is, on the whole, good. Much has to be done, however, to reduce the great nationwide variability in some aspects of the activity of Blood Centres, and still more to reach a satisfying use of technology, in particular for aspects concerning the identification of patients.