Archives of Orthopaedic and Trauma Surgery

, Volume 131, Issue 5, pp 663–667

Maximum surgical blood ordering schedules for revision lower limb arthroplasty

Authors

    • University Hospitals of Leicester, Leicester Royal Infirmary
  • Christopher Challand
    • Plymouth Hospitals NHS TrustDerriford Hospital
  • Andrew Clarke
    • Plymouth Hospitals NHS TrustDerriford Hospital
  • Jonathan Keenan
    • Plymouth Hospitals NHS TrustDerriford Hospital
Orthopaedic Surgery

DOI: 10.1007/s00402-010-1204-2

Cite this article as:
Mahadevan, D., Challand, C., Clarke, A. et al. Arch Orthop Trauma Surg (2011) 131: 663. doi:10.1007/s00402-010-1204-2

Abstract

Background

Effective utilisation of blood products is fundamental. The introduction of maximum surgical blood ordering schedules (MSBOS) for operations has been shown to improve transfusion services. A retrospective analysis was undertaken to establish an evidence-based MSBOS for revision total hip replacement (THR) and total knee revision (TKR). The impact of this schedule on blood conservation was analysed.

Methods

A retrospective analysis was undertaken on 397 patients who underwent revision THR and TKR over a 4-year period. The cross-match-to-transfusion ratio (CTR) and transfusion index (TI) were calculated. A MSBOS protocol was created based on the TIs and its’ impact on transfusion services was assessed prospectively on 125 patients by comparing CTRs.

Results

In revision THR, TI was 1.19 for elective cases, 1.55 for emergency cases and 2.35 for infected cases. There was no difference in TI for revisions of cemented and uncemented components. Single component THR revision required less transfusion. In revision TKR, TI was 0.31 for elective cases, 2.0 for emergency cases and 1.23 for cases with infection. The introduction of the MSBOS protocol had resulted in a considerable improvement in blood ordering. Reductions in the CTR were seen for all types of revision surgery, but most evident in elective revision THR (3.24–2.18) and elective revision TKR (7.95–1.2).

Conclusions

Analysis confirmed that excessive cross-matching occurred for revision lower limb arthroplasty. The introduction of our MSBOS protocol promoted blood conservation and compliance with established national guidelines.

Keywords

Revision arthroplastyHipKneeTransfusionMSBOS

Introduction

Blood has become a scarce and expensive resource. Revision total hip replacement (THR) and to a lesser degree, revision total knee replacement (TKR), can be associated with substantial blood loss and the subsequent need for transfusion [1, 2]. Increased public and medical concerns regarding transfusion-associated infections and the costs involved in blood transfusion mean that transfusions must be justified [3, 4]. More stringent eligibility criteria have also caused a reduction in the size of the donor population [5].

Increasing emphasis has been placed on effective blood conservation encompassing pre-, intra- and post-operative measures. Previous studies have demonstrated that an evidence-based approach to blood ordering has reduced the level of unnecessary cross-matching for elective orthopaedic surgery [6, 7]. The British Orthopaedic Association has recommended the use of blood ordering schedules at individual departments to facilitate an effective blood conservation programme [8].

The introduction of blood-ordering guidelines, such as the maximum surgical blood ordering schedule (MSBOS) [9] for specific operations provides standardised guidelines for effective cross-matching and can provide substantial financial savings [6, 10]. Transfusion departments use a simple formula to identify which procedures require cross-matched blood called the cross-match-to-transfusion ratio (CTR). The gold standard for the CTR is 2:1 or less [8.9]. TI establishes the likelihood of blood being given for a certain procedure by dividing the number of units transfused by the number of patients having the procedure. If TI is <0.5, then cross-matching blood is considered unnecessary [9].

The purpose of our study was to create an evidence-based MSBOS for revision THR and revision TKR and assess its impact on transfusion services. To our knowledge standardised evidence-based MSBOS for revision THR and TKR do not exist in the literature.

Materials and methods

We retrospectively analysed 397 patients who underwent revision THR and TKR over a 4-year period (290 hips, 107 knees). The blood bank database was accessed to ascertain the quantity of blood requested and actually transfused in this group of patients. The relevant CTR and TI were calculated. Patient demographics were recorded.

Revision THR and TKR were analysed separately as elective cases, emergency cases and cases for proven (or presumed) infection. Emergency cases encompassed periprosthetic fractures and expedited cases with significant symptoms from dislocations, loosening or impending fractures. For revision THR, separate analyses were performed comparing cemented and uncemented prostheses, and based on the component(s) replaced. Tourniquet use for revision TKR was variable depending on the surgeon’s preference.

The results of our initial audit were used to create a MSBOS protocol for revision THR and revision TKR. We then prospectively audited a further 125 consecutive patients undergoing revision THR and TKR to assess the impact of the MSBOS on our blood-ordering practice.

The departmental transfusion trigger throughout the study period was unchanged at a haemoglobin concentration of 7.5 g/dl. Patients did not receive other blood-conserving adjuncts during their care.

Statistical methods

Continuous data were analysed using the Student’s t test and categorical data by the Chi-square and Mann–Whitney U test (SPSS software SPSS Inc., Illinois 60606). p < 0.05 was considered to be significant.

Results

The demographics of patients analysed in the preliminary audit is described in Table 1.
Table 1

Demographics of patients prior to introduction of MSBOS for revision THR and TKR

 

Revision THR

Revision TKR

Numbers operated

290

107

Age (years)

70 (34–95)

71 (40–90)

Male

110 (38%)

46 (43%)

Female

180 (62%)

61 (57%)

Elective revision

147

64

Emergency revision

47

4

Revision for infection

96

39

Number of units ordered

1,025

262

Number of units transfused

438

73

Revision total hip replacement

The calculated TI was highest in infected cases (TI = 2.35), followed by emergency (TI = 1.55) and elective revisions (TI = 1.19). Results were similar between revisions involving cemented (TI = 1.12) and uncemented implants (TI = 1.59) (p = 0.961). TI for revisions of the acetabular component (TI = 0.88) was significantly lower than revisions of both components (TI = 1.71) (p = 0.011). The data also demonstrated a significant difference in TI between first stage (TI = 2.73) and second stage (TI = 2.08) procedures of a 2-stage THR revision for infection (p < 0.001).

Revision total knee replacement

Blood transfusions were more frequent in patients undergoing emergency surgery (TI = 2.0) and revision surgery for infection (TI = 1.23). The likelihood of requiring blood transfusion after elective revision surgery was small (TI = 0.31). There was a no significant difference in TI between stage 1 (TI = 1.67) and stage 2 (TI = 1.25) procedures for infected prostheses (p = 0.729).

Maximum surgical blood ordering schedule protocol

Based on the TIs obtained from our analysis, a blood-ordering schedule for revision THR and TKR was devised (Table 2). The blood-ordering tariff was estimated by rounding-up TI to the nearest whole number. A group and save was proposed for procedures with a TI < 0.5. The schedule was agreed through consultation between clinicians, blood bank and the hospital transfusion committee.
Table 2

Maximum surgical blood ordering schedule (MSBOS) for revision THR and TKR

Blood ordering tariff

  

Revision THR

1 component

Cross-match 1 unit

2 components

Cross-match 2 units

Emergency revision

Cross-match 2 units

Revision for infection

Cross-match 3 units

Revision TKR

Elective revision

Group and save

Emergency revision

Cross-match 2 units

Revision for infection

Cross-match 2 units

Blood-ordering post MSBOS for revision THR and TKR

Table 3 shows the demographics of patients analysed prospectively following the introduction of the MSBOS protocol.
Table 3

Demographics of patients following the introduction of MSBOS for revision THR and TKR

 

Revision THR

Revision TKR

Numbers operated

83

42

Age (years)

69 (37–95)

69 (45–87)

Male

26 (40%)

18 (54%)

Female

39 (60%)

16 (46%)

Elective revision

55

30

Emergency revision

10

4

Revision for infection

18

8

Number of units ordered

147

10

Number of units transfused

88

8

Prior to MSBOS, we found that cross matching was excessive for revision TKR and to a lesser extent revision THR with CTRs of 3.59 and 2.34, respectively. CTR was highest for elective revision TKR where approximately one unit of blood was transfused for every eight units cross-matched (CTR = 7.95). The CTR for elective revision THR was also high at 3.24. CTRs for the other groups of revisions however, were satisfactory as they fell within the gold-standard ratio of 2:1.

Following the introduction of the MSBOS, CTRs dropped for all groups of revision surgery. The overall CTR for revision THR and TKR had dropped to 1.67 and 1.25, respectively. The reductions in CTRs were most evident for elective revision TKR and elective revision THR with 85 and 33% reductions, respectively (Figs. 1, 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs00402-010-1204-2/MediaObjects/402_2010_1204_Fig1_HTML.gif
Fig. 1

Cross-match to transfusion ratios for revision THR pre- and post MSBOS

https://static-content.springer.com/image/art%3A10.1007%2Fs00402-010-1204-2/MediaObjects/402_2010_1204_Fig2_HTML.gif
Fig. 2

Cross-match to transfusion ratios for revision TKR pre- and post MSBOS

Post-operative transfusion triggers were similar for revision THR before and after the introduction of the MSBOS protocol at haemoglobin concentration of 7.5 and 7.8 g/dL, respectively. For revision TKR, the transfusion triggers were Hb 7.4 and 7.2 g/dL during this interval.

Discussion

Concerns exist regarding the availability of blood products and the cost to produce the safest possible product. Consequently, various peri-operative blood conservation strategies including autologous predonation, erythropoietin therapy, antifibrinolytic medications, cell salvage and autotransfusion are being implemented with increasing frequency requiring considerable investment and expertise. However, there are many departments around the world where only allogenic blood is available for transfusion after surgery. It is therefore important to be efficient with the use of this limited resource. Our initial audit had demonstrated that excessive amount of blood was being ordered for revision surgery that could have been used for more needy cases.

The results suggest that routine cross matching was unnecessary for elective revision TKR and a group and save may be sufficient. Patients undergoing this procedure should have their blood group established and their serum checked for antibodies. This “group and save” provision allows rapid blood delivery in an emergency. The extent to which surgery can be covered by this provision largely depends on practical, local issues such as the distance between operating theatre and transfusion department.

In revision THR, one unit of cross-matched blood is sufficient for single component revision. Our study supports the findings of previous studies which demonstrated that revision of both THR components (femoral and acetabular) is associated with increased blood loss [11, 12]. The use of cement has been shown to influence blood loss in revision hip surgery. Semkiw et al. [13] demonstrated that higher levels of postoperative bleeding occurred in patients in whom uncemented components were used in revision hip arthroplasty. The addition of bone cement reduced the raw, bleeding bone surface area and hence blood loss [13]. In revision arthroplasty, the removal of cement and all foreign material can be extremely demanding, time-consuming, and damaging to the remaining host bone [14]. We postulated that removal of well-fixed cement from bone surfaces at the time of revision surgery was likely to cause more bleeding compared to removal of uncemented components. However, our results demonstrated no significant difference in the TIs between revisions of cemented and uncemented THR.

Arthroplasty revisions for infection were undertaken as two-staged procedures. The first stage revision for infected THR had a significantly higher TI compared to the second stage. Increased operative time and more extensive surgical dissection and debridement during first stage revision surgery were likely to have increased the intra-operative blood loss. There was no statistically significant difference in the TIs for first and second stage revisions of TKR.

We also observed an increased need for blood transfusion after emergency revision arthroplasty. Fracture site bleeding in periprosthetic fractures may lower the preoperative haemoglobin concentration, therefore increasing the need for transfusion [15]. Emergencies must be dealt with prompt and consequent preoperative physiological condition of the patient may be sub-optimal. Physiological variables such as hypertension can precipitate increased operative blood loss and hence the need for transfusion [16]. Furthermore, periprosthetic fractures may precipitate more postoperative pain due to the larger zone of injury, which can influence blood loss postoperatively [17].

The evidence-based MSBOS that was created had made our transfusion service more efficient and matched established CTR gold-standards [8, 9]. The protocol can also provide substantial financial savings. The cost of a unit of blood in our department was £139. Retrospective use of the MSBOS protocol over the 4-year study period could have reduced the number of cross-matched blood by 580 units, thus potentially saving the department over £80,000.

Compliance with the new guidelines will have a significant impact on the blood transfusion service which in turn will impact upon the blood supply available to all users. Although MSBOS has improved the efficiency of blood ordering, it does not account for individual differences in transfusion requirements between different patients undergoing the same surgical procedure. The schedule has to be extended to take into account individual patient risk factors. A low preoperative haemoglobin concentration, low weight, patients over the age of 65 years and female patients are examples of factors that have been shown to predict the need for transfusion and should be considered [1822]. Furthermore, the transfusion needs of patients with significant co-morbidities are also different to the general population. Patients with ischaemic heart disease, for example, may benefit from having higher transfusion thresholds as they are at risk of anaemia-associated adverse postoperative outcomes [23]. Further adjuncts must also be considered, including pre-operative optimisation, and intra- and post-operative blood conservation methods.

We acknowledge that there are limitations to the present study. The data were collected retrospectively on consecutive patients and the study was not powered to estimate the size of the sample. The analysis did not take into account the degree of complexity of the revision surgery such as the use of endoprostheses or supplemental fixation that may influence the amount of transfusion requirements. Also, the use of tourniquets in revision TKR was not analysed in detail. However, we did not feel that this had a significant influence on transfusion rates. A recent meta-analysis has shown no significant difference in total blood loss or transfusion rates in TKRs undertaken with or without the use of tourniquets [24]. The analysis demonstrated that intra-operative blood loss was higher when tourniquets were not used, but this did influence total blood loss or transfusion levels.

This study has demonstrated the importance of auditing clinical practice in order to improve service which has both clinical and financial implications. The use of our MSBOS for revision THR and TKR is an evidence-based approach to blood ordering and promotes blood conservation.

Copyright information

© Springer-Verlag 2010