Elsevier

The Knee

Volume 10, Issue 4, December 2003, Pages 379-384
The Knee

Reduction of blood transfusion rates in unilateral total knee arthroplasty by the introduction of a simple blood transfusion protocol

https://doi.org/10.1016/S0968-0160(03)00039-5Get rights and content

Abstract

We prospectively studied blood transfusion practices within a single institution before and after the introduction of a blood transfusion protocol in consecutive patients undergoing unilateral total knee arthroplasty. Data were collected on 393 patients (group I) prior to and 295 patients (group II) after the introduction of the protocol. Following the introduction of the protocol, patients with preoperative haemoglobin of less than 11 g/dl were cross-matched prior to surgery. The criterion for postoperative transfusion was postoperative haemoglobin of less than 8.5 g/dl or a symptomatic patient with haemoglobin of greater than 8.5 g/dl. This change in practice reduced the transfusion rates from 31% in group I to 11.9% in group II. It reduced the non-utilisation of blood from 64 to 1%. There were no adverse outcomes related to the introduction of the protocol.

Introduction

Unilateral total knee arthroplasty (TKA) can result in substantial blood loss necessitating postoperative blood transfusion. Many elective surgical units use variations of the maximum surgical blood order schedule (MSBOS) [1], [2] to aid the efficiency of their blood ordering practice. However, blood ordering by this method has been shown to result in significant levels of non-utilisation of cross-matched units [3], [4], [5]. In addition, the decision to transfuse a patient postoperatively is often based on poorly defined criteria. This may, in some cases, result in inappropriate transfusion. When placed in the context of a possible health risk associated with the transmission of blood borne infective agents during allogenic blood transfusion, the reduction of unnecessary transfusion may reduce exposure of patients to potential risks.

Such a protocol depends on its ability to predict those patients at risk of transfusion. Although variants of the MSBOS [1] are commonly used, this is a basic method by which to predict transfusion requirements, as it is procedure rather than patient specific. Various patient related factors have been shown to be predictors of transfusion risk amongst surgical patients. These include total blood loss [6] type of anaesthesia [7] and the American Society of Anaesthesiologists class [7]. In total joint arthroplasty, pre operative haemoglobin has been shown to be a strong predictor of transfusion risk [6], [7], [8], [9], [10], [13], [18]. Nuttal et al. [12] showed that in patients undergoing total hip arthroplasty there was a 100% transfusion rate for patients with a pre operative haemoglobin of less than 11 g/dl.

We therefore introduced a transfusion protocol into our arthroplasty service in line with limits suggested by current Refs. [6], [7], [8], [9], [10], [12], [13], [18]. It consisted of two parts, firstly a crossmatch policy, based on the patient's pre operative haemoglobin, with the aim of ensuring that only those individuals considered at increased risk of transfusion were cross-matched preoperatively. Secondly, postoperative transfusion guidelines were adopted to ensure that patients were being transfused appropriately thereby reducing inappropriate exposure to allogenic blood transfusion amongst arthroplasty patients.

The purpose of the present study was to prospectively audit transfusion practices in patients undergoing unilateral TKA before and after the introduction of a cross matching and postoperative transfusion protocol.

Section snippets

Patients and methods

An audit nurse was employed to prospectively collect data on patients undergoing consecutive primary total knee arthroplasties between January 1995 and October 2000. The transfusion protocol was introduced in October 1998. The same 6 consultants at the same institution carried out all arthroplasties. Patients were seen 8–10 days prior to surgery in a pre assessment clinic. Their age, sex, height, weight, body mass index and American Knee Society Score were recorded (Table 1). A preoperative

Statistics

Analyses were performed using spss. Groups were compared using χ2 (with Yates correction where appropriate) for categorical data and Student t-test for continuous variables. Multiple logistic regression was used to determine the association between risk of transfusion and other variables adjusted for one another. Statistical significance was established at P<0.05.

Results

A total of 393 patients were evaluated in Group I and 295 patients in group II. The patients in these groups were comparable for age, male:female ratio, BMI, preoperative haemoglobin, and preoperative knee score as summarised in Table 1. Operative details were also similar for both groups (Table 1). Transfusion of allogenic blood was performed in 122 (31%) of the 393 patients in group I and in 35 (12%) of the 295 patients in group II. 95% confidence interval for the difference in transfusion

Discussion

The blood loss during unilateral TKA has been shown in a number of studies to be considerable. Mean haemoglobin drops of between 0.7 and 3.85 g/dl [6], [14] have been documented. As a result the extent of postoperative transfusion is not insignificant with rates of 12.5% [11] to 64% [15] being reported in the literature. Allogenic blood transfusion has potential adverse events. These range from the mild febrile allergic reactions to severe haemolytic reactions as a result of transfusion errors

References (41)

  • A. Gower et al.

    Blood utilisation in hip and knee arthroplasty: a cost minimisation study

    J R Coll Surg Edin

    (1998)
  • V.G. Hadjianastassiou et al.

    Use of the blood transfusion service in total knee replacement arthroplasty—the cost implications

    Knee

    (2002)
  • M. Roberts et al.

    A one centre prospective audit of peri- and postoperative blood loss and transfusion practice in patients undergoing hip or knee joint replacement surgery

    Ann R Coll Surg Eng

    (2000)
  • E.M. Keating et al.

    Predictors of transfusion risk in elective knee surgery

    Clin Orthop

    (1998)
  • D.A. McQueen et al.

    A comparison of epidural and non-epidural anaesthesia and analgesia in total hip or knee arthroplasty patients

    Orthopaedics

    (1992)
  • B.E. Bierbaum et al.

    An analysis of blood management in patients having a total hip or knee arthroplasty

    J Bone Joint Surg [Am] Part A

    (1999)
  • P.M. Faris et al.

    The predictive power of baseline haemoglobin for the transfusion risk in surgery patients

    Orthopaedics

    (1999)
  • G.A. Nuttal et al.

    The predictors of red cell transfusions in total hip arthroplasties

    Transfusion

    (1996)
  • A.M. Hatzidakis et al.

    Preoperative autologous donation for total joint arthroplasty

    J Bone Joint Surg Part A

    (2000)
  • B.C. Burkart et al.

    The efficacy of tourniquet release in blood conservation after total knee arthroplasty

    Clin Orthop

    (1994)
  • Cited by (26)

    • Routine pre-operative group cross-matching in total knee arthroplasty: A review of this practice in an Asian population

      2016, Knee
      Citation Excerpt :

      Our findings suggest that there is a potential to limit the use of pre-operative GXM and post-operative haemoglobin checks to those with identified risk factors. In fact, Ballantyne et al. applied a blood transfusion protocol to limit pre-operative group cross-matching to patients with pre-operative haemoglobin below 11 g/dl, and demonstrated a reduction in non-utilisation of blood from 64 to one percent [10]. The practice of measuring patient's haemoglobin level within one week of surgery allows a window for intervention should a patient be identified to have low pre-operative haemoglobin, and also aids in the selection of patients for pre-operative GXM and post-operative haemoglobin checks.

    View all citing articles on Scopus
    View full text