Autologous transfusion of drain contents in elective primary knee arthroplasty: its value and relevance

Abstract

Background

Total knee arthroplasty is associated with significant post-operative blood loss often necessitating blood transfusions. Blood transfusions may be associated with transfusion reactions and may transmit human immunodeficiency virus, hepatitis C virus and hepatitis B virus, with devastating consequences. After total knee arthroplasty, transfusion of the contents of an autologous drain is becoming common practice. The aim of our study was to look at the effectiveness of these drains in elective primary total knee arthroplasty.

Materials and methods

A prospective study was conducted including 70 non-randomised patients. A normal suction drain was used in 35 patients (group A), whereas in the other 35 patients, a CellTrans^™ drain was used (group B). All the operations were performed by four surgeons using a tourniquet with a medial parapatellar approach. Pre- and post-operative haemoglobin concentrations were recorded in both groups. A Student’s t-test was applied to determine the statistical significance of the data collected.

Results

The average fall in post-operative haemoglobin was 3.66 g/dL (SD 1.46; range, 0.6–7.0) among patients in whom the simple drain was used (group A) and 2.29 g/dL (SD 0.92; range, 0.6–5.9) among those in whom the CellTrans^™ drain was used (group B) (p<0.0001). Twenty-five units of allogeneic blood were required in group A compared to four units in group B. The rate of transfusion was 5.7% (2 patients) in the group in which CellTrans™ drain was used and 25.7% (9 patients) in the group in which a simple suction drain was used.

Discussion

Total knee arthroplasty is associated with significant post-operative blood loss despite best operative technique. Autologous reinfusion of the contents of a CellTrans^™ drain significantly reduces the rate of post-operative blood transfusion. This study indicates that the use of an autologous drain could be recommended as routine practice in primary total knee arthroplasty.

Introduction

Total knee arthroplasty is performed routinely with gratifying results. It is associated with post-operative blood loss which may be as great as 1.5 litres, making blood transfusion inevitable¹. Post-operative blood transfusions are not without risk and are associated with adverse effects in around 20% of patients and severe reactions in 0.5% of cases². Adverse reactions may range from febrile reactions, to allergic reactions, to transmission of life-threatening viral diseases.

Traditionally, simple suction drains have been used in order to avoid the development of post-operative haematomas and the collected blood is subsequently discarded. Recently, new-generation drains have been used which allow the reinfusion of collected blood in order to reduce the need of allogeneic blood transfusions and their potential associated adverse reactions. Our study aimed to evaluate the efficacy of the CellTrans^™ drain in elective primary total knee arthroplasty.

Material and Methods

This was a prospective study which included 70 non-randomised patients undergoing elective primary total knee arthroplasty between February 2007 and August 2007. There were 46 females and 34 males (male-to-female ratio 1:1.35). Patients with bleeding disorders and those undergoing revision surgery were excluded from the study.

Patients were divided into two groups according to the type of drain used. In group A (n=35), a simple suction drain was used and the blood collected was discarded. In group B (n=35), the CellTrans^™ autologous infusion drain was used and blood collected for up to 6 hours after surgery was re-infused back into the patients. All operations were performed by the same four senior consultants using a tourniquet with a medial parapatellar approach. The tourniquet was released after closure of the wound and a pressure dressing was applied in all patients. Pre-operative and post-operative haemoglobin concentrations were recorded for all the patients. The post-operative measurement of haemoglobin concentration was done on the first day after the surgical intervention. The amount of drainage was recorded and documented for all patients in both groups. In group A, the drain was removed 24–48 hours after surgery. In group B, blood collected within 6 hours of surgery was re-infused back into the patients. The CellTrans^™ drain was provided with two transfusion bags (600 mL) connected to the drain enabling the collection of up to 1,200 mL of blood. Once both transfusion bags had been used, the device functioned as a conventional low vacuum wound drainage system and was removed after 24–48 hours. Patients received a blood transfusion if the post-operative haemoglobin concentration was less than 8 g/dL, according to the standard hospital transfusion policy.

After the collection of data, a Student's t test was conducted to assess the statistical difference between the two groups.

Results

The average decrease in haemoglobin was 3.66 g/dL in group A (SD 1.47; range, 0.6–7.0 g/dL; confidence 0.48 g/dL) and 2.30 g/dL in group B (SD 0.92; range, 0.6–5.9 g/dL, confidence 0.30 g/dL) (Table I).

Table I.

Comparison of results between patients in whom the simple drain or a CellTrans^™ drain was used

	Simple drain (n=35)	CellTrans™ drain (n=35)	p-value
Mean total drain output (mL)	416.2 (60–950)	517.4 (150–1110)	0.06
Mean pre-op haemoglobin (g/dL)	13.19 (10.2–15.4)	13.33 (10.6–15.2)	0.63
Mean post-op haemoglobin (g/dL)	9.52 (6.8–13.2)	11.03 (7.6–13.6)	0.0004
Average n. of units of allogenic blood transfused	0.71/patient	0.11/patient	-
Number of patients requiring allogenic transfusion	9/35	2/35	-
Adverse transfusion reaction	2	-	-
Wound site infection	-	-	-

Nine patients in group A (25.7%) needed a postoperative blood transfusion because of low haemoglobin concentration, requiring a total of 25 units of blood (0.71 units per patient). In contrast, only two patients (5.7%) in group B needed an allogeneic blood transfusion, requiring four units (0.11 units per patient).

The average quantity of fluid collected in the drains in group A patients was 416 mL (SD 226; range, 60–950 mL; confidence 75 mL) and 518 mL in group B (SD 231; range, 150–1,110 mL; confidence 76.5 mL). A statistical analysis conducted to assess the significance of the difference in the post-operative haemoglobin decrease in the two groups showed that the difference was statistically significant (p<0.001).

Two patients in group A suffered a very mild transfusion reaction while receiving their blood transfusions. The transfusions were stopped and restarted later on the advice of a haematologist and were, thereafter, completed uneventfully. The results of our study indicate that transfusion of autologous blood collected in an appropriate drain significantly reduces the need for post-operative allogeneic blood transfusions in patients undergoing total knee arthroplasty.

Discussion

Orthopaedic hip and knee replacement surgery is associated with post-operative blood loss³. Orthopaedic patients consume 10% of UK blood and 40% of this blood is utilised in elective total knee and hip arthroplasties³. Adverse reactions related to allogeneic blood transfusions and the increasing burden of requests on blood banks for blood products have forced clinicians to think of alternative techniques of blood salvage. The clinical advantages of re-infusing autologous blood rather than transfusing allogeneic blood have been clearly documented in the literature^3,4. Primarily, autologous blood reduces the need for allogeneic blood transfusion; however, it also prevents the transmission of viral diseases (hepatitis C virus, hepatitis B virus, human immunodeficiency virus, Creutzfeldt-Jacob virus), transfusion reactions and transfusion errors^4,5. Furthermore, it reduces the burden on blood banks of providing increasingly expensive blood supplies⁶. The role of blood salvage will become increasingly important as the number of active donors may fall by 50% with the introduction of screening for Creutzfeldt-Jacob disease⁷. Recently autologous drains have been used in total knee arthroplasty in order to address this issue. Alternative techniques such as pre-operative blood donation and intra-operative cell salvage have also been used but are quite cumbersome for routine use^8,9. Compression bandages, drain clamping, fibrin sealants, and intravenous tranexemic acid have also been used with good results to minimise post-operative blood loss^10–12.

The use of drains in orthopaedic joint replacement surgery is controversial and many studies have raised considerable doubt with regards to their usage^13–17. In contrast, numerous other studies have shown reduced rates of allogeneic transfusion in association with the use of autologous drains^18–22. Our study aimed to determine the effectiveness of the autologous CellTrans^™ drain in elective primary total knee arthroplasty.

In their randomised controlled trials, Ritter et al., Wood et al. and Amin et al. found no benefit from the use of autologous drains^13,16,17. The finding were similar in the studies by Mark et al. and Adalberth et al.^14,15. In contrast, Sinha et al., in their study of 100 patients, showed an 80% decrease in the rate of allogeneic blood transfusion among patients in whom an allogeneic drain was used¹⁸. Groh et al. also showed similar results with a significant decrease in the rate of blood transfusion¹⁹. Healy et al. found a 60% decrease in the rate of allogeneic blood tranasfusions²⁰. A reduced rate of blood transfusion was also noted in the recent studies by Hendriks et al. and Rojewski et al. In our study the rate of allogeneic blood transfusions was 5.7% in the group of patients in whom a CellTrans^™ autologous drain was used and 25.7% among the group of patients in whom the simple suction drain was used. This means that there was a 78% reduction in the rate of allogeneic transfusions.

Initially, with the advent of autologous drains, serious concerns were raised about the reinfusion of the collected blood. The potential risks of re-infusing autologous blood collected in drains have been studied. It has been shown that the blood collected has a low platelet count, low pH and a reduction of clotting factors along with increased fibrin degradation products^19,20,23,24. It was found that the blood collected in the drain contains a low titre of methylmethacrylate, although this is not clinically significant^21,25.

The CellTrans^™ drain used in our study has an inbuilt Pall LipiGuard^® filter which is designed for use with post-operative salvaged blood in compliance with the policy of the American Association of Blood Banks (AABB). It has a 40 μm screen filter to reduce micro-aggregates and debris. It also has additional media which typically remove 84% of lipids and 71% of leucocytes. It decreases fat globules present, depletes the blood of leucocytes and lowers the concentrations of anaphylatoxin, C3a and its immunosuppressive metabolite (C3a des Arg77). In our study, autotransfusion was uneventful in all patients, none of whom had adverse reactions. The CellTrans™ drain also had no major effect on the measured blood parameters such as the number of red blood cells, white blood cells and platelets or haematocrit. The level of plasma free haemoglobin remained approximately one quarter of that found in banked blood with no activation of the clotting cascade or complement system. Nursing and theatre staff must be trained to use the autologous drain in accordance with the manufacturer's instructions. It is also imperative that nursing staff be proactive while using the drain in order to record the amount of fluid in the drain and its timely reinfusion.

Various hospitals have different transfusion protocols. Some studies recommend transfusion if the haemoglobin concentration drops below 10 g/dL²⁶. The National Institutes of Health Consensus Development Conference in 1988 suggested that a haemoglobin level below 8 g/dL should be the trigger for blood transfusion²⁷. It has, however, been stressed that appropriate clinical judgement is imperative to decide the need of a transfusion rather than adhering to laboratory values alone. The British Transfusion task force recommends transfusion at 7 g/dL and 8 g/dL for young and old patients, respectively²⁸. In our study we transfused patients using Claudio's criterion in most cases²⁹. Patients with a significant past medical history and multiple comorbidities with severe anaemia were exempt from the criterion. Nine patients with a simple drain required the transfusion of 25 units of allogeneic blood. In the CellTrans^™ group only two patients required four units of allogeneic blood. The CellTrans^™ drain may be of immense value in the treatment of Jehovah's witnesses since the collected blood remains in a closed loop and is not separated from the body.

Conclusion

Autologous blood transfusions using an autologous drain are a simple and effective method of reducing post-operative blood loss after primary total knee arthroplasty. In our study there was a dramatic reduction in the rate of post-operative allogeneic blood transfusions among the patients in whom an autologous drain was used. Most, importantly, the drain also saved patients from unnecessary morbidity associated with blood transfusions and completely eliminated the risk of transmission of blood-borne diseases. Based on the results of the present study, routine use of autologous drains can be recommended in elective primary total knee arthroplasty.