Printer Friendly

Changing blood transfusion practice in elective joint arthroplasty: a nursing initiative.


The use of re-infusion drains on 99 consecutive patients undergoing total knee arthroplasty surgery at a large hospital was analysed. As a control group the records of 99 patients treated without re-infusion were analysed retrospectively. The primary aim was to ascertain the cost effectiveness of the drains. Secondary aims were to assess safety of the drains, whether or not they reduced the need for allogeneic blood transfusion and whether they decreased the length of stay in hospital. The direct cost of consumables increased for the evaluation period. There was a smaller proportion of allogeneic blood transfusion (27% vs 38%) and a smaller mean number of units transfused (0.92 vs 0.54) in the re-infusion group compared to the control group. Patients benefited directly in that the mean length of stay was also significantly shorter in the re-infusion group. We anticipate more direct cost saving with experience and best practice and conclude that the use of re-infusion drains is a cost effective blood saving method in total knee joint arthroplasty.

Key Words: Allogeneic blood, re-infused blood, total knee joint arthoplasty.


Allogeneic blood transfusion (ABT) is the most common and accepted method of blood transfusion. While effective in replacing the volume of blood loss, the procedure introduces several risk factors including transmission of viral agents, specifically human immunodeficiency virus (HIV) and hepatitis, insufficient platelet count, isoimmunisation, and possible transfusion reactions (Spain, Miller, Bergamini, Montgomery, & Richardson, 1997).

Re-infusion of wound drainage has been considered a safe procedure for patients having elective surgery since the 1970s, but despite its potential to reduce the above risks, has not been widely practiced. In recent years however, due to increased public concern regarding the safety of blood transfusions, including transfusion related diseases such as HIV, hepatitis B and C and immuno-suppression, various pre, intra and postoperative techniques have been used with the goal of minimising exposure to ABT. There has been renewed interest for the use of autologous transfusion in elective orthopaedic surgery (Feagan, Wong, Johnston, & Arellano, 2002; Martin, Whiteside, Milliano, & Reedy, 1992; Munoz, Ariza, Garceran, Gomez, & Campos, 2005; Pertl & Kaltenecker, 2001; Sinardi, et al., 2005; Steinberg, Ben-Galim, Yanic, Dekel, & Menahem, 2004; Strumper, et al., 2004).

For patients undergoing total knee joint arthroplasty (TKJA), substantial blood loss is common, with most of the measured blood loss occurring during the post-operative period. This can result in post-operative anaemia. Of these patients 30-50% receive ABT (Feagan et al., 2002; Salido, Marin, Gomez, Zorilla, & Martinez, 2002; Sutton, Cresswell, Livesey, Speed & Bagga, 2004; Warner, 2001; Zarin, Grosvenor, Schurman, & Goodman, 2003).

At the hospital where the study was carried out around 300 TKJAs are performed each year. A high number of patients are aged over 65 years and ischaemic heart disease is common. There is no well established algorithm for the management of blood loss in elective knee replacement surgery. The decision to perform an allogeneic transfusion is made by each team without a formal protocol, and is based on many factors including pre and post-operative haemoglobin levels and medical co-morbidities. Often blood is given when a patient becomes symptomatic 2-3 days following surgery. Some transfusion guidelines for arthroplasty suggest that patients should receive a transfusion when the Hb falls below 10g/dL or the haemocrit falls below 30%, the 10/30 rule. However this is subject to debate (Alvarez, Hebert, & Szick, 2001; Harwin, 2004). Based on anecdotal evidence and observation, it would seem that orthopaedic nursing staff have long been concerned that patients were not being transfused in a timely manner and this was delaying discharge, some patients being unable to complete their daily rehabilitation programme each day due to low exercise tolerance and fatigue. They also believed that incidents of chest pain experienced by some patients with cardiovascular disease, which resolved following blood transfusion, were due to post-operative anaemia.

A literature review was conducted to look at alternatives to blood bank blood. Intra-operative blood salvage and auto-transfusion appeared to be a viable alternative to ABT, and are now largely viewed as safe and efficacious (Seigne, et al., 2003; Spain et al., 1997; Sutton et al., 2004; Warner, 2001; Zarin et al., 2003). However, in the setting being discussed here auto-tranfusion was a new procedure. Hence it was necessary to convince the clinical board of its efficacy.

A committee of three nurses and a doctor reviewed three types of drainage and re-infusion devices used in other New Zealand hospitals and chose the Donor Autologous Blood Re-infusion System (Donor System). This is a closed, low vacuum re-infusion drain that allows up to 800mls of post-operative wound drainage and subsequent re-infusion to the patient. It has a FDA approved filter system and is acceptable to Jehovah's Witness patients.



This was a nursing initiative to provide a safe and timely re-infusion of autologous blood based on the drainage of blood in the immediate post-operative period, i.e. up to six hours.


To determine whether (i) the use of the Donor System is cost effective, (ii) it will conserve the use of blood bank blood and (iii) it has any effect on the patient length of stay.


A cohort observational method was used. The regional Ethics Committee when approached waived the need for their approval on the ground that this was (a) an evaluation, not a randomised trial and (b) the use of drainage and re-infusion devices is well established overseas and at some hospitals in New Zealand. South Canterbury Health use the Donor System for elective orthopaedic surgery. We also approached the Maori Enrichment Services of the District Health Board (DHB) to ensure the device would be acceptable to Maori patients and received their support.

The protocol for the usage of the Donor System was obtained from the company that distributes the drains. They provided comprehensive education and made themselves available for advice until the nursing staff were practiced and confident in the use of the drains. They also provided an on call service after hours for the duration of the study.

Data Collection

Two orthopaedic nurses retrospectively audited 100 consecutive charts for patients who had undergone a primary TKJA in the previous 12 months, to determine blood transfusion practice over that period. This group of patients was treated with standard suction drains and used as a control group. One patient who received 8 units of blood was excluded from the study (leaving 99 charts audited).

New patients presenting for primary TKJA were approached at the orthopaedic pre-admission clinic and asked if they would like to take part in the study. Informed consent was obtained. Post-operative retrieval of shed blood was attempted for 99 consecutive patients undergoing primary TKJA between February and July 2005 using the donor salvage and re-infusion drains. The collection time for re-infusion was limited to 6 hours; however wound drainage was continued for at least 24 hours in all cases. Re-infusion took place after 6 hours of drainage was completed, to ensure patients were transfused as much of the collected blood as possible.

The surgery for both groups was performed by the department's orthopaedic surgeons. Although similar techniques were used for exposure and instrumentation, there were differences in the use of tourniquet, with some surgeons choosing to inflate the tourniquet at different times during the procedure. Tourniquet times were not recorded in the study, and no intra-operative blood retrieval was attempted.

Post-operatively, the drains were activated in the Post Anaesthetic Care Unit (PACU) and times recorded i.e. time of drainage commenced, amount and time of re-infusion, and amount of drainage in total. The protocol for ABT was used; this required the shed blood be checked by two nurses. Haemoglobin was recorded pre-operatively and at 24 and 72 hours post-operatively. The total length of stay was also recorded for each patient. No specific transfusion protocol was adopted for this study and the patients were transfused on a case by case basis.



Analysis of the two cohorts did not reveal significant differences in terms of age, gender, weight or pre-operative haemoglobin. There was no significant difference in the proportion of males and females for both groups. However, the control group was significantly more likely to have cardiovascular disease (70% vs 47%, p=0.001, Chi-square test) and cancer (14 vs 5 patients, p=0.02, Chi-square test).

Allogeneic blood transfusion

Ninety nine patients received salvaged blood with a mean amount of 476 mls (median 455, maximum 1160). It was anticipated that fewer patients from the re-infusion group would require ABT. The proportion of patients requiring ABT was 38% in the control group compared to 27% in the re-infusion group. This was sufficient evidence to conclude that a smaller proportion of re-infusion patients require ABT (p=0.04, one-tailed test for equal proportions). Furthermore, of those in each cohort who did receive allogeneic blood, the mean number of allogeneic units transfused was also different between the two cohorts (2.37 units in the control group vs 1.96 units in the re-infusion). This difference was sufficient to conclude that re-infusion patients did indeed require fewer units on average (p=0.02, one-tailed t-test). Putting these facts together, the re-infusion patients as a whole needed significantly fewer units of allogeneic blood on average: 0.92 units per control patient versus 0.54 units per re-infusion patient (p=0.05, one-tailed t-test).

Female patients were much more likely to require ABT than males: 49% vs 16%, a highly significant difference (p<0.001, Chi-square test). The gender gap was less in the control group (51% of females vs 24% of males requiring allogeneic blood transfusion) and more in the re-infusion group (46% of females vs 8% of males requiring allogeneic blood transfusion). This gender difference seems to be entirely explained by the lower day-1 haemoglobin for females (mean 97.1, SD 13.1) than for males (mean 111.2, SD 13.8), a statistically significant difference (p<0.001, t-test).

Length of stay

The median length of stay observed was 6 days in both cohorts. However the mean was less for the re-infused group (5.43 - 6.24 days, 95% confidence interval vs 6.00 to 7.00 days in control group). This equates to approximately half a day shorter stay on average.

Patient satisfaction

Of the 99 patients audited, all were very pleased at this innovation and happy to take part in the study. Many of the benefits are difficult to quantify. These include non delays in ambulation and patients feeling more comfortable in having their own blood re-infused. During the study there were no reported complications.


Keating and Ritter (2002) suggest that optimal blood management for elective joint arthroplasty should facilitate rapid, safe mobilisation and rehabilitation, and minimise the complications of anaemia and transfusion for a reasonable cost. Drainage and re-infusion of shed blood using the Donor System appeared to meet these requirements.

The cost savings can be assumed from the reduced length of stay and reduction in blood costs however, disappointingly, there was an increase in direct consumable costs. The total cost increase would be $6,961 for 300 TKJAs. This would be recovered if an extra 38 units of blood were saved. We anticipate this target would be reached with greater experience and development of best practice. The savings in nursing minutes were analysed. The average time for re-infusion of a unit of harvested blood was 63 minutes. The average time for ABT is 150 minutes; therefore there were savings in nursing time. When transfusing blood bank blood, our protocol demands that the nurse must remain with the patient for the first 15 minutes in case of a reaction. With re-infused blood this is unnecessary. There is also a saving in laboratory time as cross matching was not required for 73% of those re-infused. The savings in nursing minutes would equate to $6,447.21 per 300 patients. This amount would be doubled if transfusions were delayed 2-3 days post-operatively as this would require observations every 30 minutes for the duration of the transfusion. While this does not seem a large amount in monetary terms, in the light of our current nursing shortages and difficulty with recruitment any nursing time saved on a busy ward can be significant.

The reduced length of stay does not reflect dollars saved. However it does assist the DHB to fulfil their contractual requirements in terms of procedures performed. The amount received for an elective procedure is the same, regardless of whether the patient stays two days or two weeks. In order to generate income it is in our interests to discharge patients as quickly and as safely as possible. This also means a more rapid return to normal for the patients.

The effect of post-operative anaemia on rehabilitation is not well researched. Post-operative fatigue can be a factor in delayed recovery time, and post-operative anaemia could be a physiological reason for fatigue following TKJA (Keating, Ranawat, & Cats-Baril, 1999). Geier (2000) also posits that low post-operative haemoglobin levels affects early post-operative recuperative power, and the anecdotal evidence from orthopaedic nurses caring for total joint arthroplasty patients reflects this.

The nursing staff involved in theatre, PACU and the wards reported the Donor System as user friendly. The ward staff found that the system saved both time and work. We are hopeful that the use of these drains will extend to patients undergoing hip replacement and management of fractured neck of femur.


Post-operative re-infusion drains are a simple, safe and cost effective apparatus that can be used in TKJA. They also appear to be popular with patients and staff. Based on this evaluation we have been able to change the transfusion practice for patients undergoing TKJA.


Alvarez, G., Hebert, P., & Szick, S. (2001). Debate: Transfusing to normal haemoglobin levels will not improve outcome. Critical Care, 5(2), 56-63.

Feagan, B., Wong, C., Johnston, W., & Arellano, R. (2002). Transfusion practises for elective orthopaedic surgery. Canadian Medical Association Journal, 166(3), 310-314.

Geier, K. (2000). Improving outcomes in elective orthopaedic surgery: A guide for nurses and total joint arthroplasty patients. Orthopaedic Nursing, 19(Suppl), 10-21.

Harwin, S. (2004). Blood conservation in total joint arthroplasty: Blood as a drug. Orthopaedics, 27(3), 276.

Keating, E., Ranawat, C., & Cats-Baril, W. (1999). Assessment of post-operative vigor in patients undergoing elective total joint arthroplasty. A concise patient and caregiver based instrument. Orthopaedics, 22(1), 119-128.

Keating, E., & Ritter, M. (2002). Transfusion options in total joint arthroplasty. Journal of Arthroplasty, 17(4),125-128.

Martin, J., Whiteside, L., Milliano, M., & Reedy, M. (1992). Post-operative retrieval and transfusion in cementless total knee joint arthroplasty. The Journal of Arthroplasty 7(2), 205-210.

Munoz, M., Ariza, D., Garceran, M., Gomez, A., & Campos, A. (2005). Benefits of postoperative shed blood re-infusion in patients undergoing unilateral total knee replacement. Archives of Orthopaedic Trauma Surgery, 125(6), 385-389.

Pertl, D., & Kaltenecker, G. (2001). Minimizing allogeneic blood transfusion in knee prosthesis. Unfallchirurg, 104(9), 808-812.

Salido, J., Marin, L., Gomez, L., Zorilla, P., & Martinez, C. (2002). Preoperative haemoglobin levels and the need for transfusion after prosthetic hip and knee surgery: Analysis of predictive factors. Journal of Bone and Joint Surgery, 84(2), 216-221.

Seigne, R., Chacko, M., Pereira, R., Hodgson, T., Truscott, J., & Attkin, J. (2003). Reducing allogeneic blood transfusions to orthopaedic patients. Unpublished paper submitted for Quality and Innovation Awards Canterbury District Health Boards.

Sinardi, D., Marino, A., Chillemi, S., Irrera, M., Labruto, G., & Mondello, E. (2005). Composition of blood sampled from surgical drainage after joint arthroplasty: Quality of return. Transfusion 45(2), 202-207.

Spain, D., Miller, F., Bergamini, T., Montgomery, R., & Richardson, J. (1997). Quality assessment of intraoperative blood salvage and autotransfusion. The American Surgeon, 63(12), 1059-1064.

Steinberg, E., Ben-Galim, P., Yanic, Y., Dekel, S., & Menahem, A. (2004). Comparative analysis of benefits of autotransfusion of blood by a shed blood collector after total knee joint replacement. Archives of Orthopaedic Trauma Surgery, 124(2), 114-118.

Strumper, D., Weber, E., Gielen-Wijffels, S., Van Drumpt, R., Bulstra, S., Durieux, M., et al., (2004). Clinical efficacy of postoperative autologous transfusion of filtered shed blood in hip and knee arthroplasty. Transfusion 44(11), 1567-1571. Retrieved January 21, 2005, from Medline database.

Sutton, P., Cresswell, J., Livesey, J., Speed, K., & Bagga, T. (2004). Treatment of anaemia after joint replacement: A double-blind randomized controlled trial of ferrous sulphate versus placebo. Journal of Bone and Joint Surgery, 86(1), 31-33.

Warner, C. (2001). The use of orthopaedic perioperative autotransfusion system in total joint replacement surgery. Orthopaedic Nursing, 20(6), 29-32.

Zarin, J., Grosvenor, D., Schurman, D., & Goodman, S. (2003). Efficacy of intraoperative blood collection and re-infusion in revision total hip arthroplasty. Journal of Bone and Joint Surgery, 85(11), 2147-2151.

Kaaren Marcinkowski, RN, MHSc, Charge Nurse, Orthopaedic Ward, North Shore Hospital

Barry McDonald, MSc, PhD, Senior Lecturer, Massey University

Marcinkowski, K., & McDonald, B. (2006). Changing blood transfusion practice in elective joint anthroplasty: A nursing initiative. Nursing Praxis in New Zealand, 22(3), 15-21.
COPYRIGHT 2006 Nursing Praxis in New Zealand
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Marcinkowski, Kaaren; McDonald, Barry
Publication:Nursing Praxis in New Zealand
Article Type:Clinical report
Geographic Code:8NEWZ
Date:Nov 1, 2006
Previous Article:Improving health care through evaluation research: an interview with Katherine Nelson.
Next Article:The relationship between student nurse and nurse clinician: impact on student learning.

Related Articles
Establishing a successful autologous transfusion program.
Everybody wins with this autologous donor program.
More surgeons advocate autologous transfusions.
Best safety ( bar none.
Diabetes doubles arthroplasty infection rate.
Initiative to improve mothers' access to safe blood.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters