|
Reading List: Blood loss and blood management in orthopaedic trauma surgery by Professor Roger Lemaire
Sehat KR, Evans RL, Newman JH. Hidden blood loss following hip and knee arthroplasty: correct management of blood loss should take hidden loss into account. J Bone Joint Surg [Br] 2004;86-B:561-5.
This important paper clearly demonstrates that the amount of blood lost a standard orthopaedic operation is far more important than would appear from adding the estimated intra-operative bleeding and the amount of blood lost postoperatively in the drains, when used.
This was already known by those surgeons with a particular interest in blood management.
Occult blood loss simply represents extravasation of blood into the soft tissues, despite the usual efforts to achieve haemostasis. The extravasated blood undergoes haemolysis to some extent, but there is no indication that intravascular haemolysis could play any role in generating the hidden blood loss. As shown in this study, the latter may represent up to 49% of the total blood loss associated with a routine knee arthroplasty.
This should be kept in mind when planning an elective operation: considering the maximum allowable blood loss for any given patient, specific measures should be considered when it appears that the latter is likely to be smaller than the anticipated total blood loss. Allogeneic blood transfusion has traditionally been used, but an integrated blood policy may achieve the same goal while reducing the use of allogeneic blood. Patients who are anaemic before operation should have their erythrocyte stock increased by iron therapy if they are iron deficient, or by erythropoietin administration. Local application of fibrin sprays or systemic administration of tranexamic acid or aprotinin have been shown to decrease the total blood loss. Avoiding wound drainage or using delayed or intermittent drainage will also reduce blood loss and intra-or post-operative salvage of shed blood can also add to an efficient blood management policy. However, as already suggested by this study, avoiding wound drainage or using delayed or intermittent drainage will however not result in keeping in the circulation the full amount of blood that will not be retrieved in the drains, as it will probably increase the occult internal blood loss to some extent; it may however reduce bleeding through a tamponade effect, which one should strive to achieve with a relatively limited amount of extravasated blood. Fibrin sprays, tranexamic acid, compression dressings, added to careful surgical haemostasis may help to achieve this goal. Whether minimally invasive surgery will effectively reduce the occult blood loss has not yet been clearly demonstrated, as various studies have led to diverging conclusions.
This study also illustrates the fact that any study on the effect of various factors on blood loss should take into account the occult blood loss, and should include calculation of the total blood loss, rather than just comparing the amounts of blood lost in the drains, as is still often encountered. Kalairajah Y, Simpson D, Cossey AJ, Verrall GM, Spriggins AJ. Blood loss after total knee replacement: effects of computer-assisted surgery. J Bone Joint Surg [Br] 2005;87-B:1480-2.
This is an interesting although controversial paper, which concludes that, in the authors’ experience, computer assisted knee arthroplasty is associated with less peri-operative blood loss. The value of computer-assisted surgery (CAS) in improving accuracy of alignment in TKR has now been demonstrated in a number of studies, and this makes sense, but it is difficult to imagine through which mechanism computer assisted knee arthroplasty could reduce perioperative blood loss. The authors have compared the reduction in the level of haemoglobin in two groups of patients undergoing TKR, one using a conventional technique, the other with CAS, with the only difference being the use of intramedullary jigging of the femur and tibia in the first group - a bone plug being used to seal the defect created by the intramedullary rod in the femur, while no intramedullary instruments were used in the other. The authors conclude that the computer-assisted operation saves blood, reduces the risk of transfusion and may be useful in patients for whom blood products are not acceptable. It may well be however, that this conclusion only applies to their specific drainage and transfusion policy: three drains were used and were kept under aspiration for 48 hours, resulting in a mean drainage of 1351 ml in the CAS group versus 1747 ml in the other group.; all patients pre-donated autologous blood and all were routinely transfused with one or two units of blood (presumably pre-donated autologous blood), with some receiving further units (presumably allogeneic blood). It is not clearly stated how many units were pre-donated, nor how many units of allogeneic blood were transfused.
The aim was, as stated, to reduce the hidden blood loss, but there is no indication that this has really been achieved. Several studies have indeed shown that haematomas may still develop after removal of drains. Besides, the drainage policy used resulted in a mean blood loss in the drains largely in excess of 1000 ml, which is much more than many knee surgeons would be prepared to accept. Such a large external blood loss may appear undesirable to many surgeons interested in blood management, all the more as other studies have shown that non-drainage reduces total blood loss without resulting in any untoward side effects.
The indiscriminate use of autologous blood pre-donation is another point for debate: in several studies, this has resulted in a high proportion (around 50%) of pre-donated units not being used. The authors have apparently re-transfused all the pre-donated units, but this goes with their very aggressive drainage policy. Overall, the findings reported in this study are intriguing, but they are presumably linked to a specific and fairly unusual blood management policy. Tsumara N, Yoshiya S, Chin T, Shiba R, Kohso K, Doita M. A prospective comparison of clamping the drain or post-operative salvage of blood in reducing blood loss after total knee arthroplasty. J Bone Joint Surg [Br] 2006;88-B:49-53.
A number of clinical studies have compared wound healing, complications, functional results, blood loss and transfusion rates in patients undergoing THR or TKR with or without wound drainage. The overall conclusion is that wound drainage can be safely dispensed with, and that it may do more harm than good. Delayed or intermittent drain clamping has recently received more attention, as a compromise solution between drainage and non-drainage. Although delayed or intermittent drainage is used empirically by a number of surgeons, there is little data available in the literature to define an optimal procedure. It is therefore not surprising to note that some studies found no significant reduction in blood loss and transfusion requirements following delayed or intermittent clamping, while others came to an opposite conclusion. The rationale is, of course, to reduce bleeding through a tamponade effect. In this study, the authors also injected 30 ml saline with 1:500 000 adrenalin into the knee after closure, followed by drain clamping for 30 minutes, and thereafter by gradual rather than abrupt opening of the clamps. They noted a 50% reduction in the amount of drained blood as compared with the control group, in which drained blood was salvaged. Similar findings have been made in other studies without using adrenalin, but diverging findings have been reported depending on the timing and periodicity of drain clamping and opening. Although more clinical research on this point will be needed to define the optimal policy, delayed or intermittent clamping is an interesting compromise between drainage and non-drainage, in a policy aiming to reduce allogeneic blood transfusion. Helm AT, Karski MT, Parsons SJ, Sampath JS, Bale RS. A strategy for reducing blood-transfusion requirements in elective orthopaedic surgery: audit of an algorithm for arthroplasty of the lower limb. J Bone Joint Surg [Br] 2003;85-B:484-9.
A number of orthopaedic departments have become aware that they were transfusing excessive amounts of allogeneic blood, and some have questioned their traditional blood management policy. This is one such example. The authors have succeeded in reducing their transfusion rate following THR or TKR from 66% to 24%, essentially by lowering the transfusion trigger from the traditional 10 g/dl Hb to 8.5 g/dl, without detriment to the patients. It is now generally accepted that elderly patients can be allowed a Hb level of 8.0 g/dl, and younger patients as low as 7.0 g/dl before transfusion is required. This is advantageous not only to reduce costs, but also to avoid possible side effects of allogeneic blood transfusion, which are not only related with the risk of viral infection such as AIDS, HBV or HCV, but also with bacterial infection, which has been found in many studies to be increased in patients who receive allogeneic blood transfusion. Aderinto J, Brenkel IJ. Pre-operative predictors of the requirement for blood transfusion following total hip replacement. J Bone Joint Surg [Br] 2004;86-B:970-3.
In this study, the authors have identified the pre-operative haemoglobin level and the patients’ weight as significant independent factors increasing the need for transfusion after THR. Patients with a pre-operative Hb level <12 g/dl needed transfusion three times more often than the others. These findings are in line with those of similar studies, and illustrate the fact that more attention should be paid to the pre-operative erythrocyte stock of the patients. The erythrocyte stock is related to blood volume which is itself related to body weight and gender. Female patients with a body weight of less than 70 kg and a Hb level less than 12 g/dl are particularly at risk of requiring a blood transfusion. They are at the same time poor candidates for autologous blood predonation, as this will often result in a further drop in their Hb level if their haematopoietic marrow functions poorly as often happens in elderly patients. Conversely, salvage of drained blood is of particular interest in this category of elderly anaemic slender female patients with a small blood volume. It would nevertheless be preferable to anticipate the problems, and to check the Hb level of any candidate for elective THR as soon as the indication for surgery has been made: some will be found to be anaemic due to iron deficiency, and this can be corrected with iron therapy (oral or intravenous) while others may benefit from erythropoietin administration. If anaemia is diagnosed on admission for operation, it is of course too late to correct it. This is so obvious that one wonders why such an easy and logical step is not routinely taken, as it would obviate the need for allogeneic transfusion in a number of cases, while also reducing the need for such procedures as re-perfusion of drained blood, which are not exempt of risks. Walmsley PJ, Kelly MB, Hill RMF, Brenkel I. A prospective, randomised, controlled trial of the use of drains in total hip arthroplasty. J Bone Joint Surg [Br] 2005;87-B:1397-401.
This study is a further illustration of a fact which has been highlighted by a number of similar studies, particularly with respect to THR and TKR: avoiding wound drainage does not result in significant harmful side effects, while avoiding a source of retrograde infection and reducing blood loss. In fact, not draining a surgical wound probably increases the occult internal blood loss to some extent, but the overall result seems to be a reduction in the amount of blood extravasated.
In this study, the authors conclude that non-drainage was associated with a smaller drop in Hb levels and a reduced need for allogeneic blood transfusion. They did not calculate the total blood loss, but this has been done in a number of similar studies which produced similar conclusions. Diverging results have been reported with respect to the effect of not using drains on the rate of allogeneic blood transfusion; this is related to the number of variables involved, which cannot all be controlled.
Proponents of wound drainage put forward the prevention of haematomas, but other studies have shown that wound drainage does not eliminate haematoma formation, which can develop anyway after drain removal, and leaving drains in place longer than 24 hours has been shown to increase the risk of retrograde infection. There is no documented evidence that wound drainage reduces the infection rate, nor that it increases it, unless drains are left in place too long. The authors found that the use of a drain did not influence the postoperative Hb levels, the revision rates, the functional results, the length of hospital stay or the incidence of thrombo-embolism. The only drawback of not using drains appears to be the need for more frequent changes of dressings, which has been no more than a slight inconvenience in most studies, including one that we did. A recent Cochrane systematic review by Parker, Roberts and Hay came to similar conclusions.
One may therefore wonder why most surgeons keep draining surgical wounds, particularly following THR or TKR. We would tend to think that this is because the tradition of drainage is so deeply anchored, even in the patients’ minds, that the surgeons fear that the slightest problem that would occur following a non-drained THR or TKR would be poorly accepted by the patient and would automatically be related to absence of drainage, despite pre-operative information.
Non-drainage should be seen as one element among a number of others in a comprehensive blood management policy: the goal is to achieve a tamponade effect with a limited amount of extravasated blood. The probability of achieving this goal is increased if other measures are taken concurrently: local application of fibrin sprays, systemic administration of tranexamic acid or aprotinin, compressive dressings, and, of course, careful haemostasis. Booth D, Kothmann E, Tidmarsh M. Letter to the editor: a strategy for reducing blood-transfusion requirements in elective orthopaedic surgery. J Bone Joint Surg [Br] 2004;86-B:309-10.
These authors have summarised, in a letter to the editor, the essential points of a rational blood management policy. They point out that most papers available in the literature deal with just one of a number of measures to be taken and fail to develop a co-ordinated solution based on the whole patient process. This common sense reminder should be read and memorised by all orthopaedic surgeons and, most importantly, it should be implemented. The authors mention, as core measures applicable in all institutions, the early pre-operative pre-assessment of patients, to identify and treat those patients with low Hb level, peri-operative antifibrinolytic therapy, agreed transfusion triggers and blood prescribing practice, and post-operative iron supplementation, although this is questioned in other studies, and Hb follow-up in the community.
To those surgeons not familiar with the multiple facets of blood management, this brief letter to the editor is enlightening and is a useful reminder of the value of applying common sense to apparently complex medical situations. Professor Roger Lemaire, Consultant Orthopaedic Surgeon |
|
