1 Department of Colorectal Surgery, Derriford Hospital, Plymouth, UK. 2 Department of Anaesthetics, Derriford Hospital, Plymouth, UK. 3 Professor of Health Services Research, Sheffield, UK. 4 Department of Haematology, Derriford Hospital, Plymouth, UK
* Corresponding author. E-mail: kenneth.hosie{at}phnt.swest.nhs.uk
Accepted for publication August 11, 2004.
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. Consecutive patients undergoing major gastrointestinal surgery were randomized to a planned 3-unit ANH, or no ANH. Both groups underwent identical management including adherence to a transfusion protocol after surgery. Outcome measures included the number of patients receiving allogeneic blood, complications, and duration of stay.
Results. 380 patients were screened of which 160 were included in the study, median age was 62 yr (range 2390), ANH n=78, no ANH n=82. There was no significant difference between groups in the number of patients receiving allogeneic blood 22/78 (28%) vs 25/82 (30%), the total number of allogeneic units transfused (90 vs 93), complication rate, or duration of stay. Haemodilution significantly increased anaesthetic time, median 55 (range 1590) vs 40 min (range 1780) (P<0.001). Significantly fewer patients in the ANH group experienced oliguria in the immediate postoperative period 37/78 (47%) vs 55/82 (67%) (P=0.012). The most significant factors affecting transfusion were blood loss, starting haemoglobin, and age. When compared with ASA-matched historical controls, the introduction of a transfusion protocol reduced the transfusion rate in colorectal patients from 136/333 (41%) to 37/138 (27%), P=0.004.
Conclusions. In this large pragmatic study, ANH did not affect allogeneic transfusion rate in major gastrointestinal surgery. Preoperative haemoglobin, blood loss, and transfusion protocol are the key factors influencing allogeneic transfusion.
Keywords: blood, transfusion ; surgery
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Acute normovolaemic haemodilution (ANH) is an autologous technique, which has certain advantages over other techniques. There is minimal preoperative preparation for both patient and staff, which makes ANH suitable for both emergency and elective procedures. Units are collected and stored in the operating theatre incurring no storage or testing costs, and minimizing the chance of a clerical error leading to an incorrect transfusion. All blood collected is returned to the patient thereby eliminating costly blood wastage. As blood is available in theatre, patients can be group and saved allowing a lower cross match:transfusion ratio. The blood is fresh, containing functional platelets and clotting factors and the lowered blood viscosity may have benefits in terms of oxygen delivery.2 A cardinal benefit, difficult to quantify, is the peace of mind associated with knowing that they are receiving blood free from contamination with variant CJD, HIV, or hepatitis virus.3 In addition there may also be unknown transfusion transmitted diseases.
Although the technique has been used before, there is still no evidence that ANH is effective in reducing exposure to allogeneic blood.2 411
The aim of this study was to define the role of haemodilution in major gastrointestinal surgery in terms of its potential as a blood saving technique, and the effects on clinical outcomes and hospital stay.
![]() |
Methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Interventions
Standard preoperative preparation included full blood count, urea and electrolytes, cross match according to procedure, ECG and chest X-ray. Picolax® (Ferring Pharmaceuticals Ltd, Middlesex, UK) bowel preparation, if required, was administered the day before surgery. The resulting dehydration was minimized by administering 2 litres of N saline with 20 mmol potassium chloride intravenously over 12 h, commencing the evening before surgery.12
All patients received a standard anaesthetic comprising propofol induction, isoflurane maintenance, and muscle relaxation with atracurium or rocuronium. Antibiotic prophylaxis with cefuroxime and metronidazole was given shortly after induction. Epidural catheters were sited either awake or asleep depending upon the anaesthetist's preference.
Haemodilution commenced in the anaesthetic room and continued in the operating theatre if needed. The volume of blood to be withdrawn (maximum 3 units), whilst maintaining the haemoglobin above 8 g dl1, was calculated using a formula described by Gross.13 A 14-gauge cannula was sited in either the external jugular vein or a large forearm vein. Extension tubing with a three way tap and rubber bung was connected, which allowed access away from the surgical drapes. Standard blood bags containing anticoagulant citrate-phosphate-dextrose were used to passively collect the blood. The bags were placed on scales on the floor and oscillated by hand during venesection to allow adequate mixing of anticoagulant. Once filled, the blood bag tubing was clamped, the bag labelled and stored with the patient at room temperature in the operating theatre.
Warmed cell-free fluid was administered via a second cannula during blood withdrawal to maintain normovolaemia. The first litre of blood withdrawn was replaced with 1 litre of gelofusine. Subsequent blood withdrawal was replaced with Hartmann's solution at a ratio of 3 or 4:1. A Hemocue B-Haemoglobin photometer (Hemocue Ltd, Sheffield, UK) was present in theatre to allow near patient haemoglobin testing, and used at the anaesthetist's discretion.
The patient was monitored during the operation using standard equipment for the operation concerned. Minimum monitoring included continuous ECG, heart rate, and oxygen saturation. Radial artery catheters and central venous lines were used at the anaesthetist's discretion. At the end of the operation, all the autologous blood was re-transfused. If a patient required a transfusion before the completion of surgery, autologous blood was used before any allogeneic. We did not feel protocol-driven intra-operative haematocrit alone would be an accurate indicator of need for transfusion in view of the large fluid shifts present during surgery with significant blood loss. Therefore, although intra-operative transfusion usually followed protocol (Table 1), this was at the anaesthetist's discretion. After surgery patients underwent standard care including adherence to a standard transfusion protocol (Table 1).14
|
Outcomes
Primary outcome measures were the proportion of patients transfused with allogeneic red blood cells and the number of units transfused in the first 3 days after surgery. Secondary outcome measures were time taken to venesect, complications, and duration of stay.
Urine output was measured hourly for at least 24 h after surgery. Oliguria was defined as urine output less than 30 ml h1. Full blood count and urea and electrolytes were measured at 24 and 48 h with the former measured again at discharge and 30 days. Patients were followed up at 30 days by telephone.
Complications
Definitions used were standard for our unit. Pyrexia of unknown origin was defined as a temperature greater than 37.5°C over a period of 48 h on two or more occasions, with negative urine, wound, sputum and blood cultures, and requiring antibiotics. Wound infection was defined as an accumulation of pus with spontaneous discharge or requiring surgical drainage and was graded as minor or major. Cellulitis with associated pyrexia and raised white cell count was regarded as minor. Deep infection was defined as pelvic or i.p. infection as diagnosed by imaging or laparotomy. Septicaemia was defined as positive blood cultures in association with swinging high pyrexia. Chest infection was defined as productive cough, pyrexia, and abnormal CXR. Urinary tract infection was defined as a positive culture more than 1000 organisms ml1 with associated pyrexia. Deep vein thrombosis and pulmonary embolus were diagnosed clinically and confirmed by above-knee duplex scanning and CTPA (CT pulmonary angiography) or V/Q scanning, respectively. Anastamotic leak was present if diagnosed clinically and confirmed radiologically.
Randomization and blinding
Randomization was in variable sized blocks and allocation obtained by telephone to Sheffield Medical Care Research Unit. The patients were blinded to treatment allocation throughout the study period.
Sample size
A sample size of n=78 in each arm was calculated based on reducing allogeneic blood exposure from 41% (transfusion rate in our unit for all ASA I and II patients (n=333) undergoing colectomy surgery from 1997 to 1999 inclusive, identified from our colorectal database) to 20%, with a power of 80% and two-tailed significance level of 0.05.
Analysis
Continuous data were analysed using the Student's t-test or MannWhitney U-test depending on the distribution of the data. Proportional data were analysed using the 2 test or Fisher's exact test. All analysis was performed on an intention-to-treat basis with P<0.05 taken as being significant, using SPSS version 9. Logistic regression was used to identify factors associated with having a transfusion. The model was built in stages and factors significant at the 10% level were retained. Interactions among variables were examined.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
|
There was no significant difference in median (range) haemoglobin levels between groups throughout the postoperative study period (day 1, 10.5 (6.015.4) and 10.8 g dl1 (7.215.0); day 2, 10.8 (6.914.9) and 10.4 g dl1 (7.515.0); discharge, 11.8 (8.414.7) and 11.7 g dl1 (8.714.9); day 30, 12.4 (9.216.0) and 12.3 g dl1 (7.514.8); ANH and control, respectively). In addition, there was no significant difference in preoperative and day 2 calculated red blood cell (RBC) loss (ANH mean 614 ml (SD 427), control mean 616 (SD 451)).2 15 The median lowest intra-operative pre-autologous transfusion haemoglobin in patients who were haemodiluted was 7.7 g dl1 (4.811).
Venesection lasted a median of 30 min (1070), and significantly increased anaesthetic time (Table 2), median 55 (range 1590) vs 40 min (range 1780) (P<0.001). Significantly fewer patients in the ANH group experienced oliguria in the immediate postoperative period 37/78 (47%) vs 55/82 (67%) (P=0.012).
There was no significant difference in complication rate in the two groups (Table 6) 14/78 (18%) vs 22/82 (27%) P=0.179; however, the complication rate was significantly greater in patients who received allogeneic blood compared with those who did not, 16/48 (33%) vs 20/112 (18%) (P=0.03). There were no specific complications as a result of ANH.
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Logistic regression identified three factors, which affect transfusion; starting haemoglobin, blood loss, and age, which concurs with previous studies.17 Twenty-nine per cent of the 154 patients excluded were so because of a starting haemoglobin of less than 11 g dl1. Looking at the entire study population (380 patients), 29% had a starting haemoglobin level of less than 12 g dl1 and 47% of more than 13 g dl1. These figures imply that preoperative boosting of haemoglobin level may be of use in decreasing allogeneic transfusion.
When compared with ASA matched historical controls (used for sample size calculation), the introduction of a transfusion protocol reduced the transfusion rate in the colorectal patients from 136/333 (41%) to 37/138 (27%), P=0.004. This may be study- and/or transfusion-protocol related, but explains the large allogeneic savings seen in studies with historical controls.
The median volume of blood venesected was 1350 ml (01454). Twelve patients had less than 900 ml venesected (Fig. 1); five had no blood venesected, four because of failed i.v. access and the other because of consultant's decision in theatre (listed for major operation, minor performed). In total 66/78 (85%) patients had the desired greater than/equal to 900 ml venesection. The main reasons for failure included poor venous access resulting in low flow or clot formation in the tubing, and anaesthetist's discretion preventing further venesection. Excluding those patients with less than 900 ml venesection from analysis, again resulted in no significant difference in transfusion rates 20/66 (30%) vs 25/82 (30%). This illustrates that the technique is not as simple in practice as it appears to be in theory.
Haemodilution increased anaesthetic time by a median of 15 min, despite haemodiluting during insertion of epidurals, urinary catheters, and draping the patient. This delay would have cost implications, if the technique were to be adopted. Epidural local anaesthetic block may have resulted in hypervolaemic haemodilution, as i.v. fluids were administered to compensate for the vasodilation. Although this affected both groups equally, as all patients had epidurals, the results cannot necessarily be extrapolated to patients not receiving an intra-operative epidural.
It was not felt possible to blind the clinicians, investigators, or anaesthetists. It would have been possible to involve a separate clinician to decide upon appropriateness of allogeneic transfusion after surgery, however, this was not felt necessary in view of the transfusion protocol, which was adhered to in all but three patients in each group.
Although there was no significant difference in complication rate between groups, there was a significantly increased complication rate in those patients who received allogeneic blood. This occurred despite the use of leucocyte-depleted allogeneic blood. It is still unclear whether it is the events necessitating transfusion, rather than transfusion itself exerting an effect.
In this study, the largest performed in the past 25 yr, ANH did not affect allogeneic transfusion rate, complication rate, or duration of stay in major gastrointestinal surgery. The study was designed to be pragmatic and applicable widely.
In the recent study performed by Matot and colleagues,18 in patients undergoing liver resection, the median blood loss was 750 and 890 ml in the ANH and control groups, respectively. This was despite attempting to select patients with an anticipated blood loss of 70% or more of the patient's blood volume and illustrates that blood loss prediction remains difficult. They did show a reduction in allogeneic transfusion rate from 36 to 10% in patients haemodiluted, with a median venesection volume of 2020 ml and 39 patients in each arm. However, the safety of ANH, in particular to extreme levels, remains unknown. Carvalho and colleagues19 reported a case of myocardial ischaemia in a patient undergoing ANH before abdominal aortic aneurysm repair. Some other studies have also shown a reduction in transfusion rate with ANH. Wong and co-workers20 studied 145 patients, undergoing elective infra-renal aortic surgery, randomized to receive allogeneic transfusion or a combination of cell salvage and ANH. They demonstrated a significant reduction in the number of units transfused, but not the number of patients transfused. However, only a mean of 1.66 units (450 ml) of blood were venesected for ANH, and as two interventions were studied, it is impossible to know the relative effect of either one.
Boldt and colleagues21 looked at 60 patients undergoing retropubic prostatectomy allocated to three groups, ANH, controlled hypotension, and a control. They venesected a mean (SEM) 1278 (150) ml and replaced this with gelofusine at a ratio of 1:1. In the hypotensive group, they aimed to maintain MAP at 50 mm Hg using sodium nitroprusside. They found significantly fewer units of blood were transfused in the study groups, the hypotensive group having the least (21, 14, and 28, respectively). The number of patients transfused in the three groups were 9, 5, and 12, respectively. However, only a mean of 710 ml extra fluid was administered to the ANH group compared with controls, implying that patients in the ANH group were bled rather than haemodiluted.
Herregods and co-workers22 demonstrated no difference in ST segment depression in patients undergoing ANH to a haematocrit of 34% compared with controls during coronary artery bypass grafting. Although they demonstrated a significant decrease in allogeneic blood exposure with ANH, the control group had significantly greater blood loss.
Olfsanger and colleagues23 found a significantly greater number of units transfused in controls compared with patients who were haemodiluted to a target packed cell volume (PCV) of 2830% (mean 1000 ml collected) undergoing total knee replacement. However, only 30 patients were studied, randomized into three groups; two receiving ANH and one control. The control group had higher ASA grades and significantly greater postoperative blood losses.11
In a three-arm study, McGill and colleagues24 randomized patients undergoing cardiac surgery to cell salvage, ANH plus cell salvage, and control. They demonstrated a significant reduction in the number of patients and units transfused in the cell salvage group compared with control but no additional benefit with ANH, although only 10 ml kg1 blood was venesected during ANH.
Earlier studies were mostly flawed and the deficiencies have been well reviewed25 26 (inadequate numbers, differing populations of patients compared, wide variations in the volume of autologous blood withdrawn, no transfusion protocol, and most importantly, the use of historical controls).
Our study demonstrates that the widespread use of 3-unit ANH is not justified in major gastrointestinal surgery. The efficacy of extreme ANH, in cases with massive blood loss remains unknown, but the safety in these situations is questionable.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Gillon J, Thomas MJ, Desmond MJ. Consensus conference on autologous transfusion. Acute normovolaemic haemodilution. Transfusion 1996; 36: 6403[CrossRef][ISI][Medline]
3 Rutherford CJ, Kaplan HS. Autologous blood donationcan we bank on it? N Engl J Med 1995; 332: 7402
4 Brecher ME, Rosenfeld M. Mathematical and computer modeling of acute normovolemic hemodilution. Transfusion 1994; 34: 1769[CrossRef][ISI][Medline]
5 Feldman JM, Roth JV, Bjoraker DG. Maximum blood savings by acute normovolemic hemodilution. Anesth Analg 1995; 80: 10813[Abstract]
6 Singbartl G, Singbartl K, Schleinzer W. Is acute normovolemic hemodilution cost-effective? Transfusion 1996; 36: 84950[CrossRef][ISI][Medline]
7 Thomas MJ. Royal College of Physicians of Edinburgh Consensus Conference on autologous transfusion: final consensus statement. Transfus Sci 1996; 17: 32930[ISI][Medline]
8 Napier JA, Bruce M, Chapman J, et al. Guidelines for autologous transfusion. II. Perioperative haemodilution and cell salvage. British Committee for Standards in Haematology Blood Transfusion Task Force. Autologous Transfusion Working Party. Br J Anaesth 1997; 78: 76871
9 Bryson GL, Laupacis A, Wells GA. Does acute normovolemic hemodilution reduce perioperative allogeneic transfusion? A meta-analysis. The International Study of Perioperative Transfusion. Anesth Analg 1998; 86: 915[Abstract]
10 Allain JP, Akehurst RL, Hunter JM. Royal College of Physicians of Edinburgh. Autologous transfusion, 3 years on: what is new? What has happened? Transfusion 1999; 39: 9101[CrossRef][ISI][Medline]
11 Gillon J, Desmond M, Thomas MJ. Acute normovolaemic haemodilution. Transfus Med 1999; 9: 25964[CrossRef][ISI][Medline]
12 Sanders G, Mercer SJ, Saeb-Parsey K, Akhavani MA, Hosie KB, Lambert AW. Randomized clinical trial of intravenous fluid replacement during bowel preparation for surgery. Br J Surg 2001; 88: 13635[CrossRef][ISI][Medline]
13 Gross JB. Estimating allowable blood loss: corrected for dilution. Anesthesiology 1983; 58: 27780[ISI][Medline]
14 Clagett GP, Valentine RJ, Jackson MR, Mathison C, Kakish HB, Bengtson TD. A randomized trial of intraoperative autotransfusion during aortic surgery. J Vasc Surg 1999; 29: 2230[ISI][Medline]
15 Mercuriali F, Inghilleri G. Proposal of an algorithm to help the choice of the best transfusion strategy. Curr Med Res Opin 1996; 13: 46578[ISI][Medline]
16 Faust RJ, Stanhope CR. Ineffectiveness of acute normovolemic hemodilution. Anesth Analg 1995; 81: 660[ISI][Medline]
17 Tartter PI, Mohandas K, Azar P, Endres J, Kaplan J, Spivack M. Randomized trial comparing packed red cell blood transfusion with and without leukocyte depletion for gastrointestinal surgery. Am J Surg 1998; 176: 4626[CrossRef][ISI][Medline]
18 Matot I, Scheinin O, Jurim O, Eid A. Effectiveness of acute normovolemic hemodilution to minimize allogeneic blood transfusion in major liver resections. Anesthesiology 2002; 97: 794800[CrossRef][ISI][Medline]
19 Carvalho B, Ridler BM, Thompson JF, Telford RJ. Myocardial ischaemia precipitated by acute normovolaemic haemodilution. Transfus Med 2003; 13: 1658[CrossRef][ISI][Medline]
20 Wong JC, Torella F, Haynes SL, Dalrymple K, Mortimer AJ, McCollum CN. Autologous versus allogeneic transfusion in aortic surgery: a multicenter randomized clinical trial. Ann Surg 2002; 235: 14551[CrossRef][ISI][Medline]
21 Boldt J, Weber A, Mailer K, Papsdorf M, Schuster P. Acute normovolaemic haemodilution vs controlled hypotension for reducing the use of allogeneic blood in patients undergoing radical prostatectomy [see comments]. Br J Anaesth 1999; 82: 1704
22 Herregods L, Moerman A, Foubert L, et al. Limited intentional normovolemic hemodilution: ST-segment changes and use of homologous blood products in patients with left main coronary artery stenosis. J Cardiothorac Vasc Anesth 1997; 11: 1823[CrossRef][ISI][Medline]
23 Olsfanger D, Fredman B, Goldstein B, Shapiro A, Jedeikin R. Acute normovolaemic haemodilution decreases postoperative allogeneic blood transfusion after total knee replacement. Br J Anaesth 1997; 79: 31721
24 McGill N, O'Shaughnessy D, Pickering R, Herbertson M, Gill R. Mechanical methods of reducing blood transfusion in cardiac surgery: randomised controlled trial. BMJ 2002; 324: 1299
25 Stehling L, Zauder HL. Acute normovolemic hemodilution. Transfusion 1991; 31: 85768[CrossRef][ISI][Medline]
26 Triulzi DJ, Ness PM. Intraoperative hemodilution and autologous platelet rich plasma collection: two techniques for collecting fresh autologous blood. Transfus Sci 1995; 16: 3344[ISI][Medline]
|