Surgical blood conservation: Acute normovolemic hemodilution

INTRODUCTION — Acute normovolemic hemodilution (ANH) is a blood conservation technique that entails the removal of whole blood from a patient shortly after induction of anesthesia, with maintenance of normovolemia using crystalloid and/or colloid replacement fluid. The amount of blood removed typically varies between 1 and 3 units (450 to 500 mL constitutes one unit), although 3 to 4 units may be withdrawn safely in selected patients.

ANH can be used as the sole blood conservation technique, but for better results, it can be combined with other strategies to minimize or avoid transfusion, as discussed in separate topics:

●(See "Perioperative blood management: Strategies to minimize transfusions".)

●(See "Surgical blood conservation: Intraoperative blood salvage".)

RATIONALE — Objectives of ANH are to reduce transfusion of allogeneic blood by [1,2]:

●Decreasing blood concentration during the period when most surgical blood loss is occurring, thereby minimizing the effects of loss of red blood cells (RBCs), platelets, and plasma factors.

●Reinfusing the patient's own fresh whole blood (which includes RBCs, viable platelets, and plasma containing sufficient levels of clotting factors kept at room temperature in the operating room at the patient's bedside and can be re-infused when it is needed during or shortly after the surgical procedure. Since ANH whole blood is collected directly from the patient's circulation, these units have not been exposed to contact activation by extracorporeal circuits.

PATIENT SELECTION — ANH can be considered for patients with normal or high initial hemoglobin (Hgb) levels (see "Diagnostic approach to anemia in adults", section on 'Caveats for normal ranges'), who are expected to lose more than 500 to 750 mL blood during surgery. This recommendation is based on the fact that most patients can safely tolerate intraoperative blood losses up to 2 units of red blood cells (RBCs) without need for intervention (eg, blood transfusion), so neither the cost nor the labor required to implement intraoperative hemodilution would normally be justified for lesser volumes of expected blood loss.

ANH is the only autologous blood conservation strategy for patients who are unable to use other conservation methods due to factors such as or contraindications to intraoperative blood salvage. (See "Surgical blood conservation: Intraoperative blood salvage".)

ANH is also a good option for patients who decline blood transfusions, including Jehovah's Witnesses who generally agree to ANH, However, one must inquire if ANH is acceptable for an individual patient. (See "Approach to the patient who declines blood transfusion".)  

ANH is generally avoided but not necessarily contraindicated in the following settings [3]:

●Hemodynamically significant arrhythmias

●Acute infections

●Patients with impaired cardiac function (eg, low ejection fraction <45 percent, presence of significant aortic stenosis) who may have limited ability to increase cardiac output, since the main compensatory mechanism for induced anemia is a physiologic increase in cardiac output.

●Impaired renal function with oliguria, since potentially large amounts of replacement fluids ultimately need to be excreted.

●Baseline Hgb <11 g/dL.

●Presence of congenital/hereditary hemoglobinopathy

●Low concentrations of coagulant proteins or platelets, or abnormalities of coagulation or platelet function tests.

●Inadequate vascular access.

●Inability to monitor Hgb concentration, platelet number, and coagulation function with rapidly available results (ie, within 5 to 30 minutes).

TECHNICAL CONSIDERATIONS

Amount of blood to withdraw — Typically, 1 to 3 (occasionally up to 4 but no more than 4 units) units of blood are removed although 2 units and no more than 4 would render better ANH results. As blood is withdrawn, it is essential to maintain normovolemia using crystalloid and/or colloid replacement fluid. Most anesthesiologists aim for a hemoglobin (Hgb) of 8 to 9 g/dL after ANH has been completed as this maintains an appropriate safety margin in the event of acute surgical blood loss. Although evidence suggests that healthy carefully monitored patients can tolerate a Hgb as low as 7 g/dL, targeting lower Hgb concentrations (usually by means of removing more than 4 units of whole blood) will yield dilute units of blood that may not have the desired effect when they are reinfused. (See "Indications and hemoglobin thresholds for RBC transfusion in adults", section on 'Orthopedic surgery'.)

The volume of blood that may be withdrawn during ANH is derived from the following formula:

where V is the volume of blood to be removed (liters), EBV is the baseline estimated blood volume (liters) of the patient, Hgb-i is the initial Hgb (g/dL), Hgb-f is the final desired Hgb (g/dL), and Hgb-av is the average Hgb (average of Hgb-i and Hgb-f). Blood volume in adults can be crudely estimated as 70 mL/kg in lean men and 60 mL/kg in lean women [4]. However, it is important to recognize the imprecision of this estimate, and to realize that mass-based blood volume decreases in a non-linear fashion with increasing age and with increasing body mass [4-6].

As an example, in a 70 kg male patient with an estimated blood volume of 4900 mL (rounded to 5 L), an initial Hgb of 14 g/dL, and a desired final Hgb of 9 g/dL, the volume of blood to be drawn is:

The actual amount of blood removed will depend upon the patient's initial blood volume [7], hemodynamic stability, and the anticipated blood needs.

Vascular access — ANH requires adequate vascular access [1], ideally two large-bore cannulae placed into central and/or large peripheral vein(s) to allow simultaneous blood withdrawal and volume replacement.

Bladder catheterization — Placing an indwelling bladder catheter is suggested since urine output provides supplemental information regarding intravascular volume status. (See "Intraoperative fluid management", section on 'Traditional static parameters'.)

Timing — ANH is generally initiated after the induction of anesthesia and stopped before surgical bleeding becomes significant, with exact timing determined by the anesthesiologist.

Volume replacement — Replacement of intravascular volume can be accomplished using either colloid or crystalloid solutions. The volume infused depends upon the solution used. Crystalloids are distributed through the extracellular (intravascular plus interstitial) volume. As a result, the volume of crystalloid infused usually exceeds the volume of blood withdrawn, and is typically replaced on a 1.5:1.0 volume basis. In comparison, colloids are initially restricted to the vascular space. Thus, the volume initially administered should approximate the volume of blood withdrawn on a 1:1 basis. (See "Intraoperative fluid management", section on 'Crystalloid solutions' and "Intraoperative fluid management", section on 'Colloid solutions'.)

Regardless of the solution used, monitoring of vital signs and intravascular volume status is necessary since replacement formulas are not precise. The actual volume of fluid administered is adjusted depending on factors such as hemodynamic stability, urine output, and indicators of volume responsiveness, such as pulse pressure variation. (See "Intraoperative fluid management", section on 'Monitoring intravascular volume status'.)

Blood withdrawal procedure — Blood is withdrawn into a commercially available blood collection bag which typically contains citrate for anticoagulation and additives to prolong the life of the stored blood. A scale is used to determine the amount of blood removed. Overfilled whole blood bags may develop clots as the effectiveness of anticoagulation may compromised when the ratio of the preservative to anticoagulant is reduced. Once clots are formed and propagated, the blood may no longer be used. This will result in more blood loss negating the goal of ANH.

Preassembled hemodilution "kits" make ANH more convenient and may improve the safety of this procedure. Such kits contain two blood collection bags, a Y-type connector set with a Luer lock adapter, and a blood recipient identification band. Each unit of blood should be immediately labeled with the patient's name, medical record number, time of blood withdrawal, and sequential unit number (if more than one unit is removed).

Handling the blood — Bags containing the collected autologous blood should be placed on a mechanical agitator (ie, a rocker) unless gentle agitation of the blood bag can be accomplished by hand; this is particularly important during the collection period. Agitation prevents platelet aggregation thereby preserving platelet function.

It is essential that the anesthesiologist consult the local blood bank for labeling and storage standards and recommendations. According to the 5^th edition of the Association for the Advancement of Blood & Biotherapies (AABB) Standards for Perioperative Autologous Blood Collection and Administration, blood units collected by ANH can be stored in the operating room at room temperature (approximately 22°C [72°F]) for up to eight hours [8]. However, in some situations, the medical director of the intraoperative autologous blood service can prolong the room temperature storage time if necessary, for an individual patient. During reinfusion, warming the blood with a blood warmer may be appropriate to avoid hypothermia if a large number of units are transfused. (See "Perioperative temperature management", section on 'Consequences'.)

If withdrawn units of blood are not used within eight hours, they can be stored at 1 to 6°C (34 to 43°F) for up to 24 hours, provided that cold storage is begun within eight hours of initiating blood withdrawal [8]. Any units removed from the operating room must contain the statement "For autologous use only," on the label and are restricted to use for the ANH patient. Unused units cannot be added to the blood bank’s donated inventory.

Reinfusion — Generally speaking, ANH blood is infused when the patient requires a blood transfusion [9]. Reinfusion is guided by estimation of blood loss and serial Hgb determinations. The collected blood is reinfused after major blood loss has ceased, or sooner if indicated (eg, low Hgb level, hemodynamic instability).

The units of blood are usually reinfused in the reverse order of collection. Thus, the first unit, which has the highest Hgb and contains the most platelets and undiluted coagulation factors, is reinfused last. This allows final infusion of the most concentrated unit, theoretically at the time when bleeding is least. Occasionally, it may be prudent to change the sequence to ensure that the first unit(s) is (are) not wasted and are administered within eight hours of collection (eg, development of unintended hypervolemia that does not respond to diuretic administration).

Either Hgb or hematocrit levels should be monitored regularly during surgery, with the interval determined according to assessments of blood loss. Blood reinfusion is typically initiated when Hgb is approximately 7 to 8 g/dL (hematocrit approximately 21 to 24 percent) if there is no significant ongoing bleeding. However, the lowest safe Hgb (or hematocrit) level in patients undergoing ANH is unknown, and there is inter-individual variability. In some cases, reinfusion may be necessary before quantitative laboratory assessment of Hgb can be obtained (eg, when blood loss is rapid and extensive). Decision-making is based on the rate of bleeding, expected volume of ongoing blood loss, and the need to have an adequate margin of safety if additional severe bleeding is a possibility. Further discussion of evidence supporting selection of appropriate Hgb thresholds for intraoperative red blood cell (RBC) transfusion is available in other topics:

●(See "Intraoperative transfusion and administration of clotting factors", section on 'Red blood cells'.)

●(See "Indications and hemoglobin thresholds for RBC transfusion in adults", section on 'Orthopedic surgery'.)

●(See "Indications and hemoglobin thresholds for RBC transfusion in adults", section on 'Cardiac surgery'.)

EFFICACY — Challenges in demonstrating efficacy of ANH include heterogeneity among studies due to difficulties in selection of patents likely to derive benefit from the technique, inconsistencies in volume of blood removed (removal of a single unit may have little or no benefit compared with removal of 2 to 3 units), lack of standardization for patients in the control groups (who received usual care in some studies or another blood conservation method in other studies), and problems with blinding, randomization, and publication bias [10-12]. Overall, the efficacy of ANH remains unclear due to marked heterogeneity in published studies [3,13].

Potential benefits of ANH included decreased exposure to allogeneic blood, thereby minimizing or avoiding possible risks associated with transfusions (eg, infectious disease, allergic reactions, allosensitization, immunomodulation) [1,2,11,12,14-19]. Also, ANH may decrease blood loss during surgery, since hemodilution lowers the concentration of RBCs, platelets, and plasma factors in the lost blood. Furthermore, ANH provides the opportunity to transfuse patients with their own fresh whole blood containing viable platelets and adequate levels of clotting factors. In addition, ANH is generally more cost-effective compared with allogeneic blood transfusion since testing, storage, and crossmatching costs are not incurred, and additional surgical personnel are not needed for processing the withdrawn blood. Notably, in most studies, reduction in blood loss and exposure to allogeneic blood units has been only modestly reduced. However, combinations of blood conservation strategies that have only modest individual efficacy may result in significant reductions or complete avoidance of transfusion [20].

Data in cardiac and noncardiac surgical patients are summarized below:

●Cardiac surgery – A 2020 meta-analysis (22 randomized trials with 1688 patients undergoing coronary artery bypass grafting, predominantly on-pump) noted that patients in the ANH group received on average 0.6 fewer units of allogeneic RBC transfusions compared with controls (-0.24 units, 95% CI -0.96 to -0.24 units), as well as a lower RBC transfusion rate (risk ratio [RR] 0.65, 95% CI 0.52 to 0.82) and less postoperative estimated blood loss (standardized mean difference -0.53, 95% CI -0.88 to -0.17) [21]. Similarly, a 2017 meta-analysis of 29 randomized trials that included more than 2400 patients undergoing cardiac surgery noted a lower incidence of allogeneic transfusion in the ANH group compared with usual care (42 versus 56 percent; RR 0.74, 95% CI 0.62-0.87), and the average number of transfused units was reduced by approximately one unit in the ANH group [14] (see "Blood management and anticoagulation for cardiopulmonary bypass", section on 'Autologous priming techniques'). A 2015 meta-analysis that included both cardiac and noncardiac surgical procedures in 63 randomized trials with more than 3800 patients also noted that ANH reduced the risk of allogeneic RBC transfusion compared with usual care (RR 0.74, 95% CI 0.63-0.88) [11]

Also, retrospective studies in cardiac surgical patients have also noted lower risks for transfusion [15,16]. One study in patients undergoing thoracic aortic repair with cardiopulmonary bypass noted that while allogeneic RBC transfusions were not reduced, use of other allogeneic blood components (eg, fresh frozen plasma, platelets, cryoprecipitate) was reduced in patients receiving ANH versus standard-practice [22].

Furthermore, in a national registry that included more than 18,000 adult cardiac surgical patients, use of ANH resulted in lower intraoperative transfusion rates (8 to 10 percent were transfused without or with an autologous priming [AP] blood conservation technique, respectively), compared with patients who had an AP technique alone (20 percent were transfused) or those who had no blood conservation intervention (27 percent were transfused) [17].

In some cardiac surgical patients, ANH may also result in increased tissue perfusion due to decreased blood viscosity caused by hemodilution [23]. For example, in those with normal or high initial hemoglobin (Hgb) levels, the decreased blood viscosity associated with induced anemia during ANH may have a cardioprotective effect [24,25]. Also, for patients with polycythemia due to cyanotic heart disease or other pathology, sufficient reduction of Hgb and hematocrit values using ANH may prevent postoperative coagulopathy and/or thrombosis [26-28].

●Major noncardiac surgery – A prospective study that included 50 patients undergoing total knee or hip replacement surgery noted that patients in the ANH group had a lower rate of allogeneic blood transfusion (16 versus 60 percent) and experienced significantly fewer postoperative complications (28 versus 0.64 percent) compared with controls, although total estimated blood loss was similar between the groups [29].

In a retrospective study of 62 patients undergoing cerebral aneurysm clipping, the perioperative allogeneic transfusion rate did not differ among the 20 patients who had ANH compared with the 42 cohort patients who did not receive ANH. However, ANH patients had significantly higher hemoglobin concentration on postoperative days one (11.5 ± 2.5 versus 10.3 ± 2.0 g/dL) and three (12.1 ± 2.0 versus 10.7 ± 1.3 g/dL), and higher postoperative hemoglobin concentration was identified as a predictor of better outcome [30].

SAFETY — ANH is safest when used in healthy, young adults, but may be used in other populations undergoing major surgery (see 'Patient selection' above). Although no significant adverse effects of ANH have been noted in most studies, safety of the technique has not been thoroughly investigated since not all studies included in large meta-analyses and registries reported adverse events [12,14,17].

A theoretical adverse effect of ANH is dilution of coagulation factors, which may actually increase blood loss if significant surgical bleeding occurs [12,31]. One in vitro study noted that platelet function may be impaired in the blood collected and temporarily stored in standard bags containing citrate, phosphate, dextrose, and adenine (CPDA) [32]. However, reports in patients requiring cardiopulmonary bypass note that infusion of ANH blood is associated with improved coagulation (reduced activated partial thromboplastin time [aPTT] and increased fibrinogen) compared with patients not receiving ANH [22,33].

An additional concern is potential waste of the harvested blood if the expected blood loss does not occur. However, in practice, harvested blood is almost always returned in a slow and prolonged infusion. If necessary, a small dose of diuretic may be administered to reduce the volume effect. Notably, reinfusion of ANH blood should take place in the operating room or the post-anesthesia care unit (PACU). Transporting ANH units to other locations may create confusion and is a potential safety concern if mistaken for donor blood and wrongly infused into another recipient.

Finally, technical considerations for the safe use of the ANH technique require committed anesthesia personnel trained in collection of autologous blood and the requisite intensive monitoring of patients who have been significantly hemodiluted. For these reasons, ANH use has remained largely limited to institutions with trained and committed teams of anesthesiologists and surgeons. (See 'Technical considerations' above.)

SUMMARY AND RECOMMENDATIONS

●Acute normovolemic hemodilution (ANH) entails the removal of blood (typically 1 to 3 but no more than 4) units shortly after induction of anesthesia, with maintenance of normovolemia using crystalloid and/or colloid replacement fluid. (See 'Technical considerations' above.)

●Objectives of ANH are to reduce or completely avoid transfusion of allogeneic blood by (see 'Rationale' above):

•Decreasing blood concentration during the period when most surgical blood loss is occurring, thereby minimizing the effects of loss of red blood cells (RBCs), platelets and plasma.

•Reinfusing the patient's own fresh whole blood (which includes RBCs, viable platelets, and adequate levels of clotting factors kept at room temperature in the operating room) when it is needed during or shortly after the surgical procedure.

●ANH is an option for patients with normal or high initial Hgb levels who are expected to lose more than 2 units of blood (500 to 750 mL) during surgery. ANH is generally not recommended in patients with impaired cardiac or renal function, baseline Hgb <11 g/dL, low concentrations or other abnormalities of coagulant proteins or platelets, or abnormalities of coagulation or inadequate vascular access. (See 'Patient selection' above.)

●The volume of blood that may be withdrawn during ANH is derived from the following formula (see 'Amount of blood to withdraw' above):

where V is the volume of blood to be removed (liters), EBV is the estimated blood volume (liters), Hgb-i is the initial Hgb (g/dL), Hgb-f is the desired Hgb (g/dL), and Hgb-av is the average Hgb (average of Hgb-i and Hgb-f). Most anesthesiologists aim for a preoperative Hgb level between 8 to 9 g/dL (hematocrit between 24 and 27 percent).

However, actual amount removed will depend upon the patient's initial blood volume and hemodynamic tolerance of the procedure and anticipated amount of blood loss.

●Replacement of intravascular volume can be accomplished using either crystalloid, typically on a 1.5:1.0 volume basis, or colloid on a 1:1 basis. Regardless of the solution used, the volume of fluid administered is adjusted depending on factors such as hemodynamic stability and volume responsiveness. (See 'Volume replacement' above.)

●The blood is withdrawn into bags containing citrate for anticoagulation that are gently agitated and maintained at room temperature in the operating room for up to eight hours. Collected blood is reinfused during or shortly after surgery when major blood loss has ceased or Hgb is approximately 7 to 8 g/dL (hematocrit approximately 21 to 24 percent), but sooner if indicated (eg, hemodynamic instability). (See 'Handling the blood' above and 'Reinfusion' above.)

●Overall efficacy of intraoperative ANH in decreasing total loss of blood and reducing exposure to allogeneic blood transfusions has been modest in most studies. The technique is typically combined with other blood conservation strategies to minimize or avoid transfusion. Although no significant adverse effects of ANH have been noted, safety of the technique has not been thoroughly investigated since adverse events have not been reported in all studies. (See 'Efficacy' above and 'Safety' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Michael Avidan, MD, who contributed to an earlier version of this topic review.