Abstract
Background: Patients undergoing revision total hip arthroplasty
frequently require perioperative blood transfusion, increasing the risk for
blood-borne disease and anaphylactic and hemolytic reactions. The purpose of
this retrospective study was to evaluate the effect of intraoperative blood
collection and reinfusion on net blood loss in patients undergoing revision
hip arthroplasty.
Methods: The medical records of 126 patients who had had a revision
total hip arthroplasty with intraoperative blood salvage, with use of a
collection and reinfusion device, during a twenty-eight-month period were
reviewed. For comparison, the medical records of ninety-six patients who had
undergone revision hip arthroplasty without intraoperative blood salvage were
reviewed. Each of the 222 patients was categorized into a group on the basis
of the type of revision.
Results: Patients who had a revision of the femoral and acetabular
components (Group C) had significantly higher mean intraoperative and total
blood loss than did those who had a revision of the femoral component only
(Group A [p = 0.009 and p = 0.02, respectively]) or a revision of the
acetabular component only (Group B [p = 0.0001 for both]). Total blood loss
was not significantly different between Groups A and B. The mean amount of
blood reinfused intraoperatively was 356 mL for the patients in Group A, 374
mL for the patients in Group B, and 519 mL for the patients in Group C.
Regression analysis showed a significant decrease in net blood loss with
intraoperative collection and reinfusion in Groups B (p = 0.002) and C (p =
0.0001) but not in Group A.
Conclusions: Intraoperative collection and reinfusion substantially
decreased net perioperative blood loss in patients who had a revision of both
components (Group C) and in those who had a revision of the acetabular
component (Group B). The use of intraoperative blood collection and reinfusion
appears to be a valuable method of preserving blood volume in the
perioperative period.
Level of Evidence: Therapeutic study, Level III-2
(retrospective cohort study). See Instructions to Authors for a complete
description of levels of evidence.
The blood loss that accompanies a revision total joint arthroplasty can be
substantial and frequently requires perioperative blood
transfusion1,2.
The costs and potential health risks associated with allogeneic blood,
including the transmission of blood-borne infections and allergic reactions,
have altered blood management practices.
Alternatives to allogeneic blood transfusion are being used to meet the
transfusion needs of patients having elective procedures. Preoperative
donation of autologous blood has been shown to effectively reduce the need for
allogeneic blood
transfusion2-5.
Several studies have shown that patients managed with intraoperative blood
collection and reinfusion have a decreased need for allogeneic blood
transfusion, particularly among those undergoing revision hip
replacement6-10.
Wilson showed that intraoperative collection and reinfusion resulted in a 42%
reduction in the total amount of allogeneic blood transfused in ninety-eight
patients undergoing 100 total hip
revisions11. The
mean blood loss and transfusion requirements, as well as the amount of
intraoperative transfusion of autologous blood, were greater in patients who
had a revision of both components than in those with a revision of only
one11.
Evaluation of the efficacy of intraoperative blood collection and
reinfusion to decrease the need for allogeneic transfusion is complicated by
the combined effects of preoperative blood donation and postoperative salvage.
Postoperative autotransfusion has been used to increase the effectiveness of
autologous blood
transfusion12.
Postoperative blood salvage effectively reduced allogeneic transfusion
requirements in patients without available predonated autologous
units6,9.
Postoperative blood salvage significantly reduced the need for allogeneic
transfusion in a study that controlled for other risk factors such as age,
gender, preoperative hematocrit, and intraoperative blood loss (p =
0.0028)9. However,
none of the patients in that study had intraoperative blood collection and
reinfusion.
The purpose of this retrospective study was to evaluate the effect of
intraoperative blood collection and reinfusion among patients undergoing
revision total hip arthroplasty and to examine the effect according to the
type of revision. Because many patients who have a total hip replacement elect
to deposit autologous blood preoperatively and to receive postoperative blood
salvage, it was of particular interest to determine whether intraoperative
blood collection and reinfusion added any benefit in this group.
The study was approved by the Human Subjects Internal Review Board at
Stanford University. An independent chart auditor reviewed the medical records
of 222 consecutive patients who underwent unilateral, elective revision total
hip arthroplasty over a twenty-eight-month period. All patients received
postoperative blood salvage with the ConstaVac CBC Drain (Stryker, Kalamazoo,
Michigan). Intraoperative blood collection and reinfusion was performed with
use of the Cell Saver (Haemonetics, Braintree, Massachusetts) in 126 revision
total hip arthroplasties, but it was not performed in ninety-six revisions.
The decision to use the Cell Saver was made by the individual surgeon. The
operations were performed by eight surgeons between October 1997 and February
2000. Indications for revision total hip arthroplasty included mechanical
failure or loosening (216 hips), secondary reconstruction after a previous
infection with no clinical or bacteriological evidence of an ongoing infection
(three hips), and periprosthetic fracture (three hips). All operations were
performed with use of a posterolateral approach to the hip.
The 222 patient records were categorized for analysis according to the type
of revision. Group A consisted of thirty-four patients who underwent revision
of the femoral component only. There were fourteen men and twenty women, with
an average age (and standard deviation) of 67.5 ± 16.9 years. The Cell
Saver was used in twenty-two patients (65%). Group B consisted of ninety-six
patients who had a revision of the acetabular component only. There were
thirty-five men and sixty-one women, with an average age of 64.6 ± 16.8
years. The Cell Saver was used in forty-three patients (45%) in this group.
Group C consisted of ninety-two patients who underwent revision of both the
femoral and the acetabular component. This group consisted of forty-nine men
and forty-three women, with an average age of 64.1 ± 16.1 years. The
Cell Saver was used in sixty-one patients (66%) in Group C.
Intraoperatively, blood was collected by means of a heparinized
double-lumen suction catheter, was then filtered through a 180-µm filter
that removed gross debris, and was stored in a standard reservoir. The
filtered blood was washed with normal saline solution and centrifuged.
Aggregated red cells, platelets, white blood cells, fibrinogen, free
hemoglobin, and other proteins contained in the supernatant were discarded.
The remaining packed red-blood cells in saline solution were reinfused through
a 20-µm filter to the patient. Although blood collected with the Cell Saver
may be processed and reinfused in as few as six minutes, the collection of
blood in this study often continued until the completion of the operation and
the blood was processed and administered either during wound closure or in the
recovery room1. The
decision to transfuse autologous or allogeneic units postoperatively was made
by each surgeon on the basis of the clinical assessment, rather than as a
response to a preset quantitative laboratory threshold. A variety of criteria,
including age, hematocrit measurement, history of myocardial infarction or
angina, tachycardia, dizziness, and postural hypotension, contributed to this
decision1.
Statistical Analysis
Univariate analyses were used to evaluate patient demographics and surgical
procedures. In order to assess comparability between the groups, and to
measure mean volumes of blood loss and reinfusion, independent t tests were
conducted on a series of variables. Net blood loss was defined as the total
intraoperative and postoperative blood loss minus the blood volume reinfused
to the patient through intraoperative and postoperative blood salvage
techniques. Analysis of variance was used to measure differences in net blood
loss in patients with and without intraoperative blood collection and
reinfusion, while controlling for perioperative factors such as age, gender,
body-mass index, cemented or cementless fixation, femoral osteotomy,
intraoperative blood loss, operating time, and postoperative autotransfusion.
Intraoperative blood collection and reinfusion, postoperative blood salvage,
surgical technique (cemented or cementless fixation), and femoral osteotomy
were treated as dichotomous variables. Net blood loss, age, body-mass index,
preoperative hematocrit, and intraoperative blood loss were treated as
continuous variables. Linear regression was used to measure the associations
between intraoperative blood collection and reinfusion and net blood loss
while controlling for the perioperative variables above. These data were used
to construct a predictive model graph for the utility of the Cell Saver on the
basis of the type of revision.
Patient age, gender, preoperative hematocrit, and body-mass index did not
differ significantly among Groups A, B, and C. The patients in Group A
(revision of the femoral component only) had a mean operating-room time (and
standard deviation) of 186 ± 72 minutes, with a mean intraoperative
blood loss of 892 ± 557 mL and a mean postoperative blood loss of 595
± 379 mL. A mean of 378 ± 214 mL (range, 200 to 1125 mL) of
blood was collected intraoperatively in this group, and a mean of 356 mL was
reinfused. Total blood loss (intraoperative and postoperative loss less the
volume reinfused from intraoperative and postoperative blood salvage) was 1258
± 597 mL in Group A. Patients in Group B (revision of the cup only) had
a mean operating room time of 143 ± 61 minutes. The mean blood loss was
719 ± 437 mL intraoperatively and 573 ± 317 mL postoperatively.
The Cell Saver collected a mean of 403 ± 299 mL (range, 0 to 1750 mL)
of blood. On the average, 374 mL of blood collected intraoperatively was
reinfused. The mean total blood loss was 1114 ± 530 mL in Group B. The
patients in Group C (revision of both components) had a mean operating-room
time of 237 ± 101 minutes. The mean intraoperative blood loss was 1244
± 733 mL, and the mean postoperative blood loss was 711 ± 352
mL. Intraoperative blood collection included a mean of 575 ± 393 mL
(range, 125 to 1600 mL), with a mean of 519 mL reinfused. The mean total blood
loss was 1603 ± 739 mL in Group C.
The patients in Group C had a significantly longer mean operating-room time
than did the patients in Group A (p = 0.005) and Group B (p = 0.0001). The
mean operating-room times were significantly longer for Group A than for Group
B (p = 0.022). Group C also had significantly higher mean intraoperative and
mean total blood loss than both Groups A (p = 0.009 and p = 0.02,
respectively) and B (p = 0.0001 for both). The mean postoperative blood loss
was significantly higher for Group C than for Group B (p = 0.018), but it was
not significantly different from that for Group A. The mean intraoperative,
postoperative, and total blood loss were not significantly different among the
patients undergoing revision of a single component.
The results were similar with respect to intraoperative blood collection
and reinfusion. The amount of blood collected and reinfused by the Cell Saver
was significantly higher for the patients in Group C than for those in Group B
(p = 0.0001), but it was not significantly different from that for the
patients in Group A (p > 0.05). The volume of blood collected by the Cell
Saver and reinfused did not differ between the two groups that had revision of
a single component.
For the patients who had revision of the femoral component (Group A), the
mean intraoperative blood loss was 1020.45 ± 122.06 mL with the Cell
Saver and 658.33 ±133.26 without the device (p = 0.069) (see Appendix).
(Note that these calculations do not include blood transfused through use of
the Cell Saver.) For patients who had revision of the acetabular component
(Group B), the mean intraoperative blood loss was 931.55 ± 76.12 mL
with the Cell Saver and 550.94 ± 40.80 without the device (p = 0.001).
For patients who had revision of a single component, the use of the Cell Saver
resulted in the transfusion of a mean of 355.68 ± 45.85 mL of blood
during femoral revision and 373.72 ± 34.43 mL during acetabular
revision. Despite the use of the Cell Saver in more complex cases associated
with increased intraoperative blood loss, no significant difference between
patients managed with or without the Cell Saver was found with respect to the
total blood loss and the number of transfused allogeneic units.
The net blood loss for the patients in Group A was shown to be
significantly influenced by the intraoperative blood loss (p = 0.001),
postoperative autotransfusion (p = 0.013), total operating time (p = 0.048),
and the performance of a femoral osteotomy (p = 0.036)
(Fig. 1, A).
Figure 1, B shows the
predicted net blood loss for the patients in Group A who were managed without
a femoral osteotomy. The net blood loss for the patients in Group B was
significantly affected by the intraoperative blood loss (p = 0.0001),
postoperative autotransfusion (p = 0.001), and the use of the Cell Saver (p =
0.002) (Fig. 1, C).
The net blood loss for the patients undergoing both femoral and acetabular
revisions (Group C) was also significantly influenced by the intraoperative
blood loss (p = 0.0001), postoperative autotransfusion (p = 0.0001), and the
use of the Cell Saver (p = 0.0001) (Fig. 1,
D).
Autologous blood collection and reinfusion are often used perioperatively
during primary and revision total hip arthroplasty in the hope of reducing
allogeneic
transfusion1,6-8,10,13,14.
However, questions concerning the efficacy and cost of devices such as the
Cell Saver have arisen. The purpose of this study was to evaluate the efficacy
of the Cell Saver on the basis of the type and complexity of revision hip
arthroplasty while controlling for important perioperative factors.
The Cell Saver did not appear to significantly influence the net blood loss
in Group A (the patients who had a femoral revision only). The predicted
amount of blood reinfused by the Cell Saver in Group A was approximately 150
mL. Even in patients with three units of predonated autologous blood, the Cell
Saver did not provide sufficient blood to overcome the blood deficit faced by
the patient. The patients in Group A who had a revision with an extended
femoral osteotomy had an increased predicted net blood loss of approximately
325 mL; the use of the Cell Saver may have merit in these cases, especially in
patients who are unable to donate blood preoperatively. Although the Cell
Saver did not significantly decrease the predicted net blood loss on the basis
of regression analysis, clinical judgment may indicate that reinfusion of
blood to the patient is worthwhile to decrease the overall blood deficit. Our
data suggest that the decision to use the Cell Saver in patients undergoing
revision of the femoral component only should be made on the basis of the
number of predonated autologous units and whether a femoral osteotomy will be
performed.
Patients undergoing revision of the acetabular component only (Group B) may
benefit from use of the Cell Saver, as it significantly decreased the net
blood loss (p = 0.001) and would return approximately 265 mL of autologous
blood. In a patient undergoing revision of the acetabular component, two units
of predonated autologous blood and use of the Cell Saver should provide
sufficient autologous blood volume to replace the predicted net blood loss of
approximately 1100 mL.
Patients undergoing revision of both the femoral and the acetabular
component (Group C) may also benefit from the use of the Cell Saver. In this
group, the total blood loss and the intraoperative blood loss were both
significantly greater than those in Groups A and B. The increased mean volume
of reinfused blood (382 mL) significantly reduced the net blood loss. A
patient with two units of predonated autologous blood had a predicted net
blood loss of nearly 650 mL. With the addition of the Cell Saver, the
predicted net loss was reduced to approximately 250 mL. Although the finding
was not significant, the Cell Saver was associated with the need for
reinfusion of approximately one unit of blood in patients undergoing revision
of the femoral component only (Group A). One unit of blood has not
traditionally been viewed as a sufficient amount to influence the transfusion
practice clinically; however, it represents the return of approximately 3% on
the hematocrit perioperatively. Greater amounts of reinfused blood should
raise the hematocrit proportionally.
This study did not follow a prospective protocol for blood transfusion
according to preset quantitative laboratory thresholds, as the exact triggers
for transfusion are controversial and our analysis was retrospective in
nature. The decision to use the Cell Saver was made preoperatively, on the
basis of the surgeon's subjective assessment of the complexity of the
revision, which may have introduced a selection bias for the more difficult
cases. It is not surprising, therefore, that use of the Cell Saver was
associated with greater intraoperative blood loss for single component
revisions.
The cost of collecting and reinfusing blood intraoperatively is not
insubstantial. At our institution, the charges for a unit of autologous and
allogeneic blood are approximately $526 and $827, respectively. Additional
costs are incurred if special tests are necessary for difficult crossmatches
or if blood fractions are used. The costs for use of the Cell Saver are
difficult to find at our institution; however, an estimate for the charges for
expendable items and technician time is approximately $400 per patient. All of
the above charges must be balanced by the wishes and expectations of the
patient and the surgeon and by the risks and benefits of each procedure.
However, the use of intraoperative collection and reinfusion appears to be a
valuable method of preserving blood volume in the perioperative period in
difficult hip revision cases.
A table showing the mean values (and standard error) for variables with and
without the Cell Saver by group is available with the electronic versions of
this article, on our web site at
(go to the article citation and click on "Supplementary Material")
and on our quarterly CD-ROM (call our subscription department, at
781-449-9780, to order the CD-ROM).
Guerra JJ, Cuckler JM. Cost
effectiveness of intraoperative autotransfusion in total hip arthroplasty
surgery. Clin Orthop.1995;315:
212-22.315212
1995
[PubMed]
Woolson ST, Marsh JS, Tanner JB.
Transfusion of previously deposited autologous blood for patients undergoing
hip-replacement surgery. J Bone Joint Surg Am.1987;69:
325-8.69325
1987
[PubMed]
Bierbaum BE, Callaghan JJ, Galante
JO, Rubash HE, Tooms RE, Welch RB. An analysis of blood management in
patients having a total hip or knee arthroplasty. J Bone Joint Surg
Am.1999;81:
2-10.812
1999
Hatzidakis AM, Mendlick RM, McKillip
T, Reddy RL, Garvin KL. Preoperative autologous donation for total joint
arthroplasty. An analysis of risk factors for allogenic transfusion.
J Bone Joint Surg Am.2000;82:
89-100.8289
2000
[PubMed]
Thomson JD, Callaghan JJ, Savory CG,
Stanton RP, Pierce RN. Prior deposition of autologous blood in elective
orthopaedic surgery. J Bone Joint Surg Am.1987;69:
320-4.69320
1987
[PubMed]
Ayers DC, Murray DG, Duerr DM.
Blood salvage after total hip arthroplasty. J Bone Joint Surg
Am.1995;77:
1347-51.771347
1995
Elawad AA, Fredin H.
Intraoperative autotransfusion in hip arthroplasty. A retrospective study of
214 cases with matched controls. Acta Orthop Scand.1992;63:
369-72.63369
1992
[PubMed][CrossRef]
Goulet JA, Bray TJ, Timmerman LA,
Benson DR, Bargar WL. Intraoperative autologous transfusion in orthopaedic
patients. J Bone Joint Surg Am.1989;
71: 3-8.713
1989
[PubMed]
Grosvenor D, Goyal V, Goodman S.
Efficacy of postoperative blood salvage following total hip arthroplasty in
patients with and without deposited autologous units. J Bone Joint
Surg Am.2000;82:
951-4.82951
2000
Law JK, Wiedel JD.
Autotransfusion in revision total hip arthroplasties using uncemented
prostheses. Clin Orthop.1989;245:
145-9.245145
1989
[PubMed]
Wilson WJ. Intraoperative
autologous transfusion in revision total hip arthroplasty. J Bone
Joint Surg Am.1989;71:
8-14.718
1989
Semkiw LB, Schurman DJ, Goodman SB,
Woolson ST. Postoperative blood salvage using the Cell Saver after total
joint arthroplasty. J Bone Joint Surg Am.1989;71:
823-7.71823
1989
[PubMed]
Huo MH, Paly WL, Keggi KJ. Effect
of preoperative autologous blood donation and intraoperative and postoperative
blood recovery on homologous blood transfusion requirement in cementless total
hip replacement operation. J Am Coll Surg.1995;180:
561-7.180561
1995
[PubMed]
Han CD, Shin DE. Postoperative
blood salvage and reinfusion after total joint arthroplasty. J
Arthroplasty.1997;12:
511-6.12511
1997
[CrossRef]