JBJS | Article

The Journal of Bone & Joint Surgery, Volume 82, Issue 1

Articles | January 01, 2000

Preoperative Autologous Donation for Total Joint Arthroplasty. An Analysis of Risk Factors for Allogenic Transfusion*

ARMODIOS M. HATZIDAKIS, M.D.†; MICHAEL R. MENDLICK, M.D.†; TRACY McKILLIP, P.A.C.†; RAM L. REDDY, M.D.‡; KEVIN L. GARVIN, M.D.†, OMAHA, NEBRASKA

View Disclosures and Other Information

Investigation performed at University of Nebraska Medical Center/Omaha Veterans Administration Medical Center, Omaha

The Journal of Bone & Joint Surgery. 2000; 82:89-100

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

Article

Tables

References

text A A A

Abstract

Background: While autologous blood is commonly predonated to provide replacement of blood lost in orthopaedic procedures, few studies of patients managed with total joint replacement have addressed the problem of which patients are likely to benefit from an autologous blood-donation program.

Methods: A retrospective analysis of 489 consecutive patients who had had a total joint arthroplasty was performed to identify the risk factors for allogenic transfusion and to further define the indications for preoperative autologous blood donation. The operations included 247 total knee replacements (157 unilateral primary, thirty-two revision, and twenty-nine one-stage bilateral primary procedures) and 271 total hip replacements (163 primary and 108 revision procedures). Fifty-four percent (264) of the 489 patients donated a total of 527 units of blood (average, 2.0 units per patient) preoperatively.

Results: One hundred and ninety-one patients (39 percent) required a transfusion of autologous blood or allogenic blood, or both. One hundred and thirty-one patients (27 percent) received autologous blood, and eighty-two patients (17 percent) received a transfusion of allogenic blood; twenty-two patients (4 percent) received both autologous and allogenic blood. Neither form of transfusion caused serious complications. Fifty-six percent (295) of the 527 units of autologous blood were discarded.

Autologous donation significantly decreased the requirements for allogenic transfusion (relative risk, 0.1; p < 0.0001). It also caused the level of hemoglobin to decrease an average of 12.2 grams per liter from the time before donation to the time before the operation (p < 0.0001). Factors that increased the risk for allogenic transfusion were a revision knee or hip procedure or a one-stage bilateral primary knee replacement (relative risk, 5.7; p < 0.0001), an initial hemoglobin level of less than 130 grams per liter (relative risk, 5.6; p < 0.0001), and an age of sixty-five years or older (relative risk, 2.8; p = 0.02). None of the sixty-seven patients who had a primary knee or hip arthroplasty and an initial hemoglobin level of 150 grams per liter or more required an allogenic transfusion. In addition, none of the sixty-three patients who had a primary arthroplasty, an initial hemoglobin level of between 130 and less than 150 grams per liter, and an age of less than sixty-five years required an allogenic transfusion. Eighty-three percent (115) of the 138 autologous units donated by the seventy patients in these two groups were discarded. These wasted units accounted for 39 percent of the 295 discarded units for the entire study sample.

Conclusions: The efficiency of collection of autologous blood can be improved by identifying patients who have a very low risk of transfusion according to the type of arthroplasty, the initial level of hemoglobin, and age. Patients who have an initial hemoglobin level of at least 150 grams per liter or an initial hemoglobin level of between 130 and 150 grams per liter and an age of less than sixty-five years have a minimal risk of needing a transfusion during or after a primary total joint replacement. These patients should be apprised of their low risk so that they can make an informed decision regarding preoperative autologous donation.

Figures in this Article

Introduction

Introduction | Materials and Methods | Results | Discussion | References

The success of preoperative autologous donation programs in the last twenty years has established their important role in the care of patients having a total joint arthroplasty^{3-6,8,9,17,21,22,28,29,34,41,45,50,51,54-56}. Transmission of acquired immunodeficiency syndrome and viral hepatitis is a feared complication of allogenic transfusion. Despite markedly decreased viral contamination of the current allogenic blood supply, the preference of surgeons and patients for autologous blood remains very strong^5,28,29. Donation for joint replacement in the lower extremities has contributed to a marked rise in the collection of autologous blood in the United States⁵³. About one-half of this blood is not transfused^5,29. Most American blood banks and hospitals discard these units rather than incorporate them into the general blood supply^5,29,44. One reason for this waste is that surgeons are frequently inaccurate in predicting how much blood will be required for each patient^1,5,7,13,18. In addition, autologous blood has its own risks. Donation reactions (adverse events occurring as a result of blood collection) and transfusion reactions have been reported^{10,19,33,43,48,57}. As with allogenic blood, the risk of transfusion error with autologous blood can result in a hemolytic transfusion reaction^{24,27,31,47,57}. Donation-induced preoperative anemia increases the overall rates of transfusion in donors of autologous blood compared with nondonors, potentially amplifying the exposure to clerical error^16,23,57.

The National Heart, Lung, and Blood Institute Expert Panel on the Use of Autologous Blood recently recommended that patients who have less than a 10 percent likelihood of needing perioperative transfusion should not be encouraged to donate³⁷. Surprisingly, few studies of total joint replacement have determined which patients are likely to benefit from predonation of autologous blood. In the current study, we examined the perioperative variables that are independently associated with transfusion after joint arthroplasty. We hypothesized that three preoperative risk factors—the type of procedure, the initial level of hemoglobin, and the patient's age—could be analyzed both separately and in combination to identify specific groups of patients for whom transfusion was common or uncommon. The purpose of our study was to analyze how autologous donation affected the need for subsequent allogenic transfusion in these cohorts and to define more specific indications for autologous donation before total joint replacement.

*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.

†Department of Orthopaedics, University of Nebraska Medical Center/Omaha Veterans Administration Medical Center, 600 South 42nd Street, Omaha, Nebraska 68198-1080.

‡American Red Cross Midwest Regional Blood Services, 3838 Dewey Avenue, Omaha, Nebraska 68105-1196.

*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.

†Department of Orthopaedics, University of Nebraska Medical Center/Omaha Veterans Administration Medical Center, 600 South 42nd Street, Omaha, Nebraska 68198-1080.

‡American Red Cross Midwest Regional Blood Services, 3838 Dewey Avenue, Omaha, Nebraska 68105-1196.

View Larger

TABLE I: INDICATIONS FOR PRIMARY AND REVISION TOTAL JOINT ARTHROPLASTY

*The numbers in parentheses represent the number of patients who had the indication.

Type of Operation	Total No. of Patients	Indication*
Osteo- arthritis	Rheumatoid Arthritis	Post- traumatic Arthritis	Hemophilic Arthropathy	Avascular Necrosis	Congenital Dysplasia	Acute Fracture	Mechanial Failure or Loosening	Infectionm	Peri- prosthetic Fracture
Primary
Total knee arthroplasty	157	94% (147)	4% (6)	1% (2)	1% (2)	-	-	-
One-stage bilateral total knee arthroplasty	29	100% (29)	-	-	-	-	-	-
Total hip arthroplasty	163	75% (122)	2% (3)	6% (9)	-	10% (17)	7% (11)	1% (1)
Revision
Total knee arthroplasty	32	-	-	-	-	-	-	-	94% (30)	3% (1)	3% (1)
Total hip arthroplasty	108	-	-	-	-	-	-	-	89% (96)	8% (9)	3% (3)

Add to My JBJS

TABLE I: INDICATIONS FOR PRIMARY AND REVISION TOTAL JOINT ARTHROPLASTY

*The numbers in parentheses represent the number of patients who had the indication.

Type of Operation	Total No. of Patients	Indication*
Osteo- arthritis	Rheumatoid Arthritis	Post- traumatic Arthritis	Hemophilic Arthropathy	Avascular Necrosis	Congenital Dysplasia	Acute Fracture	Mechanial Failure or Loosening	Infectionm	Peri- prosthetic Fracture
Primary
Total knee arthroplasty	157	94% (147)	4% (6)	1% (2)	1% (2)	-	-	-
One-stage bilateral total knee arthroplasty	29	100% (29)	-	-	-	-	-	-
Total hip arthroplasty	163	75% (122)	2% (3)	6% (9)	-	10% (17)	7% (11)	1% (1)
Revision
Total knee arthroplasty	32	-	-	-	-	-	-	-	94% (30)	3% (1)	3% (1)
Total hip arthroplasty	108	-	-	-	-	-	-	-	89% (96)	8% (9)	3% (3)

View Larger

TABLE II: PREVALENCE OF COMORBID CONDITIONS AMONG THE FOUR HUNDRED AND EIGHTY-NINE PATIENTS

*The numbers in parentheses represent the number of patients who had the condition.

Comorbid Condition*	Prevalence
Hypertension	26% (128)
History of cardiac disease	14% (70)
Diabetes mellitus	7% (36)
History of cancer	5% (26)
Chronic obstructive pulmonary disease	5% (24)
Hypothyroidism	5% (23)
Rheumatoid arthritis	2% (11)

Add to My JBJS

TABLE II: PREVALENCE OF COMORBID CONDITIONS AMONG THE FOUR HUNDRED AND EIGHTY-NINE PATIENTS

*The numbers in parentheses represent the number of patients who had the condition.

Comorbid Condition*	Prevalence
Hypertension	26% (128)
History of cardiac disease	14% (70)
Diabetes mellitus	7% (36)
History of cancer	5% (26)
Chronic obstructive pulmonary disease	5% (24)
Hypothyroidism	5% (23)
Rheumatoid arthritis	2% (11)

View Larger

TABLE III: TRANSFUSION REQUIREMENTS ACCORDING TO THE TYPE OF OPERATION

*The values are given as the average and the standard deviation.

	All Patients (N = 489)	Total Knee Arthroplasty			Total Hip Arthroplasty
Primary (N = 157)	Revision (N = 32)	Bilateral (N = 29)	Primary (N = 163)	Revision (N = 108)
Age* (yrs.)	65.0 ± 12.5	67.0 ± 9.7	69.8 ± 10.1	71.8 ± 6.7	60.8 ± 13.8	65.2 ± 13.9
Male:female ratio	290:199 (59%/41%)	117:40 (75%/25%)	14:18 (44%/56%)	11:18 (38%/62%)	94:69 (58%/42%)	54:54 (50%/50%)
Participation in autologousblood-donation program
No. of patients	264 (54%)	72 (46%)	13 (41%)	21 (72%)	99 (61%)	59 (55%)
Total no. of units	527	132	27	46	201	121
Average no. of units	2.0	1.8	2.1	2.2	2.0	2.1
No. of units utilized	232 (44%)	50 (38%)	15 (56%)	37 (80%)	73 (36%)	57 (47%)
Predonation hemoglobin level
No. of patients available for analysis	149	45	2	13	70	19
Time that level was measured (days preop.)	22	20	15	20	22	25
Level* (g/L)	138.9 ± 14.7	141.5 ± 14.2	120.5 ± 0.7	141.1 ± 14.6	137.4 ± 15.2	138.8 ± 13.8
Preop. hemoglobin level
No. of patients available for analysis	484	156	31	29	162	106
Time that level was measured (days preop.)	6	5	5	6	6	7
Level* (g/L)	131.3 ± 16.4	137.0 ± 16.0	130.0 ± 14.0	128.0 ± 13.0	129.4 ± 17.0	127.0 ± 16.0
Type of anesthesia(no. of patients)
General	179 (44%)	53 (44%)	18 (64%)	6 (25%)	49 (35%)	53 (55%)
General and epidural	204 (50%)	50 (41%)	9 (32%)	18 (75%)	86 (61%)	41 (43%)
Epidural and intra- venous sedation	18 (4%)	11 (9%)	1 (4%)	0	5 (4%)	1 (1%)
Spinal, epidural, and intravenous sedation	8 (2%)	7 (6%)	0	0	0	1 (1%)
Estimated blood loss* (ml)	480 ± 437	185 ± 133	310 ± 308	340 ± 211	503 ± 339	844 ± 556
Use of Cell Saver
No. of patients	65 (13%)	0	2 (6%)	3 (10%)	8 (5%)	52 (48%)
Volume used* (ml)	398 ± 292	—	125 ± 35	198 ± 137	403 ± 203	419 ± 308
Drain output* (ml)	439 ± 318	491 ± 352	407 ± 311	718 ± 406	381 ± 223	401 ± 328
Transfusion performed(no. of patients)
Allogenic	82 (17%)	9 (6%)	9 (28%)	10 (34%)	18 (11%)	36 (33%)
Autologous	131 (27%)	29 (18%)	9 (28%)	17 (59%)	42 (26%)	34 (31%)

Add to My JBJS

TABLE III: TRANSFUSION REQUIREMENTS ACCORDING TO THE TYPE OF OPERATION

*The values are given as the average and the standard deviation.

	All Patients (N = 489)	Total Knee Arthroplasty			Total Hip Arthroplasty
Primary (N = 157)	Revision (N = 32)	Bilateral (N = 29)	Primary (N = 163)	Revision (N = 108)
Age* (yrs.)	65.0 ± 12.5	67.0 ± 9.7	69.8 ± 10.1	71.8 ± 6.7	60.8 ± 13.8	65.2 ± 13.9
Male:female ratio	290:199 (59%/41%)	117:40 (75%/25%)	14:18 (44%/56%)	11:18 (38%/62%)	94:69 (58%/42%)	54:54 (50%/50%)
Participation in autologousblood-donation program
No. of patients	264 (54%)	72 (46%)	13 (41%)	21 (72%)	99 (61%)	59 (55%)
Total no. of units	527	132	27	46	201	121
Average no. of units	2.0	1.8	2.1	2.2	2.0	2.1
No. of units utilized	232 (44%)	50 (38%)	15 (56%)	37 (80%)	73 (36%)	57 (47%)
Predonation hemoglobin level
No. of patients available for analysis	149	45	2	13	70	19
Time that level was measured (days preop.)	22	20	15	20	22	25
Level* (g/L)	138.9 ± 14.7	141.5 ± 14.2	120.5 ± 0.7	141.1 ± 14.6	137.4 ± 15.2	138.8 ± 13.8
Preop. hemoglobin level
No. of patients available for analysis	484	156	31	29	162	106
Time that level was measured (days preop.)	6	5	5	6	6	7
Level* (g/L)	131.3 ± 16.4	137.0 ± 16.0	130.0 ± 14.0	128.0 ± 13.0	129.4 ± 17.0	127.0 ± 16.0
Type of anesthesia(no. of patients)
General	179 (44%)	53 (44%)	18 (64%)	6 (25%)	49 (35%)	53 (55%)
General and epidural	204 (50%)	50 (41%)	9 (32%)	18 (75%)	86 (61%)	41 (43%)
Epidural and intra- venous sedation	18 (4%)	11 (9%)	1 (4%)	0	5 (4%)	1 (1%)
Spinal, epidural, and intravenous sedation	8 (2%)	7 (6%)	0	0	0	1 (1%)
Estimated blood loss* (ml)	480 ± 437	185 ± 133	310 ± 308	340 ± 211	503 ± 339	844 ± 556
Use of Cell Saver
No. of patients	65 (13%)	0	2 (6%)	3 (10%)	8 (5%)	52 (48%)
Volume used* (ml)	398 ± 292	—	125 ± 35	198 ± 137	403 ± 203	419 ± 308
Drain output* (ml)	439 ± 318	491 ± 352	407 ± 311	718 ± 406	381 ± 223	401 ± 328
Transfusion performed(no. of patients)
Allogenic	82 (17%)	9 (6%)	9 (28%)	10 (34%)	18 (11%)	36 (33%)
Autologous	131 (27%)	29 (18%)	9 (28%)	17 (59%)	42 (26%)	34 (31%)

View Larger

TABLE IV: RATES OF AUTOLOGOUS BLOOD DONATION, ALLOGENIC TRANSFUSION, AND UTILIZATION OF AUTOLOGOUS BLOOD UNITS ACCORDING TO SINGLE AND COMBINATION-RISK-FACTOR GROUPS*

*All p values that appear in the table are significant.

Risk Factor	Total No. of Patients	No. of Patients Who Donated Autologous Blood	No. of Patients Who Received Allogenic Transfusion	Rate of Utilization of Autologous Blood Units*
Type of operation
Primary	320	53% (171 of 320)	8% (27 of 320)	37% (123 of 333)
Revision or bilateral	169	55% (93 of 169)	33% (55 of 169)	56% (109 of 194)
P value*			< 0.001	<0.001
Initial hemoglobin level
= 150 g/L	85	52% (44 of 85)	1% (1 of 85)	18% (17 of 94)
130 to < 150 g/L	222	53% (117 of 222)	12% (26 of 222)	36% (84 of 233)
<130 g/L	178	57% (102 of 178)	29% (52 of 178)	64% (127 of 199)
P value for = 150 g/L vs. 130 to < 150 g/L			<0.01	0.033
P value for = 150 g/L vs. < 130 g/L			<0.001	<0.0001
Age
< 65 yrs.	193	64% (123 of 193)	6% (12 of 193)	32% (77 of 241)
=65 yrs.	296	48% (141 of 296)	24% (70 of 296)	54% (155 of 286)
P value*			<0.001	<0.001
Primary total joint arthroplasty and initial hemoglobin level of = 150 g/L	67	46% (31 of 67)	0	14% (9 of 64)
Primary total joint arthroplasty and initial hemoglobin level of 130 to < 150 g/L	150	52% (78 of 150)	5% (7 of 150)	28% (42 of 151)
Primary total joint arthroplasty, initial hemoglobin level of 130 to < 150 g/L, and age of <65 yrs.	63	62% (39 of 63)	0	19% (14 of 74)

Add to My JBJS

TABLE IV: RATES OF AUTOLOGOUS BLOOD DONATION, ALLOGENIC TRANSFUSION, AND UTILIZATION OF AUTOLOGOUS BLOOD UNITS ACCORDING TO SINGLE AND COMBINATION-RISK-FACTOR GROUPS*

*All p values that appear in the table are significant.

Risk Factor	Total No. of Patients	No. of Patients Who Donated Autologous Blood	No. of Patients Who Received Allogenic Transfusion	Rate of Utilization of Autologous Blood Units*
Type of operation
Primary	320	53% (171 of 320)	8% (27 of 320)	37% (123 of 333)
Revision or bilateral	169	55% (93 of 169)	33% (55 of 169)	56% (109 of 194)
P value*			< 0.001	<0.001
Initial hemoglobin level
= 150 g/L	85	52% (44 of 85)	1% (1 of 85)	18% (17 of 94)
130 to < 150 g/L	222	53% (117 of 222)	12% (26 of 222)	36% (84 of 233)
<130 g/L	178	57% (102 of 178)	29% (52 of 178)	64% (127 of 199)
P value for = 150 g/L vs. 130 to < 150 g/L			<0.01	0.033
P value for = 150 g/L vs. < 130 g/L			<0.001	<0.0001
Age
< 65 yrs.	193	64% (123 of 193)	6% (12 of 193)	32% (77 of 241)
=65 yrs.	296	48% (141 of 296)	24% (70 of 296)	54% (155 of 286)
P value*			<0.001	<0.001
Primary total joint arthroplasty and initial hemoglobin level of = 150 g/L	67	46% (31 of 67)	0	14% (9 of 64)
Primary total joint arthroplasty and initial hemoglobin level of 130 to < 150 g/L	150	52% (78 of 150)	5% (7 of 150)	28% (42 of 151)
Primary total joint arthroplasty, initial hemoglobin level of 130 to < 150 g/L, and age of <65 yrs.	63	62% (39 of 63)	0	19% (14 of 74)

View Larger

TABLE V: EFFECT OF AUTOLOGOUS DONATION ON RATES OF ALLOGENIC TRANSFUSION IN SINGLE AND COMBINATION-RISK-FACTOR GROUPS

*P values that are significant are shown in boldface type.

Risk Factor	Total No. of Patients	No. of Patients Who Received Allogenic Transfusion		P Value*
Nondonors	Donors
Type of operation
Primary	320	15% (22 of 149)	3% (5 of 171)	<0.001
Revision or bilateral	169	49% (37 of 76)	19% (18 of 93)	<0.001
Initial hemoglobin level
= 150 g/L	85	2% (1 of 41)	0% (0 of 44)	0.24
130 to <150 g/L	222	18% (19 of 105)	6% (7 of 117)	0.005
<130 g/L	178	49% (37 of 76)	15% (15 of 102)	<0.001
Age
<65 yrs.	193	9% (6 of 70)	5% (6 of 123)	0.36
=65 yrs.	296	34% (53 of 155)	12% (17 of 141)	<0.001
Primary total joint arthroplasty and initial hemoglobin level of =150 g/L	67	0% (0 of 36)	0% (0 of 31)
Primary total joint arthroplasty and initial hemoglobin level of 130 to <150 g/L	150	10% (7 of 72)	0 of 78	0.005
Primary total joint arthroplasty, initial hemoglobin level of 130 to <150 g/L, and age of <65 yrs.	63	0 of 24	0 of 39

Add to My JBJS

TABLE V: EFFECT OF AUTOLOGOUS DONATION ON RATES OF ALLOGENIC TRANSFUSION IN SINGLE AND COMBINATION-RISK-FACTOR GROUPS

*P values that are significant are shown in boldface type.

Risk Factor	Total No. of Patients	No. of Patients Who Received Allogenic Transfusion		P Value*
Nondonors	Donors
Type of operation
Primary	320	15% (22 of 149)	3% (5 of 171)	<0.001
Revision or bilateral	169	49% (37 of 76)	19% (18 of 93)	<0.001
Initial hemoglobin level
= 150 g/L	85	2% (1 of 41)	0% (0 of 44)	0.24
130 to <150 g/L	222	18% (19 of 105)	6% (7 of 117)	0.005
<130 g/L	178	49% (37 of 76)	15% (15 of 102)	<0.001
Age
<65 yrs.	193	9% (6 of 70)	5% (6 of 123)	0.36
=65 yrs.	296	34% (53 of 155)	12% (17 of 141)	<0.001
Primary total joint arthroplasty and initial hemoglobin level of =150 g/L	67	0% (0 of 36)	0% (0 of 31)
Primary total joint arthroplasty and initial hemoglobin level of 130 to <150 g/L	150	10% (7 of 72)	0 of 78	0.005
Primary total joint arthroplasty, initial hemoglobin level of 130 to <150 g/L, and age of <65 yrs.	63	0 of 24	0 of 39

Materials and Methods

Introduction | Materials and Methods | Results | Discussion | References

We requested the charts of 495 consecutive patients who had had a total joint arthroplasty performed at one of two institutions by one of two surgeons during a thirty-six-month period (between February 1, 1994, and January 31, 1997). Six charts were not available at the time of this study, limiting the review to 489 patients. Individual consent for participation in the study was not required because of its retrospective nature.

The operations included 247 total knee arthroplasties (157 unilateral primary, thirty-two revision, and twenty-nine one-stage bilateral primary procedures) and 271 total hip arthroplasties (163 primary and 108 revision procedures). The analyzed documents included the preoperative history and physical examination report, operative report, anesthesia record, nursing notes, transfusion records, and physician's progress notes. The preoperative data included the patient's age and gender, the indications for the operation, the presence of comorbid conditions, and whether the patient had participated in an autologous blood-donation program. The average age (and standard deviation) of the 489 patients was 65 ± 12.5 years. Two hundred and ninety patients (59 percent) were men, and 199 (41 percent) were women. The predonation and preoperative levels of hemoglobin also were analyzed. The initial hemoglobin level was defined as the first level that was measured, whether it was predonation or preoperative. We believed that use of this hemoglobin value for donors and nondonors of autologous blood would ensure that the two groups of patients were properly matched, independent of the potential phlebotomy-induced anemic effect of donation.

The indications for primary total knee arthroplasty (both unilateral and bilateral) and primary total hip arthroplasty included osteoarthritis, rheumatoid arthritis, posttraumatic arthritis, hemophilic arthropathy, avascular necrosis, congenital dysplasia, and acute fracture (Table I). The indications for revision total knee and hip arthroplasty included mechanical failure or loosening, infection, and periprosthetic fracture (Table I). Comorbid conditions included hypertension, a history of cardiac disease or cancer, diabetes mellitus, chronic obstructive pulmonary disease, hypothyroidism, and rheumatoid arthritis (Table II).

Before the operation, patients had been advised to donate blood at a local blood bank. The criteria for autologous donation included an absence of active infection or of uncontrolled cardiac disease and a preparticipation screening hemoglobin value of at least 110 grams per liter. Phlebotomy typically was performed weekly, although in some patients it was performed at three to five-day intervals. A patient was not allowed to donate if the hemoglobin value was less than 110 grams per liter at any time during the blood-acquisition process. No patient received human recombinant erythropoietin at any time¹⁴.

Two hundred and sixty-four patients donated blood for possible autologous transfusion. Oral administration of iron (ferrous sulfate) supplements, 325 milligrams three times per day, was initiated before donation. Predonation red blood-cell indices were available for 149 patients (56 percent) who donated at the Red Cross center. Both hemoglobin and hematocrit values were obtained for ninety-one (61 percent) of these patients before the first autologous donation, but only the hematocrit was determined for fifty-eight (39 percent). For these fifty-eight patients, a hemoglobin level was estimated by dividing the hematocrit value by three. The Red Cross records also were carefully reviewed for the occurrence of reactions to autologous donation.

Preoperative hemoglobin levels were available for 484 of the 489 patients. Other data that were analyzed included the type of anesthesia that was used and the estimated blood loss during the procedure. Intraoperative cell salvage (Cell Saver; Haemonetics, Braintree, Massachusetts) was performed for sixty-five patients, and reinfusion volumes were also included in the data analysis. Postoperative data included the total drain output, the indications for the transfusion, and the total transfusion requirements (Table III). In accordance with the recently established guidelines for red blood-cell transfusion^41,49, an absolute indication for transfusion was an intraoperative or postoperative hemoglobin level of less than seventy grams per liter. Most of the patients, however, received the transfusion after identification of clinically based signs and laboratory findings that were consistent with symptomatic anemia. Symptomatic anemia was defined as a low level of hemoglobin (less than 120 grams per liter) associated with at least one of the following: persistent tachycardia that was refractory to intravenous administration of fluids, orthostatic hypotension, dyspnea on exertion, and profound fatigability that precluded participation in physical therapy. Elderly patients and patients who had preexisting myocardial disease were thought to have lower physiological reserve; thus, less stringent criteria for transfusion tended to be used for them. Full informed consent for blood transfusion was obtained preoperatively.

Statistical Analysis

Logistic regression analysis was used to examine the role of several factors (the type of procedure, the patient's age and gender, whether the patient had donated blood preoperatively, the preoperative hemoglobin level, comorbid conditions, the type of anesthesia, the estimated blood loss, the volume of intraoperative cell salvage returned, and postoperative drain outputs) in determining the requirements for allogenic transfusion. For the analysis of the type of procedure, primary procedures were compared with bilateral and revision operations. For the comparison of continuous variables, the Student t test was used. For the comparison of more than two groups, analysis of variance was performed with the Student-Newman-Keuls procedure for multiple comparisons. The chi-square test for association and Fisher's exact test were used to compare categorical variables. P values that were less than 0.05 were considered significant. For p values that were greater than 0.05, a power analysis was performed. Power calculations of 80 percent or greater were considered adequate to confirm a lack of association.

Results

Introduction | Materials and Methods | Results | Discussion | References

Transfusion Requirements, Indications, and Complications

Fifty-four percent (264) of the 489 patients donated a total of 527 units of blood (average, 2.0 units per patient) for possible autologous transfusion. One hundred and ninety-one patients (39 percent) required a transfusion of autologous or allogenic blood, or both, and 229 transfusions (134 autologous and ninety-five allogenic) were performed. One hundred and nine patients (22 percent) received autologous blood only, sixty (12 percent) received allogenic blood only, and twenty-two (4 percent) required a transfusion of both forms of blood. Eighty-two patients (17 percent) had a transfusion of allogenic blood (total, 189 units). Two hundred and ninety-eight (61 percent) of the 489 patients and 132 (50 percent) of the 264 donors did not require a transfusion of either autologous or allogenic blood. Overall, 56 percent (295) of the 527 units of autologous blood were discarded. Eighty-nine percent (263) of the 295 units of discarded blood had been collected from the 132 donors who did not require a transfusion.

The hemoglobin level was available at the time of the transfusion decision for 109 (81 percent) of the 134 autologous transfusions and seventy (74 percent) of the ninety-five allogenic transfusions. The average hemoglobin level at the time of the 106 initial autologous transfusions (84.0 ± 10.5 grams per liter) was not significantly different from that at the sixty initial allogenic transfusions (83.0 ± 12.4 grams per liter) (p = 0.42, with an 80 percent power to detect a difference of five grams per liter). At the second autologous transfusion (three patients), the pre-transfusion hemoglobin level averaged 80.0 ± 2.0 grams per liter. At the second allogenic transfusion (ten patients), the pre-transfusion hemoglobin level averaged 84.5 ± 7.2 grams per liter. There was no significant difference in the hemoglobin level between the autologous and allogenic transfusions, regardless of the time when the transfusions were performed (p = 0.59, with an 80 percent power to detect a difference of six grams per liter). Clinical signs of anemia, such as tachycardia and hypotension, also were not significantly different between the allogenic and autologous transfusions, regardless of the time when the transfusions were performed (an 80 percent power to detect a difference of seven beats per minute in the pulse or a difference of eight millimeters of mercury [1.07 kilopascals] in the systolic or diastolic blood pressure).

Neither autologous nor allogenic transfusion caused serious complications in the patients who were studied. According to Red Cross and hospital records, there were no minor complications directly attributable to autologous blood donation. Postoperatively, eleven patients had a fever of more than 38.5 degrees Celsius as measured orally, within twenty-four hours after the transfusion. Six fevers occurred within twenty-four hours after an autologous transfusion and were attributable to atelectasis. Five febrile episodes occurred within twenty-four hours after an allogenic transfusion; four of the five were attributable to atelectasis, and one was diagnosed as a nonhemolytic febrile transfusion reaction. The overall rate of nonhemolytic febrile reactions to allogenic blood was therefore 1.2 percent (one of eighty-two recipients of allogenic blood). There were no bacterial, hemolytic, or sickle-cell transfusion reactions.

Effect of Blood Loss, Initial Hemoglobin Level, and Autologous Donation on Transfusion Requirements

Patients who did not require transfusion lost less blood intraoperatively (423 ± 372 milliliters) than those who required allogenic blood only (658 ± 411 milliliters; p < 0.002) or those who required both autologous and allogenic transfusion (909 ± 956 milliliters; p < 0.0001). Patients who required only autologous blood lost an average of 475 ± 370 milliliters of blood intraoperatively. Patients who did not require transfusion were younger (average age, 63.0 ± 12.5 years) than those who required allogenic blood only (average age, 73.3 ± 11.0 years; p < 0.0001) and those who required both autologous and allogenic transfusion (average age, 70.0 ± 8.7 years; p = 0.013). Drain outputs were not significantly different among the three groups (423 ± 308, 495 ± 338, and 463 ± 374 milliliters; p = 0.19 for all comparisons, with an 80 percent power to detect a difference of 225 milliliters).

Patients who did not require transfusion had a higher average preoperative hemoglobin level (137.2 ± 12.2 grams per liter; 297 patients) than those who required autologous blood only (120.6 ± 13.4 grams per liter; 108 patients), those who received allogenic blood only (125.5 ± 16.0 grams per liter; fifty-eight patients), and those who received both forms of transfusion (119.6 ± 10.9 grams per liter; twenty-one patients) (p < 0.0001 for all comparisons).

Initial predonation hemoglobin levels were available for 56 percent (149) of the 264 donors. A preoperative hemoglobin level was not available for four of these 149 patients. Hemoglobin levels decreased from 139.0 ± 14.7 grams per liter (in 149 patients) before the first donation (twenty-two days before the operation) to 126.8 ± 14.3 grams per liter (in 145 patients) six days before the operation (p < 0.0001) in the group for which the predonation hemoglobin level was available. Preoperative hemoglobin levels were lower in the 261 donors than in the 223 nondonors (127.2 ± 14.7 grams per liter compared with 136.1 ± 17.1 grams per liter; p < 0.0001). Donors nonetheless had a significantly decreased rate of allogenic transfusion than did nondonors (9 percent [twenty-three] of 264 compared with 26 percent [fifty-nine] of 225; p < 0.0001). Fifty percent (132) of the 264 donors required an autologous transfusion (a total of 234 units). Fifty percent (133) of the 264 donors were men compared with 70 percent (157) of the 225 nondonors (p < 0.001).

Logistic Regression Analysis

Multivariate analysis revealed that autologous donation significantly decreased the risk for allogenic transfusion (relative risk, 0.1; p < 0.0001). Revision and one-stage bilateral joint replacement were associated with an increased risk for allogenic transfusion (relative risk, 5.7; p < 0.0001). Other factors associated with an increased risk for allogenic transfusion included a preoperative hemoglobin level of less than 130 grams per liter (relative risk, 5.6; p < 0.0001), an age of sixty-five years or older (relative risk, 2.8; p = 0.02), and an increased estimated intraoperative blood loss (per milliliter increase in relative risk, 0.001; p < 0.0001). Neither the type of anesthesia nor the presence of a comorbid condition had a significant effect on the requirement for allogenic transfusion (p > 0.05 for both variables), although the power to detect such associations was low (an 80 percent power to detect a relative risk of 2.3 to 6.2).

Individual Risk-Factor Analysis According to Type of Procedure

Patients who had a primary total knee arthroplasty had a lower rate of allogenic transfusion (6 percent [nine of 157]) than did those who had a revision total knee arthroplasty (28 percent [nine of thirty-two]) or those who had a bilateral total knee arthroplasty (34 percent [ten of twenty-nine]) (p < 0.001 for both comparisons). Sixty-two percent (eighty-two) of the 132 autologous units donated before primary total knee arthroplasties were discarded compared with 44 percent (twelve) of the twenty-seven units donated before revision knee replacements (p < 0.041) and only 20 percent (nine) of the forty-six units donated before bilateral total knee arthroplasties (p < 0.0005). More blood was lost intraoperatively during the revision knee replacements (average, 310 ± 308 milliliters; median, 250 milliliters; p = 0.004) and the bilateral knee replacements (average, 340 ± 211 milliliters; median, 250 milliliters; p = 0.0002) than was lost during the primary knee replacements (average, 185 ± 133 milliliters; median, 150 milliliters). Patients who had a bilateral total knee arthroplasty had a significantly higher postoperative drain output (average, 718 ± 406 milliliters; median, 750 milliliters) than did those who had a revision (average, 407 ± 311 milliliters; median, 350 milliliters; p < 0.0035) and those who had a primary total knee arthroplasty (average, 491 ± 352 milliliters; median, 475 milliliters; p < 0.0062). The postoperative drain outputs after revision and primary procedures were not significantly different (p = 0.31, with an 80 percent power to detect a difference of 194 milliliters) (Table III).

The rate of allogenic transfusion in the patients who had a primary total hip arthroplasty (11 percent [eighteen of 163]) was lower than that in the patients who had a revision total hip arthroplasty (33 percent [thirty-six of 108]) (p < 0.001). One hundred and twenty-eight (64 percent) of 201 units that were donated before primary total hip arthroplasties were discarded compared with sixty-four (53 percent) of 121 units that were donated before revision hip replacements (p < 0.018). Patients who had a revision total hip arthroplasty had a significantly higher intraoperative blood loss (average, 844 ± 556 milliliters; median, 750 milliliters) than did those who had a primary total hip arthroplasty (average, 503 ± 339 milliliters; median, 400 milliliters) (p < 0.0001). There was no significant difference, between the patients who had a primary hip procedure and those who had a revision, in the rate of participation in an autologous blood-donation program (61 percent [ninety-nine] of 163 compared with 55 percent [fifty-nine] of 108; p = 0.15, with an 80 percent power to detect a difference of 17 percent) or in the postoperative drain output (average, 381 ± 223 milliliters compared with 401 ± 328 milliliters; p = 0.92, with an 80 percent power to detect a difference of ninety-six milliliters). Intraoperative cell salvage was used for 48 percent (fifty-two) of the 108 revision procedures compared with only 5 percent (eight) of the 163 primary procedures (p < 0.001) (Table III).

Combination of the data for the knee and hip replacements revealed that 8 percent (twenty-seven) of the 320 patients who had a primary procedure required an allogenic transfusion compared with 33 percent (fifty-five) of the 169 who had a revision or bilateral procedure (p < 0.001) (Table IV). Two hundred and ten (63 percent) of the 333 autologous units that were donated before primary total joint replacements were discarded compared with eighty-five (44 percent) of the 194 units that were donated before revision or bilateral total joint replacements (p < 0.001) (Table IV). The estimated blood loss during the primary joint arthroplasties (average, 364 ± 312 milliliters) was significantly lower than that during the revision or bilateral replacements (average, 675 ± 538 milliliters) (p < 0.0001). The patients who had a primary total joint replacement were slightly younger (average age, 63.9 ± 12.3 years) than those who had a revision or bilateral procedure (average age, 67.2 ± 12.5 years) (p = 0.0045). Intraoperative cell salvage was used more often in revision and bilateral procedures (used in fifty-seven [34 percent] of 169 patients) than in primary procedures (used in eight [3 percent] of 320 patients) (p < 0.001). The use of the Cell Saver in revision procedures was not associated with a significant decrease in the requirement for allogenic transfusion in either nondonors (p = 0.63) or donors (p = 0.57), but the power to detect such associations was low (an 80 percent power to detect a difference of 28 percent).

Initial Level of Hemoglobin and Age

Of the patients who had an initial hemoglobin concentration of 150 grams per liter or more, only 1 percent (one of eighty-five) required an allogenic transfusion. This rate was significantly lower than that for patients who had a hemoglobin level of 130 to less than 150 grams per liter (12 percent [twenty-six of 222]; p < 0.01) and that for those who had a hemoglobin level of less than 130 grams per liter (29 percent [fifty-two of 178]; p < 0.001) (Table IV). No patient who had an initial hemoglobin level of 130 to less than 150 grams per liter, had a primary arthroplasty, and was less than sixty-five years old required an allogenic transfusion. The autologous blood discard rate was highest in the group of patients who had an initial hemoglobin level of 150 grams per liter or more (82 percent [seventy-seven of ninety-four units]). The discard rates were lower for patients who had an initial hemoglobin level of 130 to less than 150 grams per liter (64 percent [149 of 233 units]; p = 0.033) and those who had a level of less than 130 grams per liter (36 percent [seventy-two of 199 units]; p < 0.0001) (Table IV). Men accounted for 37 percent (sixty-six) of the 178 patients who had a hemoglobin level of less than 130 grams per liter. The percentage of men was higher in the group that had a hemoglobin level of 130 to less than 150 grams per liter (65 percent [145 of 222]; p < 0.001) and in the group that had a level of more than 150 grams per liter (91 percent [seventy-seven of eighty-five]; p < 0.0001).

Patients who were less than sixty-five years old had a significantly lower rate of allogenic transfusion (6 percent [twelve of 193]) than did those who were sixty-five years old or more (24 percent [seventy of 296]; p < 0.001) (Table IV). Sixty-eight percent (164) of the 241 autologous units that were donated by patients less than sixty-five years old were discarded compared with 46 percent (131) of the 286 units that were donated by patients sixty-five years old or more (p < 0.001) (Table IV). The percentage of patients less than sixty-five years old who had a primary procedure (74 percent [143 of 193]) was greater than that of patients who were sixty-five years old or more (60 percent [177 of 296]) (p < 0.001). There were no significant differences between the two age-groups with regard to male:female ratio (p = 0.51, with an 80 percent power to detect a difference of 12 percent), estimated blood loss (p = 0.17, with an 80 percent power to detect a difference of 113 milliliters), postoperative drain output (p = 0.093, with an 80 percent power to detect a difference of eighty-three milliliters), percentage of procedures in which the Cell Saver was used (p = 0.89, with an 80 percent power to detect a difference of 10 percent), and preoperative hemoglobin level (p = 0.55, with an 80 percent power to detect a difference of five grams per liter).

Autologous blood donation significantly decreased the need for allogenic transfusion in all single-risk-factor groups except for the group with an initial hemoglobin level of 150 grams per liter or more and the group that was less than sixty-five years old. Autologous donation was particularly beneficial for patients undergoing revision hip and knee or bilateral knee arthroplasty, patients sixty-five years old or older, and patients with a hemoglobin level of less than 130 grams per deciliter (Table V).

Risk-Factor Combinations and Effect of Autologous Donation

Patients who had any of three combinations of risk factors were hypothesized to have a very low risk of requiring a transfusion. These combinations included primary arthroplasty and an initial hemoglobin level of 150 grams per liter or more; primary arthroplasty and an initial hemoglobin level of 130 to less than 150 grams per liter; and primary arthroplasty, an initial hemoglobin level of 130 to less than 150 grams per liter, and an age of less than sixty-five years. These three groups of patients had very low rates of allogenic transfusion, and a high percentage of the autologous blood units in these groups were discarded (Table IV). None of sixty-seven patients (thirty-six nondonors and thirty-one donors) who had a primary total joint replacement and an initial hemoglobin level of 150 grams per liter or more required an allogenic transfusion. None of the thirty-six nondonors required a transfusion of any type. Sixteen percent (five) of the thirty-one donors had an autologous transfusion. Eighty-six percent (fifty-five) of the sixty-four autologous units that were collected from these patients were discarded (Table IV). The thirty-one donors had a lower average preoperative level of hemoglobin (142.7 ± 11.9 grams per liter) than did the thirty-six nondonors (159.7 ± 9.1 grams per liter) (p < 0.001). Ninety-three percent (sixty-two) of the sixty-seven donors and nondonors were men.

Five percent (seven) of the 150 patients who had a primary total joint replacement and a preoperative hemoglobin level of 130 to less than 150 grams per liter required an allogenic transfusion; all seven patients were sixty-five years or older. Despite the overall low rate of allogenic transfusion, autologous blood donation significantly decreased the need for allogenic blood in this group, with 10 percent (seven) of seventy-two nondonors compared with none of seventy-eight donors requiring an allogenic transfusion (p < 0.005). This difference was negated if patients sixty-five years or older were excluded (Table V). None of the sixty-three patients who were less than sixty-five years old, had a primary arthroplasty, and had an initial hemoglobin level of 130 to less than 150 grams per liter required an allogenic transfusion. None of the twenty-four nondonors required any type of transfusion. Of the thirty-nine donors, 18 percent (seven) had an autologous transfusion. Eighty-one percent (sixty) of the seventy-four autologous units that were collected from these patients were discarded (Table IV). The thirty-nine donors had a lower preoperative hemoglobin level (130.8 ± 10.2 grams per liter) than did the twenty-four nondonors (141.6 ± 10.3 grams per liter) (p < 0.001).

Discussion

Introduction | Materials and Methods | Results | Discussion | References

Several complications are directly attributable to allogenic transfusion. Nonhemolytic febrile transfusion reactions, febrile allergic reactions, and alloimmunization reactions are most common but do not generally result in serious morbidity. More serious complications are much rarer^2,20. The most feared risk of allogenic blood, infectious contamination, has declined substantially since 1985 secondary to improved criteria for the selection of donors and the advent of sensitive screening tests for the human immunodeficiency virus (1985) and the hepatitis-C virus (1990)^{2,20,38,46,57}. The most common viral risk at this time is that of hepatitis, with a transmission rate of approximately one per 100,000 units transfused^{2,20,25,46,57}; this represents a 10,000-fold decrease from the transmission rate in 1985. The rate of transmission of the human immunodeficiency virus is also exceptionally low (one per 200,000 to two million units)^2,20,25,46. Although these viral risks are diminishing, there is the possibility that a new viral agent could contaminate the blood supply²⁰. Nonetheless, considering the diminished prevalence of viral transmission, one of the most substantial risks at present is thought to be administrative error; approximately one incorrect unit is administered for every 19,000 units transfused³⁰. Other serious complications include acute and delayed hemolytic transfusion reactions, bacterial contamination, transfusion-related lung injury, anaphylaxis, graft-versus-host disease, and Rh sensitization in women of childbearing age^11,20. Some investigators have shown that exposure to allogenic blood increases susceptibility to cancer and infection^8,15,29,36; however, others have challenged these findings^32,52.

The high rates of viral contamination of allogenic blood in the 1960s and 1970s led to the introduction of autologous blood donation before operations for the treatment of scoliosis^9,50,51. Additional investigations established the useful role of autologous blood donation in total joint replacement^{12,22,34,50,55,56}. The advantages of autologous transfusion include the elimination of viral infection, transfusion-related lung injury, anaphylaxis, graft-versus-host disease, alloimmunization, and Rh sensitization^11,21. In addition, preoperative blood donation provides a psychological benefit to patients as they feel that they are in more control of their care²⁸. Nonetheless, the collection and transfusion of autologous blood are not problem-free⁵⁷. Multiple phlebotomies can be inconvenient and time-consuming for the patient. In addition, mild donation reactions (vasovagal episodes) occur at an estimated frequency of two to five per 100 donations. Serious complications can occur but are much rarer^19,33,43,48. Popovsky et al., in a 1995 review of more than 200,000 autologous blood donations, found a one-in-16,783-unit prevalence of donation-related complications necessitating hospitalization⁴³. Vasovagal episodes accounted for most of these complications, but anginal attack, tetany, compartment syndrome, phlebitis, and vessel damage necessitating repair also occurred on occasion. Minor reactions to transfusion of autologous blood also can occur. The 1994 American Association of Blood Banks Annual Questionnaire demonstrated an overall rate of transfusion reaction of nine per 10,000 autologous transfusions⁵⁶. Although most adverse transfusion reactions are febrile nonhemolytic reactions, rare occurrences of hemolytic, septic, and sickle-cell reactions to autologous transfusion also have been reported^10,57.

Human error also applies to autologous transfusion. The Association of American Pathologists Comprehensive Transfusion Medicine Surveys of 1992 and 1993 found that forty-two (0.7 percent) of 5681 institutions reported transfusion of incorrect autologous units⁵⁷. Currently, the estimated risk of transfusion of an incorrect unit of autologous blood is one per 16,000 to 30,000 units transfused^30,31,57. We and others^{9,12,16,23,55} found that donors had a significant decrease in hemoglobin and hematocrit levels, ranging from ten to sixteen grams per liter. Donation-induced anemia can lead to higher overall rates of transfusion, theoretically amplifying the exposure of autologous donors to error^16,23,57. To our knowledge, however, there have been no reports of fatality due to the transfusion of incorrect units of autologous blood.

There is a need for increased efficiency in the collection of autologous blood^{5,21,29,37,44}. Renner et al., in a study of transfusion practices in 612 hospitals, reported an overall rate of nonutilization of autologous blood of 40 percent⁴⁴. Twenty percent (8037) of all 40,163 autologous units were donated for operations that were associated with less than a 10 percent risk that a transfusion would be needed⁴⁴. In the current series, 89 percent (263) of the 295 units of discarded blood had been collected from patients who did not require a transfusion. In a recent prospective, multicenter analysis of more than 9000 patients who had a joint replacement, Bierbaum et al. found a 45 percent rate of wastage of autologous blood (4464 of 9920 units)⁵. The current study revealed a comparably high discard rate of 56 percent (295 of 527 units). The financial burden of discarded autologous blood is borne by the patient, the hospital, and the medical system^5,7,13. This cost can potentially be reduced by better-defined indications for donation.

In a study of 299 patients who had a total hip arthroplasty, Nuttall et al. performed a stepwise regression analysis to identify those who were at high and low risk for transfusion³⁹. The authors found that the preoperative hemoglobin level, weight, estimated blood loss, and use of aspirin were factors that affected the need for allogenic transfusion. In a prospective, follow-up investigation, Nuttall et al. used these risk factors to order a more appropriate quantity of allogenic red blood cells, predicting the actual need for transfusion with greater accuracy⁴⁰. They also suggested that patients who have a primary total hip arthroplasty and a preoperative hemoglobin level of at least 150 grams per liter have only a slight likelihood of requiring a postoperative transfusion³⁹. However, Nuttall et al. did not address the impact of autologous donation on the need for transfusion^39,40. Biesma et al. showed that, for total hip replacement, donation of two units of autologous blood was as effective as donation of four units in decreasing the need for allogenic transfusion⁶. Although they recommended an overall reduction in autologous donation, the authors did not identify specific patients who had a particularly low need for transfusion. Bernstein et al. recommended that patients who had a primary total knee arthroplasty and a hemoglobin level of more than 150 grams per liter not donate blood for possible autologous transfusion⁴. Knight et al. suggested that the need for allogenic transfusion was minimum for patients who had a primary total knee arthroplasty, a preoperative hematocrit of more than 40 percent, and a weight of more than 130 pounds (fifty-nine kilograms)²⁶. It was concluded that autologous donation was of negligible benefit for these patients. Neither of these studies included patients who had a total hip arthroplasty.

Like most published reports regarding autologous blood donation before total joint replacement, the current investigation was not a prospective, randomized trial^{4,5,7,8,12,16,22,34,50,51,54-56}. Nonetheless, the number of patients was sufficient to achieve significance in the logistic regression analysis and the single-risk-factor comparisons that were performed. Predonation red blood-cell indices were available for 149 of the 264 donors, and, similar to the findings of other investigators^{3,9,12,14,16,23,38,56}, preoperative donation-induced reductions in the levels of hemoglobin were shown to be significant (p < 0.0001). The timing and amount of blood predonation were variable, with the last donation occurring anywhere from twenty-seven to four days before the operation. Blood donations within two weeks before the operation are particularly likely to cause preoperative anemia because of an insufficient erythropoietic response^24,27. Standardization of donation schedules, with the last donation occurring at least two weeks before the operation, could have minimized the rate of postdonation anemia in our study population, but this standardization would be difficult to achieve even in prospective protocols. Finally, the effect of the patient's weight, a factor used in other studies to determine the risk of transfusion^26,39,40, was not analyzed in this study. We believe that variables such as the type of procedure, the preoperative level of hemoglobin, and age are better choices for protocol parameters than weight, as they more directly reflect blood loss, oxygen-carrying capacity, and physiological reserve.

To our knowledge, this is the first study of total knee and hip arthroplasty in which patients who were at low risk for needing a transfusion were identified according to the type of procedure, the initial hemoglobin level, and age. Bierbaum et al. thought that patients with a hemoglobin level of 137 grams per liter had a lesser likelihood of requiring a transfusion in association with a total joint arthroplasty than did their more anemic counterparts⁵. The current study supports this contention, as the patients who did not require an allogenic transfusion had an average preoperative hemoglobin level of 137.2 ± 12.2 grams per liter (297 patients). The indications for autologous and allogenic blood transfusion in our study sample did not differ, as evidenced by similar levels of hemoglobin at the time of the autologous and allogenic transfusions (84.0 ± 10.5 compared with 83.0 ± 12.4 grams per liter, respectively). No patient had a recognized complication related to having or not having had a blood transfusion.

We continue to recommend autologous donation to most patients having a total joint replacement because of its beneficial effect in reducing exposure to allogenic transfusion^21,29,55. Furthermore, from an ethical and medicolegal standpoint, it is important to make autologous blood donation available to all patients having a total joint replacement, including those with a very low risk of needing a transfusion¹⁷. Nonetheless, the 50 percent wastage rate of current protocols is an issue that should be addressed and improved. In an attempt to increase the efficiency of autologous predonation programs, the National Heart, Lung, and Blood Institute Expert Panel of 1995 recommended that patients with less than a 10 percent likelihood of needing a transfusion not be urged to donate blood for possible autologous transfusion³⁷. The goal of our investigation was to determine how to identify these low-risk individuals before total joint arthroplasty.

Consideration of the type of operation, the preoperative level of hemoglobin, and age can assist the surgeon in preoperatively determining the risk of a transfusion being needed. In the current investigation, a 0 percent rate of allogenic transfusion was found in two groups of patients: those who had a primary total joint replacement and an initial hemoglobin level of at least 150 grams per liter and those who had a primary total joint replacement, an initial hemoglobin level of 130 to less than 150 grams per liter, and an age of less than sixty-five years. This minimal risk for exposure to allogenic blood in these patients was unaffected by the availability of autologous blood. We believe that these two groups of patients did not require blood because they had optimum parameters for avoiding transfusion—the lowest amount of blood loss in conjunction with the highest or a high-normal hemoglobin level and a younger age (and thus a higher cardiovascular reserve). Twenty-seven percent (130) of our 489 patients fulfilled these criteria. The discarded autologous blood that had been donated by these patients accounted for 39 percent (115) of the 295 units discarded in the entire study sample. Decreasing autologous donations from these patients affords the opportunity to improve the overall efficiency of autologous collection programs for total joint replacement—without causing an increase in exposure to allogenic blood. If a patient's proposed type of arthroplasty, initial level of hemoglobin, and age place him or her in one of these two groups, we explain to the patient that the risk of transfusion is very slight and is unlikely to be markedly affected by the donation of autologous blood. With this in mind, patients can make an adequately informed decision regarding autologous donation before a total joint replacement.

Note: The authors thank David Huberty, B.A., for his assistance with completion of this project.

References

Introduction | Materials and Methods | Results | Discussion | References

AuBuchon, J. P., and Birkmeyer, J. D.: Controversies in transfusion medicine. Is autologous blood transfusion worth the cost?. Transfusion,34: 79-83, 1994.3479 1994 [PubMed]

AuBuchon, J. P.; Birkmeyer, J. D.; and Busch, M. P.: Safety of the blood supply in the United States: opportunities and controversies. Ann. Internal Med.,127: 904-909, 1997.127904 1997

Ayers, D. C.; Murray, D. G.; and Duerr, D. M.: Blood salvage after total hip arthroplasty. J. Bone and Joint Surg.,77-A: 1347-1351, Sept. 1995.77-A1347 1995

Bernstein, L. H.; Coles, M.; and Granata, A.: The Bridgeport Hospital experience with autologous transfusion in orthopedic surgery. Orthopedics,20: 677-680, 1997.20677 1997 [PubMed]

Bierbaum, B. E.; Callaghan, J. J.; Galante, J. O.; Rubash, H. E.; Tooms, R. E.; and Welch, R. B.: An analysis of blood management in patients having a total hip or knee arthroplasty. J. Bone and Joint Surg.,81-A: 2-10, Jan. 1999.81-A2 1999

Biesma, D. H.; Marx, J. J. M.; and Van De Wiel, A.: Collection of autologous blood before elective hip replacement. A comparison of the results with the collection of two and four units. J. Bone and Joint Surg.,76-A: 1471-1475, Oct. 1994.76-A1471 1994

Birkmeyer, J. D.; Goodnough, L. T.; AuBuchon, J. P.; Noordsij, P. G.; and Littenberg, B.: The cost-effectiveness of preoperative autologous blood donation for total hip and knee replacement. Transfusion,33: 544-551, 1993.33544 1993 [PubMed]

Blumberg, N.; Kirkley, S. A.; and Heal, J. M.: A cost analysis of autologous and allogeneic transfusions in hip-replacement surgery. Am. J. Surg.,171: 324-330, 1996.171324 1996 [PubMed]

Cowell, H. R., and Swickard, J. W.: Autotransfusion in children's orthopaedics. J. Bone and Joint Surg.,56-A: 908-912, July 1974.56-A908 1974

Cregan, P.; Donegan, E.; and Gotelli, G.: Hemolytic transfusion reaction following transfusion of frozen and washed autologous red cells. Transfusion,31: 172-175, 1991.31172 1991 [PubMed]

Dzieczkowski, J., and Anderson, K.: Transfusion biology and therapy. In Harrison's Principles of Internal Medicine, edited by A. S. Fauci, E. Braumwald, K. J. Isselbacher, D. J. Wilson, J. B. Martin, D. L. Casper, S. L. Hauser, and D. L. Longo. Ed. 14, pp. 718-724. New York, McGraw-Hill, 1998.

Eckardt, J. J.; Gossett, T. C.; and Amstutz, H. C.: Autologous transfusion and total hip arthroplasty. Clin. Orthop.,132: 39-45, 1978.13239 1978 [PubMed]

Etchason, J.; Petz, L.; Keeler, E.; Calhoun, L.; Kleinman, S.; Snider, C.; Fink, A.; and Brook, R.: The cost effectiveness of preoperative autologous blood donations. New England J. Med.,332: 719-724, 1995.332719 1995

Faris, P. M.; Ritter, M. A.; Abels, R. I.; and the American Erythropoietin Study Group: The effects of recombinant human erythropoietin on perioperative transfusion requirements in patients having a major orthopaedic operation. J. Bone and Joint Surg.,78-A: 62-72, Jan. 1996.78-A62 1996

Fernandez, M. C.; Gottlieb, M.; and Menitove, J. E.: Blood transfusion and postoperative infection in orthopedic patients. Transfusion,32: 318-322, 1992.32318 1992 [PubMed]

Forgie, M. A.; Wells, P. S.; Laupacis, A.; and Fergusson, D.: Preoperative autologous donation decreases allogeneic transfusion but increases exposure to all red blood cell transfusion: results of a meta-analysis. International Study of Perioperative Transfusion (ISPOT) Investigators. Arch. Internal Med.,158: 610-616, 1998.158610 1998

Goldman, E. B.: Legal considerations for allogeneic blood transfusion. Am. J. Surg.,170 (Supplement 6A): 27S-31S, 1995.170 (Supplement 6A)27 1995

Goodnough, L. T.: The implications of cost-effectiveness for autologous blood procurement [editorial]. Arch. Pathol. and Lab. Med.,118: 333-334, 1994.118333 1994

Goodnough, L. T., and Monk, T. G.: Evolving concepts in autologous blood procurement and transfusion: case reports of perisurgical anemia complicated by myocardial infarction. Am. J. Med.,101 (Supplement 2A): 33S-37S, 1996.101 (Supplement 2A)33 1996

Goodnough, L. T.; Brecher, M. E.; Kanter, M. H.; and AuBuchon, J. P.: Transfusion medicine (first of two parts). Blood transfusion. New England J. Med.,340: 438-447, 1999.340438 1999

Goodnough, L. T.; Brecher, M. E.; Kanter, M. H.; and AuBuchon, J. P.: Transfusion medicine (second of two parts). Blood conservation. New England J. Med.,340: 525-533, 1999.340525 1999

Goulet, J. A.; Bray, T. J.; Timmerman, L. A.; Benson, D. R.; and Bargar, W. L.: Intraoperative autologous transfusion in orthopaedic patients. J. Bone and Joint Surg.,71-A: 3-8, Jan. 1989.71-A3 1989

Kanter, M. H.; van Maanen, D.; Anders, K. H.; Castro, F.; Mya, W. W.; and Clark, K.: Preoperative autologous blood donations before elective hysterectomy. J. Am. Med. Assn.,276: 798-801, 1996.276798 1996

Kickler, T. S., and Spivak, J. L.: Effect of repeated whole blood donations on serum immunoreactive erythropoietin levels in autologous donors. J. Am. Med. Assn.,260: 65-67, 1988.26065 1988

Klein, H. G.: Allogeneic transfusion risks in the surgical patient. Am. J. Surg.,170 (Supplement 6A): 21S-26S, 1995.170 (Supplement 6A)21 1995

Knight, J. L.; Sherer, D.; and Guo, J.: Blood transfusion strategies for total knee arthroplasty: minimizing autologous blood wastage, risk of homologous blood transfusion, and transfusion cost. J. Arthroplasty,13: 70-76, 1998.1370 1998 [PubMed]

Larsen, N.; Foyt, M.; and Marengo-Rowe, A.: Late donation of autologous units increases allogenic transfusion requirements. Transfusion,35 (Supplement): 24S, 1995.35 (Supplement)24 1995

Lee, S. J.; Liljas, B.; Churchill, W. H.; Popovsky, M. A.; Stowell, C. P.; Cannon, M. E.; and Johannesson, M.: Perceptions and preferences of autologous blood donors. Transfusion,38: 757-763, 1998.38757 1998 [PubMed]

Lemos, M. J., and Healy, W. L.: Current concepts review. Blood transfusion in orthopaedic operations. J. Bone and Joint Surg.,78-A: 1260-1270, Aug. 1996.78-A1260 1996

Linden, J. V.; Paul, B.; and Dressler, K. P.: A report of 104 transfusion errors in New York State. Transfusion,32: 601-606, 1992.32601 1992 [PubMed]

Linden, J. V.: Autologous blood errors and incidents. Transfusion,34 (Supplement): 28S, 1994.34 (Supplement)28 1994

McAlister, F. A.; Clark, H. D.; Wells, P. S.; and Laupacis, A.: Perioperative allogeneic blood transfusion does not cause adverse sequelae in patients with cancer: a meta-analysis of unconfounded studies. British J. Surg.,85: 171-178, 1998.85171 1998

McVay, P. A.; Andrews, A.; Kaplan, E. B.; Black, D. B.; Stehling, L. C.; Strauss, R. G.; and Toy, P. T.: Donation reactions among autologous donors. Transfusion,30: 249-252, 1990.30249 1990 [PubMed]

Mallory, T. H., and Kennedy, M.: The use of banked autologous blood in total hip replacement surgery. Clin. Orthop.,117: 254-257, 1976.117254 1976 [PubMed]

Monk, T. G.; Goodnough, L. T.; Brecher, M. E.; Pulley, D. D.; Colberg, J. W.; Andriole, G. L.; and Catalona, W. J.: Acute normovolemic hemodilution can replace preoperative autologous blood donation as a standard of care for autologous blood procurement in radical prostatectomy. Anesth. and Analg.,85: 953-958, 1997.85953 1997

Murphy, P.; Heal, J. M.; and Blumberg, N.: Infection or suspected infection after hip replacement surgery with autologous or homologous blood transfusions. Transfusion,31: 212-217, 1991.31212 1991 [PubMed]

National Heart, Lung, and Blood Institute Expert Panel on the Use of Autologous Blood: Transfusion alert: use of autologous blood. Transfusion,35: 703-711, 1995.35703 1995 [PubMed]

Nelson, C. L., and Bowen, W. S.: Total hip arthroplasty in Jehovah's Witnesses without blood transfusion. J. Bone and Joint Surg.,68-A: 350-353, March 1986.68-A350 1986

Nuttall, G. A.; Santrach, P. J.; Oliver, W. C., Jr.; Horlocker, T. T.; Shaughnessy, W. J.; Cabanela, M. E.; and Bryant, S.: The predictors of red cell transfusions in total hip arthroplasties. Transfusion,36: 144-149, 1996.36144 1996 [PubMed]

Nuttall, G. A.; Santrach, P. J.; Oliver, W. C., Jr.; Ereth, M. H.; Horlocker, T. T.; Cabanela, M. E.; Trousdale, R. T.; Bryant, S.; and Currie, T. W.: A prospective randomized trial of the surgical blood order equation for ordering red cells for total hip arthroplasty patients. Transfusion,38: 828-833, 1998.38828 1998 [PubMed]

Office of Medical Applications of Research: Consensus conference. Perioperative red blood cell transfusion. J. Am. Med. Assn.,260: 2700-2703, 1988.2602700 1988

Oishi, C. S.; D'Lima, D. D.; Morris, B. A.; Hardwick, M. E.; Berkowitz, S. D.; and Colwell, C. W., Jr.: Hemodilution with other blood reinfusion techniques in total hip arthroplasty. Clin. Orthop.,339: 132-139, 1997.339132 1997 [PubMed]

Popovsky, M. A.; Whitaker, B.; and Arnold, N. L.: Severe outcomes of allogeneic and autologous blood donation: frequency and characterization. Transfusion,35: 734-737, 1995.35734 1995 [PubMed]

Renner, S. W.; Howanitz, P. J.; and Bachner, P.: Preoperative autologous blood donation in 612 hospitals. A College of American Pathologists' Q-probes study of quality issues in transfusion practice. Arch. Pathol. and Lab. Med.,116: 613-619, 1992.116613 1992

Rollo, V. J.; Hozack, W. J.; Rothman, R. H.; Chao, W.; and Eng, K. O.: Prospective randomized evaluation of blood salvage techniques for primary total hip arthroplasty. J. Arthroplasty,10: 532-539, 1995.10532 1995 [PubMed]

Schreiber, G. B.; Busch, M. P.; Kleinman, S. H.; and Korelitz, J. J.: for the Retrovirus Epidemiology Donor Study: The risk of transfusion-transmitted viral infections. New England J. Med.,334: 1685-1690, 1996.3341685 1996

Simpson, M. B.; Georgopoulos, G.; Orsini, E.; and Eilert, R. E.: Autologous transfusions for orthopaedic procedures at a children's hospital. J. Bone and Joint Surg.,74-A: 652-658, June 1992.74-A652 1992

Spiess, B. D.; Sassetti, R.; McCarthy, R. J.; Narbone, R. F.; Tuman, K. J.; and Ivankovich, A. D.: Autologous blood donation: hemodynamics in a high-risk patient population. Transfusion,32: 17-22, 1992.3217 1992 [PubMed]

Task Force on Blood Component Therapy: Practice guidelines for blood component therapy: a report by the American Society of Anesthesiologists Task Force on Blood Component Therapy. Anesthesiology,84: 732-747, 1996.84732 1996 [PubMed]

Thomson, J. D.; Callaghan, J. J.; Savory, C. G.; Stanton, R. P.; and Pierce, R. N.: Prior deposition of autologous blood in elective orthopaedic surgery. J. Bone and Joint Surg.,69-A: 320-324, March 1987.69-A320 1987

Turner, R. S.: Autologous blood for surgical autotransfusions. In Proceedings of the American Academy of Orthopaedic Surgeons. J. Bone and Joint Surg.,50-A: 834, June 1968.50-A834 1968

Vamvakas, E. C.; Moore, S. B.; and Cabanela, M.: Blood transfusion and septic complications after hip replacement surgery. Transfusion,35: 150-156, 1995.35150 1995 [PubMed]

Wallace, E. L.; Churchill, W. H.; Surgenor, D. M.; An, J.; Cho, G.; McGurk, S.; and Murphy, L.: Collection and transfusion of blood and blood components in the United States, 1992. Transfusion,35: 802-812, 1995.35802 1995 [PubMed]

Wilson, W. J.: Intraoperative autologous transfusion in revision total hip arthroplasty. J. Bone and Joint Surg.,71-A: 8-14, Jan. 1989.71-A8 1989

Woolson, S. T.; Marsh, J. S.; and Tanner, J. B.: Transfusion of previously deposited autologous blood for patients undergoing hip-replacement surgery. J. Bone and Joint Surg.,69-A: 325-328, March 1987.69-A325 1987

Woolson, S. T., and Watt, J. M.: Use of autologous blood in total hip replacement. A comprehensive program. J. Bone and Joint Surg.,73-A: 76-80, Jan. 1991.73-A76 1991

Yomtovian, R.: Autologous transfusion complications. In Transfusion Reactions, pp. 237-280. Edited by M. Popovsky. Bethesda, Maryland, AABB Press, 1996.

View Larger

TABLE I: INDICATIONS FOR PRIMARY AND REVISION TOTAL JOINT ARTHROPLASTY

*The numbers in parentheses represent the number of patients who had the indication.

Type of Operation	Total No. of Patients	Indication*
Osteo- arthritis	Rheumatoid Arthritis	Post- traumatic Arthritis	Hemophilic Arthropathy	Avascular Necrosis	Congenital Dysplasia	Acute Fracture	Mechanial Failure or Loosening	Infectionm	Peri- prosthetic Fracture
Primary
Total knee arthroplasty	157	94% (147)	4% (6)	1% (2)	1% (2)	-	-	-
One-stage bilateral total knee arthroplasty	29	100% (29)	-	-	-	-	-	-
Total hip arthroplasty	163	75% (122)	2% (3)	6% (9)	-	10% (17)	7% (11)	1% (1)
Revision
Total knee arthroplasty	32	-	-	-	-	-	-	-	94% (30)	3% (1)	3% (1)
Total hip arthroplasty	108	-	-	-	-	-	-	-	89% (96)	8% (9)	3% (3)

Add to My JBJS

TABLE I: INDICATIONS FOR PRIMARY AND REVISION TOTAL JOINT ARTHROPLASTY

*The numbers in parentheses represent the number of patients who had the indication.

Type of Operation	Total No. of Patients	Indication*
Osteo- arthritis	Rheumatoid Arthritis	Post- traumatic Arthritis	Hemophilic Arthropathy	Avascular Necrosis	Congenital Dysplasia	Acute Fracture	Mechanial Failure or Loosening	Infectionm	Peri- prosthetic Fracture
Primary
Total knee arthroplasty	157	94% (147)	4% (6)	1% (2)	1% (2)	-	-	-
One-stage bilateral total knee arthroplasty	29	100% (29)	-	-	-	-	-	-
Total hip arthroplasty	163	75% (122)	2% (3)	6% (9)	-	10% (17)	7% (11)	1% (1)
Revision
Total knee arthroplasty	32	-	-	-	-	-	-	-	94% (30)	3% (1)	3% (1)
Total hip arthroplasty	108	-	-	-	-	-	-	-	89% (96)	8% (9)	3% (3)

View Larger

TABLE II: PREVALENCE OF COMORBID CONDITIONS AMONG THE FOUR HUNDRED AND EIGHTY-NINE PATIENTS

*The numbers in parentheses represent the number of patients who had the condition.

Comorbid Condition*	Prevalence
Hypertension	26% (128)
History of cardiac disease	14% (70)
Diabetes mellitus	7% (36)
History of cancer	5% (26)
Chronic obstructive pulmonary disease	5% (24)
Hypothyroidism	5% (23)
Rheumatoid arthritis	2% (11)

Add to My JBJS

TABLE II: PREVALENCE OF COMORBID CONDITIONS AMONG THE FOUR HUNDRED AND EIGHTY-NINE PATIENTS

*The numbers in parentheses represent the number of patients who had the condition.

Comorbid Condition*	Prevalence
Hypertension	26% (128)
History of cardiac disease	14% (70)
Diabetes mellitus	7% (36)
History of cancer	5% (26)
Chronic obstructive pulmonary disease	5% (24)
Hypothyroidism	5% (23)
Rheumatoid arthritis	2% (11)

View Larger

TABLE III: TRANSFUSION REQUIREMENTS ACCORDING TO THE TYPE OF OPERATION

*The values are given as the average and the standard deviation.

	All Patients (N = 489)	Total Knee Arthroplasty			Total Hip Arthroplasty
Primary (N = 157)	Revision (N = 32)	Bilateral (N = 29)	Primary (N = 163)	Revision (N = 108)
Age* (yrs.)	65.0 ± 12.5	67.0 ± 9.7	69.8 ± 10.1	71.8 ± 6.7	60.8 ± 13.8	65.2 ± 13.9
Male:female ratio	290:199 (59%/41%)	117:40 (75%/25%)	14:18 (44%/56%)	11:18 (38%/62%)	94:69 (58%/42%)	54:54 (50%/50%)
Participation in autologousblood-donation program
No. of patients	264 (54%)	72 (46%)	13 (41%)	21 (72%)	99 (61%)	59 (55%)
Total no. of units	527	132	27	46	201	121
Average no. of units	2.0	1.8	2.1	2.2	2.0	2.1
No. of units utilized	232 (44%)	50 (38%)	15 (56%)	37 (80%)	73 (36%)	57 (47%)
Predonation hemoglobin level
No. of patients available for analysis	149	45	2	13	70	19
Time that level was measured (days preop.)	22	20	15	20	22	25
Level* (g/L)	138.9 ± 14.7	141.5 ± 14.2	120.5 ± 0.7	141.1 ± 14.6	137.4 ± 15.2	138.8 ± 13.8
Preop. hemoglobin level
No. of patients available for analysis	484	156	31	29	162	106
Time that level was measured (days preop.)	6	5	5	6	6	7
Level* (g/L)	131.3 ± 16.4	137.0 ± 16.0	130.0 ± 14.0	128.0 ± 13.0	129.4 ± 17.0	127.0 ± 16.0
Type of anesthesia(no. of patients)
General	179 (44%)	53 (44%)	18 (64%)	6 (25%)	49 (35%)	53 (55%)
General and epidural	204 (50%)	50 (41%)	9 (32%)	18 (75%)	86 (61%)	41 (43%)
Epidural and intra- venous sedation	18 (4%)	11 (9%)	1 (4%)	0	5 (4%)	1 (1%)
Spinal, epidural, and intravenous sedation	8 (2%)	7 (6%)	0	0	0	1 (1%)
Estimated blood loss* (ml)	480 ± 437	185 ± 133	310 ± 308	340 ± 211	503 ± 339	844 ± 556
Use of Cell Saver
No. of patients	65 (13%)	0	2 (6%)	3 (10%)	8 (5%)	52 (48%)
Volume used* (ml)	398 ± 292	—	125 ± 35	198 ± 137	403 ± 203	419 ± 308
Drain output* (ml)	439 ± 318	491 ± 352	407 ± 311	718 ± 406	381 ± 223	401 ± 328
Transfusion performed(no. of patients)
Allogenic	82 (17%)	9 (6%)	9 (28%)	10 (34%)	18 (11%)	36 (33%)
Autologous	131 (27%)	29 (18%)	9 (28%)	17 (59%)	42 (26%)	34 (31%)

Add to My JBJS

TABLE III: TRANSFUSION REQUIREMENTS ACCORDING TO THE TYPE OF OPERATION

*The values are given as the average and the standard deviation.

	All Patients (N = 489)	Total Knee Arthroplasty			Total Hip Arthroplasty
Primary (N = 157)	Revision (N = 32)	Bilateral (N = 29)	Primary (N = 163)	Revision (N = 108)
Age* (yrs.)	65.0 ± 12.5	67.0 ± 9.7	69.8 ± 10.1	71.8 ± 6.7	60.8 ± 13.8	65.2 ± 13.9
Male:female ratio	290:199 (59%/41%)	117:40 (75%/25%)	14:18 (44%/56%)	11:18 (38%/62%)	94:69 (58%/42%)	54:54 (50%/50%)
Participation in autologousblood-donation program
No. of patients	264 (54%)	72 (46%)	13 (41%)	21 (72%)	99 (61%)	59 (55%)
Total no. of units	527	132	27	46	201	121
Average no. of units	2.0	1.8	2.1	2.2	2.0	2.1
No. of units utilized	232 (44%)	50 (38%)	15 (56%)	37 (80%)	73 (36%)	57 (47%)
Predonation hemoglobin level
No. of patients available for analysis	149	45	2	13	70	19
Time that level was measured (days preop.)	22	20	15	20	22	25
Level* (g/L)	138.9 ± 14.7	141.5 ± 14.2	120.5 ± 0.7	141.1 ± 14.6	137.4 ± 15.2	138.8 ± 13.8
Preop. hemoglobin level
No. of patients available for analysis	484	156	31	29	162	106
Time that level was measured (days preop.)	6	5	5	6	6	7
Level* (g/L)	131.3 ± 16.4	137.0 ± 16.0	130.0 ± 14.0	128.0 ± 13.0	129.4 ± 17.0	127.0 ± 16.0
Type of anesthesia(no. of patients)
General	179 (44%)	53 (44%)	18 (64%)	6 (25%)	49 (35%)	53 (55%)
General and epidural	204 (50%)	50 (41%)	9 (32%)	18 (75%)	86 (61%)	41 (43%)
Epidural and intra- venous sedation	18 (4%)	11 (9%)	1 (4%)	0	5 (4%)	1 (1%)
Spinal, epidural, and intravenous sedation	8 (2%)	7 (6%)	0	0	0	1 (1%)
Estimated blood loss* (ml)	480 ± 437	185 ± 133	310 ± 308	340 ± 211	503 ± 339	844 ± 556
Use of Cell Saver
No. of patients	65 (13%)	0	2 (6%)	3 (10%)	8 (5%)	52 (48%)
Volume used* (ml)	398 ± 292	—	125 ± 35	198 ± 137	403 ± 203	419 ± 308
Drain output* (ml)	439 ± 318	491 ± 352	407 ± 311	718 ± 406	381 ± 223	401 ± 328
Transfusion performed(no. of patients)
Allogenic	82 (17%)	9 (6%)	9 (28%)	10 (34%)	18 (11%)	36 (33%)
Autologous	131 (27%)	29 (18%)	9 (28%)	17 (59%)	42 (26%)	34 (31%)

View Larger

TABLE IV: RATES OF AUTOLOGOUS BLOOD DONATION, ALLOGENIC TRANSFUSION, AND UTILIZATION OF AUTOLOGOUS BLOOD UNITS ACCORDING TO SINGLE AND COMBINATION-RISK-FACTOR GROUPS*

*All p values that appear in the table are significant.

Risk Factor	Total No. of Patients	No. of Patients Who Donated Autologous Blood	No. of Patients Who Received Allogenic Transfusion	Rate of Utilization of Autologous Blood Units*
Type of operation
Primary	320	53% (171 of 320)	8% (27 of 320)	37% (123 of 333)
Revision or bilateral	169	55% (93 of 169)	33% (55 of 169)	56% (109 of 194)
P value*			< 0.001	<0.001
Initial hemoglobin level
= 150 g/L	85	52% (44 of 85)	1% (1 of 85)	18% (17 of 94)
130 to < 150 g/L	222	53% (117 of 222)	12% (26 of 222)	36% (84 of 233)
<130 g/L	178	57% (102 of 178)	29% (52 of 178)	64% (127 of 199)
P value for = 150 g/L vs. 130 to < 150 g/L			<0.01	0.033
P value for = 150 g/L vs. < 130 g/L			<0.001	<0.0001
Age
< 65 yrs.	193	64% (123 of 193)	6% (12 of 193)	32% (77 of 241)
=65 yrs.	296	48% (141 of 296)	24% (70 of 296)	54% (155 of 286)
P value*			<0.001	<0.001
Primary total joint arthroplasty and initial hemoglobin level of = 150 g/L	67	46% (31 of 67)	0	14% (9 of 64)
Primary total joint arthroplasty and initial hemoglobin level of 130 to < 150 g/L	150	52% (78 of 150)	5% (7 of 150)	28% (42 of 151)
Primary total joint arthroplasty, initial hemoglobin level of 130 to < 150 g/L, and age of <65 yrs.	63	62% (39 of 63)	0	19% (14 of 74)

Add to My JBJS

TABLE IV: RATES OF AUTOLOGOUS BLOOD DONATION, ALLOGENIC TRANSFUSION, AND UTILIZATION OF AUTOLOGOUS BLOOD UNITS ACCORDING TO SINGLE AND COMBINATION-RISK-FACTOR GROUPS*

*All p values that appear in the table are significant.

Risk Factor	Total No. of Patients	No. of Patients Who Donated Autologous Blood	No. of Patients Who Received Allogenic Transfusion	Rate of Utilization of Autologous Blood Units*
Type of operation
Primary	320	53% (171 of 320)	8% (27 of 320)	37% (123 of 333)
Revision or bilateral	169	55% (93 of 169)	33% (55 of 169)	56% (109 of 194)
P value*			< 0.001	<0.001
Initial hemoglobin level
= 150 g/L	85	52% (44 of 85)	1% (1 of 85)	18% (17 of 94)
130 to < 150 g/L	222	53% (117 of 222)	12% (26 of 222)	36% (84 of 233)
<130 g/L	178	57% (102 of 178)	29% (52 of 178)	64% (127 of 199)
P value for = 150 g/L vs. 130 to < 150 g/L			<0.01	0.033
P value for = 150 g/L vs. < 130 g/L			<0.001	<0.0001
Age
< 65 yrs.	193	64% (123 of 193)	6% (12 of 193)	32% (77 of 241)
=65 yrs.	296	48% (141 of 296)	24% (70 of 296)	54% (155 of 286)
P value*			<0.001	<0.001
Primary total joint arthroplasty and initial hemoglobin level of = 150 g/L	67	46% (31 of 67)	0	14% (9 of 64)
Primary total joint arthroplasty and initial hemoglobin level of 130 to < 150 g/L	150	52% (78 of 150)	5% (7 of 150)	28% (42 of 151)
Primary total joint arthroplasty, initial hemoglobin level of 130 to < 150 g/L, and age of <65 yrs.	63	62% (39 of 63)	0	19% (14 of 74)

View Larger

TABLE V: EFFECT OF AUTOLOGOUS DONATION ON RATES OF ALLOGENIC TRANSFUSION IN SINGLE AND COMBINATION-RISK-FACTOR GROUPS

*P values that are significant are shown in boldface type.

Risk Factor	Total No. of Patients	No. of Patients Who Received Allogenic Transfusion		P Value*
Nondonors	Donors
Type of operation
Primary	320	15% (22 of 149)	3% (5 of 171)	<0.001
Revision or bilateral	169	49% (37 of 76)	19% (18 of 93)	<0.001
Initial hemoglobin level
= 150 g/L	85	2% (1 of 41)	0% (0 of 44)	0.24
130 to <150 g/L	222	18% (19 of 105)	6% (7 of 117)	0.005
<130 g/L	178	49% (37 of 76)	15% (15 of 102)	<0.001
Age
<65 yrs.	193	9% (6 of 70)	5% (6 of 123)	0.36
=65 yrs.	296	34% (53 of 155)	12% (17 of 141)	<0.001
Primary total joint arthroplasty and initial hemoglobin level of =150 g/L	67	0% (0 of 36)	0% (0 of 31)
Primary total joint arthroplasty and initial hemoglobin level of 130 to <150 g/L	150	10% (7 of 72)	0 of 78	0.005
Primary total joint arthroplasty, initial hemoglobin level of 130 to <150 g/L, and age of <65 yrs.	63	0 of 24	0 of 39

Add to My JBJS

TABLE V: EFFECT OF AUTOLOGOUS DONATION ON RATES OF ALLOGENIC TRANSFUSION IN SINGLE AND COMBINATION-RISK-FACTOR GROUPS

*P values that are significant are shown in boldface type.

Risk Factor	Total No. of Patients	No. of Patients Who Received Allogenic Transfusion		P Value*
Nondonors	Donors
Type of operation
Primary	320	15% (22 of 149)	3% (5 of 171)	<0.001
Revision or bilateral	169	49% (37 of 76)	19% (18 of 93)	<0.001
Initial hemoglobin level
= 150 g/L	85	2% (1 of 41)	0% (0 of 44)	0.24
130 to <150 g/L	222	18% (19 of 105)	6% (7 of 117)	0.005
<130 g/L	178	49% (37 of 76)	15% (15 of 102)	<0.001
Age
<65 yrs.	193	9% (6 of 70)	5% (6 of 123)	0.36
=65 yrs.	296	34% (53 of 155)	12% (17 of 141)	<0.001
Primary total joint arthroplasty and initial hemoglobin level of =150 g/L	67	0% (0 of 36)	0% (0 of 31)
Primary total joint arthroplasty and initial hemoglobin level of 130 to <150 g/L	150	10% (7 of 72)	0 of 78	0.005
Primary total joint arthroplasty, initial hemoglobin level of 130 to <150 g/L, and age of <65 yrs.	63	0 of 24	0 of 39