Journal of Vascular Surgery
Volume 51, Issue 3 , Pages 616-621.e3, March 2010

Blood transfusion is associated with increased morbidity and mortality after lower extremity revascularization

Presented at the 2009 Vascular Annual Meeting, Denver, Colo, June 11-14, 2009.

  • Shane D. O'Keeffe, MD

      Affiliations

    • University of Kentucky Medical Center, Lexington, Ky
    • ,
  • Daniel L. Davenport, PhD

      Affiliations

    • University of Kentucky Medical Center, Lexington, Ky
    • ,
  • David J. Minion, MD

      Affiliations

    • University of Kentucky Medical Center, Lexington, Ky
    • ,
  • Ehab E. Sorial, MD

      Affiliations

    • University of Kentucky Medical Center, Lexington, Ky
    • ,
  • Eric D. Endean, MD

      Affiliations

    • University of Kentucky Medical Center, Lexington, Ky
    • ,
  • Eleftherios Sarantis Xenos, MD, PhD

      Affiliations

    • University of Kentucky Medical Center, Lexington, Ky
    • VA Medical Center, Lexington, Ky
    • Corresponding Author InformationReprint requests: Eleftherios S. Xenos, MD, PhD, University of Kentucky Medical Center, 800 Rose Street, Lexington, KY 40536

Received 2 July 2009; accepted 3 October 2009. published online 28 January 2010.

Article Outline

Background

Little is known about the significance of blood transfusion in patients with peripheral arterial disease. We queried the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database to examine the effect of intraoperative blood transfusion on the morbidity and mortality in patients who underwent lower extremity revascularization.

Methods

We analyzed data from the participant use data file containing vascular surgical cases submitted to the ACS NSQIP in 2005, 2006, and 2007 by 173 hospitals. Current procedural terminology codes were used to select lower extremity procedures that were grouped into venous graft, prosthetic graft, or thromboendarterectomy. Thirty-day outcomes analyzed were (1) mortality, (2) composite morbidity, (3) graft/prosthesis failure, (4) return to the operating room within 30 days, (5) wound occurrences, (6) sepsis or septic shock, (7) pulmonary occurrences, and (8) renal insufficiency or failure. Intraoperative transfusion of packed red blood cells was categorized as none, 1 to 2 units, and 3 or more units. Outcome rates were compared between the transfused and nontransfused groups using the χ2 test and multivariable regression adjusting for transfusion propensity, comorbid and procedural risk.

Results

A total of 8799 patients underwent lower extremity revascularization between 2005 and 2007. Mean age was 66.8 ± 12.0 years and 5569 (63.3%) were male. Transfusion rates ranged from 14.5% in thromboendarterectomy patients to 27.1% in prosthetic bypass patients (P < .05). After adjustment for transfusion propensity and patient and procedural risks, transfusion of 1 or 2 units remained significantly predictive of mortality, composite morbidity, sepsis/shock, pulmonary occurrences, and return to the operating room. The adjusted odds ratios for 30-day mortality ranged from 1.92 (95% confidence interval [CI] 1.36-2.70) for 1 to 2 units to 2.48 (95% CI 1.55-3.98) for 3 or more units.

Conclusion

In a large number of patients undergoing lower extremity revascularization, we have found that there is a higher risk of postoperative mortality, pulmonary, and infectious complications after receiving intraoperative blood transfusion. Additional studies are necessary to better define transfusion triggers that balance the risk/benefit ratio for blood transfusion.

 

The population of patients with peripheral arterial disease (PAD) is steadily advancing in age, and older patients with multiple comorbid conditions undergo lower extremity revascularization. Red blood cell (RBC) transfusion is a common event in the perioperative course of these patients. This practice is not without risk. Worse outcomes in transfused patients have been observed in various settings such as critically ill patients, elderly patients, cardiac surgery/trauma/orthopedic surgical patients, and patients with acute coronary syndrome. In these studies, patients receiving allogeneic transfusions have had higher mortality rates, higher risk of intensive care unit (ICU) admission, longer hospital and ICU stays, higher postoperative infection rates, higher risk of developing adult respiratory distress syndrome (ARDS), longer time to ambulation, higher incidence of atrial fibrillation, and higher risk of ischemic outcomes compared with nontransfused cohorts.1, 2, 3, 4, 5, 6 Little is known about the significance of RBC transfusion in patients with PAD. We queried the National Surgical Quality Improvement Program (NSQIP) database to examine the effect of intraoperative blood transfusion on the morbidity and mortality in patients who underwent lower extremity revascularization.

Back to Article Outline

Methods 

Study population 

The American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) is a reporting system designed to provide reliable, risk-adjusted surgical outcomes data to surgical services and administrators at medical centers throughout the private sector so that surgical quality can be assessed and improved on a national level. We analyzed the data from the participant use data file containing vascular surgical cases submitted to the ACS NSQIP in 2005, 2006, and 2007 by 173 hospitals throughout the United States.

Data collection 

The ACS NSQIP collects data on 135 variables, including preoperative risk factors, intraoperative variables, and 30-day postoperative morbidity and mortality outcomes for patients undergoing surgical procedures in both the inpatient and outpatient setting. Data are prospectively collected in a standardized fashion according to strict definitions by dedicated surgical clinical nurse reviewers. Patients are followed throughout their hospital course and after discharge from the hospital up to 30 days postoperatively. Nurse reviewers collect data from computerized and paper patient medical records, doctor's office records, and telephone interviews with patients. The accuracy and reproducibility of the data have been previously demonstrated.7, 8, 9 Primary procedure current procedural terminology (CPT-4) codes were used to select lower extremity procedures, which were grouped into venous graft (CPT code 35533, 35541, 35546, 35556, 35558, 35565, 35566, 35571, 35583, 35585, or 35587), prosthetic graft (CPT code 35646, 35647, 35654, 35656, 35661, 35665, 35666, or 35671) or thromboendarterectomy (CPT code 35371, 35372, or 35381). Thirty-day outcomes analyzed were (1) mortality, (2) composite morbidity (one or more of 21 adverse events uniformly defined by the ACS NSQIP, including all of the following subgroups), (3) graft/prosthesis failure (defined as “mechanical failure of an extra-cardiac graft or prosthesis…requiring return to the operating room, interventional radiology, or a balloon angioplasty.”), (4) return to the operating room for any reason within 30 days, (5) wound occurrences (superficial, deep or organ/space surgical site infection and/or wound dehiscence), (6) sepsis or septic shock, (7) pulmonary occurrences (ventilation greater than 48 hours, pneumonia and/or unplanned intubation), and (8) renal insufficiency or failure (defined as “the reduced capacity of the kidney to perform its function as evidenced by a rise in creatinine of >2 mg/dL from preoperative value, but with no requirement for dialysis”).

Intraoperative transfusion of packed RBCs was categorized as none, 1 to 2 units, and 3 or more units. Preoperative hematocrit is measured according to the value closest to the entry to the operating room. Outcome rates were compared between the transfused and nontransfused groups using the χ2 test. Significant P values for the multiple comparisons were adjusted to .005.

The risk (propensity) of intraoperative transfusion in this patient population was calculated using logistic regression of more than 55 ACS NSQIP patient risk factors in a forward stepwise fashion (P for entry <.05, for exit >.10) followed by the addition of procedure group and complexity (work relative value units [RVUs]). Patients were ranked into five equal-sized groups (quintiles) based on their transfusion propensity. Within each quintile, the numbers of transfused and nontransfused patients were counted, and the mortality rates were compared using χ2 tests (P < .01 to correct for multiple comparisons). This was performed as a high-level omnibus test to aid in visualization prior to the more powerful multivariable analysis.

The odds ratios (ORs) by transfusion category were calculated for each of the outcomes using logistic regression (P < .005 was considered significant to correct for multiple comparisons). Adjusting variables included transfusion propensity, independent patient risk factors included in a forward stepwise fashion (P for entry <.05, for exit >.10), wound class, operative duration, procedure type, and complexity. Transfusion category was added at the end to the model. (Appendix Table AI, online only).

Back to Article Outline

Results 

The ACS NSQIP database contained 8799 patients who underwent lower extremity revascularization between 2005 and 2007. Their mean age was 66.8 ± 12.0 years and 5569 (63.3%) were male. Transfusion rates varied across procedure group, ranging from 14.5% in thromboendarterectomy patients to 27.1% in prosthetic bypass patients (χ2 P < .05, Table I).

Table I. Intraoperative transfusion rates and 30-day outcomes by type of lower extremity revascularization procedure
Lower extremity revascularization proceduren, %Transfused %Mortality %Morbidity %
Bypass w/prosthetic3696,4227.13.125.4
Bypass w/vein4076,4620.12.527.1
Thromboendarterectomy1027,1214.53.419.7
Total8799,10022.42.925.5

Transfusion and morbidity rates varied significantly by procedure type, χ2 P < .05.

The amount of RBC units given by preoperative hematocrit (Hct) appears in Fig 1. The transfusion propensity model included 20 patient risk factors and three intraoperative risk factors (Appendix Table AI, online only). The 10 most significant predictors of transfusion in lower extremity revascularization patients, by order of entry into the model were (1) preoperative Hct, (2) procedure group and (3) complexity (work RVUs), (4) American Society of Anesthesiologists' physical status classification [ASA class], (5) general anesthesia versus other, (6) prior coronary intervention, (7) emergent status, (8) age, (9) alkaline phosphatase greater than 125 U/L, and (10) serum albumin (g/dL, inverse relationship).

Transfusion rates ranged from 4.4% in the lowest propensity quintile to 52.9% in the high propensity quintile. The Mantel-Haenszel common odds ratio estimate across propensity quintiles was 2.13 (95% confidence interval [CI], 1.61-2.81; P < .001). The Breslow-Day test did not support rejecting homogeneity (P = .288). The mortality rate was significantly higher in transfused patients versus nontransfused for the highest and low–middle quintiles (χ2 P < .01, Fig 2). Mortality was threefold overall in transfused patients (6.2% in transfused patients vs 1.9% in nontransfused). All unadjusted outcomes examined were significantly higher in transfused patients (χ2 P < .001, Table II).

  • View full-size image.
  • Fig 2. 

    Mortality in transfused vs nontransfused patients within groups matched by their preoperative risk (propensity) of intraoperative transfusion. *χ2 P < .01.

  • Propensity is the estimated risk based on regression of preoperative hematocrit, procedure type and complexity, ASA class, emergent status, age, and 18 other risk factors vs whether or not the patient received a transfusion intraoperatively. The propensity regression model is in the Appendix.

Table II. Thirty-day outcomes for lower extremity revascularization patients transfused intraoperatively vs not
OutcomeNot transfusedTransfusedχ2P valuePropensity and risk-adjusteda transfusion odds ratio (95% CI)Wald P value across categories
6827,77.61971,22.4 1-2 u.3+ u.
Mortality, %1.96.2<.0011.92(1.36-2.70)2.48(1.55-3.98)<.001
Morbidity, %21.738.8<.0011.37(1.19-1.58)1.81(1.46-2.24)<.001
30-d graft failure, %4.87.8<.0011.28(1.00-1.64)1.09(0.74-1.62).151
Return to OR, %15.426.8<.0011.29(1.10-1.51)1.35(1.06-1.72).002
Wound occurrence,b %10.514.4<.0011.20(0.99-1.45)1.33(1.00-1.77).059
Sepsis/shock, %5.012.9<.0011.41(1.12-1.77)2.11(1.56-2.86)<.001
Pulmonary occurrence,c %3.514.5<.0012.23(1.75-2.85)3.68(2.68-5.06)<.001
Renal insufficiency/failure, %1.03.0<.0011.17(0.71-1.94)2.44(1.36-4.40).009

CI, Confidence interval; OR, operating room.

Unadjusted rates and odds ratios adjusted for transfusion propensity, all significant ACS NSQIP preoperative patient risk factors, procedure group, and complexity.

aEach outcome regression model included transfusion propensity, all independently significant ACS NSQIP adjusting risk factors included in forward stepwise regression, procedure group and complexity, wound class, and operative duration with level of intraoperative transfusion added at the end. The mortality and overall morbidity models are shown in the Appendix. Others are not shown.

bWound occurrences included superficial, deep and organ/space surgical site infection and wound dehiscence.

cPulmonary occurrences included pneumonia, unplanned intubation, and ventilation greater than 48 hours.

Each of the outcome multivariable regression models was highly significant in predicting each outcome (χ2 P < .001) and discrimination (c-indices) ranged from 0.66 (graft failure) to 0.85 (mortality). The mortality and composite morbidity models are shown in the Appendix Table AII, online only, Appendix Table AIII, online only. After adjustment for transfusion propensity and patient and procedural risks, transfusion of 1 or 2 units remained significantly (P < .005) predictive of mortality, composite morbidity, sepsis/shock, pulmonary occurrences, and return to the operating room. Risks for these outcomes increased with level of transfusion (Table II), although the significance of the increase was not determined. The adjusted odds ratios for 30-day mortality ranged from 1.92 (95% CI, 1.36-2.70) for 1 to 2 units to 2.48 (95% CI, 1.55-3.98) for 3 or more units (Wald P < .001).

Back to Article Outline

Discussion 

Blood product transfusion is a common event during peripheral vascular operations with the goal of replacing volume and increasing blood oxygen carrying capacity.10 Dramatic improvements in reduction of transmission of infectious agents11, 12 have led to new and increased focus on other serious hazards of transfusion such as multiple system organ failure associated with cytokine release, bacterial contamination, sepsis, metabolic disturbances, circulatory overload, hemolytic reactions, and risk of old blood.

After removal from the body and with the added effect of storage, RBCs undergo changes (many irreversible) that adversely affect their viability and function. These adverse changes include oxidation and rearrangement of lipids, loss of proteins, and depletion of ATP and 2, 3-diphosphoglycerate. In storage, RBCs continuously lose their membrane through shedding vesicles and become rigid. Moreover, during storage, bioactive by-products and ions (hemoglobin, lipids, and potassium)—some with proinflammatory effects—are released from RBCs and accumulate in blood units whereby they can cause adverse reactions in a recipient. These changes are collectively called “storage lesion.” Transfusion of blood that is stored for prolonged periods (but still within the currently accepted maximum allowed storage time of 42 days) has been linked to increased risk of complications and reduced survival in patients undergoing cardiac surgery and in other patient populations.13, 14, 15, 16, 17

Studies in cardiac surgery patients associated transfusion with increased mortality, higher incidence of postoperative infection, prolonged respiratory support, higher risk of postoperative infection, and higher risk of renal failure.4, 8 Similarly, in critical care patients, transfusion has been associated with increased overall and ICU 14-day mortality rate, higher 28-day mortality rate, longer length of stay, higher risk of developing ARDS, and higher incidence of bloodstream infections.18, 19, 20, 21, 22

We focused on the risks of blood transfusion in patients following peripheral vascular operations and using the NSQIP database examined 30-day morbidity and mortality. Our data suggest that after risk adjustment, patients receiving intraoperative transfusion are at higher risk of developing morbidity and mortality.

Transfusion-associated lung injury (TRALI) is a well recognized complication and is considered to be the second leading cause of mortality from transfusion.23, 24 In our population, there was almost a threefold overall increase in the incidence of pulmonary complications in transfused patients, transfusion of 3 or more units carried an even higher risk as compared with 1 to 2 units. The pathophysiologic mechanisms of TRALI are incompletely understood and have been described as antibody-mediated and nonantibody-mediated. In both cases, activation of neutrophils plays a causal role, and these activated cells are thought to locally mediate pulmonary injury.25, 26

The threshold at which a patient receives a blood transfusion is arbitrary, depending on the culture of practice at various different institutions. The optimal range of hematocrit values that balances the associated complications of blood transfusion with complications related to anemia is unknown. Habib et al27 reported that the lowest hematocrit value on cardiopulmonary bypass (<22%) was associated with increased morbidity and was predictive of worse 0- to 6-year survival. Although Wu et al28 described a benefit for RBC transfusion in patients who were elderly and had low admission hematocrit values (<30%), they also reported that for those patients with hematocrit values of ≥36% and who received transfusions had a higher risk of death within 30 days than patients with a similar hematocrit who did not receive blood. In a randomized, controlled trial of 838 critically ill patients, Hebert29 reported similar overall 30-day mortality for a restrictive transfusion strategy (hemoglobin levels maintained between 7 and 9 g/dL) vs a more liberal strategy (hemoglobin levels maintained between 10 and 12 g/dL). Although mortality was similar in the two groups, it was significantly lower with the restrictive strategy among patients who were less acutely ill. Because we found that transfusion was not limited to patients with low Hct in our population, we carried out a propensity risk assessment based on preoperative Hct, procedure type and complexity, ASA class, emergent status, age, and 18 other risk factors vs whether or not the patient received a transfusion intraoperatively. The propensity regression showed that mortality is significantly higher in transfused vs nontransfused patients within groups matched by their preoperative risk (propensity) of intraoperative transfusion, even for low- and medium-propensity patients.

The immunologic effects of blood transfusion may be responsible for the observed increase in risk of infection in blood-transfused patients. Blood transfusions have been shown to be independent risk factor for infection.30, 31 Transfusion-related immunomodulation (TRIM) includes both alloimmunization of the host and immune activation as well as tolerance manifested as cancer recurrence, improved allograft survival, and higher rates of postoperative infections.32, 33 Leukoreduction may decrease postoperative infections,34 and today, most RBC transfusions in the US are leukoreduced; however, the cost-effectiveness of leukoreduction has yet to be proven, especially in low-risk populations.35

A limitation of our study is that the transfusion propensity was calculated based on preoperative hematocrit values possibly not accurately reflecting acute intraoperative blood loss. There are other factors not captured by the NSQIP database that can affect the outcome of the revascularization procedure as well as the need for intraoperative blood transfusion, including the skill and experience of the surgeon and technical difficulty of the procedure. We took into account in our propensity and risk models adjustment for operative duration, wound class, work RVUs, emergency status, and use of preoperative transfusion, which capture some of the variation due to skill and difficulty.

To conclude, our study in a large number of patients undergoing lower extremity revascularization indicates that allogeneic intraoperative transfusion is associated with higher postoperative morbidity and mortality. This finding is true after adjusting for propensity for transfusion, thus, the reason that transfused patients do poorly is not because they have a lower preoperative hematocrit. When do the risks of anemia outweigh the hazards of transfusion? In the absence of acute bleeding, hemoglobin levels consistent with the TRICC trial (7.0-9.0 g/dL) are well tolerated.29 There is little evidence that RBC transfusion in the nonbleeding patient with a hemoglobin concentration greater than 7.0 g/dL leads to improved outcome. Clinicians should use hemodynamic and physiologic parameters such as blood pressure, heart rate, and urine output in conjunction with hemoglobin levels to decide whether a patient needs to be transfused.

Additional prospective randomized studies are required to determine the risk/benefit of RBC transfusion in various disease states, the optimal transfusion trigger, and the effects of blood storage time and leukodepletion on clinical outcomes.

Back to Article Outline

Author contributions 


Conception and design: EX

Analysis and interpretation: EX, DD

Data collection: SO, DD

Writing the article: EX, DD, SO

Critical revision of the article: EE, DM, ES

Final approval of the article: EX, DD, EE, DM, SO, ES

Statistical analysis: DD

Obtained funding: Not applicable

Overall responsibility: EX

Back to Article Outline

 

The American College of Surgeons National Surgical Quality Improvement Program and the hospitals participating in the ACS NSQIP are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors.

Back to Article Outline

Appendices AI-AIII (online only) 

Appendix Table AI, online only. Transfusion regression model, c-index = 0.79
Entry orderVariables in the equationOdds ratio95% CI ORSig.
LowerUpper
1Preop hematocrit vs >48% .000
≤24%23.02611.43946.353.000
≤28%18.63310.94131.731.000
≤32%8.4235.11013.884.000
≤36%4.1652.5456.816.000
≤40%2.4541.5063.998.000
≤44%1.8541.1333.033.014
≤48%1.362.8092.293.245
2Procedure vs thromboendarterectomy .000
Bypass w/prosthetic.889.7031.125.329
Bypass w/vein.349.269.453.000
3Work RVUs1.1481.1331.164.000
4ASA class vs 1 .000
ASA 2.276.115.664.004
ASA 3.457.1981.055.066
ASA 4.609.2621.418.250
ASA 5.594.1702.072.414
5General anesthetic vs other2.2671.8102.838.000
6Prior coronary intervention1.4481.2831.634.000
7Emergency status1.6261.3291.990.000
8Age1.0101.0051.015.000
9Alkaline phosphatase >1251.5231.2371.875.000
10Serum albumin1.3901.2361.563.000
11Known bleeding disorder1.2281.0781.399.002
12Functional status vs independent .004
Partially dependent1.2921.1091.505.001
Fully dependent1.191.8401.689.326
13Platelets <1501.4121.1531.727.001
14Male vs female.806.716.907.000
15PT >13.21.1831.0521.331.005
16Prior periph. vasc. procedure1.1781.0491.323.006
17Hx CHF, AMI or angina1.3031.0621.600.011
18Diabetes vs not treated .019
Orally treated.833.713.972.021
Insulin treated.844.725.984.030
19Platelets >4001.2751.0491.548.015
20Preop transfusion >4 u. RBCs2.6571.1226.288.026
21EtOH >2 drinks/day1.2761.0161.603.036
22Preop renal failure.566.343.934.026
23BUN >401.2611.0181.561.034
Constant.000 .000

AMI, Acute myocardial infarction; ASA, American Society of Anesthesiologists; CHF, congestive heart failure; EtOH, alcohol; RVUs, relative value units.

Appendix Table AII, online only. Mortality regression model, c-index = 0.853
StepVariableSig.Odds ratio95% CI
LowerUpper
ForcedTransfusion propensity.073.434.1741.082
1Functional status vs independent
Partially dependent.0011.7321.2542.392
Fully dependent.0191.9661.1203.451
2On dialysis<.0013.0032.0674.363
3Age<.0011.0471.0331.061
4Emergency<.0012.0611.4053.025
5ASA class vs 1<.001Sig. across all values
ASA 2.754.692.0696.926
ASA 3.585.544.0614.846
ASA 4.9341.097.1239.772
ASA 5.7191.564.13717.841
6Serum albumin.008.736.587.922
7BMI categ. Kg/m2 vs 18.5-25
≤18.5.0032.0181.2603.231
25.1-30.0.184.805.5851.108
30.1-35.0.030.586.362.949
35.1-40.0.032.376.154.922
>40.0.963.979.4032.380
8Hx cardiovasc. disease.0031.7401.2002.523
9BUN >40.0011.9381.3252.836
10Bilirubin >1.0.0022.1941.3423.587
11Systemic inflammation vs none
SIRS.0091.7471.1522.650
Sepsis.376.638.2361.725
Septic shock.1371.923.8134.548
12Male vs female.010.691.522.915
13Coma >24 hours postop.01612.4611.61096.442
14Pulmonary compromise.0251.4641.0502.041
ForcedWound class vs clean
Clean/contaminated.3621.424.6663.046
Contaminated.5061.192.7111.997
Dirty/infected.0811.630.9412.824
ForcedOperative duration minutes.0271.0021.0001.003
ForcedProcedure type vs thromboendart.
Revasc. w/non-vein.296.765.4631.265
Revasc. w/vein.020.488.267.893
ForcedWork RVUs.0521.0401.0001.082
ForcedIntraop transfusion vs none
1-2 u.<.0011.9191.3622.704
3+ u.<.0012.4821.5483.978
Constant<.001.001

ASA, American Society of Anesthesiologists; BMI, body mass index; RVUs, relative value units; SIRS, systemic inflammatory response syndrome.

Appendix Table AIII, online only. Morbidity regression model, c-index = 0.70
StepVariableSig.Odds ratio95% CI
LowerUpper
ForcedTransfusion propensity.685.921.6181.372
1Functional status vs independent
Partially dependent<.0011.4781.2841.702
Fully dependent<.0012.2021.5853.057
2Systemic inflammation vs none
SIRS<.0011.6731.3532.069
Sepsis.0601.465.9842.180
Septic shock.1211.740.8633.507
3BMI categ. Kg/m2 vs 18.5-25.0
≤18.5.752.959.7391.245
25.1-30.0.2391.078.9511.222
30.1-35.0<.0011.4401.2391.674
35.1-40.0<.0011.6971.3732.096
>40.0.0091.4741.1041.969
4Rest pain<.0011.2941.1651.437
5ASA class vs 1<.001Sig. across all values
ASA 2.146.538.2331.241
ASA 3.290.646.2871.452
ASA 4.723.863.3821.951
ASA 5.5461.452.4324.877
6Emergency<.0011.6501.3661.994
7Male vs female<.001.755.678.840
8Pulmonary compromise.0011.2651.0981.457
9Hx stroke/TIA.0041.2061.0611.371
10SGOT >40.0071.3331.0821.643
11Creatinine >1.2.0021.1891.0631.330
12PT >13.2.0021.1871.0651.324
13Sodium <135.0061.2301.0611.426
14CNS tumor.0376.1031.11933.294
15Hx cardiovascular disease.0031.3341.1041.611
16Recent 10% weight loss.0311.4611.0352.062
17Preop transfusion 5+ u..0282.5721.1055.988
ForcedWound class vs clean.019
Clean/contaminated.0741.300.9751.733
Contaminated.0591.245.9921.562
Dirty/infected.103.797.6071.047
ForcedOperative duration minutes<.0011.0021.0011.003
ForcedProcedure type vs thrombo endarterectomy.401
Revasc. w/non-vein.1801.152.9371.416
Revasc. w/vein.3051.129.8951.425
ForcedWork RVUs.0041.0221.0071.037
ForcedIntraop transfusion vs none
1-2 u.<.0011.3691.1881.577
3+ u.<.0011.8121.4632.243
Constant<.001.090

ASA, American Society of Anesthesiologists; BMI, body mass index; RVUs, relative value units; TIA, transient ischemic attack.

Back to Article Outline

References 

  1. Murphy GJ, Reeves BC, Rogers CA, Rivzi SI, Culliford L, Angelini GD. Increased mortality, postoperative morbidity, and cost after red blood cell transfusion in patients having cardiac surgery. Circulation. 2007;116:2544–2552
  2. Rao SV, Jollis JG, Harrington RA, Granger CB, Newby LK, Armstrong PW, et al. Relationship of blood transfusion and clinical outcomes in patients with acute coronary syndromes. JAMA. 2004;292:1555–1562
  3. Jani SM, Smith DE, Share D, Kline-Rogers E, Khanal S, O'Donnell MJ, et al. Blood transfusion and in-hospital outcomes in anemic patients with myocardial infarction undergoing percutaneous coronary intervention. Clin Cardiol. 2007;30(10 Suppl 2):II49–II56
  4. Leal-Noval SR, Rincon-Ferrari MD, Garcıa-Curiel A, Herruzo-Aviles A, Camacho-Larana P, Garnacho-Montero J, et al. Transfusion of blood components and postoperative infection in patients undergoing cardiac surgery. Chest. 2001;119:1461–1468
  5. Chelemer SB, Prato BS, Cox PM, Morton JR. Association of bacterial infection and red blood cell transfusion after coronary artery bypasses surgery. Ann Thorac Surg. 2002;73:138–142
  6. Engoren MC, Habib RH, Zacharias A, Schwann TA, Riordan CJ, Durham CJ. Effect of blood transfusion on long term survival after cardiac operation. Ann Thorac Surg. 2002;74:1180–1186
  7. Khuri SF, Daley J, Henderson W, Hur K, Demakis J, Aust JB, et al The Department of Veterans Affairs' NSQIP: the first national, validated, outcome-based, risk-adjusted, and peer-controlled program for the measurement and enhancement of the quality of surgical care (National VA Surgical Quality Improvement Program). Ann Surg. 1998;228:491–507
  8. Khuri SF, Henderson WG, Daley J, Jonasson O, Jones RS, Campbell DA, et al. The patient safety in surgery study: background, study design, and patient populations. J Am Coll Surg. 2007;204:1089–1102
  9. Daley J, Khuri SF, Henderson W, Hur K, Gibbs JO, Barbour G, et al. Risk adjustment of the postoperative morbidity rate for the comparative assessment of the quality of surgical care: results of the National Veterans Affairs Surgical Risk Study. J Am Coll Surg. 1997;185:328–340
  10. Haller M, Forst H. Red cell transfusion in the critical care setting. Transfusion Sci. 1997;18:459
  11. Busch MP. Closing the windows on viral transmission by blood transfusion. In:  Stramer SL editors. Blood safety in the new millennium. Bethesda (MD): American Association of Blood Banks; 2001;p. 33–54
  12. Dodd RY, Notari EP, Stramer SL. Current prevalence and incidence of infectious disease markers and estimated window-period risk in the American Red Cross blood donor population. [see comment] Transfusion. 2002;42:975–979
  13. King KE, Shirey RS, Thoman SK, Bensen-Kennedy D, Tanz WS, Ness PM. Universal leukoreduction decreases the incidence of febrile nonhemolytic transfusion reactions to RBCs. Transfusion. 2004;44:25–29
  14. Tinmouth A, Fergusson D, Yee IC ABLE Investigators; Canadian Critical Care Trials Group. Clinical consequences of red cell storage in the critically ill. Transfusion. 2006;46:2014–2027
  15. Kriebardis AG, Antonelou MH, Stamoulis KE, Economou-Petersen E, Margaritis LH, Papasideri IS. Storage-dependent remodeling of the red blood cell membrane is associated with increased immunoglobulin G binding, lipid raft rearrangement, and caspase activation. Transfusion. 2007;47:1212–1220
  16. Ho J, Sibbald WJ, Chin-Yee IH. Effects of storage on efficacy of red cell transfusion: when is it not safe?. Crit Care Med. 2003;31(12 Suppl):S687–S697
  17. Koch CG, Li L, Sessler DI, Figueroa P, Hoeltge GA, Mihaljevic T. Duration of red-cell storage and complications after cardiac surgery. N Engl J Med. 2008;358:1229–1239
  18. Vincent JL, Baron JF, Reinhart K, Gattitoni L, Thijs L, Webb A, et al. ABC (Anemia and Blood Transfusion in Critical Care) Investigators Anemia and blood transfusion in critically ill patients. JAMA. 2002;288:1499–1507
  19. Corwin HL, Gettinger A, Pearl RG, Fink MP, Levy MM, Abraham E, et al. The CRIT Study: Anemia and blood transfusion in the critically ill–current clinical practice in the United States. Crit Care Med. 2004;32:39–52
  20. Gong MN, Thompson BT, Williams P, Pothier L, Boyce PD, Christiani DC. Clinical predictors of and mortality in acute respiratory distress syndrome: potential role of red cell transfusion. Crit Care Med. 2005;33:1191–1198
  21. Shorr AF, Jackson WL, Kelly KM, Fu M, Kollef MH. Transfusion practice and blood stream infections in critically ill patients. Chest. 2005;127:1722–1728
  22. Taylor RW, O'Brien J, Trottier SJ, Manganaro L, Cytron M, Lesko MF, et al. Red blood cell transfusions and nosocomial infections in critically ill patients. Crit Care Med. 2006;34:2302–2308
  23. Holness L, Knippen MA, Simmons L, Lachenbruch PA. Fatalities caused by TRALI. Transfus Med Rev. 2004;18:184–188
  24. Toy P, Popovsky MA, Abraham E, Ambruso DR, Holness LG, Kopko PM, et al Transfusion-related acute lung injury: definition and review. Crit Care Med. 2005;33:721–726
  25. Goldman M, Webert KE, Arnold DM, Freedman J, Hannon J, Blajchman MA. Proceedings of a consensus conference: towards an understanding of TRALI. Transfus Med Rev. 2005;19:2–31
  26. Kopko PM. Leukocyte antibodies and biologically active mediators in the pathogenesis of transfusion-related acute lung injury. Curr Hematol Rep. 2004;3:456–461
  27. Habib RH, Zacharias A, Schwann TA, Riordan CJ, Engoren M, Durham SJ, et al. Role of hemodilutional anemia and transfusion during cardiopulmonary bypass in renal injury after coronary revascularization: Implications on operative outcome. Crit Care Med. 2005;33:1749–1756
  28. Wu WC, Rathore SS, Wang Y, Radford MJ, Krumholz HM. Blood transfusion in elderly patients with acute myocardial infarction. N Engl J Med. 2001;345:1230–1236
  29. Hebert PC, Wells G, Blajchman MA, Marshall J, Martin C, Pagliarello G, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care (Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group). N Engl J Med. 1999;340:409–417
  30. Claridge JA, Sawyer RG, Schulman AM, McLemore EC, Young JS. Blood transfusions correlate with infections in trauma patients in a dose-dependent manner. Am Surg. 2002;68:566–572
  31. Malone DL, Dunne J, Tracy JK, Putnam AT, Scalea TM, Napolitano LM. Blood transfusion, independent of shock severity, is associated with worse outcome in trauma. J Trauma. 2003;54:898–905
  32. Raghavan M, Marik PE. Anemia, allogeneic blood transfusion, and immunomodulation in the critically ill. Chest. 2005;127:295–307
  33. Vamvakas EC, Blajchman MA. Transfusion-related immunomodulation (TRIM): an update. Blood Rev. 2007;21:327–348
  34. Fergusson D, Khanna MP, Tinmouth A, Hebert PC. Transfusion of leukoreduced red blood cells may decrease postoperative infections: two meta-analyses of randomized controlled trials. Can J Anaesth. 2004;51:417–424
  35. Corwin HL, AuBuchon JP. Is leukoreduction of blood components for everyone?. JAMA. 2003;289:1993–1995

 Competition of interest: none.

 Additional material for this article may be found online at www.jvascsurg.org.

 The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a competition of interest.

PII: S0741-5214(09)02095-3

doi:10.1016/j.jvs.2009.10.045

Journal of Vascular Surgery
Volume 51, Issue 3 , Pages 616-621.e3, March 2010

Article Tools

* Image Usage & Resolution