erythropoietin is equally effective as fresh-blood ......hemoglobin groups and the erythropoietin-...

7
Neurologic Critical Care Erythropoietin is equally effective as fresh-blood transfusion at reducing infarct size in anemic rats Anargyros Xenocostas, MD; Houxiang Hu, MD, PhD; Nicolas Chin-Yee; Xiangru Lu, MD; Ian Chin-Yee, MD; Qingping Feng, MD, PhD A nemia is a common condition in patients suffering from acute myocardial infarction (MI). The estimated prevalence of anemia, defined as a hemoglobin (Hb) level of 120 g/L, is approximately 15– 40% in patients with acute MI. Elderly populations show increased rates with more severe anemia (Hb 100 g/L) pre- senting in approximately 5% of individu- als (1, 2). Anemia and the resulting de- crease in the oxygen-carrying capacity of the blood could increase the potential for tissue damage during acute MI, and a number of clinical studies suggest that anemia adversely affects clinical out- comes (2–5). In the largest of these stud- ies, Sabatine et al (2) found that the mor- tality rate increased in acute coronary syndromes as Hb levels decreased below 14 g/dL, and there was an adjusted odds ratio of 1.21 (95% confidence interval, 1.12–1.30) for each 10 g/L decrement in Hb. Given the prevalence of anemia and its direct adverse effects on cardiovascu- lar function, it is important to seek effec- tive therapeutic interventions to reduce myocardial damage after acute MI (6, 7). Presently, the most common method of treating anemic patients is by red blood cell (RBC) transfusion. The effects of transfusion on acute coronary syn- dromes have also been studied, and there have been conflicting results. In a study conducted by Wu et al (1), transfusion was shown to reduce short-term mortal- ity in elderly patients after MI; however, contradictory results associating transfu- sion with higher mortality rates were ob- served in the setting of acute coronary syndromes (8). A subsequent clinical study found that although transfusion with a nadir Hb level of 80 g/L was associated with an increased mortality rate in patients experiencing acute MI, transfusion in patients with a nadir Hb level of 80 g/L showed a protective ef- fect following MI (9). Our group recently found similar results in a rat model show- ing transfusion to a Hb level of 100 g/L to be cardioprotective, but transfusion to higher Hb levels did not seem to be ben- eficial (10). If blood transfusion is to be- come an accepted treatment for myocar- dial ischemia, then the optimal Hb level for transfusion must be determined and balanced against the risks such as volume overload, transfusion-related acute lung injury, infection, and immunosuppres- sive effects associated with this interven- tion. From the Centre for Critical Illness, Research Lawson Health Research Institute (AX, HH, XL, ICY, QF); Canadian Blood Services (IC-Y); Departments of Medicine (AX, HH, IC-Y, QF), Physiology and Pharmacology (NC-Y, QF), Uni- versity of Western Ontario, London, Ontario, Canada; North Sichuan Medical College First Affiliated Hospital, Nanchong, Sichuan, P.R. China (HH). This study was supported by the CIHR, Canadian Blood Services, Hema Quebec, Bayer Partnership Fund, and the Anemia Institute for Research and Education. Dr. Feng is a Heart and Stroke Foundation of Ontario Career Investigator. Dr. Xenocostas has been a consultant for and has received educational grants and honoraria from Ortho Biotech, Canada. Dr. I Chin-Yee is the recipient of a grant from Canadian Blood Services. The remaining authors have not disclosed any potential conflicts of interest. For information regarding this article, E-mail: [email protected] or [email protected] Copyright © 2010 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e3181f17d6e Objective: We recently demonstrated that transfusion of ane- mic animals up to 100 g/L hemoglobin with fresh blood protects the heart from ischemic injuries following myocardial infarction. Erythropoietin has cardioprotective effects independent of its erythropoietic activity. The objective of this study was to compare the cardioprotective effects of erythropoietin treatment to fresh- blood transfusion in anemic rats after acute myocardial infarc- tion. Design: Randomized animal study. Setting: University laboratory. Subjects: Male Sprague-Dawley rats weighing 200 –300 g. Intervention: Myocardial infarction was induced by coronary artery ligation in 76 rats, 55 of which were anemic (80 –90 g/L) and 21 of which had normal hemoglobin levels. Animals were randomized to erythropoietin (2000 units/kg), fresh-blood trans- fusion to 100 g/L hemoglobin, or saline-treatment groups imme- diately following myocardial infarction. Measurements and Main Results: At 24 hrs after myocardial infarction, cardiac function and infarct size were determined. Myocardial apoptosis was determined by caspase-3 activity and terminal deoxynucleotidyl transferase d-UTP nick end labeling (TUNEL) assay. Infarct size was significantly decreased in anemic rats treated with erythropoietin or blood transfusion compared to those in the saline-treatment group. Cardiac function, as mea- sured by maximal positive and minimal negative first derivatives of left ventricular pressure, was better preserved in the normal hemoglobin groups and the erythropoietin- or transfusion-treated anemic animals compared to saline-treated anemic animals. Myocardial caspase-3 activity and TUNEL-positive nuclei were significantly increased in anemic rats but were decreased by erythropoietin treatment or red blood cell transfusion. Conclusions: Erythropoietin treatment is equally effective as fresh-blood transfusion in anemic rats after acute myocardial infarction at reducing infarct size, myocardial apoptosis, and improving cardiac function. (Crit Care Med 2010; 38:2215–2221) KEY WORDS: anemia; blood transfusion; myocardial infarction; erythropoietin 2215 Crit Care Med 2010 Vol. 38, No. 11

Upload: others

Post on 12-Jun-2020

7 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Erythropoietin is equally effective as fresh-blood ......hemoglobin groups and the erythropoietin- or transfusion-treated anemic animals compared to saline-treated anemic animals

Neurologic Critical Care

Erythropoietin is equally effective as fresh-blood transfusion atreducing infarct size in anemic rats

Anargyros Xenocostas, MD; Houxiang Hu, MD, PhD; Nicolas Chin-Yee; Xiangru Lu, MD; Ian Chin-Yee, MD;Qingping Feng, MD, PhD

Anemia is a common conditionin patients suffering fromacute myocardial infarction(MI). The estimated prevalence

of anemia, defined as a hemoglobin (Hb)level of �120 g/L, is approximately 15–40% in patients with acute MI. Elderly

populations show increased rates withmore severe anemia (Hb �100 g/L) pre-senting in approximately 5% of individu-als (1, 2). Anemia and the resulting de-crease in the oxygen-carrying capacity ofthe blood could increase the potential fortissue damage during acute MI, and anumber of clinical studies suggest thatanemia adversely affects clinical out-comes (2–5). In the largest of these stud-ies, Sabatine et al (2) found that the mor-tality rate increased in acute coronarysyndromes as Hb levels decreased below14 g/dL, and there was an adjusted oddsratio of 1.21 (95% confidence interval,1.12–1.30) for each 10 g/L decrement inHb. Given the prevalence of anemia andits direct adverse effects on cardiovascu-lar function, it is important to seek effec-tive therapeutic interventions to reducemyocardial damage after acute MI (6, 7).

Presently, the most common methodof treating anemic patients is by redblood cell (RBC) transfusion. The effectsof transfusion on acute coronary syn-dromes have also been studied, and therehave been conflicting results. In a study

conducted by Wu et al (1), transfusionwas shown to reduce short-term mortal-ity in elderly patients after MI; however,contradictory results associating transfu-sion with higher mortality rates were ob-served in the setting of acute coronarysyndromes (8). A subsequent clinicalstudy found that although transfusionwith a nadir Hb level of �80 g/L wasassociated with an increased mortalityrate in patients experiencing acute MI,transfusion in patients with a nadir Hblevel of �80 g/L showed a protective ef-fect following MI (9). Our group recentlyfound similar results in a rat model show-ing transfusion to a Hb level of 100 g/L tobe cardioprotective, but transfusion tohigher Hb levels did not seem to be ben-eficial (10). If blood transfusion is to be-come an accepted treatment for myocar-dial ischemia, then the optimal Hb levelfor transfusion must be determined andbalanced against the risks such as volumeoverload, transfusion-related acute lunginjury, infection, and immunosuppres-sive effects associated with this interven-tion.

From the Centre for Critical Illness, Research LawsonHealth Research Institute (AX, HH, XL, ICY, QF); CanadianBlood Services (IC-Y); Departments of Medicine (AX, HH,IC-Y, QF), Physiology and Pharmacology (NC-Y, QF), Uni-versity of Western Ontario, London, Ontario, Canada;North Sichuan Medical College First Affiliated Hospital,Nanchong, Sichuan, P.R. China (HH).

This study was supported by the CIHR, CanadianBlood Services, Hema Quebec, Bayer Partnership Fund,and the Anemia Institute for Research and Education. Dr.Feng is a Heart and Stroke Foundation of Ontario CareerInvestigator.

Dr. Xenocostas has been a consultant for and hasreceived educational grants and honoraria from OrthoBiotech, Canada. Dr. I Chin-Yee is the recipient of a grantfrom Canadian Blood Services. The remaining authorshave not disclosed any potential conflicts of interest.

For information regarding this article, E-mail:[email protected] or [email protected]

Copyright © 2010 by the Society of Critical CareMedicine and Lippincott Williams & Wilkins

DOI: 10.1097/CCM.0b013e3181f17d6e

Objective: We recently demonstrated that transfusion of ane-mic animals up to 100 g/L hemoglobin with fresh blood protectsthe heart from ischemic injuries following myocardial infarction.Erythropoietin has cardioprotective effects independent of itserythropoietic activity. The objective of this study was to comparethe cardioprotective effects of erythropoietin treatment to fresh-blood transfusion in anemic rats after acute myocardial infarc-tion.

Design: Randomized animal study.Setting: University laboratory.Subjects: Male Sprague-Dawley rats weighing 200–300 g.Intervention: Myocardial infarction was induced by coronary

artery ligation in 76 rats, 55 of which were anemic (80–90 g/L)and 21 of which had normal hemoglobin levels. Animals wererandomized to erythropoietin (2000 units/kg), fresh-blood trans-fusion to 100 g/L hemoglobin, or saline-treatment groups imme-diately following myocardial infarction.

Measurements and Main Results: At 24 hrs after myocardialinfarction, cardiac function and infarct size were determined.

Myocardial apoptosis was determined by caspase-3 activity andterminal deoxynucleotidyl transferase d-UTP nick end labeling(TUNEL) assay. Infarct size was significantly decreased in anemicrats treated with erythropoietin or blood transfusion compared tothose in the saline-treatment group. Cardiac function, as mea-sured by maximal positive and minimal negative first derivativesof left ventricular pressure, was better preserved in the normalhemoglobin groups and the erythropoietin- or transfusion-treatedanemic animals compared to saline-treated anemic animals.Myocardial caspase-3 activity and TUNEL-positive nuclei weresignificantly increased in anemic rats but were decreased byerythropoietin treatment or red blood cell transfusion.

Conclusions: Erythropoietin treatment is equally effective asfresh-blood transfusion in anemic rats after acute myocardialinfarction at reducing infarct size, myocardial apoptosis, andimproving cardiac function. (Crit Care Med 2010; 38:2215–2221)

KEY WORDS: anemia; blood transfusion; myocardial infarction;erythropoietin

2215Crit Care Med 2010 Vol. 38, No. 11

Page 2: Erythropoietin is equally effective as fresh-blood ......hemoglobin groups and the erythropoietin- or transfusion-treated anemic animals compared to saline-treated anemic animals

Potential alternatives that avoid therisks of transfusion can include cytopro-tective agents administered to blunt post-ischemic injury. Erythropoietin (EPO) isa 34-kDa glycoprotein that promotes sur-vival of erythroid progenitors and therebymaintains the circulating RBC mass. Inaddition, pharmacologic doses of EPOhave been shown to produce cytoprotec-tive effects through interaction with theerythropoietin receptor, which is locatedin a variety of tissues in the cardiovascu-lar, renal, gastrointestinal, and centralnervous systems (11, 12). In the ischemicmyocardium, EPO has both immediateand delayed protective effects includingthe inhibition of cardiomyocyte apoptosis(13–16), reduction of inflammatory re-sponse to reperfusion (17, 18), improve-ment of tissue microcirculation (19), anti-arrhythmic effects (20, 21), and stimula-tion of angiogenesis (22–25).

Currently, it is unknown whether EPOcan produce similar cardioprotective ef-fects during MI in an anemic host or howthese effects compare to improved out-comes observed with fresh-blood transfu-sion. To more clearly define the effects ofanemia and blood transfusion duringacute MI we developed an animal modelin which anemia both increased infarctsize and decreased cardiac function andanimal survival, while blood transfusionreduced infarct size and improved cardiacfunction (10). In the present study, weinvestigated the cardioprotective effectsof EPO in anemic animals compared toanimals that received transfusion to ex-plore potential alternative cardioprotec-tive treatment modalities.

MATERIALS AND METHODS

Experimental Animals and Induction ofAnemia. The experiments were conducted byusing male Sprague-Dawley rats (200–300 g).All animals were provided water and food adlibitum and housed in a temperature- andhumidity-controlled facility with 12-hr lightand dark cycles. The investigation conformedto the Guide for the Care and Use of Labora-tory Animals published by the US NationalInstitutes of Health (NIH publication No. 85-23, revised 1996). Experimental protocolswere approved by the University of WesternOntario Animal Use Subcommittee.

To induce anemia, rats were fed an iron-deficient diet (10 –20 ppm; TestDiet 5859,Richmond, IN) previously described by Strubeet al (26) in combination with phlebotomy(2–3 mL) from the jugular vein twice weeklyto reach target Hb levels of 80–90 g/L. Imme-diately following phlebotomy, rats were in-fused with an equal volume of 10% pen-

tastarch (Bristol-Myers Squibb, Montreal,Quebec, Canada).

Experimental Design and Transfusion.Rats were randomly assigned to five experi-mental groups:

Group 1. Normal Hb level at 140 –150 g/L,MI with saline (vehicle) treatment(n � 11)

Group 2. Normal Hb level at 140–150 g/L, MIwith EPO treatment (n � 10)

Group 3. Hb level at 80–90 g/L, MI with saline(vehicle) treatment (n � 20)

Group 4. Hb level at 80–90 g/L, MI with EPOtreatment (n � 18)

Group 5. Hb level at 80–90 g/L, MI with bloodtransfusion to Hb level of 100 g/L(n � 17)

To achieve cardioprotective effects, highdoses of EPO (2000–5000 units/kg IV) arecommonly used and have been shown to beeffective in rodents following MI (11, 21).Therefore, in the present study, rats in groups2 and 4 were treated with IV EPO (2000 units/kg) via jugular vein immediately after coro-nary artery ligation surgery, and 0.1 mL ofsaline (vehicle) was given to control animalsin groups 1 and 3. Transfusion was performedin group 5 immediately after coronary arteryligation surgery to raise the Hb level to 100g/L by using fresh blood (stored �4 hrs). Do-nor blood was collected from rats and storedas previously described (10). Briefly, rats wereanesthetized with pentobarbital (65 mg/kg IP),and the carotid artery was cannulated to re-trieve the maximum amount of blood. Bloodwas collected into a 20-mL syringe with 3 mLof citrate-phosphate-dextrose-adenine-1 (Bax-ter, Toronto, Ontario, Canada) and injectedinto a sterile blood-collection bag (Fenwal,Baxter). The whole blood was kept at 4°C forless than 4 hrs. Right before transfusion, thestored nonleukoreduced blood was drawnfrom the collection bag and centrifuged at500g for 5 mins at 4°C. The supernatant wasremoved to achieve a hematocrit level of 75%.Total volume of packed RBCs infused toachieve 100 g/L Hb was approximately 3 mLwithin 30 mins. Thirty minutes after transfu-sion, the Hb levels in the recipient rats weremeasured from a 10-�L blood sample obtainedfrom the saphenous vein by spectrophotome-try using a Hb assay kit (Pointe Scientific,Canton, MI).

Induction of MI. MI was induced by liga-tion of the left descending coronary artery asdescribed in our previous reports (10, 27).Briefly, rats were anesthetized with an intra-muscular injection of ketamine (60 mg/kg)and xylazine (10 mg/kg). Animals were thenintubated and artificially ventilated. A left tho-racotomy was performed to expose the heart,and the left anterior descending coronary ar-tery was ligated by positioning a 6–0 silksuture between the pulmonary artery out-flow

tract and the left atrium. The lungs werethereafter hyperinflated by using positive end-expiratory pressure, and the thorax immedi-ately closed. Animals were caged individuallyafter each surgical operation.

Hemodynamic and Infarct Size Measure-ments. Hemodynamic measurements weremade to assess in vivo cardiac function andwere performed as previously described 24 hrsafter coronary artery ligation (10). Rats werereanesthetized with ketamine and xylazine,and a polyethylene catheter (PE-50) was in-serted into the right carotid artery to recordthe arterial pressure. The catheter was thenadvanced retrograde into the left ventricle(LV) for recording of LV pressures. Pressuresignals were fed to an analog-digital converterand collected by a computer. Heart rate, arte-rial pressure, LV systolic pressure, LV end-diastolic pressure, and maximal positive andminimal negative first derivatives of LV pres-sure (�dP/dtmax and �dP/dtmin; parametersthat represent contractility and diastolic relax-ation, respectively) were analyzed by Power-Lab Chart software (ADInstruments, ColoradoSprings, CO).

Infarct size was determined according toour previous reports (10, 21). Briefly, afterhemodynamic measurements, 3 mL of Evansblue dye (2% in saline) was injected into theascending part of the aorta to stain the area ofthe myocardium perfused by the patent coro-nary arteries, thereby delineating the area atrisk by negative staining. The heart was ex-cised, and the LV was carefully isolated andsliced transversely into sections 3– 4 mmthick. Sections were weighed and incubated in5% triphenyltetrazolium chloride (TTC) inphosphate-buffered saline at 37°C for 15 minsto stain the viable myocardium. Photographsof heart slices were taken with a digital cam-era. The nonrisk area (stained by Evans blue),risk area (unstained by Evans blue), and in-farct area (unstained by TTC) were quantitatedby using an image-analysis system (SigmaScanPro, Ashburn, VA). By using volumetricanalyses, the proportions of the LV that werenormally perfused, at risk, or infarcted werecalculated and expressed as a percentage oftotal LV weight. Infarct size (%) was expressedas the weight of infarcted LV divided by theweight of LV at risk.

Caspase-3 Activity Measurement.Caspase-3 activity was measured by using acaspase-3 assay kit AK-700 (BIOMOL, Ply-mouth Meeting, PA) as we previously de-scribed (27–29). Briefly, heart tissues from theperi-infarct area were homogenized, and pro-tein concentration was determined by Lowryassay. Samples (200 �g of protein) were incu-bated at 37°C for 4 hrs in the presence of thecaspase-3 substrate. N-acetyl-Asp-Glu-Val-Asp-7-amino-4-methylcoumarin with or with-out inhibitor N-acetyl-Asp-Glu-Val-Asp-CHO.Fluorescent intensity (excitation at 360 nM,emission at 460 nM) was measured by using afluorescent microplate reader (SpectraMaxM5, Molecular Devices, Sunnyvale, CA). Data

2216 Crit Care Med 2010 Vol. 38, No. 11

Page 3: Erythropoietin is equally effective as fresh-blood ......hemoglobin groups and the erythropoietin- or transfusion-treated anemic animals compared to saline-treated anemic animals

were expressed as the amount of 7-amino-4-methylcoumarin substrate cleaved per �g pro-tein per hour.

TUNEL Staining. Terminal deoxynucleo-tidyl transferase d-UTP nick end labeling(TUNEL) staining was performed by usingan In Situ Cell Death Detection kit (Roche,Indianapolis, IN) on paraffin heart sections(5 �m) as previously described (29 –31).Hoechst 33,342 was used as a counterstain.Fifteen separate fields of peri-infarct areawere examined by using a fluorescent mi-croscope (Observer D1, Zeiss, Gottingen,Germany) at �630 magnification to quantifythe number of TUNEL-positive nuclei. Im-ages were obtained by using a laser confocalmicroscope (LSM 510 Meta, Zeiss). Datawere expressed as the average percentage ofTUNEL-positive nuclei by two independentobservers.

Statistics. Data are expressed as themeans � SEM. One-way analysis of variancewas performed followed by Bonferroni test. pvalues of �.05 were considered statisticallysignificant.

RESULTS

Body Weight and Hb. All animals hada similar body weight and Hb level at

baseline (Table 1). Animals fed an iron-deficient diet in combination with phle-botomy had significantly decreased Hblevels compared to those in the standard-diet group by the end of a 2-wk periodbefore surgery; mean Hb levels reachedthe target level of 80–90 g/L. Followingtransfusion, the Hb level increased to101.0 � 0.7 g/L in anemic animals. EPOtreatment for 24 hrs did not change Hblevels in either normal (139.0 � 3.0 vs.139.5 � 4.0 g/L, p � not significant [NS])or anemic (85.7 � 2.0 vs. 86.3 � 1.5 g/L,p � NS) groups.

Effects of EPO and Transfusion onSurvival After MI. Following MI, the24-hr survival rate was monitored. In thenormal Hb group (140–150 g/L) survivalafter MI in rats treated with or withoutEPO was 100% (10/10) and 91% (10/11),respectively (Fig. 1). In the anemicgroups (Hb 80–90 g/L), animals treatedwith saline (vehicle without EPO) dem-onstrated a significantly decreased sur-vival rate after MI 40% (8/20) comparedto the normal Hb group (p � .05). Thesurvival rate improved significantly afterEPO treatment in the Hb 80 –90 g/Lgroups (72% [13/18]) 24 hrs after MIcompared to saline-treated anemic rats(40% [8/20], p � .05; Fig. 1).

To study whether blood transfusionhad any effect on survival, packed RBCswere immediately transfused following MI.The survival rate was significantly in-creased after transfusion to raise the Hblevel from 80–90 g/L to 100 g/L (77% [13/17]) compared to anemic rats receiving sa-line (40% [8/20], p � .05; Fig. 1). Further-more, survival rates were similar in theblood-transfused (77%) and EPO-treatedgroups (72%, p � NS; Fig. 1).

Effects of EPO and Transfusion on In-farct Size After MI in Anemic Rats. Twen-ty-four hours after MI, the area at risk toLV weight ratios were not significantlydifferent between any groups, indicatingthat a similar degree of myocardial isch-emia was induced in all groups (p � NS;Fig. 2A). However, the infarct to riskweight ratios were significantly increasedin the Hb 80–90 g/L saline-treated groupcompared to the normal Hb groups (p �.01; Fig. 2E). Infarct size was significantlydecreased in the EPO-treated comparedto saline-treated rats in both the Hb 140–150 g/L and Hb 80–90 g/L groups. Fur-thermore, RBC transfusion significantlydecreased infarct size compared to the Hb80–90 g/L saline-treated group (p � .01;Fig. 2E). However, no statistical differ-ence was observed between the EPO-

Figure 1. Effects of erythropoietin (EPO) treatmentand transfusion on survival in anemic rats aftermyocardial infarction (MI). Recombinant humanerythropoietin (2000 units/kg) was injected intra-venously into the tail vein immediately after MI.Transfusion was performed in rats with hemoglo-bin (Hb) levels of 80–90 g/L immediately after MI toraise Hb levels to 100 g/L using fresh-packed redblood cells (RBCs). Survival was monitored for 24hrs following surgery. n � 10–20 rats per group, *p � .05 vs. Hb 140–150 g/L � saline or EPO; † p �.05 vs. Hb 80–90 g/L � saline.

Figure 2. Effects of erythropoietin (EPO) treat-ment and packed red blood cell (RBC) transfu-sion on infarct size after myocardial infarction inanemic rats. A, Percent weight of risk area tototal left ventricular (LV) weight shows similarischemic areas between all groups; B–D, Repre-sentative heart sections with triphenyltetrazo-lium chloride staining after myocardial infarc-tion from normal hemoglobin (Hb) � saline (B),Hb 80–90 g/L � saline (C), and Hb 80–90 g/L �EPO (D) treatment groups; E, Infarct size ex-pressed as a percentage of weight of infarct toweight of LV at risk. Data are mean � SEM from7–10 rats per group. * p � .01 vs. Hb 140–150g/L saline; † p � .01 vs. Hb 80–90 g/L saline.

Table 1. Body weight and hemoglobin at baseline and following 2 wks of treatment with iron-deficientdiet in combination with bleeding before coronary artery ligation surgery in 5 experimental groups

Group

Body Weight, g Hemoglobin, g/L

BaselineBefore

Surgery BaselineBefore

Surgery

1. Hb 140–150g/L, saline (n � 11) 182.1 � 2.3 267.8 � 2.8 142.6 � 1.2 143.3 � 1.32. Hb 140–150g/L, erythropoietin

(n � 10)178.4 � 1.8 268.9 � 3.3 146.0 � 1.1 146.6 � 0.9

3. Hb 80–90g/L, saline (n � 20) 181.8 � 4.8 269.1 � 6.7 149.9 � 1.8 84.4 � 0.84. Hb 80–90g/L, erythropoietin

(n � 18)182.1 � 1.4 271.8 � 2.8 147.1 � 1.0 83.1 � 1.2

5. Hb 80–90g/L, red blood cells(n � 17)

181.3 � 2.4 274.4 � 2.8 145.4 � 1.3 82.5 � 0.9

Hb, hemoglobin.Data are mean � standard error. Red blood cells indicate packed red blood cell transfusion to reach

Hb of 100 g/L. Body weight was not statistically different between groups. Hemoglobin levels were notsignificantly different within the normal or anemic groups.

2217Crit Care Med 2010 Vol. 38, No. 11

Page 4: Erythropoietin is equally effective as fresh-blood ......hemoglobin groups and the erythropoietin- or transfusion-treated anemic animals compared to saline-treated anemic animals

treated and transfused groups of anemicrats (p � NS; Fig. 2E).

Effects of EPO and Transfusion onCardiac Function After MI in AnemicRats. Cardiac function was determined 24hrs after MI. Heart rate was not signifi-cantly different between all groups (p �NS; Fig. 3A). The mean arterial pressurewas significantly decreased in the anemicgroups without blood transfusion com-pared to the normal Hb groups (p � .01;Fig. 3B). However, the mean arterialpressure was significantly increased inthe RBC-transfused group compared tothe saline-treated group (p � .01; Fig.3B). LV systolic pressure was decreased inthe Hb 80–90 g/L saline-treated groupcompared to the normal Hb saline-treated group (p � .01). RBC transfusionsignificantly increased LV systolic pres-sure compared to the Hb 80–90 g/L sa-line-treated group (p � .01; Fig. 4A). LVend-diastolic pressure was increased onlyin the Hb 80–90 g/L saline-treated groupcompared to the normal Hb saline- orEPO-treated group (p � .01). However,the LV end-diastolic pressure in the EPO-treated and RBC-transfused groups wasnot significantly different when com-pared to their respective controls (p �NS; Fig. 4B).

Reduction in Hb levels from 140–150to 80–90 g/L after MI correlated with asignificant decrease in LV �dP/dtmax

(contractility) and �dP/dtmin (diastolic

function) in saline-treated rats (p � .01;Fig. 5). EPO treatment in anemic ratssignificantly increased LV �dP/dtmax

(p � .01) and �dP/dtmin (p � .05) com-pared to anemic rats treated with saline.Fresh RBC transfusion to raise Hb levelsto 100 g/L significantly increased LV�dP/dtmax and �dP/dtmin compared tothe saline-treated group (p � .01; Fig. 5).There was no significant difference in LV�dP/dtmax or �dP/dtmin between theEPO-treated and RBC-transfused groups(p � NS). Central venous pressure wasmonitored by placing a PE-50 into thesuperior vena cava via the right jugularvein before and 1 hr after transfusion.Central venous pressure was not signifi-cantly altered by RBC transfusion (4.8 �1.0 vs. 5.6 � 0.6 mm Hg, n � 5, p � NS).

Effects of EPO and Transfusion onMyocardial Apoptosis After MI in AnemicRats. To assess possible mechanisms bywhich EPO and RBC transfusion improvethe outcome of these animals, myocardialapoptosis was determined. Caspase-3 ac-tivity in the peri-infarct area was signifi-cantly increased in anemic rats comparedto that in normal Hb rats (p � .05; Fig.6A). EPO treatment and fresh RBC trans-fusion significantly decreased the myo-cardial caspase-3 activity compared totheir respective controls (p � .05; Fig.6A). However, there was no significant

difference between EPO treatment andfresh RBC transfusion in anemic rats(p � NS; Fig. 6A).

Myocardial apoptosis was further ana-lyzed by using TUNEL staining. Repre-sentative images of TUNEL staining fornormal Hb and saline, Hb 80–90 g/L andsaline, and Hb 80–90 g/L and EPO treat-ment groups are shown in Figure 6B, 6C,and 6D, respectively. Our data showedthat the number of TUNEL-positive nu-clei was significantly increased in theperi-infarct area of the anemic rats com-pared to the Hb 140–150 g/L group (p �.05; Fig. 6E). EPO treatment in both nor-mal and anemic rats, and fresh RBCtransfusion to raise Hb levels to 100 g/L,significantly decreased TUNEL-positivenuclei compared to their respective salinecontrols (p � .05; Fig. 6E). Consistentwith the caspase-3 data, there was nosignificant difference between EPO treat-ment and fresh RBC transfusion in ane-mic rats (p � NS; Fig. 6E).

DISCUSSION

The present study demonstrated forthe first time in an anemic rat model thatEPO treatment independent of erythro-poiesis is cardioprotective following acuteMI, and that the beneficial effects aresimilar in magnitude compared to those

Figure 3. Effects of erythropoietin (EPO) treatmentand transfusion on heart rate (A) and mean arterialpressure (MAP) (B) 24 hrs after myocardial infarc-tion in anemic rats. Data are mean � SEM from7–10 rats per group. RBC, red blood cell. * p � .05vs. hemoglobin (Hb) 140–150 g/L � saline; † p �.01 vs. Hb 80–90 g/L � EPO.

Figure 4. Effects of erythropoietin (EPO) treat-ment and transfusion on left ventricular (LV)pressures 24 hrs after myocardial infarction inanemic rats. A, Left ventricular systolic pressure(LVSP); B, Left ventricular end diastolic pressure(LVEDP). Data are mean � SEM from 7–10 ratsper group. RBC, red blood cell. * p � .05 vs.hemoglobin (Hb) 140–150 g/L � saline; † p �.05 vs. Hb 80–90 g/L � saline.

Figure 5. Effects of erythropoietin (EPO) treat-ment and transfusion on left ventricular (LV)contractility (�dP/dt) and relaxation (�dP/dt) 24hrs after myocardial infarction in anemic rats. A,LV � dP/dtmax; B, LV � dP/dtmin. Data aremean � sem from 7–10 rats per group. RBC, redblood cell. * p � .01 vs. hemoglobin (Hb) 140–150 g/L � saline or EPO; † p � .01 vs. Hb 80–90g/L � saline.

2218 Crit Care Med 2010 Vol. 38, No. 11

Page 5: Erythropoietin is equally effective as fresh-blood ......hemoglobin groups and the erythropoietin- or transfusion-treated anemic animals compared to saline-treated anemic animals

observed after RBC transfusion in whichthe Hb level was increased by 20 g/L to atarget Hb level of 100 g/L. Both EPOtreatment and RBC transfusion increasedanimal survival and decreased myocardialapoptosis and infarct size after MI. Car-diac function as measured by rate of LVcontraction and relaxation was also sig-nificantly improved in both the EPO andtransfusion groups.

EPO induces cardioprotective effectsduring ischemia in animals with normalHb levels via a number of mechanismsindependent of erythropoeisis (11, 21,31). EPO binds to the EPO receptor andsubsequently activates the PI3-kinase/Aktpathway, leading to reduction of cardio-myocyte apoptosis (11, 13, 30, 31). EPOalso improves skeletal muscle microcir-culation in sepsis (19) and reduces ven-tricular fibrillation during myocardialischemia and reperfusion (20). We fur-ther demonstrated that the antiarrhyth-mic effects are mediated by increased ex-pression of neuronal nitric oxide synthasefollowing EPO treatment (21). Finally,the beneficial effects of EPO are also me-diated by its anti-inflammatory propertiesto reduce cytokines and oxidative stress(17, 18, 32). In the present study, wedemonstrated that EPO treatment de-creased myocardial apoptosis, thereforeincreasing the number of surviving car-diomyocytes and preserving myocardialfunction. In anemic animals without EPOtreatment this effect was absent, whichresulted in reduced myocardial function.

The magnitude of the protective ef-fects of EPO in the absence of anychanges in Hb is striking in our study,because it was similar to the effects seenwhen Hb was increased by 20 g/L. Al-though we saw beneficial effects fromboth EPO treatment and transfusion, wedid not test both treatments together todetermine if there could be additive orsynergistic effects for further improve-ment in outcomes. However, the differ-ences in outcomes between the EPO-treated group, the transfused rats, andthe control nonanemic animals weresmall. It could be argued that both treat-ments used in isolation produced near-maximal salvaging of myocardium at riskin this model of irreversible ischemia.There are limits to the amount of myo-cardium that can be salvaged after irre-versible ischemia, and it is unlikely thatcombining both treatments would be ableto improve outcomes further in thismodel of MI.

Although transfusion remains themost common therapeutic interventionfor the immediate correction of anemia,its effectiveness is controversial (1, 8, 9),and it is associated with a number of risksincluding volume overload, transfusion-related acute lung injury, infection, andimmunosuppressive effects (33). EPOtreatment avoids these risks because thevolume administered is negligible andthere is no exposure to alloantigens or

cytokines. However, potential drawbacksof EPO might include the activation ofplatelets with an increased incidence ofthrombosis (34)—an adverse effect thathas to be considered even in the contextof antiplatelet therapy used as part of thestandard treatment for acute MI.

STUDY LIMITATIONS

1. The model used was one of irreversibleischemia, and we did not examine theprotective effects of EPO followingreperfusion injury. In the clinical set-ting, patients with acute coronary syn-dromes are commonly treated bythrombolysis or angioplasty. However,the protective effects of EPO seem toextend to reperfusion models as well(14, 23, 31). Furthermore, becauseour experiment only measured out-comes 24 hrs following MI, long-termeffects were not examined. In otherstudies EPO treatment was shown toreduce infarct size, promote angiogen-esis, and improve cardiac function 4 to9 wks after MI (23, 24).

2. The anemic group was transfused toonly one Hb level of 100 g/L, becauseour previous study found that a targetHb level of 100 g/L was optimal forincreased animal survival, decreasedinfarct size, and improved cardiacfunction (10). Clinically, transfusionto a Hb level of �80 g/L failed to showany benefits and actually seemed to bedeleterious (9, 35, 36). Therefore, ex-trapolation from the current animalstudy to the human population interms of transfusion threshold shouldbe cautioned. In addition, to minimizeRBC-storage lesions, blood was storedless than 4 hrs before transfusion inthe present study. However, the use of4-hr fresh blood may not be possiblein clinical practice.

3. Central venous pressure is commonlyused to assess cardiac preload and vol-ume status (37). In the present study,volume overload was not observedbased on central venous pressure mea-surements during blood transfusionover a period of 30 mins. It is impor-tant to note that changes in centralvenous pressure are affected by thespeed of fluid infusion and complianceof the cardiovascular system (37).Thus, central venous pressure mea-surements alone may not accuratelyreflect blood volume changes (38).

4. -Adrenergic blockers have beenshown to improve coronary perfusion

Figure 6. Effects of erythropoietin (EPO) treat-ment and transfusion on myocardial apoptosis inperi-infarct area 24 hrs after myocardial infarc-tion in anemic rats. A, Myocardial caspase-3 ac-tivity; B–D, Representative confocal images ofterminal deoxynucleotidyl transferase d-UTP nickend labeling (TUNEL) staining from normal he-moglobin (Hb) � saline (B), Hb 80–90 g/L �saline (C), and Hb 80–90 g/L � EPO-treated (D)groups. TUNEL-positive nuclei stained in greenwith Hoechst (blue) as a nuclear stain. Whitearrows indicate positive signal. E, Quantitativeanalysis of TUNEL-positive nuclei. Data aremean � SEM, n � 5 rats per group. RBC, redblood cell. * p � .05 vs. Hb 140–150 g/L � saline;† p � .01 vs. Hb 80–90 g/L � saline.

2219Crit Care Med 2010 Vol. 38, No. 11

Page 6: Erythropoietin is equally effective as fresh-blood ......hemoglobin groups and the erythropoietin- or transfusion-treated anemic animals compared to saline-treated anemic animals

reserve, promote the growth of coro-nary arterioles, and facilitate regionalmyocardial perfusion in rat MI models(39). Because -blockade is a standardtreatment strategy in patients with MI,further studies using -blockers incombination with EPO or transfusionare warranted.

5. Induction of an iron-deficient state inthe anemic animals itself may haveinfluenced cardiac function (40). How-ever, our iron-deficient rats undergo-ing treatment demonstrated similargrowth and hemodynamics comparedto control rats. It is also clinically rel-evant to note that a significant propor-tion of patients with acute coronarysyndromes are anemic, and althoughthe anemia may not be due to irondeficiency, ongoing inflammatorystress can contribute to anemia ofchronic disease resulting in a func-tional iron-deficiency state (41).

CONCLUSIONS

A novel finding from our study is thatthe cardioprotective effects of EPO can beextended to the anemic host—an impor-tant consideration given the significantnumber of patients who have acute cor-onary syndrome and are anemic. Fur-thermore, EPO was equally effective asfresh-blood transfusion at reducing in-farct size and improving cardiac function,and these effects occurred in the absenceof any changes in Hb. Our findings sug-gest that use of EPO may offer a newtreatment strategy for anemic patientsexperiencing acute MI. The demonstra-tion of clinical benefits of EPO requiresfurther clinical study and awaits the re-sults of ongoing trials.

REFERENCES

1. Wu WC, Rathore SS, Wang Y, et al: Bloodtransfusion in elderly patients with acutemyocardial infarction. N Engl J Med 2001;345:1230–1236

2. Sabatine MS, Morrow DA, Giugliano RP, etal: Association of hemoglobin levels withclinical outcomes in acute coronary syn-dromes. Circulation 2005; 111:2042–2049

3. Langston RD, Presley R, Flanders WD, et al:Renal insufficiency and anemia are indepen-dent risk factors for death among patientswith acute myocardial infarction. Kidney Int2003; 64:1398–1405

4. Nikolsky E, Aymong ED, Halkin A, et al:Impact of anemia in patients with acute myo-cardial infarction undergoing primary percu-taneous coronary intervention: Analysis fromthe Controlled Abciximab and Device Inves-

tigation to Lower Late Angioplasty Compli-cations (CADILLAC) Trial. J Am Coll Cardiol2004; 44:547–553

5. Lee PC, Kini AS, Ahsan C, et al: Anemia is anindependent predictor of mortality after per-cutaneous coronary intervention. J Am CollCardiol 2004; 44:541–546

6. Gerber DR: Transfusion of packed red bloodcells in patients with ischemic heart disease.Crit Care Med 2008; 36:1068–1074

7. Napolitano LM, Kurek S, Luchette FA, et al:Clinical practice guideline: Red blood celltransfusion in adult trauma and critical care.Crit Care Med 2009; 37:3124–3157

8. Rao SV, Jollis JG, Harrington RA, et al: Re-lationship of blood transfusion and clinicaloutcomes in patients with acute coronarysyndromes. JAMA 2004; 292:1555–1562

9. Aronson D, Dann EJ, Bonstein L, et al: Im-pact of red blood cell transfusion on clinicaloutcomes in patients with acute myocardialinfarction. Am J Cardiol 2008; 102:115–119

10. Hu H, Xenocostas A, Chin-Yee I, et al: Effectsof anemia and blood transfusion in acutemyocardial infarction in rats. Transfusion2010; 50:243–251

11. Burger D, Xenocostas A, Feng Q: Molecularbasis of cardioprotection by erythropoietin.Curr Mol Pharmacol 2009; 2:56–69

12. Brines M, Cerami A: Erythropoietin-medi-ated tissue protection: Reducing collateraldamage from the primary injury response.J Intern Med 2008; 264:405–432

13. Tramontano AF, Muniyappa R, Black AD, etal: Erythropoietin protects cardiac myocytesfrom hypoxia-induced apoptosis through anAkt-dependent pathway. Biochem BiophysRes Commun 2003; 308:990–994

14. Calvillo L, Latini R, Kajstura J, et al: Recom-binant human erythropoietin protects themyocardium from ischemia-reperfusion in-jury and promotes beneficial remodeling.Proc Nat Acad Sci U S A 2003; 100:4802–4806

15. Parsa CJ, Matsumoto A, Kim J, et al: A novelprotective effect of erythropoietin in the in-farcted heart. J Clin Invest 2003; 112:999–1007

16. Cai Z, Manalo DJ, Wei G, et al: Hearts fromrodents exposed to intermittent hypoxia orerythropoietin are protected against isch-emia-reperfusion injury. Circulation 2003;108:79–85

17. Rui T, Feng Q, Lei M, et al: Erythropoietinprevents the acute myocardial inflammatoryresponse induced by ischemia/reperfusionvia induction of AP-1. Cardiovasc Res 2005;65:719–727

18. Li Y, Takemura G, Okada H, et al: Reductionof inflammatory cytokine expression and ox-idative damage by erythropoietin in chronicheart failure. Cardiovasc Res 2006; 71:684–694

19. Kao R, Xenocostas A, Rui T, et al: Erythro-poietin improves skeletal muscle microcircu-lation and tissue bioenergetics in a mousesepsis model. Crit Care 2007; 11:R58

20. Hirata A, Minamino T, Asanuma H, et al:

Erythropoietin just before reperfusion re-duces both lethal arrhythmias and infarctsize via the phosphatidylinositol-3 kinase-dependent pathway in canine hearts. Cardio-vasc Drugs Ther 2005; 19:33–40

21. Burger DE, Xiang FL, Hammoud L, et al:Erythropoietin protects the heart from ven-tricular arrhythmia during ischemia andreperfusion via neuronal nitric-oxide syn-thase. J Pharmacol Exp Ther 2009; 329:900–907

22. Nakano M, Satoh K, Fukumoto Y, et al: Im-portant role of erythropoietin receptor topromote VEGF expression and angiogenesisin peripheral ischemia in mice. Circ Res2007; 100:662–669

23. Nishiya D, Omura T, Shimada K, et al: Effectsof erythropoietin on cardiac remodeling aftermyocardial infarction. J Pharmacol Sci 2006;101:31–39

24. van der Meer P, Lipsic E, Henning RH, et al:Erythropoietin induces neovascularizationand improves cardiac function in rats withheart failure after myocardial infarction.J Am Coll Cardiol 2005; 46:125–133

25. Jaquet K, Krause K, Tawakol-Khodai M, et al:Erythropoietin and VEGF exhibit equal an-giogenic potential. Microvasc Res 2002; 64:326–333

26. Strube YN, Beard JL, Ross AC: Iron defi-ciency and marginal vitamin A deficiency af-fect growth, hematological indices and theregulation of iron metabolism genes in rats.J Nutr 2002; 132:3607–3615

27. Xiang FL, Lu X, Hammoud L, et al: Cardiom-yocyte-specific overexpression of humanstem cell factor improves cardiac functionand survival post myocardial infarction inmice. Circulation 2009; 120:1065–1074

28. Feng Q, Song W, Lu X, et al: Development ofheart failure and congenital septal defects inmice lacking endothelial nitric oxide syn-thase. Circulation 2002; 106:873–879

29. Lu X, Hamilton JA, Shen J, et al: Role oftumor necrosis factor-alpha in myocardialdysfunction and apoptosis during hindlimbischemia and reperfusion. Crit Care Med2006; 34:484–491

30. Burger D, Lei M, Geoghegan-Morphet N, etal: Erythropoietin protects cardiomyocytesfrom apoptosis via up-regulation of endothe-lial nitric oxide synthase. Cardiovasc Res2006; 72:51–59

31. Burger D, Xiang F, Hammoud L, et al: Roleof heme oxygenase-1 in the cardioprotectiveeffects of erythropoietin during myocardialischemia and reperfusion. Am J PhysiolHeart Circ Physiol 2009; 296:H84–H93

32. Feng Q: Beyond erythropoiesis: the anti-inflammatory effects of erythropoietin. Car-diovasc Res 2006; 71:615–617

33. Murphy GJ: Does blood transfusion harmcardiac surgery patients? BMC Med 2009;7:38

34. Stohlawetz PJ, Dzirlo L, Hergovich N, et al:Effects of erythropoietin on platelet reactiv-ity and thrombopoiesis in humans. Blood2000; 95:2983–2989

2220 Crit Care Med 2010 Vol. 38, No. 11

Page 7: Erythropoietin is equally effective as fresh-blood ......hemoglobin groups and the erythropoietin- or transfusion-treated anemic animals compared to saline-treated anemic animals

35. Hebert PC: The TRICC trial: A focus on thesub-group analysis. Vox Sang 2002; 83 Suppl1:387–396

36. Yang X, Alexander KP, Chen AY, et al: Theimplications of blood transfusions for pa-tients with non-ST-segment elevation acutecoronary syndromes: results from the CRU-SADE National Quality Improvement Initia-tive. J Am Coll Cardiol 2005; 46:1490–1495

37. Magder S: Central venous pressure: A useful

but not so simple measurement. Crit CareMed 2006; 34:2224–2227

38. Kumar A, Anel R, Bunnell E, et al: Pulmo-nary artery occlusion pressure and centralvenous pressure fail to predict ventricularfilling volume, cardiac performance, or theresponse to volume infusion in normal sub-jects. Crit Care Med 2004; 32:691–699

39. Dedkov EI, Christensen LP, Weiss RM, et al:Reduction of heart rate by chronic beta1-

adrenoceptor blockade promotes growth ofarterioles and preserves coronary perfusionreserve in postinfarcted heart. Am J PhysiolHeart Circ Physiol 2005; 288:H2684–H2693

40. Petering DH, Stemmer KL, Lyman S, et al:Iron deficiency in growing male rats: A causeof development of cardiomyopathy. Ann NutrMetab 1990; 34:232–243

41. Weiss G, Goodnough LT: Anemia of chronicdisease. N Engl J Med 2005; 352:1011–1023

2221Crit Care Med 2010 Vol. 38, No. 11