Intraarterial Administration of Bone Marrow Mononuclear ... · Background—Critical limb ischemia due to peripheral arterial occlusive disease is associated with a severely increased

M. Zeiher and for the PROVASA InvestigatorsSchlüter, Torsten Tonn, Florian Seeger, Stefanie Dimmeler, Edelgard Lindhoff-Last, Andreas

Dirk H. Walter, Hans Krankenberg, Jörn O. Balzer, Christoph Kalka, Iris Baumgartner, MichaelLimb Ischemia: A Randomized-Start, Placebo-Controlled Pilot Trial (PROVASA)

Intraarterial Administration of Bone Marrow Mononuclear Cells in Patients With Critical

Print ISSN: 1941-7640. Online ISSN: 1941-7632 Copyright © 2011 American Heart Association, Inc. All rights reserved.

Avenue, Dallas, TX 75231is published by the American Heart Association, 7272 GreenvilleCirculation: Cardiovascular Interventions

doi: 10.1161/CIRCINTERVENTIONS.110.9583482011;4:26-37; originally published online January 4, 2011;Circ Cardiovasc Interv.

http://circinterventions.ahajournals.org/content/4/1/26World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circinterventions.ahajournals.org//subscriptions/

is online at: Circulation: Cardiovascular Interventions Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Answer

Permissions and Rights Question andunder Services. Further information about this process is available in thepermission is being requested is located, click Request Permissions in the middle column of the Web pageClearance Center, not the Editorial Office. Once the online version of the published article for which

can be obtained via RightsLink, a service of the CopyrightCirculation: Cardiovascular Interventionsin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

at UNIV OF KANSAS SCH OF MED on September 3, 2014http://circinterventions.ahajournals.org/Downloaded from at UNIV OF KANSAS SCH OF MED on September 3, 2014http://circinterventions.ahajournals.org/Downloaded from

http://circinterventions.ahajournals.org/content/4/1/26

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circinterventions.ahajournals.org//subscriptions/

http://circinterventions.ahajournals.org/


Intraarterial Administration of Bone Marrow MononuclearCells in Patients With Critical Limb Ischemia

A Randomized-Start, Placebo-Controlled Pilot Trial (PROVASA)

Dirk H. Walter, MD; Hans Krankenberg, MD; Jorn O. Balzer, MD; Christoph Kalka, MD;Iris Baumgartner, MD; Michael Schluter, MD; Torsten Tonn, MD; Florian Seeger, MD;Stefanie Dimmeler, PhD; Edelgard Lindhoff-Last, MD; Andreas M. Zeiher, MD; for the

PROVASA Investigators

Background—Critical limb ischemia due to peripheral arterial occlusive disease is associated with a severely increasedmorbidity and mortality. There is no effective pharmacological therapy available. Injection of autologous bonemarrow-derived mononuclear cells (BM-MNC) is a promising therapeutic option in patients with critical limb ischemia,but double-blind, randomized trials are lacking.

Methods and Results—Forty patients with critical limb ischemia were included in a multicenter, phase II,double-blind, randomized-start trial to receive either intraarterial administration of BM-MNC or placebo followedby active treatment with BM-MNC (open label) after 3 months. Intraarterial administration of BM-MNC did notsignificantly increase ankle-brachial index and, thus, the trial missed its primary end point. However, cell therapywas associated with significantly improved ulcer healing (ulcer area, 3.2�4.7 cm2 to 1.89�3.5 cm2 [P�0.014]versus placebo, 2.92�3.5 cm2 to 2.89�4.1 cm2 [P�0.5]) and reduced rest pain (5.2�1.8 to 2.2�1.3 [P�0.009]versus placebo, 4.5�2.4 to 3.9�2.6 [P�0.3]) within 3 months. Limb salvage and amputation-free survival ratesdid not differ between the groups.

Repeated BM-MNC administration and higher BM-MNC numbers and functionality were the only independent predictorsof improved ulcer healing. Ulcer healing induced by repeated BM-MNC administration significantly correlated withlimb salvage (r�0.8; P�0.001).

Conclusions—Intraarterial administration of BM-MNC is safe and feasible and accelerates wound healing in patientswithout extensive gangrene and impending amputation. These exploratory findings of this pilot trial need to beconfirmed in a larger randomized trial in patients with critical limb ischemia and stable ulcers.

Clinical Trial Registration—URL: http://www.clinicaltrials.gov. Unique identifier: NCT00282646.(Circ Cardiovasc Interv. 2011;4:26-37.)

Key Words: stem cells � peripheral vascular disease � angiogenesis

Critical limb ischemia, defined as rest pain or tissuenecrosis with ulceration or gangrene, attributable to

peripheral arterial occlusive disease (PAOD) not only has amajor impact on quality of life, but also is associated with adramatically increased mortality.1 Despite recent advances ininterventional or surgical techniques, prognosis and amputation-free survival remains poor because a large number of patientswith critical limb ischemia are not candidates for suchrevascularization procedures.1,2 At present, no effective phar-macological therapy is available.3

Editorial see p 2Clinical Perspective on p 37

Recent experimental studies indicated that autologous bonemarrow-derived mononuclear cells (BM-MNC) as well asselected CD34�, CD133�, or CXCR4� cells or culturedperipheral blood-derived proangiogenic cells may be a prom-ising therapeutic option in patients with critical limb ische-mia.4 Specifically, the first landmark trial, the TherapeuticAngiogenesis by Cell Transplantation (TACT) study,5 dem-

Received June 21, 2010; accepted November 30, 2010.From the Division of Cardiology and Angiology (D.H.W., F.S., S.D., E.L.-L., A.M.Z.) and Department of Radiology (J.O.B.), University of Frankfurt,

Frankfurt, Germany; Hamburg University Cardiovascular Center: Prof Mathey, Prof Schofer (D.H.W., H.K., M.S.), Hamburg, Germany; VascularMedicine (C.K., I.B.), Inselspital Bern, Bern, Germany; and Institute for Transfusion Medicine and Immunohematology, Frankfurt Red Cross BloodDonor Service, Baden-Wuerttemberg-Hessen, Germany.

Correspondence to Andreas Zeiher, MD, or Dirk H. Walter, MD, Cardiology and Vascular Medicine, University of Frankfurt, Theodor-Stern Kai 7,60590 Frankfurt, Germany. E-mail [email protected] or [email protected]

© 2011 American Heart Association, Inc.

Circ Cardiovasc Interv is available at http://circinterventions.ahajournals.org DOI: 10.1161/CIRCINTERVENTIONS.110.958348

26 at UNIV OF KANSAS SCH OF MED on September 3, 2014http://circinterventions.ahajournals.org/Downloaded from


onstrated a significant improvement of ankle-brachial index(ABI), rest pain, and transcutaneous oxygen pressure (TCO2)after intramuscular injection of autologous BM-MNC. However,despite a number of additional smaller, nonrandomized trialsshowing promising results of cell transplantation in patients withPAOD or thrombangiitis obliterans (TAO)6–11 (for review, seeSprengers et al4), definitive proof is still not available because ofthe lack of double-blinded controls. In addition, although themajority of trials have tested a potential effect of intramuscularinjection of cells, no data are available that assessed the effectsof intraarterial administration of BM-MNC.

Therefore, we designed the Intraarterial Progenitor CellTransplantation of Bone Marrow Mononuclear Cells for Induc-tion of Neovascularization in Patients With Peripheral ArterialOcclusive Disease (PROVASA) trial in order to test the hypoth-esis that intraarterial administration of BM-MNC is associatedwith improved limb perfusion, resulting in an increase of ABIwith reduction of ischemic rest pain and improved healing oftissue necrosis as clinically relevant secondary end points.

MethodsPatients and Study DesignThe PROVASA trial is a multicenter phase II trial with a double-blindrandomized-start design. At baseline, in a double-blind fashion, 40

patients were randomly assigned (1:1) to receive BM-MNC treatment orplacebo and were followed for 3 months (randomized start). At 3months, all patients received active treatment with BM-MNC and werefollowed for another 3 months (open-label). Thus, at the end of 3months, all patients who had received placebo switched to activetreatment (BM-MNC, crossover), and patients who had received activetreatment initially received a repetitive treatment with BM-MNC. Aprespecified analysis was performed after the randomized-start phase aswell as at the end of 6 months follow-up, where the 3-month effects ofa single treatment with BM-MNC (crossover from the initial placebogroup) could be compared with potential effects of a repetitive treatmentat 3-month intervals (initial verum group). For patients with ulcers(Rutherford class 5) and evidence of delayed wound healing after 6months, the PROVASA trial was extended by an amendment in order toallow for additional serial treatments with up to 3 more intraarterialapplications of BM-MNC (verum). Timing and indication for continu-ation of BM-MNC therapy was decided according to the estimation ofthe treating physician in case of delayed wound healing. An interval ofat least 3 months was required between serial treatments. The studyoutline is shown in Figure 1.

The study was approved by the review boards of the local ethicscommittees in Frankfurt, Hamburg, and Bern, Germany, and by thenational authority, the Paul-Ehrlich-Institut, Langen, Germany. In-dependent on-site monitoring was available at each study site. Thestudy was conducted in accordance with the Declaration of Helsinki.

Inclusion CriteriaPatients aged 18 to 80 years with ischemic rest pain (Rutherford class4) or nonhealing ulcers (Rutherford class 5 or 6) due to PAOD and

Figure 1. Study flow chart of the PROVASA study.f/u indicates follow-up. *Major amputations inpatients with advanced Rutherford class 5 or 6.†Patient withdrawn from study because of coloncancer at 1 month. ‡Major amputation precedingdeath. #Extended protocol: in case of delayedwound healing, BM-MNC treatments could berepeated.

Walter et al Stem Cells and PAOD 27

at UNIV OF KANSAS SCH OF MED on September 3, 2014http://circinterventions.ahajournals.org/Downloaded from


TAO who were not candidates for interventional or surgical revas-cularization or who failed to respond to interventional or surgicalprocedures were eligible for inclusion in the study. To be able todocument a cell effect, technically successful interventional orsurgical procedures (defined as patent artery after percutaneoustransluminal angioplasty [PTA] or patent bypass graft) must havebeen performed at least 3 months before inclusion. Patients hadeither infrapopliteal vessel occlusions or chronic femoropoplitealocclusions. Written informed consent was obtained from all patientsbefore enrollment.

Exclusion CriteriaTechnically successful interventional or surgical procedures �3months before screening, a history of infectious diseases (HIV,active hepatitis) or evidence for chronic inflammatory diseases (eg,Crohn disease, rheumatoid arthritis), a history of malignancieswithout complete remission (�5 years), a history of stroke ormyocardial infarction �3 months before screening, or advancedrenal insufficiency (creatinine �2.0 mg/dL at the time of treatment)were the exclusion criteria.

End PointsThe primary end point was the change in ABI of the treated leg at 3and 6 months based on the findings of the TACT study. Secondaryend points were complete healing of all ulcers or a reduction of ulcerarea (cumulative ulcer size in square centimeters), amputation-freesurvival (major amputation above the ankle) and overall survival,and freedom from rest pain or reduction of rest pain (by at least 1point). Wound healing was assessed by documentation using digitalcameras and standardized measurement (Visitrak) to quantify totalulcer area (square centimeters). Rest pain or severity of pain wasassessed on a visual analog scale ranging from 0 to 10 points, with0 indicating the best (complete relief of pain without analgesics) and10, the worst pain.

TCO2 was measured with the TCM3 (Radiometer; Copenhagen,Denmark) at 2 prespecified sites of the lower limb (dorsal surface ofthe foot and surface of the lower calf) at 2 centers (Frankfurt, Bern)and is presented as a subgroup analysis. Quantitative analyses forulcer area and pain scale are presented for patients with completeserial measurements at all time points (baseline, month 3, and month6). Physicians or study nurses who measured end points (ABI, ulcersize, pain scale, TCO2) were blinded to the treatment.

BM-MNC IsolationProcessing of bone marrow aspirates and cell isolation procedureswere identical to the protocols used in the REPAIR-AMI (Reinfusionof Enriched Progenitor Cells and Infarct Remodeling in AcuteMyocardial Infarction) trial.12–13 In brief, 50 mL of bone marrow wasaspirated from the iliac crest into heparin-treated syringes with theuse of local anesthesia. The bone marrow aspirate, together with 20mL of venous blood used to produce the patient’s own serum, wasshipped at room temperature to the central cell-processing laboratoryat the Red Cross Blood Donation Center in Frankfurt, where patientswere randomly assigned to receive BM-MNC or placebo in therandomized-start phase. The randomization was performed for theentire study cohort at the cell-isolation center by a simple randomallocation. There was no blocking at each center.

BM-MNC were isolated and enriched using density gradientcentrifugation. After several washing steps, cells were resuspendedin X-VIVO 10 medium (a serum-free medium containingpharmaceutical-grade human components (Cambrex) supplementedwith 2 mL of the patient’s own serum. Cell analysis was performedfrom the final cell preparation. Placebo control was prepared bymixing X-VIVO 10 medium with autologous serum. All cell pro-cessing and labeling of the product were performed according togood clinical practice guidelines. The method of cell isolation hasbeen validated extensively in Frankfurt,13 and cell viability is provenuntil at least 24 hours. Cell viability was �98% in all cases, andneither viability nor functional capacity of cells was compromised byshipping to external centers. Our preclinical experiments and previ-

ous clinical trials using BM-MNC in patients with ischemic heartdisease provided the basis for the PROVASA trial.12,14–19

Cell CharacterizationThe cell suspension consisted of a heterogeneous cell population thatincluded hematopoietic, mesenchymal, and other progenitor cells aswell as mononuclear cells. The total number of mononuclear cellswas determined using a differential counter (XT-1800i; Sysmex).Cells were further characterized by flow cytometry using CD34�/CD45�/CD133� antibodies. Viability was assessed using trypanblue staining. Functional parameters of the final cell product, such asbasal or stromal-derived factor 1-induced migration, were assessedin a Boyden chamber as previously described.18 Colony forming unit(CFU) capacity assays were performed using Methocult H4534Classic without erythropoietin (Stemcell Technologies), which supportsgrowth of granulocyte, macrophage, and granulocyte-macrophageCFUs. For quantification, all 3 types of colonies were detected.

Intraarterial Catheter-Based Applicationof BM-MNCAfter arterial puncture, all patients received at least 5000 U ofheparin. Angiographies were performed by way of crossover sheath.Cells or placebo solutions were administered by hand injection intothe distal superficial femoral artery in patients with infrapoplitealdisease or into the deep femoral artery in patients with additionalchronic femoropopliteal occlusions. Low-pressure balloon occlu-sions of the superficial femoral artery (SFA) for 1�5 minutes wereused in 20 patients with faster distal runoff (eg, patients with TAOwith small vessel disease of the foot) but not in patients with slowdistal runoff in the lower limb by way of collaterals (n�20).

Data and Statistical AnalysisSample size calculation for the primary end point was based on thefindings in the TACT study5: assuming a mean increase in ABI of0.13 (SD, 0.1) after BM-MNC treatment and no increase afterplacebo, a sample size of 15 patients per group was considerednecessary to document a significant effect with a statistical power of90% (2-sided ��0.05). Taking patient dropout and death or majoramputation within 6 months into consideration led to a sample sizeof 20 patients per treatment group.

Data are expressed as counts and percentages for discrete variablesand as mean�SD for continuous variables or median values withinterquartile ranges (IQRs) or 95% CIs. Categorical variables werecompared by means of the Fisher exact test. Continuous variableswere compared by nonparametric methods (Mann-Whitney U test).Paired comparisons were performed by Wilcoxon test. Multivariateanalysis was performed using the Cox regression analysis on SPSSversion 15.0 software, including fixed covariates. Statistical signif-icance was assumed to be 2-sided at P�0.05. Long-term clinicaloutcome and associations with wound healing or amputation werecompared by Kaplan-Meier survival curves, and the corresponding Pvalue was obtained from the log-rank test. Overall survival andamputation-free survival were counted as the duration before ampu-tation or death, and the patients still alive were marked as censoredat the date of last follow-up.

ResultsPatients were enrolled from October 2005 to January 2009 at3 centers. Mean follow-up was 30.2 months (median, 28months; range, 6 to 57 months). All surviving patients havean ongoing follow-up �1 year. The patient characteristics arelisted in Table 1. There were no clinically relevant differencesbetween the study groups. Before inclusion into the study, therate of previous interventional or surgical procedures washigher in the placebo group. However, according to thedesign of the study, these procedures were performed �3months (mostly �6 months) before screening in all cases andhad failed to induce wound healing. Therefore, the long

28 Circ Cardiovasc Interv February 2011



run-in phase excluded a potential bias between interventionand surgery or potential cell effects.

Seventy-three bone marrow aspirations and catheter pro-cedures (52 BM-MNC and 21 placebo applications) wereperformed until 6 months; 1 patient in the placebo grouprefused the second procedure at 3 months (and had noimprovement of rest pain since). Fourteen additional BM-MNCadministrations were performed between 6 and 18 months in12 patients. There were no complications during a total of 87

bone marrow aspiration procedures. There were no cell-related adverse events observed in all intraarterial catheter-based BM-MNC applications. In 1 patient, a small thrombusdeveloped after low-pressure balloon occlusion in the distalSFA within a preexisting stent. The thrombus could beaspirated, resulting in an uneventful clinical course. Therewas 1 puncture-related hematoma and 1 pseudoaneurysm inthe groin, which healed without sequelae.

After 6 months, 12 patients with incomplete healing ofulcers qualified and gave separate informed consent forcontinuation of cell therapy in the extended protocol. Ofthose, 6 belonged to the initial verum group and 6 to theplacebo group during the randomized-start phase of the trial.Eleven patients received 1 additional BM-MNC application;1 patient obtained 3 repeated BM-MNC therapies �6 months.Cells are characterized in Table 2.

ABIThere were no significant differences in ABI changes be-tween the BM-MNC and the placebo groups (Figure 2).During the double-blind, randomized-start phase of the studyuntil 3 months, the median ABI slightly, but non signifi-cantly, increased from 0.66 (IQR, 0.12 to 0.85) at baseline to0.75 (IQR, 0.57 to 0.89; P�0.4) in the BM-MNC group andremained unchanged in the placebo group (baseline median,0.64; IQR, 0.32 to 0.83; 3-month median, 0.66; IQR, 0.43 to1.00; P�0.2). At 6 months, ABI had increased to 0.85 (IQR,0.46 to 0.95; P�0.6 versus baseline) in the initial BM-MNCgroup receiving an additional BM-MNC administration at 3months and to 0.70 (IQR, 0.43 to 1.08; P�0.1 versus baseline)in the initial placebo group receiving a single BM-MNC treat-ment at 3 months. Excluding patients with TAO and mediascle-rosis, which frequently results in false-positive distal pressurerecordings, did not alter the results. At baseline, the 8 patientswith TAO had higher ABI values (median, 0.67; IQR, 0 to 0.7)compared to the 32 patients with PAOD (median, 0.33; IQR,0.53 to 1.03; P�0.037 versus TAO).

Death, Limb Salvage, andAmputation-Free SurvivalThere were no differences in clinical outcome with respect todeath and limb salvage during the randomized-start, placebo-controlled, double-blind phase I of the study. One patient inthe verum group and none in the placebo group died duringthe first 3 months. At the end of phase II (6 months), anadditional 5 deaths had occurred, 2 in the initial verum groupand 3 in the initial placebo group (Figure 1, Table 3). Allpatients who died had atherosclerotic PAOD, whereas allpatients with TAO survived. Cause of death was suddencardiac death in 2, heart failure in 1, sepsis after amputationin 1, stroke in 1, and metastatic colon cancer in 1. Thus,overall cumulative survival was 86% at 6 months and 84% at1 year. The patient with advanced colon cancer was discov-ered to have cancer 1 month after inclusion in the study,indicating that the cancer must have been present but yetunknown before cell therapy. No further tumors were ob-served during a mean follow-up of 30.2 months.

As summarized in Table 3, 4 patients (1 in the initialplacebo group, 3 in the verum group) underwent amputation

Table 1. Patient Characteristics

BM-MNC InitialTreatment

(n�19)

Placebo InitialTreatment

(n�21) P

Risk factor

Age, y 64.4�15 64.5�16 0.97

Male sex 16 (84) 13 (62) 0.31

History of smoking 9 (47) 11 (52) 1.0

Current smoker 4 (21) 1 (4.8) 0.17

Hypertension 14 (74) 14 (67) 0.74

Diabetes 10 (53) 10 (48) 1.0

IDDM 9 (47) 6 (29) 0.33

Hyperlipidemia 15 (79) 16 (76) 1.0

Kidney dysfunction(creatinine �1.4 mg/dL)

7 (37) 6 (29) 0.74

History of CAD 10 (53) 10 (48) 1.0

History of myocardialinfarction

6 (32) 4 (19) 0.47

Heart failure EF �35% 3 (16) 1 (4.8) 0.33

Chronic venous insufficiency 2 (11) 3 (14) 1.0

PAOD-specific details

Infrapopliteal Disease 13 (68) 14 (67) 1.0

Femoropopliteal occlusions 6 (32) 7 (33) 1.0

Buergers disease (TAO) 3 (16) 5 (24) 0.69

Mediasclerosis 5 (26) 3 (14) 0.44

Fontaine class 3 4 (21) 6 (29) 0.72

Fontaine class 4 15 (79) 15 (71) 0.72

Rutherford class 4 (rest pain) 4 (21) 6 (29) 0.72

Rutherford class 5 (minortissue loss)

12 (63) 14 (67) 1.0

Rutherford class 6 (gangrene) 3 (16) 1 (4.8) 0.33

Highly resistant bacteria(MRSA, ESBL, or Pseudomonas)

11 (58) 9 (43) 0.71

History of PTA�3 mo 3 (16) 10 (48) 0.046

History of peripheral bypassoperation

3 (16) 9 (43) 0.089

Successful �3 mo 1 (4.8)

History of minor amputation 3 (16) 2 (10) 0.65

Second BM-MNC therapyat 3 mo

14 (74) 19 (90) 0.23

Extended protocol: patients withcell applications �6 mo

6 (32) 6 (29) 1.0

Data are presented as no. (%) or mean�SD. CAD indicates coronary arterydisease; EF, ejection fraction; ESBL, extended-spectrum �-lactamase; IDDM,insulin-dependent diabetes mellitus; MRSA, methicillin-resistant Staphylocco-cus aureus.




above the ankle during the randomized-start phase, and anadditional 2 patients (1 in the placebo group and 1 in theinitial verum group) underwent amputation above the ankleduring the second phase of the study until 6 months. Impor-tantly, all patients with Rutherford class 6 at inclusion into thestudy (3 in the verum group, 1 in the placebo group)underwent amputation above the ankle within 3 monthsduring the randomized-start phase, regardless of treatmentallocation. Two patients with advanced Rutherford class 5underwent amputation at 4 months. Thus, the overall cumu-lative limb salvage rate was 86% at 6 months, and 6-monthsamputation-free survival was 76%. There was 1 additionalamputation during the extended study period until 1 year,resulting in a limb salvage rate of 84% at 1 year and a 1-yearamputation-free survival of 73%. There were no significantdifferences in amputation-free survival for the 2 groupsinitially randomized to placebo or verum during the startphase of the trial. Clinical outcome dependent on the initialRutherford class is shown in Table 4.

Ulcer HealingAt inclusion into the randomized-start phase, 30 patients hadnonhealing ulcers (15 in the BM-MNC group, 15 in theplacebo group). As illustrated in Figure 3A, for patients withserial measurements at all time points (excluding patientswith amputations), ulcer area significantly (P�0.014) de-creased in the BM-MNC group during the randomized-startphase of the study, whereas no changes in ulcer area wereobserved in the placebo group up to 3 months follow-up. Inboth groups, 3 patients had complete ulcer healing after thedouble-blind, randomized-start phase. In the initial placebogroup that switched to BM-MNC at 3 months, ulcer area wassignificantly decreased at 6 months, whereas in the initialBM-MNC group receiving a second BM-MNC administra-tion at 3 months, ulcer area continued to decrease until6-month follow-up (Figure 3A). To demonstrate the effect ofa single BM-MNC treatment, ulcer area data of the BM-MNCpatients of the randomized-start phase was combined withulcer area data of the initial placebo group patients duringphase II after switching to BM-MNC treatment at 3 monthsand revealed that within 3 months after a single BM-MNCadministration, ulcer area was significantly (P�0.003) re-duced from 3.21�4.5 cm2 to 1.75�3.2 cm2 (n�19), whereasulcer area remained constant 3 months after placebo administra-tion (2.92�3.5 cm2 to 2.89�4.1 cm2; P�0.5) (n�12) (Figure3B). Overall, the median relative change in ulcer area was

Table 3. Major Adverse Events

Event

�3 Months InitialTreatment All BM-MNC

BM-MNC Placebo 4–6 Months 7–12 Months �12 Months

Majoramputation

3 1* 2 1 2

Death 1 . . . 5* . . . 2

Cancer 1 (4 weeks after) . . . . . . . . .

Data are presented as counts.*One patient with major amputation and death �6 months.

Table 2. Cell Characteristics

CharacteristicFirst BM-MNC

Treatment (n�19)First Placebo

Treatment (n�21)*Second BM-MNC

Treatment at 3 mo (n�33)Third BM-MNC

Treatment at �6 mo (n�12)

Cells (Mio.) 153�78 148�117 155�76 165�77

CD45�/CD34� per WBC, % 1.76�0.6 1.73�0.8 1.78�0.6 1.62�0.5

CD45�/CD34� absolute, �106 2.93�2.1 2.9�3.1 2.9�2 2.7�1.9

CD45�/CD133� per WBC, % 1.0�0.4 1.0�0.6 1.0�0.5 1.0�0.4

CD45�/CD133� absolute, �106 1.79�1.5 1.59�1.6 1.6�1.2 1.6�1

CD133� per CD45�/CD34�, % 62�19 61�23 60�21 55�24

CD133� per CD45�/CD34� absolute, �106 1.8�1.5 1.6�1.6 1.7�1.2 1.3�0.9

CD45�/KDR�, % 0.33�0.17 0.42�0.3 0.33�0.19 0.53�0.3

CD133�/KDR�, % 0.11�0.07 0.15�0.17 0.11�0.1 0.15�0.2

Viability, % 98.6�0.5 98.7�0.4 98.7�0.5 98.5�0.5

Sterile, % 100 100 100 100

Data are presented as mean�SD. All P are nonsignificant.*Cells not infused.Mio.��106.

Figure 2. ABI during the randomized-start phase and until 6months follow-up (open-label phase). ABI values are shown asmedian and IQR for 26 patients with complete serial measure-ments at all time points. BL indicates baseline; M3, month 3;M6, month 6. *P�0.6 at 3 months BM-MNC versus placebo andP�0.7 BM-MNC versus placebo after switching to BM- MNC.




�72% (IQR, �100% to �28%) within 3 months after BM-MNC administration compared to a minor change of �23%(IQR, �92% to �12%) after placebo. In the extended study thatincluded 12 patients with a third or fourth BM-MNC treatment,10 had completely healing wounds, 1 underwent amputation,and 1 had a small residual ulcer at last visit, further support-ing that serial treatments are necessary for a clinical response.

We divided the ulcer area data into 2 groups according to themedian of 2.3 cm2, resulting in 14 with an ulcer area �2.3 cm2

versus 16 with larger wounds �2.3 cm2. Among the 30 patientswith ulcers, 12 (85%) of 14 had healing wounds in the groupwith smaller ulcers versus only 8 (50%) of 16 in the group withlarger ulcers (P�0.038). The absolute reduction in ulcer areaafter 2 BM-MNC treatments was higher than after 1 BM-MNCtreatment and significantly higher versus placebo (Figure 3C).

During the entire study observation period, there wascomplete ulcer healing in 20 (66.6%) of the 30 patients withulcers at baseline. Ulcer healing occurred at a mean time of10.9 months after inclusion in the study. Fourteen patientshad healing ulcers after �2 BM-MNC treatments; 3 hadhealing ulcers after 1 BMC application, whereas 3 smallulcers healed after placebo. Repeated BM-MNC administra-tion (�2) was significantly correlated with complete ulcerhealing (log-rank P�0.017 for �2 BM-MNC therapies ver-sus 1-time BM-MNC). Interestingly, 1 patient’s ulcers healedtwice after BM-MNC treatments (after the first and againafter the third cell application) while a novel ulcer intermit-tently occurred at 8 months due to progression of atheroscle-rosis requiring additional intervention of a newly occludedSFA. Taken together, these findings support the concept thatserial treatments are frequently necessary to stimulate ulcerhealing in patients with critical ischemia.

Pain AnalysisDuring the double-blind, randomized-start phase of the trial,there was a significantly greater reduction of pain in theBM-MNC group compared to the placebo group (Figure 4A).During the open-label phase from 3 to 6 months, rest painalso was significantly reduced in the initial placebo groupreceiving BM-MNC at 3 months, whereas rest pain continuedto improve in the BM-MNC group, which received a secondBM-MNC administration at 3 months.

At 6 months, 20 (60%) of the 33 patients with rest pain atinclusion showed an improvement in the pain score, with 14(58%) being free of rest pain and 6 having reduced analgesicconsumption. Of those 14 patients, 12 were free of rest painafter BM-MNC treatments compared to only 2 after placebo.

Combining pain data of the BM-MNC patients of therandomized-start phase with pain data of the placebo grouppatients after crossover to BM-MNC therapy at 3 monthsrevealed that within 3 months after a single BM-MNCadministration, pain score decreased significantly from4.4�2.4 to 1.9�1.7 (P�0.001) in 26 patients in the BM-MNC group, whereas no significant change was observed inthe placebo group (4.5�2.4 to 3.9�2.6; P�0.3; n�15).

The absolute reduction in pain scale after 2 BM-MNCtreatments is higher than after 1 BM-MNC treatment, andsignificantly higher versus placebo (Figure 4B).

TCO2TCO2 measurements were obtained at only 2 study sites(Frankfurt and Bern) and performed in 20 patients (16 withserial measurements) (Figure 5). During the randomized-startphase, TCO2 at the lower-calf position increased from31.6�24 mm Hg to 40.5�23 mm Hg in the verum group(P�0.058; n�6), but decreased in the placebo group(46.9�11 mm Hg to 39.7�17 mm Hg; P�0.032; n�10). Atthe end of the open-label phase of the trial at 6 months, TCO2

was slightly, but nonsignificantly, further increased in theinitial BM-MNC group (to 47�12 mm Hg; P�0.17 versusbaseline), but now increased significantly from39.7�17 mm Hg to 53.8�11 mm Hg; P�0.018) in the initialplacebo group receiving BM-MNC at 3 months.

Combining TCO2 measurements of the BM-MNC patientsof the randomized-start phase with those of the placebo grouppatients after crossover to BM-MNC revealed that within 3months after a single BM-MNC administration, TCO2 signif-icantly increased from 37.6�19 mm Hg to 48�19 mm Hg(P�0.001; n�14) in the BM-MNC-treated patients, whereasa significant decrease was observed in the placebo group(46.9�11 mm Hg to 39.7�17 mm Hg; P�0.032; n�10).

Determinants of Ulcer Healing and Limb SalvageSeveral factors influenced ulcer healing (30 patients withulcers at baseline). Younger age, better ejection fraction,smaller ulcer size, better renal function (creatinine �1.4mg/dL), a higher number of administered BM-MNC, andrepeated BM-MNC administration positively correlated withulcer healing. To disclose a potential association between thenumber and functionality of the administered BM-MNC andulcer healing, we compared total BM-MNC number and theirmigratory capacity (marker of cell retention) as well as theirCFU capacity (marker for intrinsic progenitor cell activity) inpatients with healing versus nonhealing ulcers. Importantly,the 20 patients with healing ulcers received significantly greaternumbers of total BM-MNC (178�113�106 versus 87�29�106

BM-MNC; P�0.003) as well as of CD34�/CD45� BM-MNC(3.57�1.7�106 versus 1.78�1.7�106; P�0.033). Likewise,the migratory capacity (87�46 cell counts versus 51�28 cellcounts; P�0.016) as well as the CFU capacity (27.7�14.6versus 18.2�9.4; P�0.048) of the administered cells weresignificantly better in patients with subsequently healing ulcers(Table 5).

Most importantly, on multivariate analysis (Table 6),including all relevant clinical parameters that individuallycorrelated with ulcer healing, repeated BM-MNC administra-

Table 4. Major Adverse Events and Rutherford Class DuringLong-Term Follow-Up (Mean, 30.2 Months)

Rutherford Class

4 (n�10) 5 (n�26) 6 (n�4)

Responder 6 20 0

Major amputation 1 5 3

Death 2 4 2*

*One patient with major amputation and death �6 months.




Figure 3. A, Ulcer area during the randomized-start phase and until 6 months (open label) presented as mean�SE in patients withserial measurements. B, Serial measurements of ulcer area in the placebo group until 3 months compared with the effect after 1 singleBM-MNC treatment. Ulcer area data of the BM- MNC patients were combined for the randomized-start phase with ulcer area data ofthe initial placebo group patients after crossover to BM-MNC treatment (data are presented as mean�SD). C, Absolute decrease inulcer size according to treatment groups (data are presented as mean�SE).




tion as well as the number and functionality of administeredBM-MNC were significant independent predictors for completeulcer healing. Ulcer healing induced by repeated BM-MNCadministration significantly correlated with limb salvage(r�0.8; P�0.001). Importantly, all 20 patients with healingwounds experienced subsequent limb salvage.

In addition, we measured cell surface markers of theBM-MNC by fluorescence-activated cell sorter analysis. Asseen in Table 5, the number of absolute CD45�/CD34� cellswere significantly higher in 20 patients with healing woundsversus nonresponders (n�10). Likewise, the absolute numberof CD45�/CD133� was higher, although the difference didnot achieve statistical significance (see Table 5). In contrast,the number of BM-MNC expressing the endothelial cellmarker kinase insert domain receptor (KDR) did not differbetween responders and nonresponders. These data suggestthat the number of hematopoietic CD34� cells have the majorimpact on the functional improvement.

Finally, to identify clinical responders to therapy defined byeither partial or complete healing of ulcers or freedom from restpain or reduction in pain medication consumption, potentialdeterminants were analyzed in all 40 patients. Twenty-six (65%)of 40 patients had evidence for either healing ulcers or less restpain during follow-up. All 8 (100%) patients with TAO wereresponders, whereas 18 (56%) of 32 patients with atheroscleroticPAOD clinically improved by BM-MNC treatments (P�0.02TAO versus PAOD). Patients with gangrene (Rutherford class6) did not respond at all. On multivariate analysis, repeatedBM-MNC administration as well as the number and functional-ity of administered BM-MNC were the only independentlyassociated predictors of clinical improvement.

AngiographiesAngiographic follow-up at 6 months in the first 10 patientsdid not show any significant differences in angiographicallyvisible collateral formation on gross inspection. Therefore,

Figure 4. A, Pain scale during the randomized-start phase and until 6 months (open label) inpatients with serial measurements (data arepresented as mean�SE). *P�0.06 at 3 monthsBM-MNC versus placebo and P�0.26 at 6months. B, Absolute reduction in pain scaleaccording to treatment (data are presented asmean�SE).




routine control angiography at 6 months was abandonedunless it was necessary for a repeat BM-MNC application.

DiscussionTo our knowledge, the present study is the first randomized,double-blind, placebo-controlled multicenter phase II trial toassess potential effects of intraarterial BM-MNC administra-tion in patients with critical limb ischemia due to PAOD. Theresults of the double-blind, placebo-controlled randomized-start phase suggest that BM-MNC administration does notalter ABI, but accelerates ulcer healing and pain reduction inpatients with stable ulcers or rest pain within 3 months.However, critically ill patients with impending amputationdue to Rutherford 6 did not derive any benefit from BM-MNCadministration. Additionally, the unique design of the trialwith a randomized-start phase followed by a switch of theplacebo patients to BM-MNC administration and repetitiveBM-MNC administration in the initial verum group at 3months enabled us to address a number of clinically criticalquestions with respect to the safety and potential efficacy ofBM-MNC administration in patients with critical limb ische-mia but stable ulcers or rest pain. Successful ulcer healing

associated with improved limb salvage requires repeatedBM-MNC administration as well as functionally competentBM-MNC in sufficient numbers. Moreover, in line withprevious results,20 patients with TAO appeared to demon-strate improved responsiveness to BM-MNC administrationcompared with patients with atherosclerotic PAOD. How-ever, these exploratory findings in the present pilot trial needto be confirmed in a larger randomized trial in patients withcritical limb ischemia and stable ulcers.

In retrospect, the choice of change in ABI as the primaryend point of the double-blind, randomized-start phase repre-sents the major drawback of the present trial. Although ABIreflects distal perfusion pressures in patients with PAOD,changes in ABI values do not correlate well with ulcerhealing and limb salvage, the clinically most relevant thera-peutic goals in the treatment of patients with critical limbischemia. Moreover, in patients with TAO, who respondedmost favorably to BM-MNC administration, ABI values arespuriously high and do not reflect the degree of distalischemia and tissue necrosis. Changes in ABI values do notpredict clinical outcome or are suitable for comprehensively

Table 6. Multivariate Analysis: Association With CompleteWound Healing

Factor PExp (B)Hazard 95% CI

Age at baseline 0.68 1.009 0.96–1.05

Ulcer size at baseline 0.2 0.83 0.62–1.1

EF 0.16 1.07 0.97–1.18

Creatinine �1.4 mg/dL 0.34 2.91 0.32–26.3

Cell no. 0.001 1.03 1.01–1.04

Repeated BM-MNC (�2 applications) 0.005 0.116 0.025–0.53

Cell function (migration) 0.036 1.025 1.002–1.05

Colony forming units 0.36 1.027 0.97–1.087

Multivariate analysis obtained with Cox regression analysis. Creatininevalues, repeated BM-MNC applications were categorized.

Figure 5. TCO2 during the randomized-startphase and until 6 months (open label) in 16patients with serial measurements. Data arepresented as mean�SE.

Table 5. Cell Parameters and Functional Characteristics ofBM-MNC in Responders Versus Nonresponders

HealingWounds(n�20)

NonhealingWounds(n�10) P

CD45�/CD34� absolute, �106 3.57�1.7 1.78�1.7 0.033

CD45�/CD133� absolute, �106 2.0�1.5 1.39�1.5 0.14

BM-MNC cell no., �106 178�113 87�29 0.003

CD45�/KDR�, % 0.3�0.17 0.31�0.17 0.9

CD133�/KDR�, % 0.11�0.09 0.14�0.08 0.6

Migration, cell no. 87�46 51�28 0.016

CFUs 27.7�14.6 18.2�9.4 0.048

Data are presented as mean�SD.




assessing potential beneficial therapeutic effects on ulcerhealing and rest pain in patients with critical limb ischemia.21

Therefore, it is not surprising that the beneficial effects ofBM-MNC treatment on ulcer healing and rest pain wereobserved despite no effect on the primary end point of changein ABI during the placebo-controlled, double-blindrandomized-start phase of the present study. Moreover, be-cause there is substantial uncertainty about the best end pointto measure therapeutic benefit in critical limb ischemia, thepresentation of composite end points instead of choosing 1primary end point and several secondary end points would bean adequate option in designing future trials, according to theoutcome of this study.

Previous uncontrolled trials using intramuscular injectionof BM-MNC or blood-derived mononuclear cells reported6-month limb salvage rates of 71% (TACT)20 and 59% (BoneMarrow Outcome Trial).22 Likewise, a recent randomizedcontrolled gene therapy trial using a plasmid-encoding fibro-blast growth factor reported limb salvage rates of 75% at 1year, whereas ulcer healing did not differ from placebotreatment at 6 months.23 One-year limb salvage rate (84%)and amputation-free survival (73%) in the present studycompare favorably with these trials. However, it should bekept in mind that the severity of the underlying diseaseaccompanied by advanced necrosis or infections will affectthe necessity for major amputations. Indeed, all 4 patientswith extensive gangrene with impending amputation (Ruth-erford class 6) at inclusion in the study had to undergoamputation above the ankle already during the initial 3-monthrandomized-start phase, regardless of treatment allocation.Despite these seemingly disappointing observations in pa-tients with stable ulcers at inclusion, complete ulcer healingoccurred at a mean time of 10.9 months after BM-MNCadministration, although significant improvements in ulcersize and rest pain had occurred within 3 months afterBM-MNC administration. Thus, assessing potential clinicalbenefits of cell therapy in patients with critical limb ischemiawill require follow-up periods of at least 18 months.

The identification of cell number and cell functionality asindependent predictors of subsequent ulcer healing and clin-ical improvement is an intriguing finding of the present studybecause it indeed may indicate a cause-and-effect relation-ship. Similar observations have been reported in applyingBM-MNC to patients with acute myocardial infarction24 andchronic heart failure.18 Although we cannot fully exclude thathigher numbers and better functionality of the BM-MNCretrieved from bone marrow aspirates simply reflect lessseverely ill patients with a better natural ulcer healingcapacity, the pivotal role of preserving cell functionality toenhance blood flow recovery is firmly established in animalmodels of limb ischemia.25 Thus, future studies using cellenhancement strategies and retrieving larger amounts of bonemarrow aspirates to increase the number of BM-MNC to beadministered are necessary not only to finally document acause-and-effect relationship for BM-MNC therapy in pa-tients with critical limb ischemia, but also to improve thera-peutic efficacy in patients with severely impaired BM-MNCfunctionality.

Given the long-standing chronic disease process leading tocritical limb ischemia in PAOD, it is not surprising that ulcerhealing appears to be significantly promoted by repeatedBM-MNC administration. The excellent procedural safetyprofile of bone marrow harvest and intraarterial BM-MNCadministration documented in the present study provides thenecessary framework for a repeated treatment strategy thatcan be easily implemented into clinical guidance of patientswith critical limb ischemia.

Regarding mechanistic aspects of a cell effect, angio-graphic follow-up at 6 months in the first 10 patients did notshow any significant differences in angiographically visiblecollateral formation. We speculate on the basis of a body ofevidence from animal models that neovascularization mostlikely involves vessels of the microvasculature that weresmaller than the angiographic resolution. Indeed, using aninducible suicide approach to eliminate previously adminis-tered BM-MNC in an experimental model of myocardialinfarction, we could demonstrate that the beneficial effects ofBM-MNC administration are paralleled by increased capil-lary and arteriolar vessel density.26

LimitationsAlthough the design of the present study with a randomized-start phase during the first 3 months followed by a switch toactive treatment in the initial placebo group or repetitiveactive treatment in the initial verum group during the secondphase from 3 to 6 months provided for the unique opportunityto address a potentially beneficial effect of repeated BM-MNC administration, it precluded the comparison of activetreatment with placebo administration throughout the entirestudy period. However, we believed it to be unethical toextend the placebo phase �3 months in these patients withcritical limb ischemia. In addition, the delayed orrandomized-start design as used in the PROVASA trial is oneway of evaluating disease-modifying treatments, and therehas been regulatory support for this trial design, as previouslydiscussed in an editorial.27 Moreover, if there had been nomeasurable effect of active treatment on ulcer size or painwithin 3 months after BM-MNC administration, it would behighly unlikely that such treatment may be capable ofmodifying the disease process in these critically ill patients.Finally, the repetitive treatment group could be used toevaluate potential additive disease-modifying effects of re-peated BM-MNC administration in this chronically progress-ing disease.

The number of patients included into the trial is small. Thepresent study was designed as a pilot trial, and the sample sizewas based on the number of patients included in the TACTtrial. Thus, the inclusion of 40 patients (19 BM-MNC, 21placebo) is essentially comparable to the TACT trial5 withregard to sample size. Nevertheless, during the placebo-controlled crossover phase of the study, a total of 73 proce-dures (52 BM-MNC administrations, 21 placebo treatments)were performed until 6 months followed by an additional 14BM-MNC administrations during the extended study period.Thus, in combination with the mean follow-up time of 30.2months, the size of the study population appears to be




sufficiently large to derive clinically relevant conclusionswith respect to the safety of intraarterial BM-MNC adminis-tration and its effects in patients with stable ulcers and restpain.

The enrollment phase of the study lasted 3.5 years. How-ever, screening processes were frequently delayed becausethe inclusion criteria required at least a 3-month interval afterPTA or bypass procedures before cell therapy in potentialcandidates in order to achieve stable baseline conditions andto be able to document a cell effect without conflicting biasby other interventions.

The present study did not address the question of whetherintraarterial cell administration is superior to intramuscularinjections. Given that experimental studies demonstrated thatthe administered BM-MNC needs to persist for at least 3weeks to improve cardiac function associated with increasedcapillary and arteriolar vessel density,26 we used the intraar-terial route of application based on the assumption thatintraarterially applied cells will only home to tissue withpreserved nutrient blood supply.

Finally, we cannot exclude that selected cell populationsmight be superior to the heterogeneous BM-MNC used in thepresent study. Indeed, a recently reported, but not yet pub-lished study by Losordo and colleagues (2010) demonstratedthat intramuscular injection of granulocyte colony-stimulatingfactor-mobilized peripheral blood-derived CD34� cells wasassociated with a significantly reduced amputation rate in pa-tients with critical limb ischemia.

ConclusionIn patients with critical limb ischemia, intraarterial adminis-tration of BM-MNC does not increase ABI but promotesulcer healing and reduces rest pain. Successful ulcer healingassociated with improved limb salvage requires repeatedadministration of functionally competent BM-MNC. How-ever, critically ill patients with extensive gangrene andimpending amputation (Rutherford class 6) did not derive anybenefit, whereas patients with TAO in general responded verywell. Thus, large-scale randomized trials are warranted toassess the clinical effect of repeated BM-MNC administrationin patients with critical limb ischemia and stable ulcers andrest pain.

AcknowledgmentsWe acknowledge the expert technical assistance of our staff in thecatheterization and angiography laboratories; the Red Cross BloodDonation Center in Frankfurt, Germany; and the Department ofMolecular Cardiology, Frankfurt University. We thank A. Kopalla,Tina Rasper, Nicola Schuly, Christiane Vetter, Nadine Sorg, HeikeWagner, and J. Vogele. We appreciate the cooperation with vascularsurgeons (W. Tigges and D. Theodosiou) from the AsklepiosHospital in Hamburg-Rissen, Germany, for patient recruitment, andthe Department of Angiology (B. Linnemann) and Department ofDiagnostic and Interventional Radiology (A. Thalhammer and T.Vogl) at Frankfurt University.

Sources of FundingThe study was supported in part by the Deutsche Forschungsgemein-schaft (WA 1461-2).

DisclosuresDrs Zeiher and Dimmeler are cofounders and advisors to t2cureGmbH. Drs Walter, Krankenberg, Balzer, Kalka, Baumgartner,Schluter, Tonn, Seeger, and Lindhoff-Last have no disclosures to report.

References1. Hirsch AT, Haskal ZJ, Hertzer NR, Bakal CW, Creager MA, Halperin JL,

Hiratzka LF, Murphy WR, Olin JW, Puschett JB, Rosenfield KA, SacksD, Stanley JC, Taylor LM Jr, White CJ, White J, White RA, Antman EM,Smith SC Jr, Adams CD, Anderson JL, Faxon DP, Fuster V, Gibbons RJ,Hunt SA, Jacobs AK, Nishimura R, Ornato JP, Page RL, Riegel B.ACC/AHA 2005 Practice Guidelines for the management of patients withperipheral arterial disease (lower extremity, renal, mesenteric, andabdominal aortic). Circulation. 2006;113:e463–e654.

2. Wolfe JH, Wyatt MG. Critical and subcritical ischaemia. Eur J VascEndovasc Surg. 1997;13:578–582.

3. Hiatt WR. Medical treatment of peripheral arterial disease and claudi-cation. N Engl J Med. 2001;344:1608–1621.

4. Sprengers RW, Lips DJ, Moll FL, Verhaar MC. Progenitor cell therapy inpatients with critical limb ischemia without surgical options. Ann Surg.2008;247:411–420.

5. Tateishi-Yuyama E, Matsubara H, Murohara T, Ikeda U, Shintani S,Masaki H, Amano K, Kishimoto Y, Yoshimoto K, Akashi H, Shimada K,Iwasaka T, Imaizumi T. Therapeutic angiogenesis for patients with limbischaemia by autologous transplantation of bone-marrow cells: a pilotstudy and a randomised controlled trial. Lancet. 2002;360:427–435.

6. Huang P, Li S, Han M, Xiao Z, Yang R, Han ZC. Autologous transplan-tation of granulocyte colony-stimulating factor-mobilized peripheralblood mononuclear cells improves critical limb ischemia in diabetes.Diabetes Care. 2005;28:2155–2160.

7. Huang PP, Yang XF, Li SZ, Wen JC, Zhang Y, Han ZC. Randomisedcomparison of G-CSF-mobilized peripheral blood mononuclear cellsversus bone marrow-mononuclear cells for the treatment of patients withlower limb arteriosclerosis obliterans. Thromb Haemost. 2007;98:1335–1342.

8. Durdu S, Akar AR, Arat M, Sancak T, Eren NT, Ozyurda U. Autologousbone-marrow mononuclear cell implantation for patients with Rutherfordgrade II–III thromboangiitis obliterans. J Vasc Surg. 2006;44:732–739.

9. Van Tongeren RB, Hamming JF, Fibbe WE, Van Weel V, Frerichs SJ,Stiggelbout AM, Van Bockel JH, Lindeman JH. Intramuscular orcombined intramuscular/intra-arterial administration of bone marrowmononuclear cells: a clinical trial in patients with advanced limb ische-mia. J Cardiovasc Surg (Torino). 2008;49:51–58.

10. Miyamoto K, Nishigami K, Nagaya N, Akutsu K, Chiku M, Kamei M,Soma T, Miyata S, Higashi M, Tanaka R, Nakatani T, Nonogi H,Takeshita S. Unblinded pilot study of autologous transplantation of bonemarrow mononuclear cells in patients with thromboangiitis obliterans.Circulation. 2006;114:2679–2684.

11. Kawamoto A, Katayama M, Handa N, Kinoshita M, Takano H, Horii M,Sadamoto K, Yokoyama A, Yamanaka T, Onodera R, Kuroda A, Baba R,Kaneko Y, Tsukie T, Kurimoto Y, Okada Y, Kihara Y, Morioka S,Fukushima M, Asahara T. Intramuscular transplantation of G-CSF-mobilized CD34(�) cells in patients with critical limb ischemia: a phaseI/IIa, multicenter, single-blinded, dose-escalation clinical trial. StemCells. 2009;27:2857–2864.

12. Schachinger V, Erbs S, Elsasser A, Haberbosch W, Hambrecht R, Hol-schermann H, Yu J, Corti R, Mathey DG, Hamm CW, Suselbeck T,Assmus B, Tonn T, Dimmeler S, Zeiher AM. Intracoronary bone marrow-derived progenitor cells in acute myocardial infarction. N Engl J Med.2006;355:1210–1221.

13. Seeger FH, Tonn T, Krzossok N, Zeiher AM, Dimmeler S. Cell isolationprocedures matter: a comparison of different isolation protocols of bonemarrow mononuclear cells used for cell therapy in patients with acutemyocardial infarction. Eur Heart J. 2007;28:766–772.

14. Heeschen C, Lehmann R, Honold J, Assmus B, Aicher A, Walter DH,Martin H, Zeiher AM, Dimmeler S. Profoundly reduced neovascular-ization capacity of bone marrow mononuclear cells derived from patientswith chronic ischemic heart disease. Circulation. 2004;109:1615–1622.

15. Assmus B, Schachinger V, Teupe C, Britten M, Lehmann R, Dobert N,Grunwald F, Aicher A, Urbich C, Martin H, Hoelzer D, Dimmeler S,Zeiher AM. Transplantation of Progenitor Cells and RegenerationEnhancement in Acute Myocardial Infarction (TOPCARE-AMI).Circulation. 2002;106:3009 –3017.




16. Seeger FH, Rasper T, Koyanagi M, Fox H, Zeiher AM, Dimmeler S.CXCR4 expression determines functional activity of bone marrow-derived mononuclear cells for therapeutic neovascularization in acuteischemia. Arterioscler Thromb Vasc Biol. 2009;29:1802–1809.

17. Assmus B, Honold J, Schachinger V, Britten MB, Fischer-Rasokat U,Lehmann R, Teupe C, Pistorius K, Martin H, Abolmaali ND, Tonn T,Dimmeler S, Zeiher AM. Transcoronary transplantation of progenitorcells after myocardial infarction. N Engl J Med. 2006;355:1222–1232.

18. Assmus B, Fischer-Rasokat U, Honold J, Seeger FH, Fichtlscherer S,Tonn T, Seifried E, Schachinger V, Dimmeler S, Zeiher AM.Transcoronary transplantation of functionally competent BMCs is asso-ciated with a decrease in natriuretic peptide serum levels and improvedsurvival of patients with chronic postinfarction heart failure: results of theTOPCARE-CHD Registry. Circ Res. 2007;100:1234–1241.

19. Schachinger V, Aicher A, Dobert N, Rover R, Diener J, Fichtlscherer S,Assmus B, Seeger FH, Menzel C, Brenner W, Dimmeler S, Zeiher AM.Pilot trial on determinants of progenitor cell recruitment to the infarctedhuman myocardium. Circulation. 2008;118:1425–1432.

20. Matoba S, Tatsumi T, Murohara T, Imaizumi T, Katsuda Y, Ito M, SaitoY, Uemura S, Suzuki H, Fukumoto S, Yamamoto Y, Onodera R, Ter-amukai S, Fukushima M, Matsubara H. Long-term clinical outcome afterintramuscular implantation of bone marrow mononuclear cells (Thera-peutic Angiogenesis by Cell Transplantation [TACT] trial) in patientswith chronic limb ischemia. Am Heart J. 2008;156:1010–1018.

21. Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA, FowkesFG, Bell K, Caporusso J, Durand-Zaleski I, Komori K, Lammer J, LiapisC, Novo S, Razavi M, Robbs J, Schaper N, Shigematsu H, Sapoval M,

White C, White J, Clement D, Creager M, Jaff M, Mohler E III, Ruth-erford RB, Sheehan P, Sillesen H, Rosenfield K. Inter-Society Consensusfor the Management of Peripheral Arterial Disease (TASC II). Eur J VascEndovasc Surg. 2007;33(suppl 1):S1–S75.

22. Amann B, Luedemann C, Ratei R, Schmidt-Lucke JA. Autologous bonemarrow cell transplantation increases leg perfusion and reduces ampu-tations in patients with advanced critical limb ischemia due to peripheralartery disease. Cell Transplant. 2009;18:371–380.

23. Nikol S, Baumgartner I, Van Belle E, Diehm C, Visona A, CapogrossiMC, Ferreira-Maldent N, Gallino A, Wyatt MG, Wijesinghe LD, FusariM, Stephan D, Emmerich J, Pompilio G, Vermassen F, Pham E, Grek V,Coleman M, Meyer F. Therapeutic angiogenesis with intramuscularNV1FGF improves amputation-free survival in patients with critical limbischemia. Mol Ther. 2008;16:972–978.

24. Britten MB, Abomaali N, Assmus B, Lehmann R, Honold J, Schmitt J,Vogl TJ, Martin, Schachinger V, Dimmeler S, Zeiher AM. Infarctremodeling following intracoronary progenitor cell treatment in patientswith acute myocardial infarction (TOPCARE-AMI): mechanistic insightsfrom serial contrast enhanced magnetic resonance imaging. Circulation.2003;108:2212–2218.

25. Dimmeler S, Leri A. Aging and disease as modifiers of efficacy of celltherapy. Circ Res. 2008;102:1319–1330.

26. Yoon CH, Koyanagi M, Iekushi K, Seeger F, Urbich C, Zeiher AM,Dimmeler S. Mechanism of improved cardiac function after bone marrowmononuclear cell therapy: role of cardiovascular lineage commitment.Circulation. 2010;121:2001–2011.

27. D’Agostino RB Sr. The delayed-start study design. N Engl J Med.2009;361:1304–1306.

CLINICAL PERSPECTIVEInjection of autologous bone marrow-derived mononuclear cells (BM-MNC) is a promising therapeutic option in patientswith critical limb ischemia, but double-blind, randomized trials are lacking. The present study is the first randomized,placebo controlled trial showing that intraarterial BM-MNC administration accelerates wound healing and induces painreduction until 3 months in patients with critical limb ischemia with stable ulcers but not in patients with extensivegangrene. Ulcer healing induced by repeated BM-MNC administration significantly correlated with limb salvage.Successful ulcer healing required repeated applications of functionally competent BM-MNC. These exploratory findingsof this pilot trial need to be confirmed in a larger randomized trial in patients with critical limb ischemia and stable ulcers.




Intraarterial Administration of Bone Marrow Mononuclear ... · Background—Critical limb ischemia due to peripheral arterial occlusive disease is associated with a severely increased

Documents