Abstract
Transplantation of autologous bone marrow mononuclear cells (BMMCs) has been proven safe in animal and human studies. However, there are very few studies in stroke patients. In this study, intra-arterial autologous BMMCs were infused in patients with moderate to severe acute middle cerebral artery infarcts. The subjects of this study included 20 patients with early or late spontaneous recanalization but with persistent deficits, in whom treatment could be initiated between 3 and 7 days after stroke onset. Mononuclear cells were isolated from bone marrow aspirates and infused at the proximal middle cerebral artery of the affected hemisphere. Safety analysis (primary endpoint) during the 6-month follow-up assessed death, any serious clinical events, neurological worsening with ≥ 4-point increase in National Institutes of Health Stroke Scale (NIHSS) scores, seizures, epileptogenic activity on electroencephalogram, and neuroimaging complications including new ischemic, hemorrhagic, or neoplastic lesions. Satisfactory clinical improvement (secondary endpoint) at 90 days was defined according to the pretreatment NIHSS scores as follows: modified Rankin Scale score of 0 in patients with NIHSS <8, modified Rankin Scale scores of 0–1 in patients with NIHSS 8–14, or modified Rankin Scale scores 0–2 in patients with NIHSS >14. Good clinical outcome was defined as mRS ≤2 at 90 days. Serial clinical, laboratory, electroencephalogram, and imaging evaluations showed no procedure-related adverse events. Satisfactory clinical improvement occurred in 6/20 (30%) patients at 90 days. Eight patients (40%) showed a good clinical outcome. Infusion of intra-arterial autologous BMMCs appears to be safe in patients with moderate to severe acute middle cerebral artery strokes. No cases of intrahospital mortality were seen in this pilot trial. Larger prospective randomized trials are warranted to assess the efficacy of this treatment approach.
Introduction
Ischemic stroke is one of the leading causes of death and is the leading cause of disability in Western countries (38). Intravenous thrombolysis remains the only proven therapy for acute ischemic stroke (37). However, even in developed countries, only a small minority of stroke patients currently receive this therapy. In addition, several neuroprotective strategies have failed to show any definite benefit after a stroke (12). Acute ischemia causes irreversible damage to neurons and glial cells, leading to functional deficits and chronic sequelae with variable degrees of spontaneous recovery of function. Strokes related to large-vessel occlusion, in particular, have high rates of morbidity and mortality. Indeed, the mortality rate for acute infarcts due to proximal middle cerebral artery (MCA) occlusions has been estimated at between 27% and 78% (15,17).
Cell therapy with bone marrow-derived stem cells has been shown to have beneficial effects in animal models of stroke (11,16,41). Although the mechanisms involved are still subject to debate, it has been suggested that the injected cells migrate to the lesion site where they release cytokines and trophic factors, and can modulate neuronal death and inflammation in the penumbra area (26). In addition, bone marrow mononuclear cells (BMMCs) also contain endothelial progenitor cells, which have been reported to contribute to revascularization of ischemic tissues and repair of injured endothelium (23,40,42). These observations encourage the evaluation of bone marrow-derived stem cells as candidates in cell therapies. Indeed, the utilization of BMMCs has been shown to be safe in heart disease (1,19,28, 31,34,36) and in peripheral artery disease (4,35,43). In addition, Bang et al. (6) have shown that intravenous administration of bone marrow-derived mesenchymal cells seems to be safe in stroke patients. Moreover, a recent study has indicated that the intra-arterial injection of BMMCs appears to be feasible and safe in subacute stroke patients (9).
We hypothesized that intra-arterial (IA) infusion of autologous BMMCs (ABMMCs) would lead to better outcomes in acute stroke patients who have suffered a large-vessel occlusion. The aim of the current study was to describe the safety and feasibility of IA ABMMC infusion in a prospective cohort of patients with moderate to severe MCA strokes.
Materials and Methods
All patients were treated and investigated according to the American Heart Association and European Stroke Organisation acute stroke guidelines (2,13). Informed consent was obtained from the patients, and the study protocol was approved by the Brazilian National Ethical and Research Committee. Besides the IA ABMMC infusion, the treatment received by the study patients was the standard of care according to our institutional protocol (Cerebrovascular Division, São Lucas Hospital, PUC-RS).
Study Population
Patients met the following inclusion criteria: 1) age between 18 and 80 years old; 2) computed tomography (CT)-proven or magnetic resonance imaging (MRI)-proven MCA infarct; 3) presentation for treatment initiation between 3 and 7 days after onset of stroke symptoms; 4) persistent disabling deficits at the time of treatment, with National Institutes of Health Stroke Scale (NIHSS) >8; 5) spontaneous recanalization of the MCA prior to the procedure, as assessed by MRI angiography or transcranial Doppler examination. Patients were excluded if they met any of the following criteria: 1) lacunar syndrome; 2) hematological causes of stroke; 3) history of malignancy, hepatic or renal dysfunction, or any other severe comorbidity; 4) evidence of clinical or neurological instability measured by an increase of 4 points or more in the NIHSS score over the previous 24 h; 5) pregnancy; 6) participation in other clinical trials.
Bone Marrow Aspiration, Cell Separation, and Transplantation
Bone marrow aspiration and subsequent cell preparation were accomplished on the same day as the IA ABMMC infusion, and lasted 2–4 h. The collection was performed under local anesthesia, through a puncture and repeated aspirations at the posterior iliac crest region. A total of 50 ml of bone marrow aspirate was collected, and after removal of the bony and fatty residues, the mononuclear cells were isolated by density gradient on Ficoll-Paque Plus (Amersham Biosciences, São Paulo, Brazil), washed three times in saline, resuspended in saline with 5% human albumin, and filtered in a 100-μm nylon filter. Approximately 50 μl of this solution was used for viability test, culture, and colony formation studies; a cell viability greater than 90% was required in order to ensure the quality of the cells. A total of 15 ml of the ABMMC solution was injected over a 30-min period through a microcatheter positioned within the MCA origin of the affected hemisphere. Angiographic evaluation was performed immediately before and after the infusion, in order to confirm pre-and posttreatment vessel patency.
Clinical Variables and Measurement of Outcome
Patient demographics, medical history, vital signs, and routine laboratory levels were entered on standardized clinical report forms. NIHSS was obtained at baseline and at 1, 3, 7, 30, 60, 90, and 180 days after the procedure. Modified Rankin Scale scores (mRS) were obtained 90 and 180 days after the procedure. CT brain imaging and electroencephalogram (EEG) were performed at baseline, up to 7 days, and at 30, 60, 90, and 180 days after cell transplantation. CT and EEG were also performed at any time if a patient's neurological status declined. Brain MRI including diffusion weighted imaging (DWI), T2, and fluid attenuated inversion recovery (FLAIR) as well as pre- and post-gadolinium T1 sequences were performed at baseline, 7 days, and 6 months after the procedure.
The primary endpoints focused on safety, and included: 1) clinically significant procedural complications (defined as a decline of ≥ 4 points in the NIHSS score, or death); 2) symptomatic intracranial hemorrhage (defined as a decline of ≥ 4 points in the NIHSS score, with any blood products identified on head CT scan); 3) new ischemic lesions on CT and/or DWI MRI at day 7 postprocedure; 4) clinical seizures and/or epileptic discharges on serial EEG at any time during the hospitalization; and (5) development of intracranial neoplasia in the long-term follow-up. All imaging data were reviewed at a central core laboratory blinded to the procedure results. Secondary outcomes included the rates of satisfactory clinical improvement and good clinical outcomes at 24 h and at 90 days. Satisfactory clinical improvement at 24 h was defined as an improvement of 4 or more points in the NIHSS; and at 90 days, it was defined according to the pretreatment NIHSS scores as follows: mRS of 0 in patients with NIHSS <8, mRS of 0–1 in patients with NIHSS 8–14, or mRS 0–2 in patients with NIHSS >14 (3). A good clinical outcome was defined as mRS <2 at 90 days.
Imaging Analysis
The normalized volumes of the ischemic lesions were calculated from the CT images using the ImageJ software (National Institutes of Health; http://rsb.info.nih.gov/ij). The density of the lesion was measured in Hounsfield units (HU). In order to adjust for individual variations in parenchymal density and allow a more appropriate comparison among the imaging scores of different patients, we obtained a variable (AHU) that correlated the tissue density in the ischemic lesion with the corresponding contralateral hemisphere. AHU was calculated by subtracting the average density obtained from the corresponding nonaffected side from the average density of the area of maximum hypodensity on the affected side (29). All results were reviewed by an experienced neuroradiologist.
Statistical Analysis
Normally distributed continuous data were analyzed by two-sample t-test, and data that were not approximately normal in distribution were analyzed by nonparametric techniques. The Friedman test was used to analyze repeated measures in paired samples, and the Mann-Whitney test was used for independent samples. Means, SDs, medians, and ranges were calculated for all continuous data. A value of p < 0.05 was considered significant. All computations were performed with the aid of SPSS software (version 13.0; SPSS, Inc., Chicago, IL).
Results
Twenty patients with MCA infarcts were prospectively enrolled in this study over a 12-month period. Their baseline characteristics and the general characteristics of the brain ischemic events are summarized in Table 1. The mean age was 63 years (range 30–78), 14 (70%) were males, and the right hemisphere was the affected side in 12/20 patients (60%). The mean baseline NIHSS score was 17 ± 5.6 (median 15.5; range 9–28). Figure 1 shows CTs from all patients before BMMC transplantation. The mean time from stroke onset to treatment was 6 ± 1.8 days (range 3–10). The mean cell count in the infused solution was 22.08 × 107 cells (range 5.1 × 107-60 × 107).

Computed tomography scan before transplantation of bone marrow mononuclear cells (patients 1 to 20, from top left to bottom right).
Clinical Characteristics
TOAST: Trial of Org 10172 in Acute Stroke Treatment.
There were no serious adverse events related to the experimental procedure. Specifically, none of the primary endpoints were met. Serial follow-up CT and MRI evaluations did not show any significant anatomical or structural changes that could indicate adverse events. In particular, there was no evidence of any new ischemic, hemorrhagic, or neoplastic lesions. Three patients (15%) developed pneumonia, two (10%) had urinary tract infections, and another two (10%) had deep venous thrombosis. All patients were discharged home. There were no episodes of clinical seizures or recurrent ischemia during the follow-up period, which lasted 6 months. Similarly, there were no laboratory, radiological, or electroencephalographic changes that could not be explained by the baseline clinical picture and were attributable to the experimental treatment. There were two deaths (10%) during the follow-up. Patient 4 was discharged in good condition but died from an acute myocardial infarct 43 days after the procedure. Patient 18 developed hemorrhagic transformation of his infarct before the IA ABMMC infusion, and underwent hemicraniectomy due to the infarct extension and mass effect 2 days after the IA infusion. He responded favorably to decompressive surgery and was discharged home 30 days later. However, he died 61 days after the IA ABMMC infusion from infectious complications related to an elective cranioplasty.
Analysis of the CT variables obtained during the first week after the IA ABMMC infusion (mean 2.6 ± 2 days; range 0–7) showed a mean stroke volume of 115.887 ± 95 cm3 (median 89; range 11–381) with a mean density of the ischemic lesions of 23.24 HU (Table 2). The mean ΔHU was 6.98.
Tomographic Variables
CT, computerized tomography; ΔHU, Delta Hounsfield units.
Table 3 summarizes the evolution of the neurological deficit according to NIHSS. There was a significant reduction in the median scores of NIHSS between the pretreatment and the 180-day follow-up period (p < 0.001). Figure 2 illustrates the evolution of the patients' neurological deficit according to the NIHSS. Six patients (30%) achieved satisfactory clinical improvement in functional recovery at 90 days according to our prespecified criteria: four patients with baseline NIHSS 8–14 achieved a mRS ≤1, and another two patients with baseline NIHSS >14 achieved a mRS ≤2. A total of eight patients (40%) achieved a mRS ≤2 at 90 days. Similarly, no significant difference in terms of infarct volumes, infarct densities, or ΔHU was found between groups with and without clinical improvement (p = 0.76, 0.97, and 0.73, respectively).

Evolution of neurological deficit according to the National Institutes of Health Stroke Scale (NIHSS).
Evolution of Neurological Deficit According to NIHSS
NIHSS, National Institutes of Health Stroke Scale; mRS, modified Rankin Scale. Bold type indicates scores of the patients who reached any efficacy outcome (satisfactory clinical improvement in 24 h or in 90 days). Note: in the NIHSS we used the score 42 to indicate death.
NIHSS (day 0) ≥16.
Reached satisfactory clinical improvement in 24 h
Reached satisfactory clinical improvement in 90 days.
Discussion
This study included a group of moderate to severe acute MCA infarct patients. These patients typically have a poor chance of good functional recovery and high mortality rates. Indeed, National Institute of Neurological Disorders and Stroke (NINDS) trial patients with an NIHSS greater than 16 had a 92% chance of death or severe sequelae (37). The primary objective of our study was successfully met, with no adverse events related to the IA ABMMC infusion.
The incidence of infections and thrombotic complications in our study was similar to those previously reported in studies that evaluated the clinical course of this severe subtype of stroke (5,18). The two deaths that occurred during follow-up were considered to be unrelated to the procedure or to the effects of cell therapy. Complications of the stroke itself and its comorbidities were most likely the cause of these deaths. This mortality rate was below the expected level compared to those reported in similar populations (14,18). Even though this study was not designed to show the efficacy of IA ABMMC infusion, its results are very encouraging. Our rates of good outcomes are in the same range of what was seen in the Prolyse in Acute Cerebral Thromboembolism II (PROACT-II) trial, which also had a mean NIHSS of 17 and included only MCA strokes (15). In our pilot study, 40% of the patients achieved a mRS ≤2 at 90 days, which was similar to the active group (40%) and superior to the placebo group (25%) of the PROACT-II trial. Moreover, our results compare favorably with other intra-arterial recanalization studies that had only slightly higher baseline NIHSS scores (27, 32,33). However, since our sample is small, we cannot rule out the possibility that these results are simply explained by chance.
ABMMC therapy has many advantages over other modalities of stem cell treatment, including greater simplicity of the techniques involved, absence of cell rejection, lack of major ethical concerns, and reduced risks of infection. In the current protocol, we also avoided issues related to high contamination risks and reduced cell survival typically seen with ex vivo cell culture expansion (6), by isolating the desired cells and performing the transplantation only a few hours later. In addition, this strategy makes it possible to deliver the cells during the optimal stroke time window, because no additional time was required for culture expansion.
The route of transplantation has been a major concern in cell therapy for neurological diseases. The advantages of the intra-arterial method over stereotactic implantation include not only its less-invasive nature but also a higher likelihood of delivering the transplanted cells in a more diffuse manner, which could potentially lead to better clinical outcomes (25). The intra-arterial route also seems advantageous compared to less-invasive routes such as intravenous infusion, because it results in higher cell counts at the target site (21,22,24,39). Although these aspects were not addressed or confirmed in the present study, previous studies have shown that BMMCs migrate to the lesioned hemisphere with the ischemic region in both acute and subacute patients (7,8,10), and these results support the method that we have chosen. The major theoretical concerns related to our protocol included the risk of embolization of the injected vessels, worsening or additional ischemic injury, hemorrhagic transformation, cerebral edema, epileptogenic activity, or the appearance of neoplastic lesions. These aspects were carefully evaluated during follow-up.
Compared to the few other clinical studies evaluating cell therapy in stroke patients (9,26), this is the only one that tested cell therapy as an acute-phase treatment. This design was chosen considering some experimental evidence of better outcomes in earlier transplants (11,20) and the rationale of targeting the penumbra area, conceptually vulnerable in the acute phase. Data showing a decrease in cell death in the penumbra area in the acute ischemic phase also account for this choice (16).
The main limitations of the current study include its small sample size and the lack of a control group. In addition, we did not define the cell phenotypes present in the infused material or apply techniques to label the infused cells. There is controversy about the importance of the permanence and effective implantation of the administered cells, because functional effectiveness has been seen even in cases of reduced cellular implantation (30,41). This again raises the possibility that any benefit is at least partly related to the release of trophic factors.
In conclusion, transplantation of ABMMCs to patients with moderate to severe infarcts involving the MCA territory is feasible and seems to be safe. Further randomized clinical trials are necessary to establish the efficacy of this procedure.
Footnotes
Acknowledgments
This study was partially supported by a grant from the Ministry of Health and Ministry of Science and Technology of Brazil to Rosalia Mendez-Otero (grant number 552201/2005-7). The authors declare no conflicts of interest.
