Esta referencia está disponible actualmente en inglés. La versión en español está en preparación.

Bone Marrow Aspirate Concentrate and Adipose SVF: Composition

Por Benn Bluestein-Veyra, M.Sc. en Química Orgánica · Actualizado el 21 de octubre de 2026

Bone marrow aspirate concentrate (BMAC) is the nucleated cell and platelet fraction of aspirated marrow, collected after density centrifugation. Schäfer and colleagues measured enrichment of total nucleated cells, platelets, CFU-F colonies and several stromal cell subsets after concentration.^[1] Adipose stromal vascular fraction (SVF) is the cell pellet released from fat tissue, defined by a joint IFATS and ISCT statement.^[2]

This reference describes what each preparation contains, how preparation steps shape that content and which assays count the cells. It covers nucleated cell counts, the fibroblast colony-forming unit (CFU-F) assay and flow cytometry panels.^[2][3]

Understanding the cell composition of bone marrow aspirate

Marrow aspirate is a mixed suspension of red cells, platelets, myeloid and lymphoid nucleated cells, CD34+ hematopoietic progenitors and a small population of adherent stromal cells.^[1][3] Pittenger and colleagues isolated stromal cells from the marrow aspirates of volunteer donors and showed that single cells expanded into colonies retained adipocytic, chondrocytic and osteocytic potential.^[4]

CD34 is a highly glycosylated single-pass membrane protein encoded at chromosome 1q32.2, and it may help stem cells attach to marrow matrix or stromal cells.^[5] Stromal progenitors are rare: Muschler and colleagues counted a mean of 36 alkaline phosphatase-positive colony-forming units per million nucleated cells in iliac crest aspirates.^[6]

In a later series of 57 donors, the same group harvested a mean of 64 million nucleated cells per aspirate, with a mean prevalence of 55 colonies per million nucleated cells.^[7] Colony counts also depend on how marrow is collected and how cells are selected before plating (Fig. 1).^[8][9]

Quirici and colleagues reported a median CFU-F frequency of 35 per million low-density mononuclear cells. In the fraction selected with antibodies to the low-affinity nerve growth factor receptor, the median rose to 1,584 per million.^[9]

Aspiration volume and site shape marrow aspirate quality

Peripheral blood enters the syringe as aspiration continues, so the first milliliters drawn from a site carry the most marrow. Muschler and colleagues found that raising aspiration volume from one to four milliliters increased total colony-forming units while lowering their concentration by about 50% (Fig. 2).^[6]

Hernigou and colleagues compared bilateral aspirates from 30 adults and reported progenitor concentrations about 300% higher with a 10-milliliter syringe than with a 50-milliliter syringe.^[10] Cuthbert and colleagues likewise reported a several-fold fall in CFU-F and CD271-bright cells per milliliter as aspirated volume grew.^[3]

• Site: Pierini and colleagues reported a 1.6-fold greater yield of colony-founding progenitors from the posterior iliac crest than from the anterior crest, with comparable phenotype and differentiation of the cultured cells.^[11]
• Insertion technique: Oliver and colleagues compared single-site aspiration with needle redirection against multiple-site aspiration and found no significant difference in cell ratios or stromal cell numbers.^[12]
• Anticoagulant: Ferencakova and colleagues found a higher cell yield and more CFU colonies in heparin tubes than in EDTA tubes, with no significant change in the molecular characteristics of the cultured cells.^[8]
• Donor: Muschler and colleagues attributed 70% of the variation in progenitor prevalence to differences among donors, and in a later series nucleated cells per aspirate declined with age.^[6][7]

Density separation concentrates the buffy coat into BMAC

Centrifugation layers marrow aspirate by density into plasma, a thin buffy coat of nucleated cells and packed red cells. Bone marrow aspirate concentrate is the buffy coat collected with part of the plasma (Fig. 3).^[1][13]

Schäfer and colleagues processed 120-milliliter aspirates from 10 donors on two point-of-care systems. Both systems enriched total nucleated cells, platelets and CD45-CD73+ and CD45-CD73+CD90+ cells, with cell viability near 90% in all groups.^[1]

Concentration also raised CFU-F numbers, platelet-derived growth factor, vascular endothelial growth factor, macrophage colony-stimulating factor and VCAM-1. Neither system concentrated red cells, hematopoietic stem cells or bone morphogenetic proteins.^[1]

Separation systems yield different products

Dragoo and Guzman compared three systems on combined blood and marrow from 10 donors. The systems differed in recovery of CFU-F, CD34+ cells, leukocytes and platelets, and one system raised platelet concentration about 11 times above baseline.^[13]

Hegde and colleagues compared three systems in two paired series of bilateral aspirates from 40 donors and found different numbers and concentrations of connective tissue progenitors.^[14] Cassano and colleagues found that two systems both increased colony-forming units over the starting aspirate, and that one concentrate held more interleukin 1 receptor antagonist than platelet-rich plasma prepared with the same system from the same donors.^[15]

Characterizing BMAC with counts, colony assays and flow cytometry

An automated hematology analyzer reports total nucleated cells, platelets and red cells in the aspirate and the concentrate, and the ratio gives the fold enrichment.^[1][13] Flow cytometry adds CD34+ cell enumeration, and Cuthbert and colleagues noted donor ranges compatible with standard ISHAGE protocols.^[3]

The CFU-F assay plates nucleated cells at low density and counts the fibroblast colonies that form. It serves as the functional reference for stromal progenitor frequency.^[1][3]

Cuthbert and colleagues enumerated CD45-negative or low, CD271-bright cells with counting beads in a single-platform assay. These cells co-expressed CD73, CD105 and CD90, and their numbers correlated with manual CFU-F counts (R = 0.81).^[3] CD271, also called p75NTR, is encoded by the NGFR gene at 17q21.33, and its extracellular domain holds four cysteine-rich repeats that contain the nerve growth factor binding domain.^[16]

Cultured marrow stromal cells are described against the ISCT minimal criteria: plastic adherence, CD105, CD73 and CD90 expression without hematopoietic markers, and trilineage differentiation in vitro.^[17] Busser and colleagues enriched native marrow stromal cells by immunomagnetic selection for MSCA-1, SUSD2 or CD271, and also found a distinct CD34+ fresh marrow subset.^[18]

Adipose stromal vascular fraction is released by enzymatic or mechanical processing

Zuk and colleagues described processed lipoaspirate cells in 2002, a stromal population from fat tissue that differentiated toward osteogenic, adipogenic, myogenic and chondrogenic lineages in culture.^[19] The SVF is the cell pellet left after adipocytes float away, and it holds stromal, perivascular, endothelial and blood-derived cells (Fig. 4).^[2][20]

Enzymatic isolation digests the extracellular matrix with collagenase before centrifugation. Aronowitz and colleagues reviewed studies showing that enzymatic digestion separates cells with greater efficiency, while mechanical methods yield more blood mononuclear cells and fewer progenitor cells.^[21]

Yoshimura and colleagues characterized cells from both the fatty and fluid portions of liposuction aspirates. Multicolor flow cytometry showed blood-derived CD45+ cells, CD34+ stromal cells and CD146+ pericytes, and the adherent cells from both portions converged in phenotype after plating.^[20]

Microfragmented adipose tissue preserves the vascular stroma

Mechanical fragmentation can also reduce adipose tissue to submillimeter clusters without enzyme. Bianchi and colleagues described such clusters with preserved vascular stroma, a higher percentage of pericytes and lower hematopoietic content than enzymatically digested lipoaspirate.^[22]

Vezzani and colleagues compared matched samples and found microfragmented tissue enriched in pericytes relative to adventitial cells. In culture, the clusters released more growth factors and cytokines into the medium than isogenic SVF.^[23]

Understanding the IFATS and ISCT marker panel for SVF cells

Bourin and colleagues, writing for IFATS and ISCT, identified SVF stromal cells as CD45-negative, CD235a-negative, CD31-negative and CD34-positive. They proposed a viability marker and CD13, CD73, CD90 and CD105 as added value, and the CFU-F assay to measure progenitor frequency.^[2]

Cultured adipose stromal cells share CD90, CD73, CD105 and CD44 with other MSCs and stay negative for CD45 and CD31. The statement distinguishes them from marrow-derived MSCs by CD36 positivity and CD106 negativity.^[2]

Zimmerlin and colleagues mapped perivascular cells in intact fat. Inner pericytes were CD146+ alpha-SMA+ CD34- CD31-, outer adventitial stromal cells were CD146- CD90+ CD34+ CD31-, and capillary endothelial cells were CD31+ CD34+ CD90+.^[25] Corselli and colleagues reviewed how such flow cytometry gates allow prospective purification of perivascular stromal cells.^[26]

Mitchell and colleagues measured CFU-F in fresh SVF at a frequency of 1 in 32 cells.^[24] Busser and colleagues found that CD34 selection isolated stromal cells from adipose tissue, while SUSD2 and MSCA-1 selection did not.^[18]

Comparing marrow and adipose preparations by composition

The two preparations start from different tissues and carry different cell mixtures (Fig. 5). Marrow concentrate holds hematopoietic cells, platelets and rare stromal progenitors, and SVF holds a larger stromal and perivascular fraction alongside blood-derived cells.^[1][2][24]

Donor characteristics also shape composition. In marrow aspirates, nucleated cells per aspirate declined with age, and in women colony-forming units per aspirate also declined with age.^[7]

Locating these preparations in the federal HCT/P framework

Federal regulation states that “An HCT/P is regulated solely under section 361 of the PHS Act and the regulations in this part if it meets all of the following criteria:” The first criterion reads “(1) The HCT/P is minimally manipulated;”^[27] FDA sets out how it reads these terms in its guidance “Regulatory Considerations for Human Cells, Tissues, and Cellular and Tissue-Based Products: Minimal Manipulation and Homologous Use,” dated July 2020.^[28]

The 361 HCT/P versus 351 biologic reference on this site quotes all four criteria of 21 CFR 1271.10 in full. It is the place to read the regulatory framework alongside this page.^[27]

Where to go from here

The PRP preparation variables reference explains the centrifugation physics behind buffy coat separation, and the platelet granule reference describes the growth factors that platelets in a concentrate carry. The growth factor receptors reference follows those signals to their target cells, and the platelet study reporting reference lists the details a complete methods section includes.

The MSC source tissues and ISCT minimal criteria references describe the cultured stromal cells that marrow and fat give rise to, and the HCT/P reference sets out the federal framework. Our team is glad to discuss the characterization methods described on this page.