A flow cytometry report lands on a desk and, for most people who are not hematologists, it reads like a decoded transmission from another discipline — columns of percentages, cryptic CD numbers, gates that look like abstract art. That gap matters clinically, because misreading a single value can push a patient toward an unnecessary biopsy or, worse, delay a referral that should have happened last week. This walkthrough takes a standard CD34+ report, the kind issued after a peripheral blood or bone marrow collection, and explains each major panel in plain language. The goal is not to replace specialist interpretation but to give patients and generalist physicians enough vocabulary to ask the right questions in the room. Stem Lab Zone's team reviewed anonymized report formats from multiple accredited South African hematology laboratories and found that the three panels causing the most confusion were CD34 enumeration, the lymphocyte scatter gate, and the viability readout — so those get the most space here.
What a flow cytometer is actually measuring ¶
A flow cytometer pushes cells single-file through a laser beam and records two things: how much light each cell scatters and how much fluorescent signal it emits from any antibody tags attached to it. Scatter tells you about size and internal complexity. Fluorescence tells you whether a specific protein — a CD marker — is sitting on or inside that cell's surface. The instrument does not diagnose anything on its own. It produces a count of events, usually between 50,000 and 500,000 cells per run, and the technologist draws gates, which are simply geometric boundaries on a dot-plot, to separate populations of interest from debris, dead cells, and irrelevant events. Every percentage on the report is a percentage of a gated population, not of all cells collected. That distinction is the single most important thing to understand before reading any column of numbers.
The CD34 enumeration panel: what the percentage really tells you ¶
CD34 is a glycoprotein expressed on haematopoietic stem and progenitor cells. In a healthy adult peripheral blood sample, you would typically expect CD34+ cells to make up somewhere between 0.001% and 0.01% of total mononuclear cells — a vanishingly small fraction. After mobilisation with granulocyte colony-stimulating factor (G-CSF), that figure can jump a hundredfold or more, which is exactly the pharmacological point of mobilisation before an apheresis collection. The report will usually express the CD34 count in two ways: as a percentage of gated mononuclear cells, and as an absolute count in cells per microlitre. The absolute count is the clinically actionable number. A result of 0.5% CD34+ cells means very little without knowing what white cell count it was derived from. Transplant programs in South Africa, following JACIE-aligned standards, typically require a minimum yield of 2 x 10⁶ CD34+ cells per kilogram of recipient body weight for a successful engraftment. Whether a single apheresis run will hit that target depends on the absolute peripheral blood CD34 count measured the morning of collection, not the percentage alone.
Reading the viability gate and why dead cells distort everything ¶
Every reputable CD34 enumeration follows the ISHAGE (International Society of Haematotherapy and Graft Engineering) sequential gating protocol, which includes a viability dye — most commonly 7-AAD or DAPI. Dead cells take up these dyes because their membranes are compromised; live cells exclude them. A report that omits a viability marker should prompt an immediate question to the laboratory, because dead CD34+ cells still stain positive for CD34 and will inflate the count. Acceptable viability for a cell product intended for transplantation is generally above 70%, though most quality collection centres aim for above 90%. If the viability column on your report reads below 70%, the clinical team needs that information before any yield calculation is presented to a recipient.
The lymphocyte scatter gate and the CD45 axis ¶
CD45 is expressed on virtually all leucocytes, but at different intensities depending on cell type. Lymphocytes show the brightest CD45 signal. Monocytes are intermediate. Granulocytes are dim. Blasts and stem cells are very dim. On the standard ISHAGE CD34 plot, the laboratory uses a combination of low side scatter and dim CD45 expression to isolate the blast and progenitor region of the plot before applying the CD34 antibody gate. When a generalist physician sees a dot-plot and notices a cloud of events sitting in that dim-CD45, low-scatter quadrant and asks whether it means the patient has blasts, the answer is: it might, or it might simply be the normal progenitor population. Context, clinical history, and a morphology review are required. Flow cytometry alone, without a concurrent differential count and clinical picture, cannot confirm leukaemia.
Co-expression markers: CD38, CD133, and what they add ¶
Some extended CD34 panels include co-expression markers. CD38 is relevant because the most primitive long-term repopulating stem cells are CD34+/CD38-low, while more committed progenitors are CD34+/CD38-high. If a report breaks the CD34+ population into these two subsets, the CD34+/CD38-low fraction is a rough proxy for 'true' stem cell content, though clinical transplant decisions are still based on total CD34+ count because the CD38-low assay is not yet standardised across laboratories. CD133 (also called prominin-1) is another early progenitor marker sometimes included in neural stem cell or ophthalmology-trial panels. Seeing CD133 on a report does not mean the sample was tested for the same purpose as a CD34 haematopoietic panel. Always check the clinical indication listed on the front page of the report.
Numbers that look alarming but often are not — and vice versa ¶
A CD34+ percentage of 3% post-mobilisation is often read as low by patients who expect to see something closer to 100%. It is not low — 3% of a mobilised peripheral blood mononuclear cell fraction can represent an excellent absolute count, depending on total white cell numbers and the volume collected during apheresis. Conversely, a viability of 95% on a sample that sat at room temperature for 18 hours before processing should raise a sample-handling question, because 7-AAD viability can be artificially maintained in cells that have not yet undergone full membrane disruption. The South African National Health Laboratory Service recommends processing CD34 samples within 24 hours of collection at 18–22°C, and most accredited private laboratories follow similar pre-analytic guidelines. When in doubt, ask the laboratory for the time-from-draw-to-analysis recorded in the run log.
Questions worth asking before accepting a flow report as final ¶
Four questions cover the majority of interpretive problems seen in practice. First: what was the gating strategy, and does the report include the actual dot-plots rather than just the tabular output? Tabular numbers without plots cannot be audited. Second: was the ISHAGE protocol used for CD34 enumeration, and was a viability dye included in the panel? Third: what is the absolute CD34+ count per microlitre, and what was the total white cell count used to calculate it? Fourth: was the sample processed within the validated pre-analytic window? A laboratory that cannot answer all four of these questions promptly is worth flagging to the requesting clinician. Flow cytometry is an extraordinarily powerful tool when the pre-analytic handling is controlled and the gating is performed by an experienced technologist. The report is the last step, not the only step.
Flow cytometry reports are not designed to be self-explanatory, and the gap between what they say and what clinicians and patients hear is a genuine problem in haematology practice. Asking a laboratory for the underlying dot-plots, checking the pre-analytic handling notes, and verifying whether an ISHAGE-compliant protocol was used costs nothing and can prevent misinterpretation. Stem Lab Zone's team is available to walk through individual report questions with referring physicians — bring the actual PDF, not just the summary table.