Anti-Human Haemoglobin ELISA and Modern Immunoassays

In the sandwich ELISA for human haemoglobin, the capture antibody is adsorbed onto the walls of a polystyrene microplate well. The sample extract is added to the well; any human haemoglobin in the extract binds to the capture antibody. The well is washed to remove unbound material. Then the detection antibody, conjugated to an enzyme (usually horseradish peroxidase), is added. It binds to the captured haemoglobin from a different epitope. After another wash to remove unbound detection antibody, the enzyme substrate is added. The enzyme catalyses a colour reaction, and the microplate reader measures the optical density at the appropriate wavelength.

The optical density of the sample well is compared to a standard curve constructed from wells containing known haemoglobin concentrations (typically spanning from 0 to several hundred nanograms per millilitre). Interpolating the sample OD against the curve gives a concentration in nanograms per millilitre of the extract, which can then be converted to an estimate of blood volume using the known haemoglobin content of whole blood (approximately 14-16 grams per 100 millilitres).

The limit of detection for a well-optimised anti-human haemoglobin sandwich ELISA is typically in the range of 0.1 to 1 nanogram of haemoglobin per millilitre of extract. Given that a microlitre of whole blood contains approximately 140 micrograms of haemoglobin, this means a positive ELISA result can in principle be obtained from a bloodstain equivalent to a fraction of a nanolitre of whole blood, an amount invisible to the eye and barely detectable by a swab.

This sensitivity advantage is not always practically relevant. In most casework, the stain is large enough that a lateral-flow test gives a clear positive, and quantitation adds little to the investigation. ELISA sensitivity matters most at the edges: very small or heavily diluted stains, samples where the lateral-flow band is absent but the investigative context suggests blood is likely, and cases where blood volume estimation is required.

The cross-reactivity profile of an anti-human haemoglobin ELISA depends on the specific monoclonal or polyclonal antibodies used. Antibody manufacturers and forensic validation studies typically test extracts from a range of common and uncommon animal species and report the response as a percentage of the response to pure human haemoglobin at the same protein concentration. A cross-reactivity below 5% is generally considered acceptable for a test claiming human specificity.

The pattern is consistent with what is seen in lateral-flow assays. Great apes (gorilla, chimpanzee, orangutan) cross-react significantly because their haemoglobin amino acid sequences are closely homologous to human haemoglobin. Ferret haemoglobin shows partial cross-reactivity in most validation studies. Dog, cat, horse, pig, cow, sheep, deer, rat, mouse, and rabbit do not show significant cross-reactivity with well-validated anti-human Hb antibodies.

The quantitative output of ELISA is its most distinctive contribution relative to lateral-flow methods. A haemoglobin concentration in the stain extract can be used to estimate how much blood was deposited, which can be relevant to case interpretation. A stain containing the equivalent of several millilitres of blood is harder to explain by accidental secondary contact than a stain consistent with a microlitre or less.

Quantitation also matters when assessing whether a negative lateral-flow result is truly negative or simply below the assay's threshold. If ELISA detects haemoglobin at a concentration below the lateral-flow LOD, the analyst can report a trace detection that the lateral-flow card missed. This is useful in cases where the biological interpretation of a very small stain is contested.

• Volume estimation: concentration in extract, divided by extraction efficiency, converted using blood haemoglobin content (~150 mg/mL whole blood), gives an approximate blood volume range rather than an exact figure.
• Stain ranking: when multiple stains from a scene are tested, relative haemoglobin concentrations can guide prioritisation for DNA typing (higher haemoglobin usually means better DNA yield).
• Below-threshold detection: ELISA can confirm blood in stains where the lateral-flow test showed no band, providing a more sensitive second-line confirmation path.

Method selection in forensic serology is driven by the specific question the investigation requires answering: blood confirmation, human origin, quantity, or species. Each method addresses one or more of those questions with different capability and output format.

In most contemporary forensic workflows, the lateral-flow test (HemaTrace or RSID-Blood) is the front-line confirmatory test for human blood because it is fast, sensitive, and simple. ELISA steps in when quantitation is needed or sensitivity must be pushed lower. Classical precipitin tests are most used today for species identification in non-human blood contexts. Crystal tests remain valid second confirmations, particularly when the result will be challenged on the basis of antibody kit quality.

Reporting an ELISA result in a forensic context requires more documentation than a lateral-flow read because the output is quantitative and can be misinterpreted without context. The report should specify: the assay kit and antibody lot number used, the extraction method and solvent volume, the standard curve data (at least the R-squared value and the range), the OD of the sample, the calculated haemoglobin concentration, the extraction efficiency assumption used for any blood volume estimate, and the cross-reactivity statement for any relevant species.

If a blood volume estimate is reported, the uncertainty range must be stated. A haemoglobin concentration of 0.5 micrograms per millilitre in an extract prepared from a fabric swab, using an assumed 50% extraction efficiency, does not translate to a precise 1 microlitre of blood. It translates to a range, and the forensic report says so. Courts have repeatedly accepted ELISA results from accredited laboratories where the uncertainty was properly disclosed; they have also excluded or challenged results where the uncertainty was hidden or the cross-reactivity matrix was not provided.