A recovery study asks how much of an analyte is represented after a specified analytical process. Peptide assays may involve transfers, extraction or other preparation steps that affect the amount reaching measurement. A meaningful recovery result identifies the process being tested, the comparison used and whether the calculation separates material loss from changes in signal response.
Identify which recovery is being reported
The term recovery can refer to different comparisons. Relative spike recovery concerns how much of a known addition appears in the measured result. Extraction recovery concerns material surviving an extraction stage. A response comparison against a neat solution can include both processing losses and matrix-related signal changes.
Matuszewski and colleagues' primary HPLC-MS/MS study explicitly separated recovery assessment from matrix effects. That distinction matters because a reduced signal can arise without a corresponding reduction in analyte amount.Matuszewski and colleagues — Matrix effects and recovery in HPLC-MS/MS (opens in a new tab)
| Comparison | Main interpretation question |
|---|---|
| Spiked versus unspiked sample | How much of the known addition is measured? |
| Addition before versus after processing | What does the intervening processing change? |
| Processed matrix versus neat reference | Does the matrix alter response? |
| Whole process versus neat reference | What combined effects contribute? |
Subtract the amount already present
Eurachem's validation guide expresses relative spike recovery using the difference between mean spiked and unspiked results, divided by the known addition. Both results must refer to a consistent concentration or amount basis.Eurachem — Fitness for purpose of analytical methods (opens in a new tab)
In an original example, the unspiked sample result is 40.0 µg/mL. A matched preparation with a known added contribution of 10.0 µg/mL gives 49.0 µg/mL. Net recovery of the addition is (49.0 − 40.0)/10.0 × 100 = 90.0%. The stated addition already accounts for any preparation-volume change.
Dividing 49 by 50 instead gives 98%. That compares total measured concentration with the expected total, allowing the analyte already present to dominate the calculation. It is not the same as recovering 90% of the added contribution.
Repeated preparations would provide information about variability. A single illustrative difference does not establish the precision of a real recovery estimate, especially when a small addition is calculated by subtracting two relatively large measured values.
Test the process that matters to the result
Imagine that a standard added before a sample-cleanup stage gives a lower response than an equivalent addition after cleanup, with the final matrix and volume controlled. The comparison can reveal a loss associated with the intervening process. If the post-cleanup addition instead goes into a different solvent, response differences complicate that interpretation.
An addition after all transfers does not test losses during those earlier transfers. Likewise, a free peptide added to a protein-containing sample does not necessarily test how completely the protein releases that peptide during digestion. The experiment's chemical form and addition point define its reach.
Keep the result tied to concentration and matrix
Bienvenu and colleagues' primary study of a multi-analyte urine method found that sample background and the selected addition concentration affected matrix and recovery assessment. Its numerical design is not a universal peptide protocol, but it supports examining the actual concentration and matrix context.Bienvenu and colleagues — Matrix, recovery and internal-standard assessment (opens in a new tab)
A result above 100% does not mean the procedure created extra target molecules. Bias, background attribution, interference or variability can affect the calculated percentage. Investigate the cause rather than silently truncating the result to 100%.
A documented recovery study names the comparison, addition levels, sample types and variability. It also states whether the assay applies a correction and how that correction's uncertainty is handled. The result then helps explain quantitative performance without being mistaken for a universal purity percentage.
Sources and further detail
- Matuszewski and colleagues — Matrix effects and recovery in HPLC-MS/MS (opens in a new tab)
Anal Chem 75, 3019–3030 (2003), DOI 10.1021/ac020361s. Primary distinction between matrix effects and recovery; its drug-specific instrument comparison is not extrapolated to peptides.
- Eurachem — Fitness for purpose of analytical methods (opens in a new tab)
Third edition (2025), section 5.6.2 and equation 3 for relative spike recovery. The numerical example is original.
- Bienvenu and colleagues — Matrix, recovery and internal-standard assessment (opens in a new tab)
Anal Chem 89, 7560–7568 (2017), DOI 10.1021/acs.analchem.7b01383. Primary study applied to phthalate metabolites in urine, not a validated peptide procedure.
Sources checked 19 September 2026. Worked examples are illustrative unless a supplied report is explicitly identified. This article has not undergone independent scientific peer review.