BioDeviceHub

Percent Recovery Calculator

Calculate the percent recovery of a known amount of analyte added to a sample (a spike), or of a standard of known value, with the bias relative to 100%.

Formula

recovery (spike)=cspiked−cunspikedcadded×100\text{recovery (spike)} = \dfrac{c_{\text{spiked}} - c_{\text{unspiked}}}{c_{\text{added}}}\times 100
recovery (standard)=cmeasuredcknown×100\text{recovery (standard)} = \dfrac{c_{\text{measured}}}{c_{\text{known}}}\times 100
bias=recovery−100\text{bias} = \text{recovery} - 100
spiked\text{spiked}
measured result in the sample with the spike added
unspiked\text{unspiked}
measured result in the same sample without it
added\text{added}
the concentration the spike adds to the sample

How it works

Recovery tests whether a method measures all of an analyte in a real sample. A known amount is added to the sample (a spike) and measured; the difference between the spiked and unspiked results, divided by what was added, is the fraction recovered. A recovery below 100% suggests loss or suppression by the matrix, and above 100% suggests interference or contamination.

It is a measure of accuracy in the sample matrix, which calibration in clean solvent cannot show. A recovery is one measurement with its own uncertainty, so it is best repeated at several levels. Acceptance limits depend on the concentration and the method.

Worked example

A sample reads 12.1 without a spike, and 98.5 after adding 100 units of analyte.

  1. Recovered = 98.5 − 12.1 = 86.4 units of the 100 added.
  2. Recovery = 86.4 / 100 × 100 = 86.4%.

86.4% recovery, a bias of −13.6%, which points to some loss or suppression in this matrix.

These are the values the calculator opens with, so you can check its output against this example.

Assumptions

  • The spike is in the same form as the analyte, and mixes fully with the sample.
  • The amount added is accurately known, allowing for its dilution of the sample.
  • The unspiked and spiked samples are measured by the same method together.

Common mistakes

  • Forgetting the dilution of the sample by the spike volume when calculating the amount added.
  • Spiking at a level far above the natural amount, which hides a matrix effect at normal levels.
  • Taking a single recovery as a pass or a fail.

Related tools

Limit of Detection and Quantitation CalculatorLoD and LoQ from a calibration curve (ICH) or blanks (CLSI)Open calculatorReplicate Measurement Statistics (SD, %RSD, SE, CI)Mean, SD, %RSD, standard error and 95% CI of replicatesOpen calculatorCalibration Curve Calculator (Standard Curve)Linear fit, residuals and an unknown with its uncertaintyOpen calculatorDilution Calculator (C1V1 = C2V2)Stock and diluent volumes for any dilutionOpen calculator

Related equipment

Service documentation, failure modes and parts for the instruments this calculation is used with.