Taking a Closer Look at "Reduction"

As we continue to explore each of the 3 R’s as they relate to preclinical research, we move now to the second R as defined by Russell and Burch – Reduction. They state that the goal of Reduction is to minimize the number of laboratory animals used in a study while still obtaining statistically meaningful results.

The latter part of that definition is of upmost importance, researchers must ensure that in Reducing the numbers of animals used does not compromise the validity of the data. Otherwise, they themselves, or others, may feel the need to repeat the study, ultimately utilizing more animals to confirm the outcome of the study.

Another way to look at Reduction is also to think about both the quality and any additional information that may be obtained from the laboratory animal used in a study, which was not initially the goal of the study, but would be beneficial to the researcher or others since the study will already be underway.

When considering Reduction, there are a few approaches that scientists will consider:

• Improvements to experimental design
  • Standardization of all procedures related to and on the periphery of the study, to help minimize the variability between animals within the study. For example:
    • Surgical techniques
    • Method of acquiring and recording data
    • Time of day that any interactions take place
    • Husbandry practices
    • Etc.
• Stratification of disease severity, or entry criteria, prior to enrolling the subjects into a study. For example:
  • • Tightness of the stenosis in a transverse aortic constriction (TAC) model
    • Confirmed presence and size of lesion in a transgenic tumor model.
    • Selection of appropriate laboratory animal species and strain to best mimic the clinical presentation of the disease being studied.
    • Inbred or immune-compromised mouse strains may help reduce variability but may also remove some of the crucial biological systems and mechanisms which effect clinical outcomes.
    • Some species more closely reflect the clinical disease at a specific stage, for example at onset, or in progression to morbidity.
• Sample size calculations and statistical analysis techniques
  • When considering the required sample size for a study, it is necessary to examine the expected variability as well as the level of change which is expected, and which would be considered biologically significant not just statistically significant.
  • It is important to include an appropriate number of spare animals in each study group, as inevitably some loss will occur. The number of spare animals will need to be determined for the specific study being considered, and the techniques to be used, along with the skill of those conducting the study.

• Using advanced technologies 
  • Imaging may be used to non-invasively image the same animal over the course of a longitudinal study. 
    • One animal is used as its own control, and provides data at each imaging timepoint, rather than a different animal being required at baseline, and for each time point. This can dramatically reduce the number of animals required for a study. 
    • Additional animals may be use if post-mortem techniques are desirable at each time-point, for example histological or immunohistochemical staining, but significantly fewer animals would be needed when imaging is used and correlated with the results. 

• • Non-imaging techniques may be needed to help in the stratification of study subjects into groups, or to initiate them into the study, as described previously.