You are:

Bookmark this page

Laboratory Implementation

Great Ormond Street Hospital for Children NHS Foundation Trust was one of the first clinical laboratories to implement GLI reference equations into clinical practice. Full implementation took over a year to do, but this was mainly due to equipment specifications rather than resistance from the clinical team. Lessons were learnt on the way, and most laboratories can expect to spend a couple of months ensuring a smooth transition to the new equations (once they are confident their equipment is capable of accurately processing GLI equations).

Some steps to ensure a smooth transition to GLI include:

  • Regular departmental meetings

  • Circulating literature

  • Respiratory teaching sessions – inviting the broader clinical team

  • Parents evening: We were concerned our CF population would be most affected since they were used to reviewing their trend reports. We held a parents evening to explain the limitations in the old equations and the benefits of switching to the new equations.

  • Clinician and Patient leaflets were generated, and are available to download from the links on the right

  • Change in report design to overcome the different predicted VC value (i.e. do not sure predicted value just show the range of normality).

  • All users emails were sent a month, a week and a day before changing the equations

  • Attend multi-disciplinary team meetings to explain spirometry results.

How do we mix reference equations on the reports?

There is concern that since the GLI equations are only available for spirometry the mixed reference module (ECSC for lung volumes and TLCO) will be confusing with some people suggesting they will wait to new equations are available for all outcomes before changing. This approach is not feasible. The outcomes and inclusion criteria included in new reference equations are usually at the discretion of the investigators and very few will measure multiple outcomes, hence it is unrealistic to expect a single reference equation which encompasses everything to appear. Even the ECSC equations are a combination of various studies (TLCO was not included on the original ECSC dataset), hence the way forward is working with the manufacturers to develop appropriate “prediction modules” which represents the appropriate reference equation for each outcome (as is the case currently). The possible discrepancy for predicted VC across TLCO and spirometry can be overcome if VC is shown as an absolute number with no accompanying predicted value for TLCO or lung volumes, and predicted FVC is only displayed with spirometry outcomes.

“The interpretation of discordant results (i.e. VC in normal range in GLI and outside normal range in ECSC) requires careful clinical judgement, rather than inappropriate application of out-dated reference equations.” (P.Quanjer 2014).

GLI Result Table Structure


GLI Note for Reports


The image above is an example of a report which has mixed reference equations (GLI for spirometry and ECSC for TLCO).

Note that VIN is reported as a quality control check for TLCO technique and the predicted columns are empty.


Will changing to GLI alter the results?

  • Slight differences in predicted values will occur in patient groups which were poorly represented in previous recommended reference equations:

  • Children (particularly early childhood and puberty)

  • Transition (at 18yr switch from paediatric reference data to adult reference data (and assumed to be 25))

  • Adult women (not included in original ECSC data)

  • Elderly: ECSC is extrapolated at 75yrs

  • Non-Caucasian subjects: Ethnic differences previously estimated 10-15%.


  • Association for Respiratory Technology & Physiology
    • Tel: 0845 226 3062
    • Fax: 0121 355 2420
    • Limited Company No. 5987719
    • Registered Charity No. 1117470
    • Connect with us