To assess the correlation between ATP measurements and ACC’s we conducted measurements on multiple pre-defined identical fomites. Reproducibility was assessed by comparing two measurements on the same fomite.
There was a strong correlation between two ATP measurements on different sites on the same fomite and between two ACC measurements. Furthermore, we found a low correlation between the amount of ATP measured and the amount of microbial contamination.
An explanation for this low correlation could be that other biological materials are measured with ATP measurement. The amount of bacterial/fungal contamination on a surface could be low, while contamination with other biological material (e.g. food scraps, skin flakes, etc.) is higher. ATP measurement is frequently used to quantify surface cleanliness and determining problematic areas. Contamination of surfaces with organic materials could provide a nutritional source for bacteria and other pathogens. By using ATP measurements for analyzing surface contamination, potential nutritional sources for microbes can be traced and disposed of. A decrease in microbial growth, contributing to a clean hospital/nursing home environment, can reduce the risk of microbial transmission.10
Another point of discussion is the correlation between the amount of ATP measured and the expressed RLU. Omidbakhsh et al. (2014) did further research on this point; they conclude that there is a strong positive correlation between true concentrations of ATP and RLU readings, however this correlation is best when the concentration of ATP is higher. The same conclusion, as described above, is true for the correlation between dilutions of Staphylococcus aureus and RLU readings. For the 3M ATP luminometer an R2 of 0.9228 was found for the correlation between concentrations of ATP and RLU readings, and an R2 of 0.9746 for the correlation between dilutions of Staphylococcus aureus and RLU readings.11 This indicates that there is a good correlation between the amount of ATP/Staphylococcus aureus measured and RLU readings. This kind of lab-controlled research gives a better indication of the effect of bacterial load on RLU readings, however is not applicable in daily practice. Contamination with other biological materials can influence RLU readouts, even when bacterial contamination is low. In a sub analysis within this study we did not find a significant difference in the correlation between RLU values and ACC’s with RLU ≥ 1000.
Limitation of ACC are the incubation temperature and selection of growth medium. Various microbes grow with different temperatures and on different culture media.12 There is a possibility that a certain amount of microbes is missed with the ACC. Thus, giving a lower estimation of microbial contamination than is present on surfaces in reality. By culturing TSA agar plates for 48h at 35°C, slow growing microbes and/or those demanding different nutrients could have been missed.
Concluding, the correlation between RLU values and ACC seems to be low. ATP measurement give insight into surface cleanliness and should mainly be used for assessing surface contamination. We found a good correlation between two ATP measurements on the same fomite, indicating that ATP measurement has good reproducibility. Therefore, ATP measurement seems a reliable method for measuring surface contamination.
ATP measurement can be used to give a quantifiable outcome for the rating of cleanliness in health care facilities, however it should not be used to interpret microbial contamination directly.