Given the high estimated prevalence and financial burden of occupational asthma (OA),1 ,2 prevention is a major concern. However, in the absence of a regulatory screening protocol, the respiratory sensitising potential of a chemical is usually only apparent when it has caused a human case of OA.

The purpose of primary prevention is to detect hazardous agents before the occurrence of the issue, in order to avoid exposure of workers.3

Agius et al4 have developed and validated a computer based asthma hazard prediction model to predict the potential of low molecular weight (LMW) organic agents to cause asthma due to sensitisation.5 ,6 Based on the hazard index (HI) and using a cut-point of 0.5, the model showed good discriminative properties (sensitivity 79%, specificity 93%, negative predictive value 91–100%).6

The aim of secondary prevention is the early detection of symptomatic workers exposed to a sensitising agent.3 Secondary prevention in OA therefore requires knowledge of new sensitisers reported in the literature.

More than 400 agents are known as respiratory sensitisers able to cause OA.7 New agents are frequently reported in the literature, as shown by Pralong et al8 in their recent review, 41 new LMW agents having been reported between 2000 and 2010.

We present in table 1 all the new organic LMW agents published between 2000 and 2012 and their HI. Metals, wood dusts and chemicals with a molecular weight greater than 1000 Da are excluded, due to the inability of the model to calculate a HI for these categories.

Thirteen of the compounds shown in table 1 have their asthmagenic potential corroborated by a HI greater than 0.5. Among the five compounds with HI less than 0.5, peracetic acid showed a HI of 0.03 but for the authors it is unclear whether the mechanism inducing OA is sensitisation or irritation.9 This Quantitative Structure-Activity Relationship model (QSAR) model is not applicable to OA due to irritation, although the difficulties in distinguishing clinically between irritant and sensitisation mechanisms have recently been described.10 Moreover, specific inhalation challenge (SIC) was conducted with a mixture of peracetic acid and hydrogen peroxide, so either could have been the cause of OA.

In the four remaining compounds with a HI<0.5 (namely sevoflurane, lasamide precursor, eugenol and fluazinam), no precise explanation can be postulated yet. However, these compounds could be included in a future expansion of the QSAR learning set so as to further improve the model by iteration.

We highlighted here the QSAR model's ability to correctly predict the asthmagenic potential of new organic LMW agents.