| Front page | | Contents | | Previous | | Next |

Health effects of predatory beneficial mites and wasps in greenhouses

9 Discussion and conclusions

The present study has been an attempt to describe the possible health effects of introducing a new technology with the goal to replace chemical pesticides with more ecologically friendly agents. The question is whether the introduction of beneficial animals poses a risk to the persons employed in the greenhouse or to the consumers who are exposed to the plants after sale. Originally, the beneficial animals have been introduced without any risk assessment and it is thus not known if there is any measurable risk to be considered and need for regulation of the use of the technology.

The beneficial animals have been introduced in an environment with a considerable number of allergens and the problem is firstly, whether there is a measurable risk of the animals and secondly to compare it with those of competing technologies.

Table 9-1 shows an overview of the different exposure related effects found throughout Chapter 4 to 7.

Table 9-1. Overview of the effects of exposure to the four different predators. 0 denotes no significant effects, - denotes not measured. (A) the group of applying persons.

Estimating the allergic effects of beneficial animals in a complex environment poses many problems. The assessment of health effects caused by the house dust mite is an example of the work done to solve these problems. Standardised extracts of specific components have been developed as well as measures of exposure in the environment (beds, mattresses, carpets) both by measuring the number of vital animals and the amount of antigen. Finally standardized solutions of antigen for prick tests have been developed for these mites as well as some of the storage mites.

For the beneficial animals in the present studies no such extracts were available and therefore the first step was to produce these in sufficient amount and purity. It was a problem that animals were fed on a mixture of wheat bran and mites used for fodder (Tetranychus and Tyrophagus species). This has inevitably decreased the specificity of the extracts and thereby decreased any dose-response relations between exposure estimates and sensitization. On the other hand the rate of sensitization did not differ from the two other studies of beneficial animals and sensitization (Groenewoud et al., 2002a; Kronqvist et al., 2005).

Another problem in the study was the characterization of exposure. This study was made on observations and sampling of blood done several years before and the study was not originally set up to study beneficial animals, but microbiological pest management. Therefore we had to make retrospective investigation with interview of the greenhouse owners about the use of the individual beneficial animals during the observation period some years before. The information in the questionnaires from the participants in the study was obtained at the actual time and probably more reliable but only for those who had been applying the animals themselves. However, a good agreement between the two different measures was seen for about 70 % of the material.

The pilot study on Amblyseius cucumeris described in chapter 8 did partly show measureable airborne antigen. A series of technical problems decreased the sensitivity of the assay. A development of the method may increase the sensitivity, but as it relies on serum from persons with proved sensitization a method using other techniques may be preferred if used in a larger scale. This may be combined with methods for counting viable mites in the environment.

The employees who actually handled the animals were probably the most exposed and therefore at the largest risk of being sensitized or developing symptoms. However, no indication of either increased sensitization or respiratory symptoms was seen in this group. On the contrary this group typically had the lowest prevalence of the exposure groups. One of the cases, however, clearly reported symptoms in relation to the handling of Amblyseius cucumeris. Generally, however, the present study was not aimed at the detailed description of symptoms in relation to specific processes.

The reason for low prevalence of symptoms may be “healthy workers selection” where only the most healthy and fit persons continue being exposed while persons with symptoms formally or informally are transferred to other jobs or out of the work place. Persons with atopic disposition, shown by sensitization to one or more of the common allergens, are in elevated risk of developing allergy. However there were no indications of decreased number of atopics in the group who were applying any of the beneficial animals. An alternative hypothesis may be that the amount of time the individual person was exposed during handling was too short to increase the risk of sensitization and symptoms.

The cross sectional analyses have limitations with reporting bias and selection bias. At the present stage this probably has a minor importance as the beneficial animals were not stated as the primary goal of the study. On the other hand the three follow-up years were only a short period of a working life. Therefore a selection might already have taken place and therefore a healthy group has been investigated. The rate of self reported asthma of 7 %, however, is at the same level or higher than that of a referent population in five Danish counties including Funen County (Skadhauge et al., 2005). On the other hand the frequency of bronchial hyperreaction was very low, so persons with active asthma might not be able to work in the greenhouse athmosphere.

Among the specific animals the wasp Aphidius colemani showed a sensitization on the same level as the mites; however, in contrast to the mites no exposure relation was seen. On the other hand there was a correlation between exposure to Aphidius colemani and the incidence of upper air way symptoms which was in contrast to the mites which showed some relation to lower airway symptoms.

Among the mites Amblyseius cucumeris was the most frequently used and showed a clear relation between exposure estimates and sensitization by the histamine release reaction. The relation between sensitization and symptoms was seen in wheeze and chest tightness at work, symptoms most related to inflammatory diseases, while cough, normally related to smoking or dust exposure, also was seen in the present study. In the Dutch study, relation was mainly seen to upper air way symptoms and nasal allergies(Groenewoud et al., 2002a). Why the two studies are different may be explained by the upcoming measures in the present study.

Exposure to Phytoseiulus persimilis caused a clear sensitization, although in comparison with Amblyseius cucumeris the level in the unexposed group was higher. This pattern of sensitization is in agreement with the Swedish study (Kronqvist et al., 2005).

According to Hypoaspis miles the level of sensitization was lower and only a single person had a high IgE and the inhibition studies indicated that this IgE probably was the least reliable of the four measures. On the other hand a relation of the same size between symptoms and sensitization was seen in this mite although it was mostly seen in the symptom “Cough at work”.

Tetranychus urticae showed a relation between chest tightness at the start of work in combination with wheeze. It was not different from the pattern of the other mites.

The relation between symptoms and sensitization in the different species were very much alike. This could suggest that the relation is caused by a common background factor. However exposures in the population were different and there were only a partial overlap between those having IgE to the different species. This supports partially the hypothesis that the mites, both the beneficial and Tetranychus urticae have at least some specific biological effects.

| Front page | | Contents | | Previous | | Next | | Top |

Version 1.0 August 2007, © Danish Environmental Protection Agency