When we used NRL glove extract SPT to screen 3,269 children admitted for inhalant or food allergy testing and confirmed the diagnosis with a standardized NRL allergen SPT, latex RAST and glove use test, the prevalence of NRL allergy was found to be 1%. Although children with multiple operations are a well-known risk group for NRL allergy, two-thirds of the present 42 NRL-allergic children had no history of multiple operations. Moreover, a third of the NRL-allergic children had no history of symptoms though they had undergone operations in hospital or been in contact with balloons and other NRL products at home. Most of the non-operated (97%) and multioperated (83%) NRL-allergic children were atopics. These results show that routine skin prick testing, at present by using standardized reagents, is a valuable method of finding NRL-allergic subjects among children admitted for inhalant or food allergy testing. The glove use test is easy to perform and a high-allergenic glove brand can be recommended for use to confirm the clinical reactivity in the children who present with atypical or negative history of NRL allergy. It should be known, however, that certain NRL glove brands may contain moderate or minute amounts of cow´s milk casein. In addition to NRL allergic children we also found 5 children with cow´s milk allergy who presented with a "false" positive NRL glove use test reaction, i.e., contact urticaria, which was due to casein and not to NRL eluting from the gloves. Allergological data from the non-operated, NRL-allergic children is scanty. Accordingly, we compared SPT reactivity to NRL and IgE antibody findings between the non-operated and multioperated NRL-allergic children. The findings in both groups of children were, however, similar with regard to SPT reactivity to glove eluate and commercial NRL allergen (Stallergènes), and IgE antibody levels in latex RAST. The IgE antibody frequencies and levels were also measured for the purified major NRL allergens using ELISA. Between 58% and 86% of the non-operated and multioperated children had IgE antibodies to Hev b 6.01 (prohevein) and 6.02 (hevein), which are therefore major NRL allergens in these two groups of children. The multioperated children significantly more frequently had IgE antibodies to Hev b 1 (rubber elongation factor, 67% vs. 27%) and Hev b 5 (33% vs. 3%) than the non-operated children, suggesting that IgE response to these two NRL allergens could be due to sensitization during multiple operations. None of the present NRL-allergic children had IgE antibodies to Hev b 7, a patatin-like NRL allergen, showing that this is not an important NRL allergen for children. To examine the outcome of NRL allergy, 32 children were followed for a mean of 2.8 years. Despite careful instructions on avoidance of NRL, 22 (69%) NRL-allergic children were exposed to balloons and other NRL products at home. Ten (31%) children exhibited symptoms, 8 of them local contact symptoms and 2 children systemic symptoms. These results show that more attention should be paid to informing both non-operated and multioperated children and their parents on the risk of NRL allergy at home and in the everyday environment. The children followed did not show any change in SPT reactivity and IgE levels to NRL allergens, which supports an on-going NRL exposure in the home environment. On the other hand, the frequency of SPT reactivity to cross-reactive fruits (banana, avocado, kiwi fruit) as well as to potato remained high during the follow-up. Continuous exposure of the NRL-allergic children to cross-reactive fruits and vegetables could possibly also maintain increased IgE levels to NRL allergens, and especially to hevein, because class I endochitinases in these fruits are known to contain a hevein-like domain. In conclusion, the prevalence of NRL allergy is 1% among atopic children and it frequently affects also non-operated children. A third of the NRL-allergic children gave negative history of allergy symptoms. A combination of SPT with standardized allergen, latex RAST and NRL glove use test can be recommended for NRL allergy diagnosis in children. The major NRL allergens in children are hevein (Hev b 6.02), prohevein (Hev b 6.01) and also REF (Hev b 1) for multioperated children. During the follow-up two-thirds of the NRL-allergic children still had contacts with balloons, rubber boots or household gloves and could exhibit systemic or local symptoms, due to which the avoidance of NRL products in the home environment needs more attention.