[A three year survey of the practical application of pollen monitoring in specific allergen immunotherapy].

Specific allergen immunotherapy should be modified according to sensitivity of the patient and the time interval between injections and seasonal allergen exposition. The aim of the study was to check the effectiveness of the multinomial logistic regression models predicting the pollen concentration during the pollen season in the immunotherapy trial in patients treated with grass and birch allergens. The study was performed in Krakow in 2011-2013. Models were validated for 2012 and 2013. The effectiveness of the total correct predictions slightly differed depending on the time series, in case of birch pollen the similar percentage of correct predictions was found in both study year, while in case of grass pollen, the predictions were more correct in 2012. A group of patients treated with grass and birch allergens filled in the diary cards during the pollen season. After the 2011 season 14 diary cards were analysed, while 18 and 19, in 2012 and 2013, respectively. Because of manifested symptoms, the injection dose was reduced during the season in 12 patients in 2011, in 9 patients in 2012 and in 6 patients in 2013. No visits were delayed because of medical indications. In some cases patients got the injection in time of the high pollen occurrence (2 cases, in 2011 and 2012). In 2013 in 10/17 patients the high pollen exposure was avoided thanks the information from pollen monitoring, in opposite to 1 and 8 patients in 2011 and 2012, respectively. Patients used antihistaminic drugs on request. The regional pollen monitoring data and satisfied co-operation with patients makes the possibility of closer control of the injection doses administration during immunotherapy in the pollen season.