Ambrosia artemisiifolia, also known as common or short ragweed, is an invasive annual flowering herbaceous plant that has its origin in North America. Nowadays, ragweed can be found in many areas worldwide. Ragweed pollen is known for its high potential to cause type I allergic reactions in late summer and autumn and represents a major health problem in America and several countries in Europe. Climate change and urbanization, as well as long distance transport capacity, enhance the spread of ragweed pollen. Therefore ragweed is becoming domestic in non-invaded areas which in turn will increase the sensitization rate. So far 11 ragweed allergens have been described and, according to IgE reactivity, Amb a 1 and Amb a 11 seem to be major allergens. Sensitization rates of the other allergens vary between 10 and 50%. Most of the allergens have already been recombinantly produced, but most of them have not been characterized regarding their allergenic activity, therefore no conclusion on the clinical relevance of all the allergens can be made, which is important and necessary for an accurate diagnosis. Pharmacotherapy is the most common treatment for ragweed pollen allergy but fails to impact on the course of allergy. Allergen-specific immunotherapy (AIT) is the only causative and disease-modifying treatment of allergy with long-lasting effects, but currently it is based on the administration of ragweed pollen extract or Amb a 1 only. In order to improve ragweed pollen AIT, new strategies are required with higher efficacy and safety.

Hemofiltration National Registry is one of the patient registries implemented lately in Romania, currently in use, in response to increased clinical and research needs. The registries of patients with extracorporeal support of vital functions were developed with the support of Romanian Society of Anesthesia and Intensive Care. The registry contains data on over 200 hemofiltration procedures that were per-formed in the last 3 years in multiple Romanian hospitals. A sample of data containing records of 2018 was analyzed by K-means clustering, revealing patterns that are potentially useful for healthcare improvement. Among the 6 clusters identified, 3 contain patients with a high mortality rate (90-100%), 1 is defined by intermediate mortality (72%) and 2 by a lower mortality rate (62%). Further research is needed in order to refine the clustering criteria, by using a larger number of cases and potentially examining more outcomes.