Background: Aluminium is associated with disorders and is the commonly used vaccine adjuvant. Understanding the mechanisms of how Al is transported, metabolized or of its toxicity depends on the knowledge of Al-interactions with bioligands, i.e. Al-species. Al-speciation in serum is difficult because of low concentration and the risk of exogenous Al contamination. Furthermore, Al-measurements may be hampered according to various interferences. This study aims for developing quality controlled protocols for reliable Al- and Al-species determination and for investigating probable differences in Al (-speciation) after Al-containing subcutaneous immunotherapy (SIT).Methods: Sample donors were recruited either for the control group ("class-0", they never had been treated with SIT containing an Al-depot extract) or for the SIT-group ("class-1", they previously had been treated with SIT for insect venom allergy with an Al-depot extract). Blood was drawn for medical reasons and serum prepared. Additionally, some sample donors collected 24-h-urine. They had been informed (and they consented) about the scientific use of their samples. The study was approved by the ethic committee of the "Medical Association Westphalia-Lippe" and of the University of Munster, evaluating the study positively (No. 2013-667-f-S).We applied quality controlled sample preparation and interference-free Al detection by ICP sectorfield-mass spectrometry. Al-species were analysed using size-exclusion-chromatography-ICP-qMS.Findings: Al-concentrations or speciation in urine samples showed no differences between class-0 and class-1. Al-citrate was the main uric Al-species. In serum elevated Al-concentrations were found for both classes, with class-1 samples being significantly higher than class-0 (p = 0.041), but class-0 samples being approximately 10-fold too high compared to reference values from non-exposed persons. We identified gel-monovettes as contamination source. In contamination-free samples from HNO3-prewashed gel-free monovettes (n = 27) there was no difference in the serum Al concentration between the two patient groups (p = 0.669)Interpretation: Thorough cleaning of sample preparation ware and use of gel-free monovettes is decisive for an accurate Al analysis in serum. Without these steps, wrong analysis and wrong conclusions are likely. We conclude that gel-monovettes are unsuitable for blood sampling with subsequent Al-analysis. Whether Al in serum is elevated after SIT treatment containing an Al-depot extract, or not, remains inconclusive as the non-contaminated sample size was small.