The thermal degradation products of polyurethanes (PURs) and exposure to isocyanates were studied by stationary and personal measurements in five different occupational environments. Isocyanates were collected on glass fibre filters impregnated with 1-(2-methoxyphenyl)piperazine (2MP) and in impingers containing n-dibutylamine (DBA) in toluene, connected to a glass fibre postfilter. The derivatives formed were analysed by liquid chromatography: 2MP derivatives with UV and electrochemical detection and DBA derivatives with mass spectrometric detection. The release of aldehydes and other volatile organic compounds into the air was also studied. In a comparison of the two sampling methods, the 2MP method yielded about 20% lower concentrations for 4,4′-methylenediphenyl diisocyanate (MDI) than did the DBA method. In car repair shops, the median concentration of diisocyanates (given as NCO groups) in the breathing zone was 1.1 µg NCO m−3 during grinding and 0.3 µg NCO m−3 during welding, with highest concentrations of 1.7 and 16 µg NCO m−3, respectively. High concentrations of MDI, up to 25 and 19 µg NCO m−3, respectively, were also measured in the breathing zone during welding of district heating pipes and turning of a PUR-coated metal cylinder. During installation of PUR-coated floor covering, small amounts of aliphatic diisocyanates were detected in the air. A small-molecular monoisocyanate, methyl isocyanate, and isocyanic acid were detected only during welding and turning operations. The diisocyanate concentrations were in general higher near the emission source than in the workers' breathing zone. A sampling strategy to evaluate the risk of exposure to isocyanates is presented.