In 1996, an extensive exposure of Brazilian hemodialysis patients at a dialysis center, using a municipal water supply water contaminated with cyanotoxins, provided the first evidence for acute lethal human poisoning from the cyclic peptide hepatotoxins called microcystins. During this outbreak, 100 of 131 patients developed acute liver failure and 52 of these victims were confirmed to have been exposed to lethal levels of microcystins. Detection and quantitation of microcystins in these biological samples posed some analytical challenges since there were no well-established and routine analytic methods to measure total microcystins in tissue or sera samples. At the time of the 1996 exposure we used analytic methods that combined the use of enzyme linked immunosorbant assay (ELISA), analytical high performance liquid chromatography (HPLC), electrospray ionization ion-trap mass spectroscopy (ES-ITMS) and matrix assisted laser desorption ionization-time of flight spectroscopy (MALDI-TOF). In the intervening years these methods have been improved and others developed that allow a more quantitative and critical analysis of microcystin contaminated tissue and sera. For these reasons, and to see how storage with time might effect the detection and stability of microcystins in these matrices, we reanalyzed selected liver tissues and sera from the Caruaru victims in Brazil. We developed and validated a procedure to measure total microcystins in Caruaru human sera and liver tissue using a combination of ELISA, liquid chromatography and liquid chromatography-mass spectrometry (LC/MS), GC/MS and MS/MS techniques. GC/MS and LC/MS were followed by MS/MS to obtain a fingerprint fragment spectra for the microcystins. The validity of the extraction procedure for free microcystins was confirmed by recovery experiments with blood sera spiked with microcystin-LR. We removed proteins with the Microcon Centrifugal Filter prior to LC/MS and ELISA analysis. A solid phase extraction (SPE) procedure was used for analysis of protein bound microcystins by conversion of ADDA to erythro-2-methyl-3-methoxy-4-phenylbutyric acid (MMPB) combined with GC/MS. We found that the GC/MS method yielded a higher concentration of microcystin than that obtained by ELISA and LC/MS. We hypothesize that this difference is due to better GC/MS detection of the covalently bound form of microcystins in human liver tissue. We also concluded that microcystins are very stable when stored under these conditions for periods of almost 10 years.