To identify the causal agent of the massive intoxication occurred in February 2022 in Buenos Aires, with 98 hospitalized victims and 24 deaths. The first victims presented the following signs: miosis, shock, sensory depression, respiratory distress (bradypnea), psychomotor excitement, seizures, and cardiorespiratory arrest. Some showed a typical opiomimetic sign: “wood chest”. Given the parasympathetic signs, the victims were treated with Naloxone at high doses, responding favourably and with a continuous infusion drip of 10 mg in 250 mL at a flow rate of 11 mL/h. The latter to prevent them from deteriorating when they waked up. All the cases had in common having inhaled a line of cocaine acquired from the same dealer. Seized powder samples were preliminarily analysed by HPTLC and UPLC-DAD (Acquity System-Waters Corp, USA). A cocaine standard (LGC Std, Lot. 979337, USA) was used. Subsequently, for the detailed analysis of components, a fractionation was made by extracting the soluble components in chloroform and the soluble and insoluble fractions were analysed by 1H and 13 C NMR on a Bruker A Neo 500 spectrometer (1H 500 MHz, 13C 125 MHz). The chloroform insoluble fraction was further analysed by LC-MS with exact mass detection on a Bruker micrOTOF-Q II spectrometer coupled to an Agilent 1200 HPLC. Structure confirmation was performed by MRM (multiple reaction monitoring), and MS/MS fragmentation. UPLC-DAD analysis showed a purity of 51.6% for COC, the rest being mostly sorbitol as cutting material (confirmed by NMR, as noted below). In the soluble fraction, where the spectra were measured in deuterochloroform solution, the NMR spectra showed the characteristic signals of cocaine hydrochloride and small amounts of cinnamyl cocaine (CIN) that usually accompanies cocaine hydrochloride and truxillines. (The latter very important to infer the origin of cocaine). In the insoluble fraction the main component was identified as sorbitol from its 1H and 13C NMR spectra (in DMSO-d6). A small remnant of cocaine and cinnamyl cocaine (CIN) was also observed. On the other hand, this fraction was analysed by LC-MS, observing 7 relevant peaks: cocaine, benzoylecgonine, cinnamyl cocaine (cis and trans), truxillines and a minor component of m/z 395.2316 that corresponds to the formula C24H31N2O3 (calculated 395,2329) coincident with the [M + H]+ quasi-molecular ion of carfentanil. Structure confirmation was performed by MRM, and fragmentation by MS/MS of the M + H ion, observing the characteristic fragments of the carfentanil structure including the two confirmatory ions of m/z 335.2123 (calculated for C22H27N2O+ 335,2118) and m/z 246.1489 (calculated for C15H20NO2+ 246,1489). Quantitative analysis was performed by NMR for the COC/CIN ratio and from LC-MS for the CIN/Carfentanil ratio. Giving an estimated result of 30 μg/100 mg of pure cocaine. The toxic dose for humans is unknown, although would be less than 1 mg based on animal studies and known fentanyl lethal cases (estimated lethal dose is 20 μg) (Casale et al., 2017). The therapeutic treatment followed by the victims and the results observed allowed us to focus on opiomimetics, as the most relevant factor in the cause of the 24 deaths. The high-resolution methods applied allowed us to confirm cocaine in a percentage of purity close to half of the total powder. Carfentanil was confirmed by MRM analysis and in an amount of approximately 30 μg/100 mg of pure cocaine. Based on the data collected in the case, the consumption of cocaine was about 200 mg, therefore, in the proportion of purity, each line contained about 30 μg of carfentanil. So, for 400 mg of powder consumed, carfentanil would be found at 60 μg. Therefore, we estimated the toxic dose for this episode between 30–60 μg.