Estima-se atualmente que mais de 5% da população mundial vem fazendo uso recreativo de algum tipo de substância psicoativa, sendo que o direito a esse uso é tema recorrente da sociedade contemporânea. Por apresentar riscos associados à saúde e a segurança das populações, o uso abusivo dessas substâncias tem instigado a toxicologia social na busca de respostas, com as quais se possa caracterizar, analisar e gerenciar esses riscos. Drogas de grande consumo no Brasil são a anfetamina, cocaína e Cannabis sativa. Esta tese desenvolveu uma nova metodologia para detectar e quantificar anfetamina, cocaína e tetrahidrocanabinol em sangue total, com uso de microextração em fase líquida via fibra de polipropileno (HF-LPME), seguida de cromatografia gasosa acoplada a espectrometria de massa (GC-MS). Trata-se de uma técnica que apresenta vantagens sobre as tradicionais, uma vez que demanda quantidades menores de solvente orgânico, diminuindo riscos e custos de processo. Também propôs um estudo com a aplicação dos métodos em 69 amostras de sangue de vivos e de post mortem, as quais foram obtidas por convênio com a superintendência da polícia técnica científica de São Paulo (SPTC/SP). Os métodos desenvolvidos foram validados de acordo com diretrizes internacionais de interesse forense. Como resultado da validação, os métodos desenvolvidos se mostraram precisos e exatos para anfetamina e cocaína. O limite de detecção da cocaína foi de 5 ng . mL-1 e o limite de quantificação de 10 ng . mL-1. Quanto a anfetamina, os limites de detecção e de quantificação foram de 5 ng . mL-1. A técnica de HF-LPME não foi aplicável ao tetraidrocanabinol (Δ9-THC). Como resultado da análise das amostras, 40% delas apresentaram resultados positivos para cocaína. Desses positivos, 35% foram oriundos das matrizes de sangue de vivos e 64% oriundos de sangue post mortem. Nenhuma delas apresentou resultado quantificável para anfetamina. / It is currently estimated that more than 5% of the world\'s population has been doing recreational use of some kind of psychoactive substances and the legal right to such use is a recurring theme debated by contemporary society. Due to the risks associated with populations health and safety, the abusive use of these substances has been instigating by social toxicology to search for answers to characterize, analyze and manage these risks. Drugs of great consumption in Brazil are, amphetamine cocaine and marijuana. This thesis proposes to develop a new methodology to detect and quantify psychoactive drugs in whole blood with the use of liquid phase microextraction by polypropylene fiber (HFLPME), followed by gas chromatography coupled to mass spectrometry (GC-MS). It is a technique that presents advantages compared with traditional ones, because of the smaller amounts demands of organic solvent, reducing risks and process costs. It also proposes a study with 69 blood samples taken from living persons and post mortem blood samples, which were obtained by agreement with the Superintendency of São Paulo Scientific Technical Police (SPTC / SP). The methods developed were validated according to international guidelines of forensic interest. As a result of the validation, the methods developed were precise and accurate for amphetamine and cocaine. The limit of cocaine detection was 5 ng . mL-1 and the limit of quantification was 10 ng . mL-1. As for amphetamine, the limits of detection and quantification were 5 ng . mL-1. The HF-LPME technique was not applicable to tetrahydrocannabinol (Δ9-THC). As a result of the sample analysis, 40% of them presented positive results for cocaine. Of these, 35% were from blood samples taken from living persons and 64% from the post mortem blood samples. None of the samples presented quantifiable results for amphetamine.

Os antidepressivos pertencem a uma importante classe de medicamentos investigados na toxicologia forense. Em casos de amostras provenientes de cadáveres, o intervalo entre o óbito e a obtenção da espécie biológica pode proporcionar a redistribuição postmortem destes fármacos. Com o objetivo de elucidar esse fenômeno, métodos analíticos foram desenvolvidos e aplicados utilizando sangue total (ST), humor vítreo (HV) e fígado. Para as amostras de ST e HV, o método de extração escolhido e validado foi a microextração em fase líquida (LPME) trifásica. Fibras ocas constituídas de polipropileno, com a extensão de 8 cm cada, foram tratadas com o solvente orgânico dodecano (fase orgânica), resultando em um membrana com permeabilidade seletiva. No lúmen destas fibras, adicionou-se ácido fórmico 0,1 mol/L (fase aceptora). Em frasco de fundo chato com 5 mL de capacidade, pipetou-se 3,5 mL de NaOH 0,1 mol/L (fase doadora) e 0,5 mL de ST ou HV. Ao término da extração, as amostras foram introduzidas no GC-MS, sem a necessidade de reações de derivatização. O estudo com ST contemplou os antidepressivos amitriptilina (AMI), nortriptilina (NTR), imipramina (IMI), desipramine (DES), clomipramina (CLO), desmetilclomipramina (DMC), fluoxetina (FLU) e norfluoxetina (NFL). Os limites de quantificação para estas substâncias ficaram inferiores aos níveis terapêuticos (20 ng/mL). As médias dos coeficientes de variação intradia e interdia foram, respectivamente, de 9,7 e 9,8%. As curvas de calibração apresentaram linearidade entre as concentrações de 20 até 1200 ng/mL. A validação do parâmetro integridade da diluição assegurou a mensuração de quantidades superiores ao limite apresentado na curva de calibração. O método foi aplicado em sete amostras reais postmortem e em apenas um caso foi observada uma diferença significativa (300%) entre os valores quantificados no ST periférico e central. Os antidepressivos tricíclicos AMI, NTR, IMI e DES foram avaliados no HV e o efeito matriz foi detectado para os dois últimos analitos. O método foi otimizado e validado utilizando solução salina adicionada de AMI e NTR. O limite de detecção igual a 5 ng/mL, foi obtido com a redução da voltagem da fonte de íons do espectrômetro de massa para 50 eV. Coeficientes de variação foram inferiores a 15%. Os procedimentos validados foram aplicados em seis amostras reais de HV. A relação encontrada entre os valores obtidos no ST periférico e HV foi de aproximadamente 0,1. A extração acelerada por solvente (ASE) e, posteriormente, a extração em fase sólida (SPE) foram as técnicas de separação dos analitos da matriz fígado. Ao término das citadas extrações, os antidepressivos foram analisados no GC-MS. Para esta matriz sólida, são necessários mais estudos, pois os valores encontrados nos ensaios analíticos estão em desacordo com as diretrizes utilizadas na validação dos métodos. / Antidepressants belong to an important class of drugs investigated in forensic toxicology. In cases of samples from corpses, the interval between death and obtaining the biological specimens can provide the postmortem redistribution of these drugs. Aiming to elucidate this phenomenon, analytical methods were developed and applied using whole blood (WB), vitreous humor (VH) and liver. For samples of WB and HV, the extraction method chosen and validated was the three-phase liquid phase microextraction (LPME). Hollow fibers consist of polypropylene, with a length of 8 cm each were treated with dodecane organic solvent (organic phase) resulting in a membrane with selective permeability. Into the lumen of these fibers was added formic acid 0.1 mol/ L (acceptor phase). In the vial containing 3.5 mL of NaOH 0.1 mol / L (donor phase) was spiked 0.5 ml of biological fluids (WB or VH). Subsequently, the samples were injected in GC-MS without derivatization reactions. The study of the ST included antidepressants amitriptyline (AMI), nortriptyline (NTR), imipramine (IMI), desipramine (DES), clomipramine (CLO), desmethylclomipramine (DMC), fluoxetine (FLU) and norfluoxetine (NFL). The quantification limits for these substances were below the therapeutic levels (20 ng / ml). The mean coefficients of variation and separate intradays were respectively 9.7 and 9.8%. The calibration curves showed linearity between concentrations of 20 to 1200 ng / mL. The validation of the integrity of the dilution parameter assured measurement higher than the limit shown in the calibration curve quantities. The method was applied to seven real postmortem samples and in one case a significant difference (300%) between the measured values in the peripheral and central ST was observed. The tricyclic antidepressants AMI, NTR, IMI and DES were evaluated in VH and the matrix effect was detected in the last two analytes. The method was optimized and validated using saline spiked AMI and NTR. The limit of detection (5 ng/ml) was obtained by reducing the voltage of the ion source of the mass spectrometer 50 eV. Coefficients of variation were below 15%. The procedures were validated in six real samples of HV. The relationship found between the values obtained in the peripheral ST and HV was approximately 0.1. Accelerated solvent extraction (ASE) and subsequently the solid phase extraction (SPE) were the techniques of separation of analytes liver matrix. At the end of the cited extractions, antidepressants were analyzed in GC-MS. To this solid tissue, further studies are needed, because the values found in the analytical tests were not in accordance with the guidelines used in the validation of the methods.

Bioanalytical chemistry is a challenging field, often involving complex samples, such as blood, plasma, serum or urine. In many applications, sample cleanup is the most demanding and time-consuming step. In the work underlying this thesis a novel dynamic miniature extractor, known as a hollow-fibre liquid-phase microextractor (HF-LPME), was designed, evaluated and studied closely when used to clean plasma samples. Aqueous-organic-aqueous liquid extraction, in which the organic liquid is immobilised in a porous polypropylene membrane, was the principle upon which the extractor was based, and this is discussed in all the papers associated with this thesis. This type of extraction is known as supported-liquid membrane extraction (SLM). The aim of this work was the development of a dynamic system for SLM. It was essential that the system could handle small sample volumes and had the potential for hyphenations and on-line connections to, for instance, LC/electrospray-MS. The design of a miniaturised HF-LPME device is presented in Paper I. The extraction method was developed for some weakly acidic pesticides and these were also used for evaluation. In the work described in Paper II, the method was optimised on the basis of an experimental design using spiked human plasma samples. Paper III presents a detailed study of the mass-transfer over the liquid membrane. The diffusion through the membrane pores was illustrated by a computer-simulation. Not surprisingly, the more lipophilic, the greater the retention of the compounds, as a result of dispersive forces. The main focus of the work described in Paper IV was to make the HF/LPME system more versatile and user-friendly; therefore, the extractor was automated by hyphenation to a SIA system. / At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 4: Manuscript.