Diagnostic odor recognition

Many diseases, toxic ingestions, and intoxications have characteristic odors. These odors may provide diagnostic clues that affect rapid treatment long before laboratory confirmation or clinical deterioration. Odor recognition skills, similar to auscultation and palpation skills, require teaching and practical exposure. Dr. Goldfrank and colleagues recognized the importance of teaching odor recognition to emergency service providers. They proposed the "sniffing bar" method for odor recognition training. OBJECTIVES: (1) To identify the recognition rates of medically important odors among emergency care providers. (2) To investigate the effectiveness of teaching odor recognition. Hypothesis: The recognition rates of medically important odors will increase after teaching exposure. METHODS: The study exposed emergency care providers to 11 tubes of odors. Identifications of each substance were recorded. After corrective feedback, subjects were re-tested on their ability to identify the odors. Analysis of odor recognition improvement after teaching was done via chi-square test. RESULTS: Improvement in identification after teaching was seen in all odors. However, the improvement was significant only in the lesscommon substances because their initial recognition was especially low. Significant changes may improve with a larger sample size. Subjects often confuse the odors of alcohol with acetone, and wintergreen with camphor. CONCLUSIONS: The recognition rates are higher for the more-common odors, and lower for the less-common odors. Teaching exposures to the less well-known odors are effective and can significantly improve the recognition rate of these substances. Because odor recognition may affect rapid diagnosis and treatment of certain medical emergencies such as toxic ingestion, future studies should investigate the correlation between odor recognition and the ability to identify corresponding medical emergencies.

Distribution of amitriptyline and nortriptyline in blood: role of alpha-1-glycoprotein.

To interpret blood levels of tricyclic antidepressants, we studied the distributions of amitriptyline and nortriptyline in human blood and explored their control by plasma factors. Each compound (300 ng/ml) was added to whole adult blood and to cord blood with decreased alpha-1-glycoprotein (AGP). Drugs (250 ng/ml) were also added to washed erythrocytes (RBCs) resuspended in autologous plasma or saline (hematocrit = 0.4) with or without AGP, albumin, or tris(2-butoxyethyl) phosphate (TBEP), used to displace AGP-bound drugs. Plasma AGP was determined in all adult blood donors (n = 17). With adult blood, plasma amitriptyline was 393 +/- 52 ng/ml, RBC amitriptyline was 184 +/- 33 ng/ml. Plasma and RBC nortriptyline were 199 +/- 28 and 288 +/- 39 ng/ml, respectively. With saline, cellular amitriptyline and nortriptyline were 81 +/- 10 and 88 +/- 6%, respectively. With plasma, cellular amitriptyline and nortriptyline were 25 +/- 8 and 49 +/- 10%, respectively. The corresponding cord blood values were 52 +/- 12 and 62 +/- 6%. Graded increments of AGP in saline reproduced the distribution pattern seen with increasing concentrations of plasma. Albumin did not influence drug distribution. TBEP markedly increased erythrocyte amitriptyline in adult but not in cord blood. Plasma AGP correlated positively (p = 0.031) with the RBC/plasma ratio of amitriptyline. Amitriptyline is predominantly distributed in plasma, nortriptyline in RBCs. This differential distribution is dose dependent and reflects the higher binding of amitriptyline to AGP when compared with nortriptyline. Interpretation of tricyclic antidepressant blood levels is clarified by obtaining assays from RBCs and plasma.

Red cell and plasma concentrations of fluoxetine and norfluoxetine.

To study the distribution of fluoxetine and norfluoxetine in blood compartments we determined their concentrations in red cells and plasma after the addition of 500 ng/ml of each compound to human blood in vitro. Red cell and plasma fluoxetine concentrations were 493 +/- 79 ng/ml and 454 +/- 53 ng/ml, respectively (P > 0.1). To assess the potential implications of this distribution on routine monitoring of these compounds in plasma, we determined fluoxetine and norfluoxetine concentrations in red cells and plasma in 6 patients receiving various doses of fluoxetine. While in 4 patients the concentrations of fluoxetine and norfluoxetine in red cells and plasma were comparable, 2 patients had higher concentrations of both compounds in red cells. Variations in the distribution of fluoxetine and norfluoxetine in blood compartments are relatively small. Plasma levels may reflect the drug concentration in whole blood more reliably for fluoxetine and norfluoxetine than for tricyclic antidepressants.