| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Centre for Forensic Imaging, Institute of Forensic Medicine, University of
Bern, Bern, Switzerland.
2 Present address: Institute of Forensic Medicine, University of Lausanne, Rue
du Bugnon 21, 1005 Lausanne, Switzerland.
Received August 21, 2007;
accepted after revision November 12, 2007.
Performed in collaboration with the Federal Customs Administration (FCA),
Drugs Unit, Bern, Switzerland.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Studies were performed on three wine bottles containing cocaine solutions that were confiscated at the Swiss border. Reference values were obtained by scans of different sorts of commercially available wine and aqueous solutions of dissolved sugar. All bottles were scanned using MDCT, and data evaluation was performed by measuring the mean peak of Hounsfield units. To verify the method, simulated testing was performed.
RESULTS. Using measurements of the mean peak of Hounsfield units enables the detection of dissolved cocaine in wine bottles in a noninvasive and rapid fashion. Increasing opacity corresponds well with the concentration of dissolved cocaine. Simulated testing showed that it is possible to distinguish between cocaine-contaminated and uncontaminated wine bottles.
CONCLUSION. The described method is an efficacious screening method to detect cocaine-contaminated bottles that are hidden between untreated bottles in cargo. The noninvasive examination of cargo allows a questionable delivery to be tracked without arousing the suspicion of the smugglers.
Keywords: cocaine • CT • drug screening • drug smuggling • Virtopsy
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
However, there are many other ways to smuggle drugs—for instance, by hiding them inside luggage or cargo [6, 7]. Cocaine has been found concealed in various imaginative ways, as reports of cases of cocaine concealed in wax [8], baseball caps [9], wooden globes [10], beer cans [11], milk cans [12], or in cocaine-impregnated cloth [13] show. Methods to detect these hidden drugs at the border are important and must keep pace with the increasingly imaginative ideas of the smugglers [13–15].
Reported cases have shown the successful application of techniques such as fluoroscopy [6, 7], backscatter radiography [6, 7], conventional radiography [1, 2, 6, 7], CT [1, 2, 6, 7], and sonography [14] in detecting drug-filled foreign bodies inside humans, luggage, parcels, and cargo containers. But if a drug such as cocaine is dissolved in a liquid, it cannot be detected by these scanning methods (Fig. 1A, 1B). Several cases have shown that this kind of smuggling does exist, especially with cocaine [16]. To find a dissolved drug at the border, an immunologic test using a drug-test panel (e.g., RapidCHECK10 panel Multi Drug Professional Card, Craig Medical Distribution) must be performed. Such panels are the same as the clinical urine-stick tests and are based on the same principle as a pregnancy quick test. Thus, a control sample of the cargo must be opened to perform the test. Because it is not possible to open all boxes, and smugglers hide the drug-containing boxes between boxes containing legitimate goods, contaminated cargo can be overlooked. Investigators prefer not to arouse the smugglers' suspicion when checking the cargo.
|
|
To determine whether dissolved cocaine is detectable with MDCT, we performed MDCT of these bottles. We noticed that significant mean density differences could be measured between the bottles with and the ones without cocaine. On the basis of this observation we developed a rapid method to detect cocaine solutions inside bottles without opening them. To test our new screening method, we performed simulated testing using cocaine-spiked wine bottles hidden between normal bottles of red wine.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CT Study
Studies were performed on three wine bottles that were confiscated at the
Swiss border. The bottles were filled with a solution of water, citron acid,
red watercolor, and cocaine. The content of one of these bottles was decanted
into an open measuring cup before imaging.
Reference values were obtained scanning the following specimens: two bottles of commercial red wine (Portal de Serrandos Valencia, Tinto, 2005; Los Vascos, Cabernet Sauvignon, 2003), one bottle of unfiltered red wine (Sulin, Barbera del Monferrato, 2002), one bottle of red wine with known tartar (Tedeschi Valpolicella Classico, 2002), one bottle of dessert wine (Isola del Vento, Moscato di Pantelleria, 2002), two bottles of white wine (Veneto, Pinot Grigio, 2005; Aigle, 2004), one bottle of rosé wine (Dôle, Blanch de Fully, 2005), one bottle filled with 0.7 L of tap water, and two bottles filled with 30 and 90 g of sugar, respectively, dissolved in 0.7 L of tap water.
MDCT was performed with a Somatom Sensation 6 scanner (Siemens Medical Solutions). Scanning parameters were 0.5-mm detector collimation, 0.63-mm slice width, a reconstruction increment of 0.5 mm, and a B30 kernel. Data evaluation was performed on 2D and 3D reconstructions.
Two-dimensional evaluation—Cross-sectional images were assessed for density measurements. To minimize image noise and for more accurate measuring, new slices of 10-mm thickness were reconstructed with an increment of 5 mm. A slice approximately in the middle of the bottle was chosen to measure the X-ray attenuation. A circular region of interest (ROI) was then placed in the middle of the cross slice to measure the mean attenuation and the SD in Hounsfield units. The mean attenuation in Hounsfield units was used to detect cocaine contamination.
Three-dimensional evaluation—The data set was loaded in a volume-rendering program on the Siemens Leonardo workstation (InSpace or 3D). By adapting the window settings, the bottle could be made translucent to detect possible precipitation. Therefore, choosing a metal reconstruction was most suitable.
Chemical Analysis
Chemical analysis of the three confiscated bottles that were suspected of
containing cocaine was performed by column liquid chromatography using the
model 2695 separations module (Waters liquid chromatographic system equipped
with an automatic sampling system). A Waters 2996 Photodiode Array Detector
connected to a Compaq PC (Windows XP and Waters Empower Software version 5.0)
was used to monitor the eluent from the column at 220 nm, and full spectra
were recorded over the wavelength range 210–400 nm. The solvent was
delivered with a gradient from 100% 0.025 mol/L of triethylammonium phosphate
(TEAP) buffer to 30% 0.025 mol/L of TEAP buffer and 70% acetonitrile over 20
minutes onto a LiChrospher 60 RP-Select B 5 µm (Merck 50963) guard column
connected to a LiChroCart 125–4 LiChrospher 60 RP-Select B 5-µm
(Merck 50829) column. Limits of detection (LODs) and limits of quantitation
(LOQs) were 0.5% and 1.5%, respectively. The linearity of the calibration
curves was good, with r2 values of less than 0.999
(concentration range, 0.025–0.5 mg/L).
Simulated Testing
Twelve bottles of commercially available red wine (Sangiovese 2004) were
used for this study. Six of those bottles were spiked with cocaine in
different concentrations (10, 30, 50, 70, 90, and 120 g) by the chemists in
the following way:
Cocaine hydrochloride, pharmaceutical grade, purity > 99% (according to European Pharmacopoeia II), was purchased from Siegfried Ltd. All chemicals used were analytic grade or better, and all solvents were high-performance liquid chromatography (HPLC) grade (methanol, acetonitrile, ultrapure water, 1 mol/L of TEAP). To prepare the stock solution, 5 g of wine was diluted in 50 mL of methanol. The samples for the HPLC analysis were prepared by further dilutions with methanol of 1:10 and 1:20, respectively. The injection volume was 10 µL.
|
|
|
|
CT was performed of the whole box in an upright position using the parameters described previously. The imaging team was not allowed to open the carton. Data were analyzed on a cross-section image of the whole carton. In each bottle an ROI was set, and mean attenuation in Hounsfield units was used for comparison. Bottles were listed according to their mean density and were mapped, showing the assumed localization of the cocaine-containing bottles. These bottles were also listed according to the estimated cocaine concentration. Finally, the imaging results were compared with the solution scheme of the chemistry department.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Three-dimensional evaluation—Viewing the 3D model in different window settings gave a rapid overview as to whether precipitated material was in the bottle. The tested cocaine solutions from the three confiscated bottles showed precipitation in the 3D reconstruction (Fig. 2A) in the open measuring cup and in the two bottles. None of the tested commercial wine bottles showed precipitation in the 3D reconstruction.
Chemical Analysis
The three confiscated bottles contained in all 2,140 g of liquid with 13%
cocaine hydrochloride, or 12% cocaine base, respectively. This corresponds to
a total of 278 g of cocaine hydrochloride, or 257 g of cocaine base,
respectively.
Simulated Testing
Two-dimensional evaluation—The mean attenuation peaks on the
cross-section images varied from 31 to –4 H. Although six bottles showed
the mean peak between 0 and –4 H, one bottle displayed 2 H and five
bottles showed clearly positive attenuation peaks between 7 and 31 H
(Fig. 3). The six bottles with
attenuation peaks between 0 and –4 H were considered to be the six
bottles without cocaine. The positive attenuation peaks from 2 to 31 H were
thought to derive from the cocaine–wine solution, and the bottles were
listed according to the mean attenuation peak. Bottle 11 was therefore assumed
to contain the highest concentration of cocaine because its mean attenuation
was 31 H. Bottle 4 was considered to contain little cocaine because the peak
was marginally above 0 H.
To compare our results with the bottle location map from the chemistry department, we marked the bottles that were assumed to contain cocaine with a red star on the cross-section image. Comparison of the two carton maps showed 100% concordance (Fig. 4A, 4B). The estimation of the cocaine dose according to the mean attenuation showed that all bottles were listed in the correct order (Table 2) so that a semiquantitative analysis was possible.
|
|
|
Three-dimensional evaluation—Volume-rendering reconstruction of the MDCT scans did not show any precipitation in the bottles (Fig. 5B).
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
As the results show, the method is suitable for detecting cocaine as a dissolved substance in a liquid because cocaine shows X-ray attenuation. When a carton or a group of wine bottles contains the same wine, the bottles will have more or less the same mean attenuation on the cross-section images. The scan of our carton showed that the six uncontaminated bottles displayed a mean attenuation between –3.8 and 0.3 H. It is suspicious when the attenuation of some bottles differs much from the rest. To be absolutely certain, these bottles can be opened and tested using the immunologic method. The chance of detecting contaminated bottles is much higher than when arbitrarily chosen bottles are tested.
As our experiment shows, the method works better in detecting higher drug concentrations. Bottles containing 30 g or more of cocaine show significant differences, whereas the bottle with 10 g of cocaine is not as clearly suspicious, and the mean attenuation peak of 1.7 H was within the SD rate. However, we do not believe this poses a problem in daily practice because experience has shown that if cocaine is dissolved in a liquid, the concentration is higher—for example, 80 g—as in our described case.
Our study shows that it is not possible to give a critical mean attenuation value that makes a wine suspicious for containing dissolved substances. Although most commercially available wines show mean attenuation peaks between –18 and –10 H, some are beyond these values—for example, our tested dessert wine, which had a mean attenuation of 23.6 H, and the unfiltered Barbera, with a mean attenuation of –6 H. This fact shows that it is necessary to compare single bottles in one shipment with one another or to compare the bottles with a reference bottle that is sure to be free of the dissolved drug.
In our study we tested different kinds of wine with cocaine as a dissolved substance. Principally, the method can detect any dissolved substance in any liquid, because the dis solved substance leads to a change of X-ray attenuation in the liquid, which can be measured because of the altered mean attenuation peak. For example, we changed the X-ray attenuation of water from –3 to 14 H by dissolving sugar. With an increasing amount of the dissolved substance, an increasing mean attenuation peak will be reached. As the example of water and sugar shows, 30 g of sugar dissolved in water leads to a mean attenuation of 14 H, and 90 g of sugar produced a mean attenuation of 42 H.
The method we describe seems also to be applicable to other sorts of smuggling. As articles [10, 12, 16] show and as the Swiss FCA declares, cocaine and other drugs are also smuggled inside small sculptures and hollowed fruits or dissolved in other liquids and vessels. Using CT and measuring the mean opacity of the content, differences in hidden drugs can be detected without destroying the carrier. If the carrier is expensive, this aspect will be especially important because in cases of an incorrect suspicion the owner of the destroyed or opened object must be reimbursed.
To apply our method, collaboration between police or customs officials and a medical department is necessary. The scanning can be performed in every hospital that has a CT scanner, and forensic departments also use this radiologic tool.
If a forensic chemistry department must analyze a confiscated shipment that is suspected of containing dissolved cocaine or other drugs, the screening scan can be helpful to get a first overview to sort out suspicious vessels, so that the more expensive quantitative chemical analyses can be performed on selected items.
In conclusion, our study describes a rapid screening method for detecting dissolved substances having X-ray attenuation, such as sugar or cocaine, in liquids. Because of its noninvasive fashion and fast practicability, this method is suitable for the examination of large cargos or to confirm suspicions without changing the package.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
