In the pursuit of quicker and easier procedures for labs that conduct routine analyses or quality control measures, one target area is extraction of the sample. It can be the most time-consuming operation of the overall method, as well as requiring significant periods of analyst intervention. Of the many improvements in extraction, solid-phase microextraction (SPME), has proven to be an efficient and popular method for liquid, gaseous and solid samples.

In conventional SPME, a specially coated fibre is held in a closed space above the sample or, in the case of gases and liquids, can be immersed directly in the sample. Volatile and semi-volatile compounds are adsorbed on the fibre, which is removed and inserted into the analytical instrument, often a gas or liquid chromatograph. The whole procedure is frequently conducted manually but it can be automated to reduce operator involvement and increase throughput.

A spin-off method which is especially suitable for biological fluids is in-tube SPME. Here, as the title implies, the fibre is replaced by capillary tubing, typically a section of a GC capillary column. The tube is coated with a stationary phase then positioned between the sample vials of an autosampler and an HPLC column. The sample is subjected to repeated draw/ejection cycles until sufficient analytes are deemed to be adsorbed on the tube coating. A desorption solvent is then swept through the tube to transfer the analytes to the chromatograph for analysis.

In fibre-based SPME, the analytes are desorbed during injection and this can lead to peak broadening if desorption is slow. In contrast, with in-tube SPME, desorption and injection are decoupled so that the analytes are completely desorbed prior to injection, eliminating peak broadening. However, a solvent peak is often seen at the beginning of the chromatogram.

One good illustration of in-tube SPME coupled with HPLC was provided recently by scientists at the University of Sao Paulo, Brazil, led by senior reporter Maria Eugenia Costa Queiroz. They developed a protocol for the simultaneous extraction and analysis of six nontricyclic antidepressant drugs in human plasma in order to monitor their therapeutic levels following administration: mirtazapine, citalopram, paroxetine, duloxetine, fluoxetine and sertraline.

For method optimisation, aqueous standard solutions of the drugs plus an internal standard (clomipramine) were employed, then for validation, blank plasma was spiked with the drugs. The extraction tube was a section of fused silica 80 cm x 250 µm i.d. coated with the popular GC phase OV-1701. The extraction parameters were varied for optimisation, extraction efficiencies being determined by HPLC with UV detection at 230 nm.

Equilibrium partition was achieved for the majority of the drugs with a draw/eject volume of 100 µL and a total of 15 draw/eject cycles. The draw-eject flow rate was set to 315 µL/min which was a compromise between the longer extraction times at lower rates and the formation of bubbles inside the capillary at higher rates. The waiting time between drawing and ejecting had no effect on sensitivity over the range 30-120 s. Sample pH was adjusted to 9.0 at which the drugs, having pK_a values of 7.1-9.9, existed mainly in the non-ionic form.

With these optimised conditions, the limits of quantification were 20-50 ng/mL (20-50 ppb), with coefficients of variation less than 10%. The calibration curves were linear at 50-500 ng/mL with good correlation coefficients.

The method was applied to plasma from elderly patients who had been under therapy for at least 2 weeks with duloxetine (Cymbalta®, 60 mg/d), fluoxetine (Prozac®, 20 mg/d) and sertraline (Zoloft®, 150 mg/d). Their plasma samples were treated with acetonitrile to precipitate out the proteins which could otherwise be absorbed in the extraction tube and reduce its efficiency. The tube was rinsed with water to remove any residual proteins and prevent them reaching the HPLC column.

The drugs were extracted and quantified with no endogenous interferences from other plasma components. Their concentrations were 406, 314 and 445 ng/mL for duloxetine, fluoxetine and sertraline, respectively, which are within the respective therapeutic levels for the individual drugs.

The good method performance, combined with the ease of automation that lessens analyst intervention and increases throughput, demonstrates the value of in-tube SPME coupled to HPLC for routine analyses.

Related links:

• Journal of Chromatography B 2008 (Article in Press): "In-tube solid-phase microextraction coupled to liquid chromatography (in-tube SPME/LC) analysis of nontricyclic antidepressants in human plasma"

Article by Steve Down

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.