Analytical Communications, January 1996, Vol33 ( I 9-20) 19 Determination of Poly(ethy1ene Glycol)-400 in Urine by Fourier Transform-Infrared Spectrometry Lale Ersoy, Sedef Atmaca, Serap Saklik and Sedat Imre Faculty of Pharmacy, Department of Analytical Chemistry, University of Istanbul, 34452 Beyazit, Istanbul, Turkey A simple and rapid method for the analysis of poly(ethy1ene glycol)-400 (PEG-400) (used in the study of intestinal physiology) in human urine is described using Fourier transform-infrared spectrometry.The quantitative measurements were carried out by using the absorbance value of the C-0-C band (wavenumber 1100 cm-1). The lower limit of determination of PEG-400 in human urine was 0.8 mg cm-3. Analytical recovery of PEG-400 added to urine controls was 99.6%. The results of the analyses of one urine sample were compared with those obtained by using high-performance liquid chromatography.Limitations and advantages of the two systems are discussed. Poly(ethy1ene glyco1)s (PEGs) have been widely used diagnos- tically to determine intestinal permeability1 as they are not metabolized by colonic bacteria and their toxicity is low.* PEGs and their derivatives are also extensively used in various products including drugs, foods, cosmetics and detergents.Several gas chromatographic (GC) and high-performance liquid chromatographic (HPLC) methods have been developed to determine PEGs in biological fluids, especially in urine. Liquid chromatographic analyses of PEG were performed without derivatization by using a refractive index detector34 or after derivatization by using an ultraviolet (UV) detector.'^^ Gas chromatographic methods also require derivatization of PEG prior to anal~sis.9~10 Kumar has developed a Fourier transform- infrared spectrometric (FT-IR) method for the determination of PEG-400 in high density polyethylene resin. The sample was directly melt-pressed into very thin plaques and the integrated absorbance (1000-1 170 cm-1) per unit thickness was measured.1 1 In this paper an FT-IR method for estimating urinary concentration of PEG-400 is described. The results of one urine sample analysed by both FT-IR and HPLC were compared statistically. Experimental Materials and Reagents PEG-400 was purchased from Sigma (St. Louis, MO, USA). Other chemicals and solvents were obtained from Merck (Darmstadt, Germany).All reagents used were of analytical- reagent grade. Apparatus For IR measurements a Perkin-Elmer 1600 FT-IR spectrometer with a lithium tantalate detector was used. The spectra were recorded by using fixed pathlength (0.1 mm) NaCl liquid sampling cells. For each determination, 64 scans were added at a 4 cm-1 resolution.For HPLC, a model 6000A solvent delivery system, a U6K universal injector and a Model 440 absorbance detector (Waters, Milford, MA, USA) at 280 nm was used. The detector was connected to a strip-chart recorder (Linear 355). The column used was a 10 pm Bondpak C18 300 mm X 3.9 mm (Waters). The mobile phase was methanol-water (95 + 5 ) at a flow rate of 0.8 cm3 min-I. Stock and Standard Solutions PEG-400 (500 mg) was accurately weighed into a 10 cm3 calibrated flask and made up to the mark with trichloromethane. A series of standard solutions containing 1-7 mg cm-3 of PEG- 400 were prepared by pipetting 0.2-1.4 cm3 of stock solution into a 10 cm3 calibrated flask and adding trichloromethane.The IR absorbance spectra of these solutions were obtained to establish a calibration graph.For recovery studies, urine samples containing known amounts of PEG-400 were also prepared. For this purpose, appropriate volumes of standard solutions were transferred into test-tubes. After the evaporation of the solvent, 2 cm3 of drug-free urine was added. To obtain a calibration graph for the HPLC method, 1.2-6 cm3 of PEG-400 solution (1 mg cm-3) were evaporated and derivatized as described.Extraction PEG-400 was extracted from urine samples using a slightly modified previously reported method.12 After 2 g of ammonium sulfate was added into 2 cm3 of urine sample and mixed for 20 s, PEG-400 was extracted with 5 cm3 of dichloromethane for 2 min on a vortex mixer. The mixture was then centrifuged at 3000g for 3 min. After the aqueous phase was discarded and the organic phase was dried over anhydrous Na2S04, 3 cm3 of dichloromethane phase was transferred into another tube. The solvent was evaporated under nitrogen at 50 "C.The residue was then either dissolved in 1 cm3 trichloromethane for FT-IR analysis or derivatized with benzoyl chloride for HPLC analysis. FT-IR Trichloromethane and solutions containing PEG-400 (standard solutions or extraction residue in trichloromethane) were injected into the NaCl cell, respectively.After the transmittance spectrum was recorded and converted into absorbance units, the chloroform spectrum was automatically subtracted from that of the PEG-400 solution. The absorbance value of the analytical band at 1100 cm-1 was recorded for quantitative measure- ments.HPLC For comparison, Kinahan and Smith's HPLC method7 was used. PEG-400 in urine samples was analysed under the20 Analytical Communications, January 1996, Vol33 modified chromatographic conditions after esterification. PEG- 400 extracted from urine was esterified with benzoyl chloride in pyridine as previously described? After extraction with di- chloromethane, the evaporation residue of the organic phase dried over anhydrous Na2S04 was dissolved in 1 cm3 of mobile phase and 10 mm3 of this solution was injected into the column. The peak heights at 4.5 min were measured to determine the PEG-400 concentration in the samples.Application of the Method Following oral administration of 5.6 g of PEG-400 in 200 cm3 of water, a urine sample was collected after 6 h.The specimen was mixed thoroughly and its total volume measured. About a 50 cm3 aliquot of this was stored at -20 "C until analysis. Results and Discussion Urine samples were analysed by FT-IR spectrometry after extraction to assay PEG-400 which was consumed to determine g 0.0000 _I 1200 1100 1000 1200 1100 1000 Wavenumberkm-' Fig. 1 (A) FT-IR original absorbance spectra and (B) difference absorbance spectra of PEG-400 standard solutions (a, 1.0; b, 2.0; c, 3.0; d, 4.0; e, 5.0; f, 6.0 and g, 7.0 mg ~ m - ~ ) Table 1 Analysis of PEG-400 in urine (n = 3) Amount Amount found Mean added/ mean/ recovery Sample no.mg ~ m - ~ mg cm-3 sr (%) (%I 1 1.503 1.502 2.92 99.9 2 3.007 2.997 2.16 99.7 3 4.5 10 4.47 1 1.89 99.1 Table 2 Determination of PEG-400 in urine Statistical value FT-IR HPLC X 2.92 2.96 S 0.08 1 0.077 s r (%I 2.77 2.60 n 5 5 t-test of significance F-test of significance t = 0.80 F = 1.11 t = 2.31 (p = 0.05) F = 6.39 (p = 0.05) the intestinal permeability.Since PEG-400 shows an intense C- 0-C ether linkage band at 1100 cm-1,11 the absorbance value of this band was used for quantitative measurements. The absorbance spectra of PEG-400 solutions are shown in Fig.1(A). To measure PEG-400 at 1100 cm-1 is impossible owing to the strong absorption of trichloromethane at this wav- enumber. In order to resolve the relatively weaker C-0-C absorption band from the much stronger trichloromethane band, the spectral subtraction approach was used. The trichloro- methane subtracted spectra of the standard solution are shown in Fig.l(B). The relationship between absorbance units of the standard solutions at 1100 cm-1 and concentrations was linear for the range 1-7 mg cm-3. The regression equation was A = 0.0214~ -0.0021 (r = 0.9993). The limit of determination was 0.8 mg for 1 cm3 of urine sample. The recovery of PEG-400 from urine was determined by analysing the added urine samples at three different concentrations.The average recovery was 99.6% (Table 1) i.e., it can be extracted quantitatively from urine and there is no interference in urine. When using HPLC, the concentration range of 0.4-2.0 mg cm-3 was studied. In this range the relationship between concentration of PEG-400 and peak height was linear, A = 1.065~ - 0.04 (r = 0.9995). The determination limit of this method was 0.02 mg of PEG-400 for 1 cm3 of urine sample.Ten urine samples collected after 6 h were analysed by both methods. Average percentage urinary excretions of the administered dose of 5.6 g were 26.2 and 24.9% using FT-IR and HPLC, respectively. These results were found to be related ( r = 0.9965). Furthermore, one unknown urine sample was analysed by both FT-IR and HPLC to compare the results statistically.There is no significant difference between mean values and standard deviations (Table 2). Although the limit of determi- nation by FT-IR is higher than that of HPLC, it is acceptable for medicinal research studies. However, the determination of PEG-400 by FT-IR without any derivatization after extraction is faster and simpler.It is also possible to analyse urine samples directly without extraction. This would be particularly useful in routine analysis. References 1 2 3 4 5 6 7 8 9 10 11 12 Chadwick, V. S., Phillips, S. F., and Hofmann, A. F. , Gastro- enterology, 1977, 73, 241. F.D.A. Drug Bull., 1982, 12, 25. Young, G. 0. , Ruttenberg, D., and Wright, J. P., Clin. Chem. (Winston-Salem, N.C.), 1990, 36, 1800. Trathnigg, B, Thamer, D., Yan, X., and Kinugasa, S., J. Liq. Chromatogr., 1993, 16, 2439. Ryan, C. M., Yarmush, M. L., and Tompkins, R. G., J. Pharm. Sci., 1992, 81, 350. Delahunty, T., and Hollander, D., Clin. Chem. (Winston-Salem, N.C.), 1986, 32, 351. Kinahan, I. M., and Smyth, M. R., J. Chromatogr., 1991, 565, 297. Murphy, R., Selden, A. C., Fisher, M., Fagan, E. A., and Chadwick, V. S. J. Chromatogr., 1981, 211, 160. Bouska, J. B., and Phillips, S. F., J. Chromatogr., 1980, 183, 72. Sivakumaran, T., Jenkins, R. T., Walker, W. H. C., and Goodacre, R. L., Clin. Chem. (Winston-Salem, N.C.), 1982, 28, 2452. Kumar, T., Analyst, 1990, 115, 1597. Schwertner, H. A., Patterson, W. R.,Cissik, J. H., and Wilson, K. W., J. Chromatogr., 1992, 578, 297. Paper 510621 40 Received September 20, I995 Accepted November 17, I995