328 Analyst, April, 1979, Vol. 104, $9. 328-333 Oxidative Determination of Dextromoramide (Palfium) in Body Fluids B. Caddy and R. ldowu Forensic Science Unit, Department of Pharmaceutical Chemistry, School of Pharmaceutical Sciences, University of StrathcZyde, GZasgow, G1 1XW The oxidation of dextromoramide to benzophenone with alkaline potassium permanganate and measurement of its ultraviolet absorbance is advocated for the determination of this drug in urine and serum over the concentration range 5-40 pg ml-l. Keywords : Dextrornoramide determination ; urine ; serum ; plasma ; ultraviolet spectrop hotometry Although the drug dextromoramide (I) has been used for some years as an analgesic, there is a paucity of information on methods for its assay when present in biological fluids.This gap has caused problems in drug treatment centres1 and in a forensic2 context where it has been necessary to identify and deterrnine the drug. Interpretation of any results obtained in these circumstances would also be impossible in view of the lack of information available on the urine and blood levels to be expected after normal therapeutic use. This paper describes an oxidative procedure that may be applied to the determination of this drug and/or metabolites containing the diphenylmethyl moiety, following their extraction from an alkaline solution. Reagents n WN - 0 CH;! - Q. It C H - C - C- I I Experimental r L N All reagents were of analytical-reagent grade unless otherwise stated. Instrumentation photometer, using a silica cell with a path length of 2 cm.All ultraviolet spectra were recorded on a Cecil CE505, double-beam ultraviolet spectro- Preparation of Calibration Graphs for Dextromoramide in Aqueous Solution A stock solution containing the equivalent of 5OOpgml-l of dextromoramide base was prepared by dissolving 69.1 mg of dextromoramide tartrate in 100 ml of water. A 20-ml volume of this solution was diluted to 100 ml in order to give a working solution containing 100 pg ml-l of drug base. Aliquots (0.05-0.5 ml) of this working solution were taken and an oxidising solution, prepared by mixing 10 ml of 4% m/V potassium permanganate with 15 ml of 9 M sodium hydroxide solution, was added, followed by 5ml of hexane. The mixture was refluxed for 40 min in a water-bath thermostatically controlled at 70-80 "C.After cooling in air for 15 min, the condenser was rinsed with water to wash any oxidation product into the flask and the mixture transferred into a glass-stoppered test-tube. An aliquot of the hexane layer was removed and its absorbance over the range 220-280nmCADDY AND IDOWU 329 recorded using hexane, which had been separately refluxed with the oxidising mixture, as reference. This procedure was repeated six times for statistical evaluation. The above oxidations with alkaline permanganate were repeated using various strengths of sodium hydroxide solution in the range 3-14 M, the last concentration being that described by Caddy et aL3 for the oxidation of methadone. Preparation of Calibration Graphs for Dextromoramide in Urine Standard solutions of the drug were prepared as detailed above, using drug-free urine in place of water.Aliquots (1-10 ml) of the urine solution (containing 5-50 pg of drug base) were pipetted into 100-ml glass-stoppered test-tubes and solid sodium carbonate and sodium hydrogen carbonate (1 + 1 m/m) added until a saturated solution was obtained. To the alkaline urine an amount of 2,2,4-trimethylpentane - pentan-1-01 (20 + 1 V/V) was added such that the ratio of the volume of urine sample to that of the extracting solvent was not less than 1 : 4. The tubes were shaken for 15min using a tilt shaker and set aside until the phases separated. The extract was evaporated to dryness under reduced pressure and the residue subjected to the oxidation procedure detailed above. The absorbance of the hexane layer was recorded over the range 220-280 nm using hexane from a similarly treated, drug-free urine sample as reference.This procedure was repeated three times for statistical evaluation. Preparation of Calibration Graphs for Dextromoramide in Serum A 2-ml volume of serum was taken in a glass-stoppered test-tube and 0.1-0.8 ml of an aqueous solution containing the equivalent of 100 pg ml-1 of dextromoramide base was added followed by 5 ml of 1 M sodium hydroxide solution and 10 ml of 2,2,4-trimethyl- pentane (reagent grade). Any emulsion formed was completely cleared by adding 5 ml of 2,2,4-trimethylpentane - pentan-l- 01 (20 + 1 V/V). The organic layer was separated and evaporated to dryness under reduced pressure and the residue oxidised with alkaline permanganate as detailed for dextro- moramide in aqueous solutions.Following oxidation, the absorbance of the hexane layer was recorded over the range 220-280 nm using a hexane extract from drug-free serum that had been similarly treated, as the reference solution. The procedure was repeated nine times for statistical evaluation. The mixture was shaken for 15 min using a tilt shaker. Oxidation of Dextromoramide with Other Oxidising Agents Solutions of dextromoramide base containing 20 pg of the drug were oxidised with aqueous acidic dichromate s~lution,~ non-aqueous acidic (glacial acetic acid - concentrated sulphuric acid) chromium(V1) oxide solution5 and acidic cerium(1V) sulphate solution,6 the last procedure being modified by replacing hydrochloric acid with sulphuric acid.Dextro- moramide (20 pg) was oxidised with barium peroxide in 66% rn/V sulphuric acid and 9 M sodium hydroxide solution as described by Wallace et a1.' Drug Regime and Biological Sampling stered to a patient. intervals over a period of 24 h and one sample at 48 h. the alkaline permanganate oxidation procedure detailed above. A single oral therapeutic dose of Palfium (dextromoramide tartrate, 5mg) was admini- Serum (10 ml) and urine (25 ml) samples were obtained at hourly These samples were analysed by Results and Discussion Oxidation of Dextromoramide with Alkaline Permanganate Caddy et aL3 reported that a concentration of 1 4 ~ sodium hydroxide solution is the optimum for the oxidation of dipipanone and methadone with alkaline permanganate.As dextromoramide (I) is structurally similar to both dipipanone (11) and methadone (111), it might be expected that the same conditions would be applicable to its oxidation to benzophenone with alkaline permanganate. This was found to be so, but reproducible quantitative results were not given by this procedure. However, a problem associated with330 - H - l H 3 CADDY AND IDOWU: OXIDATIVE DETERMINATION OF Analyst, vd. 104 Q CH2 -C - 0 C - CH2 - CH3 II H3C\N / H3C 8 - CH2 - CH3 II Ill the use of sodium hydroxide solution at this concentration is that the decomposition of alkaline permanganate to manganate occurs with great facility. This can be explained in terms of the reversible reaction MnO, + HO- + Mn0,2- + HO' proposed by Symons*s@ and by Jezowska-Trezebiatowska et aL1O as the initiating reaction in a multi-step mechanism proposed for the decomposition of alkaline permanganate to manganate and oxygen.An increase in concentration of sodium hydroxide will therefore facilitate the decomposition of permanganate. When using 14 M sodium hydroxide solution, it was observed that the oxidising mixture turned green (indicating the presence of the manganate species) before the oxidation of dextromoramide was completed. The use of 9~ sodium hydroxide solution and an increase in the ratio of permanganate to alkali solution from 1 + 3 (V/V) to 2 + 3 (VlV) was found to be optimum for the oxida- tion of dextromoramide (Fig. 1). This concentration of reagents also ensured that there was an excess of permanganate in the mixture at the end of the oxidation procedure.Further, it was observed that the hexane layer from mixtures that were green owing to the presence of manganate had an unusually high absorbance at 247 nm compared with hexane from mixtures that still had an excess of permanganate after the oxidation. 0.20 - o) 0.15- C m G SI a 0.10- a 0.05- I I 220 240 250 260 270 Wavelength/nrn Fig. 1. Typical spectra of the oxidation product in hexane obtained from 30 pg of dextromoramide originally present in: A, aqueous solution; B, urine; and C, serum; together with D, serum blank; and E, urine blank.April, 1979 DEXTROMORAMIDE (PALFIUM) IN BODY FLUIDS 331 The probable explanation for this observation is that the permanganate was reduced by the organic drug molecule to give the hypomanganate (Mn043-), which in the strongly alkaline medium disproportionated to manganate and manganese dioxide : A fine suspension of manganese dioxide in the hexane layer scattered the light during the ultraviolet measurements, thereby giving rise to the apparent high values of absorbance observed.In the presence of excess of permanganate, any hypomanganate formed was instantane- ously oxidised and the formation of manganese dioxide prevented. Preparation of Calibration Graphs for Dextromoramide in Aqueous Solution and in Urine The choice of extraction solvent is dictated by the need to avoid those solvents (e.g., chloroform) which, even in trace amounts, might interfere by reacting with the oxidant, thereby lowering the yield of benzophenone from the drug.The most effective solvent for this purpose was found to be heptane or 2,2,4-trimethylpentane containing a small amount of pentan-1-01, 2,2,4-trimethylpentane being preferred for urine as heptane gave high blank values. At the solvent evaporation stage it is also most important to remove the last traces of pentan-1-01. The use of a solid mixture of sodium carbonate and sodium hydrogen carbonate to adjust the pH of the urine was found to be advantageous in avoiding emulsification during the extraction procedure. Typical ultraviolet spectra of the oxidation product, which also show appropriate blanks, read against hexane are given in Fig. 1. Statistical evaluation of the graphs obtained from both aqueous (Table I) and urinary (Table 11) samples by comparison of their variances (F = 0.004) shows there is no signifi- cant difference between the two graphs and therefore the extraction of the drug from urine is complete. This finding is consistent with simple weighing experiments using 50-mg samples, which show an extracting efficiency of 97%.A value of 114y0 was obtained from the ratio of the slopes of the two regression lines but it is not possible to calculate the standard deviation of this value. Calibration graphs of absorbance of the hexane layer at 247 nm against concentration of the drug in micrograms per millilitre in water have good linearity and reproducibility over the range 5-50 pg ml-l and can be used in the determination of the drug in urine. Preparation of Calibration Graphs for Dextromoramide in Serum The choice of extraction solvent was determined by the same considerations as stated for TABLE I RELATIONSHIP BETWEEN THE CONCENTRATION OF AQUEOUS SOLUTIONS OF DEXTROMORAMIDE AND THE ABSORBANCE OF ITS ALKALINE PERMANGANATE OXIDATION PRODUCT I N HEXANE Amount of dextromoramide base/pg* 5 10 20 30 40 50 Absorbance of hexane Standard layer a t 247 nmt deviation 0.058 0.013 0.087 0.013 0.176 0.037 0.230 0.032 0.287 0.026 0.373 0.014 Correlation coefficient = 0.975 & 0.015 (95% confidence limits).Regression equation: y = 0.024 + O.O07x, where y = absorbance of hexane layer a t 247 nm and x ( p g ) = amount of dextromoramide base oxidised. Standard error of y = 0.004. * Actual amount of drug oxidised. t All values are the means of six determinations.332 CADDY AND IDOWU : OXIDATIVE DETERMINATION OF TABLE I1 Analyst, VoZ.104 RELATIONSHIP BETWEEN THE CONCENTRATION OF DEXTROMORAMIDE IN URINE AND THE ABSORBANCE OF ITS OXIDATION PRODUCT I N HEXANE Amount of dextromoramide base/pg* 5 10 20 30 40 50 Absorbance of hexane layer a t 247 nmt 0.033 0.094 0.173 0.264 0.315 0.414 Standard deviation 0.032 0.034 0.003 0.048 0.019 0.010 Correlation coefficient = 0.965 f 0.025 (95% confidence limits). Regression equation: y = 0.002 + 0 . 0 0 8 ~ ; x and y as stated in Table I. Standard error of y = 0.009. * Actual amount of drug oxidised. t All values are the means of three determinations. the preparation of the calibration graph for dextromoramide in urine. The 2,2,4-trimethyl- pentane - pentan-1-01 mixture used for the extraction of urine was unsuitable for use in the extraction of serum as it gave rise to high blank values in the final hexane solution and recourse was made to 2,2,4-trimethylpentane alone.The recovery of dextromoramide from serum using this solvent was found to be 36% as measured from the ratio of the slopes of the regression lines. This value is lower than would be expected from simple weighing experiments using 50-mg samples of drug extracted from plasma (68%) and may be either a reflection of the efficiency of oxidation for this type of sample or more simply because of the reduction in solvent polarity that results from the omission of the pentan-1-01. How- ever, the extract was clean and results were reproducible. The analytical results are given in Table 111. A typical ultraviolet spectrum of the oxidation product produced from the oxidation of an extract of serum, together with an appropriate blank, read against hexane, is shown in Fig.1. TABLE I11 RELATIONSHIP BETWEEN THE CONCENTRATION OF DEXTROMORAMIDE IN SERUM AND THE ABSORBANCE OF ITS OXIDATION PRODUCT I N HEXANE Concentration of dextromoramide base*/ Absorbance of hexane Standard d-' layer a t 247 nmt deviation 5 10 20 30 40 0.022 0.049 0.094 0.146 0.208 0.0062 0.0103 0.0192 0.041 0 0.025 Correlation coefficient = 0.94 -& 0.03 (95% confidence limits). Regression equation: y = 0.005 1 + 0.0026%; x and y as in Table I. Standard error of y = 0.0035. * Original concentration of drug in a 2-ml sample of serum. t All values are means of nine determinations. Serum and Plasma Levels Found in a Patient Administered a Single Therapeutic Dose of Palfium The methods described above have been applied successfully to the determination of dextromoramide and/or its metabolites containing the diphenylmethyl moiety present in the urine and serum of a patient administered a single oral therapeutic dose of the drug.The levels found are given in Table IV. The time of maximum serum concentration is 4 h and that for urine 5 h. Interestingly the concentration in the serum is still higher thanA@+?, 1979 DEXTROMORAMIDE (PALFIUM) IN BODY FLUIDS TABLE IV URINE AND SERUM LEVELS OF DEXTROMORAMIDE FOUND OVER A 48-h PERIOD FOLLOWING THE INGESTION OF A SINGLE THERAPEUTIC DOSE The results also include any metabolites containing the diphenylmethyl group.333 Time/h . . .. .. . . 1 2 3 4 5 6 7 8 24 48 Serum concentration/ig in 10 ml . . 21 23 28.5 29.5 29 24.5 22.8 21 18.5 18 Urine concentration/pg in 10 ml . . 13.5 14 14.5 14.9 20 8.5 4.1 5 6.5 8 that in the urine even after 48 h. paper. It is hoped to present more excretion data in a further Oxidation of Dextromoramide with Other Oxidising Agents The use of reagents other than alkaline permanganate for the oxidation of drugs containing either the diphenylmethyl grouping or an aromatic ring with a suitable side-chain have been reported by several workers.4-7 As many of these compounds, especially in a forensic context, may be present in body fluids together with dextromoramide, it is necessary to establish if this latter drug is oxidised to benzophenone under the conditions used for the oxidative assay of these other drugs.Other reagents, such as acidified dichromateJ4 chromium(V1) oxide5 and acidified cerium(1V) sulphate,6 used for the oxidation of diphen- hydramine4s5 and dexamphetamine,6 respectively, and barium peroxideJ7 used for the oxida- tive assay of methadone, failed to oxidise dextromoramide. Although this failure to oxidise dextromoramide (I) to benzophenone with acidic oxidising agents can be explained by the absence of the requisite structural features, it is difficult to account for the lack of formation of benzophenone when dextromoramide is oxidised with barium peroxide, especially in view of its structural similarity to methadone (111). As barium peroxide has rarely been employed as an oxidant in organic chemistry, the answer may lie in the mode of action of this reagent, which is at present unknown.Use can be made of the selectivity of barium peroxide as oxidant for methadone in an assay for this drug in admixture with dextromoramide. Using the conditions detailed above, a methadone - dextromoramide mixture was first oxidised with barium peroxide and then with alkaline permanganate. The difference between the absorbances of the benzo- phenone in the hexane solution was found to be a measure of the amount of dextromoramide present. Conclusion Oxidation of dextromoramide with alkaline permanganate and subsequent measurement of the absorbance of the benzophenone obtained provides a satisfactory method for the assay of the drug and/or its metabolites containing the diphenylmethyl moiety in biological fluids. Satisfactory calibration graphs were obtained for the drug over the concentration ranges reported. Other oxidation methods investigated were of no analytical value in the oxidative deter- mination of dextromoramide, as the different oxidising agents failed to convert dextro- moramide to benzophenone. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. References Berry, D. J., personal communication. Clatworthy, A. J., personal communication. Caddy, B., Fish, F., Tranter, J . , and Mullen, P. W., J. Forens. Sci. SOC., 1973, 13, 127. Caddy, B., Fish, F., and Tranter, J., Analyst, 1974, 99, 565. Vessman, J . , Hartwig, P., and Stromberg, S., Acta Pharm. Suec., 1970, 7, 373. Wallace, J . E., Biggs, J . D., and Ladd, S. L., Analyt. Chem., 1968, 40, 2207. Wallace, J . E., Hamilton, H. E., Payte, J. T., and Blum, K., J. Pharm. Scz., 1972, 61, 1397. Symons, M. C. R., J . Chem. SOC., 1953, 3956. Symons, M. C. R., J . Chem. SOC., 1954, 3676. Jezowska-Trezebiatowska, B., Nawojska, M., and Wronska, M., Bull. Acad. Pol. Sci., Cl. 111, 1953, Received ApriE 24th, 1978 Accepted October 19th, 1978 1, 311; Chem. Abstr., 1954, 48, 97952.