491
Clinica Chimica Acta, 69 (1976) 491-495 @ Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands
CCA 7836
A MODIFIED ULTRAVIOLET SPECTROPHOTOMETRIC METHOD FOR THE DETERMINATION OF THEOPHYLLINE IN SERUM IN THE PRESENCE OF BARBITURATES
JOHN VASILIADES
and TAMELA TURNER
Department of Pathology, 35233 (U.S.A.)
The
University
of Alabama
in Birmingham,
Birmingham,
Ala.
(Received December 22,1975)
Summary Ultraviolet spectrophotometry is the most commonly used technique for the determination of theophylline levels for therapeutic monitoring. Common interferences in most methods are barbiturates and xanthines. A modified method which eliminates interferences from barbiturates and most xanthines is presented. Theophylline is extracted from serum with chloroform/isopropanol at pH 7.4 back extracted into dilute hydrochloric acid and then the solution is made alkaline withlsodium hydroxide. Barbiturate interferences are eliminated. Interference from caffeine, uric acid, 7-( 2,3-dihydroxypropyl)theophylline, xanthine, and hypoxanthine are not observed. Theobromine, and the metabolite 3-methylxanthine interfere. Interference is not observed from quinidine, diazepam, salicylate, glutethimide, methyprylon, propranolol, methaqualone, dilantin and ethchlorvynol. Sulfanilamide, procainamide and chlordiazepoxide interfere.
Introduction Theophylline (1,3_dimethylxanthine) is a powerful bronchodilator which relaxes the smooth muscle fibers of the bronchi [ 11. This property constitutes the basis of therapy for a large number of individuals with obstructive lung disease. Therapeutic concentrations of theophylline range from 5 to 15 I.cg/ml [2, 31. Toxicity is usually observed above 20 pg/ml. A large variation in theophylline half life and rates of elimination of the drug among patients makes therapeutic monitoring to achieve and maintain a certain concentration necessary. Methods presently used for the determination of theophylline in serum are usually modifications of the original ultraviolet method involving a basic extraction of theophylline from the organic solvent [4] . Gas liquid chromatography (GLC) and high pressure liquid chromatography (HPLC) have been used
492
for the determination of theophylline [ 5,6,12] . However, a spectrophotometric method is more suitable for routine monitoring of acute asthmatic emergencies as well as chronic care because of its simplicity, A modified ultraviolet spectrophotometric method for serum theophylline is presented which eliminates many of the common interferences of previous methods. The procedure involves extraction of theophylline at pH 7.4 into chloroform/isopropanol followed by back extraction from the organic phase into dilute 1 M hydrochloric acid. The acidic solution is made basic with sodium hydroxide and the ultraviolet spectrum scanned. Interferences from barbiturates are eliminated as well as other drugs extracted at basic pH. Interferences from hypoxanthine, caffeine, and other xanthines are also eliminated. Materials and methods Instrumentation A Beckman ACTA IV spectrophotometer (Beckman Instruments, Fullerton, California) was used for these studies with quartz cuvettes (l-cm path length). The inh.um-d has a photometric accuracy of 0.003 absorbance units at 1.0 A. Reagents All reagents were AR grade. Chloroform/isopropanol(20 : 1, by volume). Hydrochloric acid 1 M. Sodium hydroxide 4.4 M. Hypoxanthine, 3_methylxanthine, xanthine, and 7-(2,3-dihydroxypropyl)theophylline were obtained from Adams Chemical Company, P.O. Box 236 Round Lake, Ill. Theophylline, and theobromine were obtained from Analabs Incorporated, 80 Republic Drive, North Haven, Conn. Theophylline stock standard: 1.25 mg/ml. Dissolve 125 mg of theophylline in 100 ml of distilled water. Theophylline working stock: 500 E.cg/ml. Dilute 20 ml of 1.25 mg/ml stock standard to 50 ml with water. Theophylline working standards: 10 and 20 pg/ml. Take 2 and 4 ml of working stock 500 lug/ml and dilute to 50 ml with water. Standards of 10 and 20 pug/ ml are prepared in serum daily prior to use by reconstituting lyophilized serum with the 10 and 20 (ug/ml theophylline aqueous standards. All other xanthine solutions were prepared by dissolving the pure substance in water. Procedure Place 3 ml of serum sample, serum standard and serum blank into 50 ml round bottomed centrifuge tubes. Add 30 ml of 20 : 1 chloroform/isopropanol mixture. Shake for 5 min at low speed. Centrifuge for 5 min. Aspirate and discard aqueous layer. Transfer 25 ml of organic solvent to 50 ml round.bottomed centrifuge tubes. Add 4 ml of 1.0 M HCI. Shake for five minutes. Centrifuge for 3 min. To 3 ml of acid extract (top layer) add 1 ml of 4.4 M sodium hydroxide. Run the UV spectrum of each sample against the blank. Read the absorption maximum of each sample at 275 nm. Calculate the concentration using the closest standard. All samples and standards are run in duplicate.
493
Results and discussion Spectral scans of a 10 and 20 pg/ml standard are shown in Fig. 1. The wavelength of maximum absorption (X,,,) is at 275 nm. A standard curve prepared by taking known theophylline concentrations in serum through the procedure is presented in Fig. 2. Excellent linearity is observed in the concentration range of study 0.0-100 pg/ml. The large range of linearity makes the procedure useful in terms of normal monitoring as well as overdose cases. Replicate determinations of a 5 pug/ml serum sample showed a C.V. of 9.5% with a 105% recovery. A 10 pg/ml serum sample determined six times on the same day gave a mean of 9.42 f 0.61 pg/ml. The within day C.V. is 6.5% with a 94% recovery at the 10 pug/ml level. A pooled (15 E.cg/ml) serum control determined 17 times on separate days in duplicate over a 3 month period gave a mean of 13.6 pg/ml with a standard deviation of kO.9 pg/ml. The between day C.V. is 6% with a 91% recovery. Recovery of theophylline from serum or plasma depends on the level of the drug. Using chloroform/isopropanol as the extracting solvent recoveries of theophylline from blood have been reported to vary from 96 to 110% at the 0.6-10 E.cg/ml level, with a C.V. at 5 pg/ml of 6.5% [4]. Using similar extraction procedures recoveries of 90 and 85% have been reported [ 5,7,8] . By extracting at pH 7.4 interferences from acidic drugs, low pK,, such as salicylic acid and basic drugs, high pK, can be eliminated. The extractability of the drug is a function of the pK, of the drug and its solubility in the extracting solvent. Drugs which are extracted at neutral pH such as barbiturates would be expected to interfere. There is a great deal of confusion in the literature as to what substances will interfere in the different modified ultraviolet methods. Schack’ and Waxler showed that theophylline was extracted better using chloroform/isopropanol than chloroform [4]. Using this procedure, extracting at neutral pH and back extracting with sodium hydroxide (basic pH), uric acid was shown not to interfere. Caffeine and barbiturates are simultaneously extracted with theophylline
0.0 240
250
260
270
280
Wavelength
290
300
310
(nm)
Fig. 1. Ultraviolet spectral scan of theophylline in serum taken through extraction procedure; (A) 20 pg/ ml standard in serum. (B) 10 fig/ml standard in serum. Five ml of sample used.
494
I
0
Theophyifine
Concentration
(pg/m!}
Fig. 2. Plot of theophylline standards in serum (0.0-100 fig/ml). Serum samples of known concentration were extracted using this procedure and the absorbance measured at 275 nm.
theophyliine
and are real interferences. Extracting theophylline at pH 7.4 with chloroform/ isopropanol and back extracting with dilute sodium hydroxide eliminates the interference from caffein since its pK, is greater than that of theophylline [B] . However, interferences from barbiturates would be expected. Extracting at pH 7.4 with chloroform and back extracting with hydrochloric acid, Gupta and Lundberg [7] showed that b~bitura~ interference could be eliminated. Theophylline has a p& of 8.8 [9] while the pfl, of barbiturates range from 7.3-8.0 [ lO,ll]. Thus, by back extracting at an acidic pH, barbiturate interference is eliminated. The same authors [ 71 used differential spectrophotometry to eliminate interferences by adding sodium dihydrogen phosphate to the reference side. This would eliminate any in~rferen~e present in the serum sample provided the drug does not show spectral changes with a change in pH. Using the above method in this laboratory a cloudy solution was observed on the reference side, making reproducibility difficult. However, using a modified extraction procedure and not using differentia1 spectrophotometry accurate and reproducible results are obtained. The following drugs investigated do not interfere with this procedure at the 20 pg/ml level; amobarbital, phenobarbital, glutethimide, methyprylon, propranolol, meprobamate, dilantin, methaqualone, and ethchlorvynol. No significant interference was observed from chloramphenicol and diazepam (10 pig/ ml), quinidine (6 fig/ml), and sodium salicylate (250 E;lg/ml).Interference was observed from sulfanilamide (50 @g/ml) which was equivalent to 4 @g/ml of theophylline at 275 nm. Librium (chlordiazepoxide) showed complete interference at the 20 yg/ml level. Procainamide interferes in the procedure and is observed by a shift of the wavelength maximum from 275 to 265 nm. Interferences from xanthines using this procedure were investigated. Xanthine, hypoxanthine and 7-(2,3-dihydroxypropyl)~theophylline do not interfere at the 20 pg/ml level. Similar results have been reported for xanthine and hypoxanthine at lower levels [ 81. No interference was observed from uric acid at the
495
200 pg/ml level. Interferences would not be expected from the metabolites 1,3-dimethyluric acid and 1-methyluric acid since they are not extracted by a similar procedure [2] . A sample containing 100 pg/ml of caffein showed an interference equivalent of 6 pg/ml of theophylline. A slight shift of the wavelength maximum (275 nm) to shorter wavelengths was observed. Theobromine and the metabolite 3-methylxanthine will interfere with the present procedure. Theobromine (20 pg/ml) shows complete interference while 3-methylxanthine (20 pg/ml) shows 50% interference equivalent to 10 pg/ml of theophylline. The serum level of this metabolite is from 3 to 4 pg/ml at normal therapeutic theophylline levels [6]. Thus, a small error will occur due to interference from the 3-methyxanthine metabolite using this procedure. Using the present procedure for routine monitoring of theophylline levels patients should be off all interfering drugs and on xanthine-free diet where the interfering xanthines are eliminated. Interferences due to xanthines or other drugs at 275 nm can be detected by scanning the spectra. References 1
Ritchie, man,
J.M.
A.,
2
Jenne.
3
Piafsk;.
4
Schack,
5
Shah,
(1970)
eds.),
J.W..
pp.
Wvze.
K.M.
Thompson,
7
Gupta,
R.C.
8
J&low.
P. (1975)
9
Welling, Gilman,
11
R.D.,
A.,
Sunshine.
S.H.
Lundberg, Clin.
Lyons,
S.K.
F.S.
RII.
(1970)
eds.), I. (ed.)
pp.
H.T.
Chem. L.L..
(1969)
MacDonald, J. Pharmacol. J.W.
Anal.
(1972) 292,
Exp.
Sci.
4th
edn.,
(Goodman,
Sitar.
D.S.,
Piafsky,
K.M.,
Gil-
Clin.
Pharmacol.
The?.
13,
349
Ther.
97,
283
(1974)
J. Lab.
Clin.
Med.
84,
584
45.2403
1518 W.A.
McMillan,
Rango.
and
1218
and
Trochta.
New
G.A.
Basis
of
(1975)
Clin.
Therapeutics.
PhannacoL 4th
edn..
Ther.
and
L.S.
and
York
of Analytical R.E.
17,475
(Goodman,
Toxicology.
Chemical
Rubber
Company,
Ohio 12
L.S.
63.1283
Chem.
Pharmacological
Handbook
F.M.
J. Med.
J. Pharm.
and Jenne,
21,
of Therapeutics.
York
N. Engl.
(1973)
Craig,
in The 109.
and
(1949)
G.D.
Basis
New
(1975)
S. (1974)
Nagasawa,
and
P.G.,
Sharpless,
Rood.
Riegelman.
6
10
E..
and Waxier,
and
Pharmacological McMillan,
;nd’Ogivie,
J.A. V.P.
in The 358-370,
Ogilvie,
R1.
(1975)
Clin.
Chem.
21,
1774
Cleveland.
Clinica Chimica Acta, 69 (1976) 491-495 @ Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands
CCA 7836
A MODIFIED ULTRAVIOLET SPECTROPHOTOMETRIC METHOD FOR THE DETERMINATION OF THEOPHYLLINE IN SERUM IN THE PRESENCE OF BARBITURATES
JOHN VASILIADES
and TAMELA TURNER
Department of Pathology, 35233 (U.S.A.)
The
University
of Alabama
in Birmingham,
Birmingham,
Ala.
(Received December 22,1975)
Summary Ultraviolet spectrophotometry is the most commonly used technique for the determination of theophylline levels for therapeutic monitoring. Common interferences in most methods are barbiturates and xanthines. A modified method which eliminates interferences from barbiturates and most xanthines is presented. Theophylline is extracted from serum with chloroform/isopropanol at pH 7.4 back extracted into dilute hydrochloric acid and then the solution is made alkaline withlsodium hydroxide. Barbiturate interferences are eliminated. Interference from caffeine, uric acid, 7-( 2,3-dihydroxypropyl)theophylline, xanthine, and hypoxanthine are not observed. Theobromine, and the metabolite 3-methylxanthine interfere. Interference is not observed from quinidine, diazepam, salicylate, glutethimide, methyprylon, propranolol, methaqualone, dilantin and ethchlorvynol. Sulfanilamide, procainamide and chlordiazepoxide interfere.
Introduction Theophylline (1,3_dimethylxanthine) is a powerful bronchodilator which relaxes the smooth muscle fibers of the bronchi [ 11. This property constitutes the basis of therapy for a large number of individuals with obstructive lung disease. Therapeutic concentrations of theophylline range from 5 to 15 I.cg/ml [2, 31. Toxicity is usually observed above 20 pg/ml. A large variation in theophylline half life and rates of elimination of the drug among patients makes therapeutic monitoring to achieve and maintain a certain concentration necessary. Methods presently used for the determination of theophylline in serum are usually modifications of the original ultraviolet method involving a basic extraction of theophylline from the organic solvent [4] . Gas liquid chromatography (GLC) and high pressure liquid chromatography (HPLC) have been used
492
for the determination of theophylline [ 5,6,12] . However, a spectrophotometric method is more suitable for routine monitoring of acute asthmatic emergencies as well as chronic care because of its simplicity, A modified ultraviolet spectrophotometric method for serum theophylline is presented which eliminates many of the common interferences of previous methods. The procedure involves extraction of theophylline at pH 7.4 into chloroform/isopropanol followed by back extraction from the organic phase into dilute 1 M hydrochloric acid. The acidic solution is made basic with sodium hydroxide and the ultraviolet spectrum scanned. Interferences from barbiturates are eliminated as well as other drugs extracted at basic pH. Interferences from hypoxanthine, caffeine, and other xanthines are also eliminated. Materials and methods Instrumentation A Beckman ACTA IV spectrophotometer (Beckman Instruments, Fullerton, California) was used for these studies with quartz cuvettes (l-cm path length). The inh.um-d has a photometric accuracy of 0.003 absorbance units at 1.0 A. Reagents All reagents were AR grade. Chloroform/isopropanol(20 : 1, by volume). Hydrochloric acid 1 M. Sodium hydroxide 4.4 M. Hypoxanthine, 3_methylxanthine, xanthine, and 7-(2,3-dihydroxypropyl)theophylline were obtained from Adams Chemical Company, P.O. Box 236 Round Lake, Ill. Theophylline, and theobromine were obtained from Analabs Incorporated, 80 Republic Drive, North Haven, Conn. Theophylline stock standard: 1.25 mg/ml. Dissolve 125 mg of theophylline in 100 ml of distilled water. Theophylline working stock: 500 E.cg/ml. Dilute 20 ml of 1.25 mg/ml stock standard to 50 ml with water. Theophylline working standards: 10 and 20 pg/ml. Take 2 and 4 ml of working stock 500 lug/ml and dilute to 50 ml with water. Standards of 10 and 20 pug/ ml are prepared in serum daily prior to use by reconstituting lyophilized serum with the 10 and 20 (ug/ml theophylline aqueous standards. All other xanthine solutions were prepared by dissolving the pure substance in water. Procedure Place 3 ml of serum sample, serum standard and serum blank into 50 ml round bottomed centrifuge tubes. Add 30 ml of 20 : 1 chloroform/isopropanol mixture. Shake for 5 min at low speed. Centrifuge for 5 min. Aspirate and discard aqueous layer. Transfer 25 ml of organic solvent to 50 ml round.bottomed centrifuge tubes. Add 4 ml of 1.0 M HCI. Shake for five minutes. Centrifuge for 3 min. To 3 ml of acid extract (top layer) add 1 ml of 4.4 M sodium hydroxide. Run the UV spectrum of each sample against the blank. Read the absorption maximum of each sample at 275 nm. Calculate the concentration using the closest standard. All samples and standards are run in duplicate.
493
Results and discussion Spectral scans of a 10 and 20 pg/ml standard are shown in Fig. 1. The wavelength of maximum absorption (X,,,) is at 275 nm. A standard curve prepared by taking known theophylline concentrations in serum through the procedure is presented in Fig. 2. Excellent linearity is observed in the concentration range of study 0.0-100 pg/ml. The large range of linearity makes the procedure useful in terms of normal monitoring as well as overdose cases. Replicate determinations of a 5 pug/ml serum sample showed a C.V. of 9.5% with a 105% recovery. A 10 pg/ml serum sample determined six times on the same day gave a mean of 9.42 f 0.61 pg/ml. The within day C.V. is 6.5% with a 94% recovery at the 10 pug/ml level. A pooled (15 E.cg/ml) serum control determined 17 times on separate days in duplicate over a 3 month period gave a mean of 13.6 pg/ml with a standard deviation of kO.9 pg/ml. The between day C.V. is 6% with a 91% recovery. Recovery of theophylline from serum or plasma depends on the level of the drug. Using chloroform/isopropanol as the extracting solvent recoveries of theophylline from blood have been reported to vary from 96 to 110% at the 0.6-10 E.cg/ml level, with a C.V. at 5 pg/ml of 6.5% [4]. Using similar extraction procedures recoveries of 90 and 85% have been reported [ 5,7,8] . By extracting at pH 7.4 interferences from acidic drugs, low pK,, such as salicylic acid and basic drugs, high pK, can be eliminated. The extractability of the drug is a function of the pK, of the drug and its solubility in the extracting solvent. Drugs which are extracted at neutral pH such as barbiturates would be expected to interfere. There is a great deal of confusion in the literature as to what substances will interfere in the different modified ultraviolet methods. Schack’ and Waxler showed that theophylline was extracted better using chloroform/isopropanol than chloroform [4]. Using this procedure, extracting at neutral pH and back extracting with sodium hydroxide (basic pH), uric acid was shown not to interfere. Caffeine and barbiturates are simultaneously extracted with theophylline
0.0 240
250
260
270
280
Wavelength
290
300
310
(nm)
Fig. 1. Ultraviolet spectral scan of theophylline in serum taken through extraction procedure; (A) 20 pg/ ml standard in serum. (B) 10 fig/ml standard in serum. Five ml of sample used.
494
I
0
Theophyifine
Concentration
(pg/m!}
Fig. 2. Plot of theophylline standards in serum (0.0-100 fig/ml). Serum samples of known concentration were extracted using this procedure and the absorbance measured at 275 nm.
theophyliine
and are real interferences. Extracting theophylline at pH 7.4 with chloroform/ isopropanol and back extracting with dilute sodium hydroxide eliminates the interference from caffein since its pK, is greater than that of theophylline [B] . However, interferences from barbiturates would be expected. Extracting at pH 7.4 with chloroform and back extracting with hydrochloric acid, Gupta and Lundberg [7] showed that b~bitura~ interference could be eliminated. Theophylline has a p& of 8.8 [9] while the pfl, of barbiturates range from 7.3-8.0 [ lO,ll]. Thus, by back extracting at an acidic pH, barbiturate interference is eliminated. The same authors [ 71 used differential spectrophotometry to eliminate interferences by adding sodium dihydrogen phosphate to the reference side. This would eliminate any in~rferen~e present in the serum sample provided the drug does not show spectral changes with a change in pH. Using the above method in this laboratory a cloudy solution was observed on the reference side, making reproducibility difficult. However, using a modified extraction procedure and not using differentia1 spectrophotometry accurate and reproducible results are obtained. The following drugs investigated do not interfere with this procedure at the 20 pg/ml level; amobarbital, phenobarbital, glutethimide, methyprylon, propranolol, meprobamate, dilantin, methaqualone, and ethchlorvynol. No significant interference was observed from chloramphenicol and diazepam (10 pig/ ml), quinidine (6 fig/ml), and sodium salicylate (250 E;lg/ml).Interference was observed from sulfanilamide (50 @g/ml) which was equivalent to 4 @g/ml of theophylline at 275 nm. Librium (chlordiazepoxide) showed complete interference at the 20 yg/ml level. Procainamide interferes in the procedure and is observed by a shift of the wavelength maximum from 275 to 265 nm. Interferences from xanthines using this procedure were investigated. Xanthine, hypoxanthine and 7-(2,3-dihydroxypropyl)~theophylline do not interfere at the 20 pg/ml level. Similar results have been reported for xanthine and hypoxanthine at lower levels [ 81. No interference was observed from uric acid at the
495
200 pg/ml level. Interferences would not be expected from the metabolites 1,3-dimethyluric acid and 1-methyluric acid since they are not extracted by a similar procedure [2] . A sample containing 100 pg/ml of caffein showed an interference equivalent of 6 pg/ml of theophylline. A slight shift of the wavelength maximum (275 nm) to shorter wavelengths was observed. Theobromine and the metabolite 3-methylxanthine will interfere with the present procedure. Theobromine (20 pg/ml) shows complete interference while 3-methylxanthine (20 pg/ml) shows 50% interference equivalent to 10 pg/ml of theophylline. The serum level of this metabolite is from 3 to 4 pg/ml at normal therapeutic theophylline levels [6]. Thus, a small error will occur due to interference from the 3-methyxanthine metabolite using this procedure. Using the present procedure for routine monitoring of theophylline levels patients should be off all interfering drugs and on xanthine-free diet where the interfering xanthines are eliminated. Interferences due to xanthines or other drugs at 275 nm can be detected by scanning the spectra. References 1
Ritchie, man,
J.M.
A.,
2
Jenne.
3
Piafsk;.
4
Schack,
5
Shah,
(1970)
eds.),
J.W..
pp.
Wvze.
K.M.
Thompson,
7
Gupta,
R.C.
8
J&low.
P. (1975)
9
Welling, Gilman,
11
R.D.,
A.,
Sunshine.
S.H.
Lundberg, Clin.
Lyons,
S.K.
F.S.
RII.
(1970)
eds.), I. (ed.)
pp.
H.T.
Chem. L.L..
(1969)
MacDonald, J. Pharmacol. J.W.
Anal.
(1972) 292,
Exp.
Sci.
4th
edn.,
(Goodman,
Sitar.
D.S.,
Piafsky,
K.M.,
Gil-
Clin.
Pharmacol.
The?.
13,
349
Ther.
97,
283
(1974)
J. Lab.
Clin.
Med.
84,
584
45.2403
1518 W.A.
McMillan,
Rango.
and
1218
and
Trochta.
New
G.A.
Basis
of
(1975)
Clin.
Therapeutics.
PhannacoL 4th
edn..
Ther.
and
L.S.
and
York
of Analytical R.E.
17,475
(Goodman,
Toxicology.
Chemical
Rubber
Company,
Ohio 12
L.S.
63.1283
Chem.
Pharmacological
Handbook
F.M.
J. Med.
J. Pharm.
and Jenne,
21,
of Therapeutics.
York
N. Engl.
(1973)
Craig,
in The 109.
and
(1949)
G.D.
Basis
New
(1975)
S. (1974)
Nagasawa,
and
P.G.,
Sharpless,
Rood.
Riegelman.
6
10
E..
and Waxier,
and
Pharmacological McMillan,
;nd’Ogivie,
J.A. V.P.
in The 358-370,
Ogilvie,
R1.
(1975)
Clin.
Chem.
21,
1774
Cleveland.
