JOURNAL OF PHARMACEUTICAL SCIENCES
Ama November 1992 Volume 81, Number 11
A publication of the American Pharmaceutical Association
ARTICLES
Nicotine Absorption after Pulmonary Instillation DAVIDR. HERRMA”’, KEITH M. OLSEN’,
AND F. CHARLES
HILLER+’
Received September 23,1991, from the ‘Division of Pulmonary and Critical Care Medicine, 4301 W. Markham, Slot 555, and *Department of Pharmacy Practice, University of Arkansas for Medical Sciences, Little Rock, AR 72205. Accepted for publication March 11, 1992. Abstract 0 Levels of nicotine in plasma were determined by gas chromatography in eight mongrel dogs after instillation of 0.5 mg of nicotine in 100 pL of normal saline at three levels of the tracheobronchial tree: the trachea, a subsegmental bronchus of the right middle lobe, and a subpleural location of the right middle lobe (“distal”).An equivalent dose was given intravenously (iv). Peak of nicotine concentrations in plasma were significantlylower after instillationat the trachea (11.5 2 4.4 ng/mL) and the subsegmental bronchus (1 8.2 f 5.0 ng/mL) than after an iv dose (30.32 10.7ng/mL); p < 0.05 for each comparison. In addition, the peak concentration after instillation at the trachea was significantly lower than that after instillation at the distal site (22.1 f 6.2ng/mL, p < 0.05).Time to peak concentration was significantly longer after tracheal instillation (5.3f 3.0min) than after subsegmental instillation (2.0 2 0.0 min) or iv infusion (2.0 2 0.0 min); p < 0.05 for each comparison. Total drug absorbed, half-life, and clearance were equivalent from all four sites. This study demonstrated that quantitative absorption of nicotine from the described lung sites is equivalent to that after an iv dose, with slower absorption and lower peak concentrationsfrom the tracheal site.
The lung is an effective surface for absorbing substances into the body. The gas exchange region is thought to be the most efficient absorbing membrane,’ but there is a paucity of data documenting this impression.2 Studies of whole-lung absorption, in which the entire lung is filled ex vivo with a solution of the chemical of interest, have the advantages of a defined absorption membrane (the whole lung) and a known concentration of the chemical.3-4 Disadvantages are that absorption cannot be studied a t various levels in the respiratory tract and that the technique does not mimic any practical situation. Studies of drug levels in blood after drug delivery by aerosol have not shown quantitative absorption as a fraction of lung deposition because of the difficulties in measuring deposition quantitatively.bl1 With aerosol delivery, it is not possible to target a specific airway or even a lung region with precision. For these reasons, measurements of absorption from any specific anatomical location must use methods that will accurately place the drug a t the site of interest. Even with instillation techniques, localization of a dose to a small area is not possible. Tracheal instillation10-16 of small volumes of the chemical of interest delivers a precise dose, but the solution spreads to poorly defined levels of the 0022-3549/92/1100-1055$02.50/0 0 1992, American Pharmaceutical Association
tracheobronchial tree. We are aware of only one study in which absorption was measured a t defined levels using a measured dose.” In this study, 3sS-labeled sulfuric acid was instilled directly into either the nares or, by bronchoscope, into a second- or a seventh-order bronchus of three beagle dogs (one site per dog). Blood samples revealed rapid absorption from the seventhorder bronchus, with slower and less complete absorption from the second-order bronchus during the 30-min sampling period. Pharmacokinetic analysis of the data was not performed. The purpose of our study was to compare the quantitative absorption and pharmacokinetics of nicotine from the trachea, subsegmental airways, and distal regions of the lung. A dog model was used for these measurements. We chose nicotine because, despite its widespread use by cigarette smokers, its regional absorption characteristics are not well defined and because of the interest in our laboratory8 in the potential use of a nicotine aerosol as an aid in smoking cessation.
Experimental Section Animals-Eight mongrel dogs (18.2-25.6 kg), supplied by a United States Department of Agriculture licensed dealer, were used in the study. They were free from pulmonary infections and parasites. Preparation of Nicotine Solution-L-Nicotine (Kodak) was dissolved in normal saline to a concentration of 130 mg/mL and then diluted in normal d i n e to a final concentration of either 5 or 0.1 mg/mL (depending on the site of delivery, vide infra). The pH was adjusted to 7.5 with HCl. The solution for intravenous (iv) administration was filtered through a O.2-pm-pore-sizefilter prior to infusion. Nicotine Dosing-Nicotine absorption was measured after instillation at three lung sites (trachea, subsegmental airway, and a distal site) and compared with absorption after an iv dose. Nicotine was introduced into all eight dogs by instillation at the three lung sites and by iv administration. Food and water were withheld for 12 h prior to the procedure. On the morning of the procedure, an iv catheter was placed in a foreleg vein. The catheter was maintained by 2-mL heparin flushes (100 unita/mL) after each use. Each dog was prernedicated with atropine (0.05mgkg, subcutaneous),butorphanol(O.11mgkg, subcutaneous), and 4% thiamylal sodium (17.5-mgkg iv loading dose and then titrated to adequate sedation).Dogs were placed in a prone position, Journal of Pharmaceutical Sciences / 1055 Vol. 81, No. 1 1, November 1992
and a n endotracheal tube was inserted; the cuff was inflated prior to nicotine administration by random selection by anatomical site. The nicotine solution was injected into the lung from a catheter passed to the desired location via the suction channel of the bronchoscope (internal volume of catheter, 600 pL). A three-way stopcock was attached to the catheter, and to the stopcock were affixed 10-mL and 100-pL syringes. The catheter tip was placed in the nicotine solution and filled first with the 10-mL syringe after which the 100-pL syringe was filled. The tip was wiped off, and the catheter was inserted into the bronchoscope. For dosing, the 100-pL syringe was injected so that 100 pL are delivered a t the distal site. To confirm the accuracy of this method, several 100-pL volumes of nicotine solution were injected into a container and weighed, and the delivered volume was determined to be 100.2 ? 1.7 pL [mean ? standard deviation (SD)]. Distal Site-A fiberoptic bronchoscope (Olympus BF4B2) was inserted, and a lateral subsegment of the right middle lobe was identified. Because of differences in endobronchial anatomy among animals, the same subsegment may not have been used in every dog. A 1.7-mm 0.d. (5F) angiographic catheter (Multipurpose A-1, Cordis Corporation) was inserted through the suction channel of the bronchoscope. The catheter was advanced to a location immediately subpleural, and the position was confirmed by fluoroscopy (this position is referred to as “distal” in the remainder of the paper). Nicotine solution (5.0 mg1mL; 100 pL; total dose, 0.5 mg) was then instilled rapidly. The catheter was withdrawn through the bronchoscope, and then the bronchoscope was withdrawn. When the animals were sufficiently awake, the endotracheal tube was removed. In three dogs, after completion of the study and sacrifice of the dogs, the angiographic catheter was again passed to the distal location by using the technique just described. Radiographic contrast material (2 mL) was injected through the catheter. Figure 1 is a photograph of a representative radiograph produced in this manner. The photograph indicates that some of the contrast material reached the alveoli. Because we did not directly visualize all nicotine injection sites, we cannot be sure of the exact anatomical injection site. It is possible that some portions of the 100-pL nicotine aliquot were placed into terminal bronchioles, respiratory bronchioles, alveoli, or a combination of these sites in the various dogs. For this reason, we refer to this location as the distal site.
Figure 1-Photograph showing detail of a radiograph made after injection of 2 mL of radiographic contrast material through a catheter placed in a subpleural location of the right middle lobe, as described in the protocol. Alveolanzation of the contrast material confirms delivery to gas-exchange regions. The volume of nicotine solution delivered during the study was 100 pL. 1056 I Journal of Pharmaceutical Sciences Vol. 81, No. 1 1 , November 1992
Subsegmentnl Bronchus Site-On a different day, the same procedure was performed, and the right middle lobe of the lung was identified. The catheter tip was placed in the ostium of a lateral segment of the right middle lobe and advanced 2 cm (this position is referred to as “subsegmental” in the remainder of the paper).Nicotine solution (5.0 mg/mL; total dose, 0.5 mg) was then instilled rapidly. Because of differences in endobronchial anatomy among animals, the same segment may not have been used in every dog. In addition, the distance from the ostium of a segment to ita pleural surface varies markedly; therefore, absolute consistency in the site of deposition for this subsegmental location may not have been achieved in every dog. Tracheal Site-On a third day, the procedure was repeated, except that the tip of the bronchoscope was placed in the trachea and the tip of the angiographic catheter was positioned 10 cm proximal to the carina. The same volume of the same nicotine solution was instilled in this location. a fourth day, the dogs were Intravenous Administration-On prepared as described earlier, except that a second iv catheter (maintained by heparin flushes) was placed in a hind-leg vein. No endotracheal tube was put in place, and no bronchoscopy was performed. Nicotine solution (0.1 mglmL; 5 mL; total dose, 0.5 mg) was injected through the hind-leg vein over a period of 1 min. All dogs remained sternal and sedated for a minimum of 20 min, after which they were allowed to awaken and move about ad lib. No significant coughing was noted in any dog during instillation or subsequent blood draws. A minimum of 48 h elapsed between each of the four sessions of nicotine administration. Blood Sampling-Blood samples (10 mL per sample in vacuum tubes coated with lithium heparin) were obtained prior to nicotine administration and a t 2, 5, 7, 10, 15, 20, 40, 80, and 160 min aRer administration. For iv dosing, these times were from the beginning of the 1-min infusion. Blood was stored on ice during collection. ARer collection of all 10 samples, the blood was centrifuged at 2000 x g for 10 min, and the plasma was removed. Stability studies completed in our laboratory indicate that nicotine is stable when collected and centrifuged in this manner. The plasma was divided into two aliquota and stored a t -20 “C until analyzed. All samples were analyzed in duplicate. Nicotine Extraction-Nicotine in plasma was assayed by the method of Feyerabend and Russell18 with N-ethylnornicotine (ENN) as internal standard, as described by Jacob et al.10 Previous studies from this laboratory have used the same assay.8.20 Briefly, each plasma sample was thawed, and 2 or 3 mL of plasma was t r a n s f e d to a large screw-capped glass test tube. ENN (1.0 pg/mL) was added a t 25 uUmL of alasma followed bv 2 mL of 5 N NaOH and 3 mL of diethyl ether. The tube was capp&, mixed by vortex for 2 min, and then centrifuged for 10 min. The supernatant was transferred to a second tube and evaporated to -200 pL under nitrogen. To this concentrated supernatant, 100 pL of 2 N HCl was added. The tube was capped, mixed by vortex for 2 min, and centrifuged for 5 min. The ether layer was removed and discarded. Diethyl ether (500 4)was added, and the tube was mixed by vortex for 1 min and centrifuged for 5 min. The ether layer was removed and discarded. The residual ether was then removed by evaporation under nitrogen for 75 8. The remaining solution was transferred to a Dreyer tube, 200 pL of 5 N NaOH and 50 pL of n-butyl acetate were added, and the tube was sealed. The tube was mixed by vortex for 5 min and then allowed to stand for at least 1 h before analysis. A standard curve was prepared by using 3 mL of calf Berum to which had been added 0, 36,72, 108, and 144 pL of nicotine solution (1.0 CrglmL), with final nicotine concentrations in serum of 0, 12, 24, 36, and 48 ng/mL. ENN was added to these samples as described earlier, and extraction of nicotine was performed as for the dog plasma samples. Standard curves prepared by using calf serum and dog plasma were not significantly different; therefore, because of the availability calf serum and the volume required, all standard curves were prepared by using calf serum. Nicotine Assay-Nicotine was analyzed with gas chromatograph (Tracor model 560, Tracor Inc., Austin, TX) equipped with a nitrogen-phosphorus detector. For analysis, 4 pL of the top layer (n-butyl acetate) of the Dreyer tubes was injected into the chromatograph. The peak height ratio (peak height of nicotine:peak height of ENN) was determined for each standard-curve sample and plotted versus nicotine concentration in plasma. The peak height ratio was then calculated for each experimental sample, and the nicotine concentration was determined from the standard curve. Standard curves
were determined daily. The sensitivity of the assay is 0.1 ng/mL.17120 The mean intra- and interday coefficients of variation for the assay were 9 and 7%, respectively. Pharmacokinetic Analysis-Data of nicotine concentrations in plasma versus time were fitted to selected polyexponential equations by using the nonlinear least-squares fitting computer program ESTRIPZ1 and PCNONLIN (Statistical Consultants, Inc., Lexington, KY ). Pharmacokinetic parameters were calculated from computergenerated coefficients and exponents, as described by Wagner.= Each data set was best described by a two-compartment, open, linear model. The area under curve (AUC)of nicotine concentration in plasma versus time was determined up to the end of the data collection period and extrapolated to infinity (AUC,). Absolute bioavailability (F)from the three pulmonary administration sites was calculated by comparing the AUC, values after instillation at these sites with that after iv administration: F, % = LAUC, pulmonary site/AUC, iv route] x 100). Peak nicotine concentration in plasma (C,,,-) and the time to reach peak nicotine concentration Urnax) were determined by examination of the concentration in plasma-time curve. Clearance (CL) was estimated by dividing the administered dose by AUC. Statistics-Data were analyzed by one-way analysis of variance (ANOVA).= When the results of ANOVA were significant,the t test was used to evaluate differences between data s e t s . 2 4 All data are presented as mean 2 SD, and p < 0.05 was taken as the level of significance for all analyses.
ResuI ts Mean concentration-time curves and pharmacokinetic parameters for all routes of administration are shown in Figure 2 and Table I, respectively. Analysis of concentration-time data indicates that a biexponential function best describes nicotine disposition during the postdose period. No statistical differences were noted among AUC, and F (the parameters used for comparing bioavailability), half-life, and CL values for all routes of administration. C,,, values after instillation at tracheal and subsegmental sites were lower than after iv administration. The C,, after instillation at the distal site was not different from that after iv administration. In addition, t,, was higher after tracheal delivery compared with either iv or subsegmental delivery. The t,,, after instillation a t the distal site was not different from those after instillation at the other sites.
Discussion This work describes the pharmacokinetics of nicotine absorption from various levels of the respiratory tract. The results depend in part on injection of nicotine in the intended locations. Injection in the distal location was accomplished via fluoroscopic guidance. The alveolarization of constant media
0
20
40
60 Time (minutes)
Flgure 2-Mean concentration of nicotine in plasma (ng/mL) versus time (min) after introduction of nicotine by instillation at (0)tracheal, (W) subsegmental, and (A)distal sites or by (*) iv administration. (SDs are given in Table I).
Table CPharmacoklnetlc Parameters for Nlcotlne after lnstlllation at Three Pulmonary Sltes and Iv Admlnlstratlon'
Trachea Subsegmental Distal
iv
660 (132) 635 (197) 936 (371) 922 (340)
0.78 (0.23) 0.85 (0.41) 1.11 (0.51)
-
84.8 (36.6) 91.2 (29.0) 85.0 (56.5) 83.6 (33.3)
753 (159) 785 (214) 539 (268) 549 (199)
11.5 (4.4)'*d,e 18.2 (5.0)' 22.1 (6.2)
5.3
30.3
(3.0)C.d 2.0 (0.0) 2.6 (1.8) 2.0
(10.7)
(0.0)
All data are expressed as mean (SD). Half-life. 'Significantly different from result after iv administration (p < 0.05). dSignificantly different from result after instillation at subsegmental site (p < 0.05). Significantlydifferent from result after instillation at distal site (p < 0.05).
injected a t the distal location in several animals (Figure 1) suggests that the injection at the distal site was quite peripheral, as intended. It is possible that a portion of the nicotine instilled in the subsegmental zone moved either proximally or distally during the early part of the absorption period. The small volume instilled and the absence of significant coughing among the animals during the first 20 min after instillation should have minimized such problems. The absorption of nicotine from various levels of the tracheobronchial tree was compared with that from the iv route. C,,, and t,, after instillation at the distal lung region were very similar to those after iv administration. This result suggests that the pattern of nicotine absorption from the distal lung region is similar to that from the iv route (Figure 2). C,, values after instillation at tracheal and subsegmental levels were lower. Absorption from the trachea was slower, with t,, longer than that after iv administration or instillation at subsegmental sites. The t,,, was calculated as the mean of the t,, for each individual animal. However, because of the very high nicotine concentration in plasma in two dogs a t 2 min, the mean concentration was highest at 2 min (Figure 2). The amounts of drug reaching the systemic circulation from the four sites, as reflected by AUC, were statistically equivalent. This result suggests complete absorption of nicotine from each site. Alternatively, the difference between AUC values after instillation at tracheal and subsegmental levels and those after instillation at the distal level and iv administration may not have reached significance because of the large SDs of the results for the distal and iv sites or because of the small number of animals studied. The AUC values calculated by the pharmacokinetic programs used in this study are based on curves of level in plasma versus time, which the program extrapolates to infinity. The AUC value derived is thus very dependent on the nicotine levels in plasma obtained during the terminal elimination phase (20-160 rnin). Small amounts of imprecision in the measurement of the values might have a large effect on the AUC value calculated by PCNONLIN and, therefore, lead to the relatively large SDs seen for the AUC values. Wangensteen et al.3 and Taylor e t al.4 examined the absorption of chemicals with a n isolated, perfused mammalian lung. In these studies, the lungs were filled with a solution of the chemical of interest, and the rate of disappearance of the chemical from the solution was measured. This procedure allowed precise definition of the administered dose, the amount absorbed, and the rate of absorption. An important contribution of this technique is the comparison of absorption kinetics for different chemical substances. However, the ex vivo model makes extrapolation of the data to in vivo systems difficult, and the delivery of the chemical to the Journal of Pharmaceutical Sciences I 1057 Vol. 81, No. 1 1 , November 1992
entire lung does not allow measurement of differential absorption h m different levels of the lung. In other studies, the chemical of interest is delivered to intact lungs either by aerosoF-11 or by tracheal instillation.10-16 Studies of absorption after aerosol delivery suffer from deposition and absorption from the oropharynx when tracheostomies or endotracheal tubes are not used. Determination of the precise dose deposited is difficult, and cumbersome techniques are required to measure both the delivered and the exhaled quantities of the chemical. In addition, deposition occurs to varying degrees throughout the lung, a situation making the study of regional absorption difficult. Tracheal instillation, in which the chemical solution is instilled through an indwelling tracheal cannula, overcomes the difficulties of oropharyngeal absorption and allows delivery of a precise quantity of the chemical to the lung. Although some studies have shown that this method favors deposition in certain lobes,14 tracheal injection usually involves volumes sufficiently large that the injected solution spreads over a fairly large and poorly defined lung surface and possibly even settles at gas-exchange regions. Characterization of abaorption from specific levels of the lung is thus not possible. The current study represents an effort to measure quantitatively drug absorption from specific airway regions by direct deposition of the drug within these regions. The quantity deposited was well defined (0.5 mg), and the volume of solution was sufficiently small (100pL) to minimize spreading to distant lung regions. Total amounts of nicotine absorbed were similar from all three sites. Nicotine absorption from distal lung units was rapid and comparable with that h m an iv dose, whereas absorption from the tracheal region was slower.
References and Notes 1. Mayer, S.E.; Melmon, K. L.; Gilman, A. G. In The Phurmacological Basis of Thempeutics, 6th ed.; Gilman, A. G.; Goodman, L. S.; Gilman, A. Eds.; Macmillan, New York, 1980; pp 1-27. 2. Schanker. L. S. Biochem. Pharmacol. 1978.27.381385. 3. Wanqeneken, 0. D.; Wittmere, L. E.; Johneon, J. A. Am. J . Phvswl. 1969.216(4). 719-727. 4. Taylor, A. E.;. Guyton, A. C.; Bishop, V. S.Circ. Res. 1965, 16,
1058 I Journal of Pharmaceutical sciences Vol. 81, No. 7 7 , November 7992
353462. 5. Byron, P. R.; Roberta, S. R.; Clark, A. R. J . Phurm. Sci. 1986,75, 168-171. 6. Mentz, W. M.; Brown, J. B.; Friedman, M.; Stutta, M. J.; Gatzy J. T.; Boucher, R. C. Am. Rev. Respir. Dis. 1986,134, 938-943. 7. Brown, R. A.; Schanker, L. S. Drug Metub. Dispa. 1983, 11, 355-360. 8. Burch, S. G.; Erbland, M. E.; Gann, L. P.; Hiller, F. C. Am. Rev. Respir. Dis. 1989,140, 955-957. 9. Sepkovic, D. W.; Colosimo, S. G.; Axelrad, C. M.; Adams, J. D.; Haley, N. J. Am. J . Public Health 1986, 76, 1343-1344. 10. Schanker, L. S.; Mitchell, E. W.; Brown, R. A. Drug Metub. Dispos.1986,14,79-88. 11. Tmovec, T.; Durisova, M.; Bezek, S.; Kalla , Z.; Navarova, J.; Toncikova, 0.;Kettner, M.; Faltue, F.; Erichreb, M. Drug Metub. Dispos. 1984,12,641-644. 12. Brazzell, R. K.; Smith, R. B.; Kostenbauder, H. B. J . Phurm. Sci. i 9 8 2 , 7 i , 1268-1274. 13. Hemberger, J. A.; Schanker, L. S. Am. J . Physiol. 1978, 234, C191-Cl97. 14. Enna, S.J.; Schanker, L. S. Am. J . Physwl. 1972, 223, 12271231. 15. Enna, S. J.; Schanker, L. S. Am. J . Physiol. 1972,222,409-414. 16. Schanker, L. S.; Mitchell, E. W.; Brown, R. A. Phurmacology 1986,32, 176-180. 17. Dahl, A. R.; Felicetti, S.A.; Muggenburg, B. A. Fund. Appl. Toxicol. 1983,3, 293-297. 18. Feyerabend, C.; Rueeell, M. A. H. J.Pharm. Phurmacol. 1979,31, 73-76. 19. Jacob, P.; Wilson, M.; Benowitz, N. L. J . Chromtogr. 1981,222, 61-70. 20. Ebert, R. V.; McNabb, M. E.; McCusker, K. T.; Snow, S.L. JAMA 1983,250,2840-2842. 21. Brown, R. D.; Manna, J. F. J . Phurm. Sci. 1978,67,1687-1691. 22. Wagner, J. G. J . Phurmacokinet. Biophrm. 1976,4,443-467. 23. Glantz, S. A. In Primer o Biostatistics, 2nd ed.;Glantz, S.A.,Ed.; McGraw-Hill: New Yorl, 1987; pp 30-63. 24. Glantz, S. A. In PrimerofBiostatistics,2nded.; Glantz, S.A., Ed.; McGraw-Hill: New York, 1987; pp 64-100.
Acknowledgments We thank Cindy Ha e for secretarial assistance and Cristal McQuary and Cynthia f k b s , D.V.M., for technical assistance. We also thank Richard V. Ebert, M.D., for helpful advice and assistance with the nicotine assa and Paula Anderson, M.D., and Bill J. Gurley, Ph.D., for review of t i e manuscript.
Ama November 1992 Volume 81, Number 11
A publication of the American Pharmaceutical Association
ARTICLES
Nicotine Absorption after Pulmonary Instillation DAVIDR. HERRMA”’, KEITH M. OLSEN’,
AND F. CHARLES
HILLER+’
Received September 23,1991, from the ‘Division of Pulmonary and Critical Care Medicine, 4301 W. Markham, Slot 555, and *Department of Pharmacy Practice, University of Arkansas for Medical Sciences, Little Rock, AR 72205. Accepted for publication March 11, 1992. Abstract 0 Levels of nicotine in plasma were determined by gas chromatography in eight mongrel dogs after instillation of 0.5 mg of nicotine in 100 pL of normal saline at three levels of the tracheobronchial tree: the trachea, a subsegmental bronchus of the right middle lobe, and a subpleural location of the right middle lobe (“distal”).An equivalent dose was given intravenously (iv). Peak of nicotine concentrations in plasma were significantlylower after instillationat the trachea (11.5 2 4.4 ng/mL) and the subsegmental bronchus (1 8.2 f 5.0 ng/mL) than after an iv dose (30.32 10.7ng/mL); p < 0.05 for each comparison. In addition, the peak concentration after instillation at the trachea was significantly lower than that after instillation at the distal site (22.1 f 6.2ng/mL, p < 0.05).Time to peak concentration was significantly longer after tracheal instillation (5.3f 3.0min) than after subsegmental instillation (2.0 2 0.0 min) or iv infusion (2.0 2 0.0 min); p < 0.05 for each comparison. Total drug absorbed, half-life, and clearance were equivalent from all four sites. This study demonstrated that quantitative absorption of nicotine from the described lung sites is equivalent to that after an iv dose, with slower absorption and lower peak concentrationsfrom the tracheal site.
The lung is an effective surface for absorbing substances into the body. The gas exchange region is thought to be the most efficient absorbing membrane,’ but there is a paucity of data documenting this impression.2 Studies of whole-lung absorption, in which the entire lung is filled ex vivo with a solution of the chemical of interest, have the advantages of a defined absorption membrane (the whole lung) and a known concentration of the chemical.3-4 Disadvantages are that absorption cannot be studied a t various levels in the respiratory tract and that the technique does not mimic any practical situation. Studies of drug levels in blood after drug delivery by aerosol have not shown quantitative absorption as a fraction of lung deposition because of the difficulties in measuring deposition quantitatively.bl1 With aerosol delivery, it is not possible to target a specific airway or even a lung region with precision. For these reasons, measurements of absorption from any specific anatomical location must use methods that will accurately place the drug a t the site of interest. Even with instillation techniques, localization of a dose to a small area is not possible. Tracheal instillation10-16 of small volumes of the chemical of interest delivers a precise dose, but the solution spreads to poorly defined levels of the 0022-3549/92/1100-1055$02.50/0 0 1992, American Pharmaceutical Association
tracheobronchial tree. We are aware of only one study in which absorption was measured a t defined levels using a measured dose.” In this study, 3sS-labeled sulfuric acid was instilled directly into either the nares or, by bronchoscope, into a second- or a seventh-order bronchus of three beagle dogs (one site per dog). Blood samples revealed rapid absorption from the seventhorder bronchus, with slower and less complete absorption from the second-order bronchus during the 30-min sampling period. Pharmacokinetic analysis of the data was not performed. The purpose of our study was to compare the quantitative absorption and pharmacokinetics of nicotine from the trachea, subsegmental airways, and distal regions of the lung. A dog model was used for these measurements. We chose nicotine because, despite its widespread use by cigarette smokers, its regional absorption characteristics are not well defined and because of the interest in our laboratory8 in the potential use of a nicotine aerosol as an aid in smoking cessation.
Experimental Section Animals-Eight mongrel dogs (18.2-25.6 kg), supplied by a United States Department of Agriculture licensed dealer, were used in the study. They were free from pulmonary infections and parasites. Preparation of Nicotine Solution-L-Nicotine (Kodak) was dissolved in normal saline to a concentration of 130 mg/mL and then diluted in normal d i n e to a final concentration of either 5 or 0.1 mg/mL (depending on the site of delivery, vide infra). The pH was adjusted to 7.5 with HCl. The solution for intravenous (iv) administration was filtered through a O.2-pm-pore-sizefilter prior to infusion. Nicotine Dosing-Nicotine absorption was measured after instillation at three lung sites (trachea, subsegmental airway, and a distal site) and compared with absorption after an iv dose. Nicotine was introduced into all eight dogs by instillation at the three lung sites and by iv administration. Food and water were withheld for 12 h prior to the procedure. On the morning of the procedure, an iv catheter was placed in a foreleg vein. The catheter was maintained by 2-mL heparin flushes (100 unita/mL) after each use. Each dog was prernedicated with atropine (0.05mgkg, subcutaneous),butorphanol(O.11mgkg, subcutaneous), and 4% thiamylal sodium (17.5-mgkg iv loading dose and then titrated to adequate sedation).Dogs were placed in a prone position, Journal of Pharmaceutical Sciences / 1055 Vol. 81, No. 1 1, November 1992
and a n endotracheal tube was inserted; the cuff was inflated prior to nicotine administration by random selection by anatomical site. The nicotine solution was injected into the lung from a catheter passed to the desired location via the suction channel of the bronchoscope (internal volume of catheter, 600 pL). A three-way stopcock was attached to the catheter, and to the stopcock were affixed 10-mL and 100-pL syringes. The catheter tip was placed in the nicotine solution and filled first with the 10-mL syringe after which the 100-pL syringe was filled. The tip was wiped off, and the catheter was inserted into the bronchoscope. For dosing, the 100-pL syringe was injected so that 100 pL are delivered a t the distal site. To confirm the accuracy of this method, several 100-pL volumes of nicotine solution were injected into a container and weighed, and the delivered volume was determined to be 100.2 ? 1.7 pL [mean ? standard deviation (SD)]. Distal Site-A fiberoptic bronchoscope (Olympus BF4B2) was inserted, and a lateral subsegment of the right middle lobe was identified. Because of differences in endobronchial anatomy among animals, the same subsegment may not have been used in every dog. A 1.7-mm 0.d. (5F) angiographic catheter (Multipurpose A-1, Cordis Corporation) was inserted through the suction channel of the bronchoscope. The catheter was advanced to a location immediately subpleural, and the position was confirmed by fluoroscopy (this position is referred to as “distal” in the remainder of the paper). Nicotine solution (5.0 mg1mL; 100 pL; total dose, 0.5 mg) was then instilled rapidly. The catheter was withdrawn through the bronchoscope, and then the bronchoscope was withdrawn. When the animals were sufficiently awake, the endotracheal tube was removed. In three dogs, after completion of the study and sacrifice of the dogs, the angiographic catheter was again passed to the distal location by using the technique just described. Radiographic contrast material (2 mL) was injected through the catheter. Figure 1 is a photograph of a representative radiograph produced in this manner. The photograph indicates that some of the contrast material reached the alveoli. Because we did not directly visualize all nicotine injection sites, we cannot be sure of the exact anatomical injection site. It is possible that some portions of the 100-pL nicotine aliquot were placed into terminal bronchioles, respiratory bronchioles, alveoli, or a combination of these sites in the various dogs. For this reason, we refer to this location as the distal site.
Figure 1-Photograph showing detail of a radiograph made after injection of 2 mL of radiographic contrast material through a catheter placed in a subpleural location of the right middle lobe, as described in the protocol. Alveolanzation of the contrast material confirms delivery to gas-exchange regions. The volume of nicotine solution delivered during the study was 100 pL. 1056 I Journal of Pharmaceutical Sciences Vol. 81, No. 1 1 , November 1992
Subsegmentnl Bronchus Site-On a different day, the same procedure was performed, and the right middle lobe of the lung was identified. The catheter tip was placed in the ostium of a lateral segment of the right middle lobe and advanced 2 cm (this position is referred to as “subsegmental” in the remainder of the paper).Nicotine solution (5.0 mg/mL; total dose, 0.5 mg) was then instilled rapidly. Because of differences in endobronchial anatomy among animals, the same segment may not have been used in every dog. In addition, the distance from the ostium of a segment to ita pleural surface varies markedly; therefore, absolute consistency in the site of deposition for this subsegmental location may not have been achieved in every dog. Tracheal Site-On a third day, the procedure was repeated, except that the tip of the bronchoscope was placed in the trachea and the tip of the angiographic catheter was positioned 10 cm proximal to the carina. The same volume of the same nicotine solution was instilled in this location. a fourth day, the dogs were Intravenous Administration-On prepared as described earlier, except that a second iv catheter (maintained by heparin flushes) was placed in a hind-leg vein. No endotracheal tube was put in place, and no bronchoscopy was performed. Nicotine solution (0.1 mglmL; 5 mL; total dose, 0.5 mg) was injected through the hind-leg vein over a period of 1 min. All dogs remained sternal and sedated for a minimum of 20 min, after which they were allowed to awaken and move about ad lib. No significant coughing was noted in any dog during instillation or subsequent blood draws. A minimum of 48 h elapsed between each of the four sessions of nicotine administration. Blood Sampling-Blood samples (10 mL per sample in vacuum tubes coated with lithium heparin) were obtained prior to nicotine administration and a t 2, 5, 7, 10, 15, 20, 40, 80, and 160 min aRer administration. For iv dosing, these times were from the beginning of the 1-min infusion. Blood was stored on ice during collection. ARer collection of all 10 samples, the blood was centrifuged at 2000 x g for 10 min, and the plasma was removed. Stability studies completed in our laboratory indicate that nicotine is stable when collected and centrifuged in this manner. The plasma was divided into two aliquota and stored a t -20 “C until analyzed. All samples were analyzed in duplicate. Nicotine Extraction-Nicotine in plasma was assayed by the method of Feyerabend and Russell18 with N-ethylnornicotine (ENN) as internal standard, as described by Jacob et al.10 Previous studies from this laboratory have used the same assay.8.20 Briefly, each plasma sample was thawed, and 2 or 3 mL of plasma was t r a n s f e d to a large screw-capped glass test tube. ENN (1.0 pg/mL) was added a t 25 uUmL of alasma followed bv 2 mL of 5 N NaOH and 3 mL of diethyl ether. The tube was capp&, mixed by vortex for 2 min, and then centrifuged for 10 min. The supernatant was transferred to a second tube and evaporated to -200 pL under nitrogen. To this concentrated supernatant, 100 pL of 2 N HCl was added. The tube was capped, mixed by vortex for 2 min, and centrifuged for 5 min. The ether layer was removed and discarded. Diethyl ether (500 4)was added, and the tube was mixed by vortex for 1 min and centrifuged for 5 min. The ether layer was removed and discarded. The residual ether was then removed by evaporation under nitrogen for 75 8. The remaining solution was transferred to a Dreyer tube, 200 pL of 5 N NaOH and 50 pL of n-butyl acetate were added, and the tube was sealed. The tube was mixed by vortex for 5 min and then allowed to stand for at least 1 h before analysis. A standard curve was prepared by using 3 mL of calf Berum to which had been added 0, 36,72, 108, and 144 pL of nicotine solution (1.0 CrglmL), with final nicotine concentrations in serum of 0, 12, 24, 36, and 48 ng/mL. ENN was added to these samples as described earlier, and extraction of nicotine was performed as for the dog plasma samples. Standard curves prepared by using calf serum and dog plasma were not significantly different; therefore, because of the availability calf serum and the volume required, all standard curves were prepared by using calf serum. Nicotine Assay-Nicotine was analyzed with gas chromatograph (Tracor model 560, Tracor Inc., Austin, TX) equipped with a nitrogen-phosphorus detector. For analysis, 4 pL of the top layer (n-butyl acetate) of the Dreyer tubes was injected into the chromatograph. The peak height ratio (peak height of nicotine:peak height of ENN) was determined for each standard-curve sample and plotted versus nicotine concentration in plasma. The peak height ratio was then calculated for each experimental sample, and the nicotine concentration was determined from the standard curve. Standard curves
were determined daily. The sensitivity of the assay is 0.1 ng/mL.17120 The mean intra- and interday coefficients of variation for the assay were 9 and 7%, respectively. Pharmacokinetic Analysis-Data of nicotine concentrations in plasma versus time were fitted to selected polyexponential equations by using the nonlinear least-squares fitting computer program ESTRIPZ1 and PCNONLIN (Statistical Consultants, Inc., Lexington, KY ). Pharmacokinetic parameters were calculated from computergenerated coefficients and exponents, as described by Wagner.= Each data set was best described by a two-compartment, open, linear model. The area under curve (AUC)of nicotine concentration in plasma versus time was determined up to the end of the data collection period and extrapolated to infinity (AUC,). Absolute bioavailability (F)from the three pulmonary administration sites was calculated by comparing the AUC, values after instillation at these sites with that after iv administration: F, % = LAUC, pulmonary site/AUC, iv route] x 100). Peak nicotine concentration in plasma (C,,,-) and the time to reach peak nicotine concentration Urnax) were determined by examination of the concentration in plasma-time curve. Clearance (CL) was estimated by dividing the administered dose by AUC. Statistics-Data were analyzed by one-way analysis of variance (ANOVA).= When the results of ANOVA were significant,the t test was used to evaluate differences between data s e t s . 2 4 All data are presented as mean 2 SD, and p < 0.05 was taken as the level of significance for all analyses.
ResuI ts Mean concentration-time curves and pharmacokinetic parameters for all routes of administration are shown in Figure 2 and Table I, respectively. Analysis of concentration-time data indicates that a biexponential function best describes nicotine disposition during the postdose period. No statistical differences were noted among AUC, and F (the parameters used for comparing bioavailability), half-life, and CL values for all routes of administration. C,,, values after instillation at tracheal and subsegmental sites were lower than after iv administration. The C,, after instillation at the distal site was not different from that after iv administration. In addition, t,, was higher after tracheal delivery compared with either iv or subsegmental delivery. The t,,, after instillation a t the distal site was not different from those after instillation at the other sites.
Discussion This work describes the pharmacokinetics of nicotine absorption from various levels of the respiratory tract. The results depend in part on injection of nicotine in the intended locations. Injection in the distal location was accomplished via fluoroscopic guidance. The alveolarization of constant media
0
20
40
60 Time (minutes)
Flgure 2-Mean concentration of nicotine in plasma (ng/mL) versus time (min) after introduction of nicotine by instillation at (0)tracheal, (W) subsegmental, and (A)distal sites or by (*) iv administration. (SDs are given in Table I).
Table CPharmacoklnetlc Parameters for Nlcotlne after lnstlllation at Three Pulmonary Sltes and Iv Admlnlstratlon'
Trachea Subsegmental Distal
iv
660 (132) 635 (197) 936 (371) 922 (340)
0.78 (0.23) 0.85 (0.41) 1.11 (0.51)
-
84.8 (36.6) 91.2 (29.0) 85.0 (56.5) 83.6 (33.3)
753 (159) 785 (214) 539 (268) 549 (199)
11.5 (4.4)'*d,e 18.2 (5.0)' 22.1 (6.2)
5.3
30.3
(3.0)C.d 2.0 (0.0) 2.6 (1.8) 2.0
(10.7)
(0.0)
All data are expressed as mean (SD). Half-life. 'Significantly different from result after iv administration (p < 0.05). dSignificantly different from result after instillation at subsegmental site (p < 0.05). Significantlydifferent from result after instillation at distal site (p < 0.05).
injected a t the distal location in several animals (Figure 1) suggests that the injection at the distal site was quite peripheral, as intended. It is possible that a portion of the nicotine instilled in the subsegmental zone moved either proximally or distally during the early part of the absorption period. The small volume instilled and the absence of significant coughing among the animals during the first 20 min after instillation should have minimized such problems. The absorption of nicotine from various levels of the tracheobronchial tree was compared with that from the iv route. C,,, and t,, after instillation at the distal lung region were very similar to those after iv administration. This result suggests that the pattern of nicotine absorption from the distal lung region is similar to that from the iv route (Figure 2). C,, values after instillation at tracheal and subsegmental levels were lower. Absorption from the trachea was slower, with t,, longer than that after iv administration or instillation at subsegmental sites. The t,,, was calculated as the mean of the t,, for each individual animal. However, because of the very high nicotine concentration in plasma in two dogs a t 2 min, the mean concentration was highest at 2 min (Figure 2). The amounts of drug reaching the systemic circulation from the four sites, as reflected by AUC, were statistically equivalent. This result suggests complete absorption of nicotine from each site. Alternatively, the difference between AUC values after instillation at tracheal and subsegmental levels and those after instillation at the distal level and iv administration may not have reached significance because of the large SDs of the results for the distal and iv sites or because of the small number of animals studied. The AUC values calculated by the pharmacokinetic programs used in this study are based on curves of level in plasma versus time, which the program extrapolates to infinity. The AUC value derived is thus very dependent on the nicotine levels in plasma obtained during the terminal elimination phase (20-160 rnin). Small amounts of imprecision in the measurement of the values might have a large effect on the AUC value calculated by PCNONLIN and, therefore, lead to the relatively large SDs seen for the AUC values. Wangensteen et al.3 and Taylor e t al.4 examined the absorption of chemicals with a n isolated, perfused mammalian lung. In these studies, the lungs were filled with a solution of the chemical of interest, and the rate of disappearance of the chemical from the solution was measured. This procedure allowed precise definition of the administered dose, the amount absorbed, and the rate of absorption. An important contribution of this technique is the comparison of absorption kinetics for different chemical substances. However, the ex vivo model makes extrapolation of the data to in vivo systems difficult, and the delivery of the chemical to the Journal of Pharmaceutical Sciences I 1057 Vol. 81, No. 1 1 , November 1992
entire lung does not allow measurement of differential absorption h m different levels of the lung. In other studies, the chemical of interest is delivered to intact lungs either by aerosoF-11 or by tracheal instillation.10-16 Studies of absorption after aerosol delivery suffer from deposition and absorption from the oropharynx when tracheostomies or endotracheal tubes are not used. Determination of the precise dose deposited is difficult, and cumbersome techniques are required to measure both the delivered and the exhaled quantities of the chemical. In addition, deposition occurs to varying degrees throughout the lung, a situation making the study of regional absorption difficult. Tracheal instillation, in which the chemical solution is instilled through an indwelling tracheal cannula, overcomes the difficulties of oropharyngeal absorption and allows delivery of a precise quantity of the chemical to the lung. Although some studies have shown that this method favors deposition in certain lobes,14 tracheal injection usually involves volumes sufficiently large that the injected solution spreads over a fairly large and poorly defined lung surface and possibly even settles at gas-exchange regions. Characterization of abaorption from specific levels of the lung is thus not possible. The current study represents an effort to measure quantitatively drug absorption from specific airway regions by direct deposition of the drug within these regions. The quantity deposited was well defined (0.5 mg), and the volume of solution was sufficiently small (100pL) to minimize spreading to distant lung regions. Total amounts of nicotine absorbed were similar from all three sites. Nicotine absorption from distal lung units was rapid and comparable with that h m an iv dose, whereas absorption from the tracheal region was slower.
References and Notes 1. Mayer, S.E.; Melmon, K. L.; Gilman, A. G. In The Phurmacological Basis of Thempeutics, 6th ed.; Gilman, A. G.; Goodman, L. S.; Gilman, A. Eds.; Macmillan, New York, 1980; pp 1-27. 2. Schanker. L. S. Biochem. Pharmacol. 1978.27.381385. 3. Wanqeneken, 0. D.; Wittmere, L. E.; Johneon, J. A. Am. J . Phvswl. 1969.216(4). 719-727. 4. Taylor, A. E.;. Guyton, A. C.; Bishop, V. S.Circ. Res. 1965, 16,
1058 I Journal of Pharmaceutical sciences Vol. 81, No. 7 7 , November 7992
353462. 5. Byron, P. R.; Roberta, S. R.; Clark, A. R. J . Phurm. Sci. 1986,75, 168-171. 6. Mentz, W. M.; Brown, J. B.; Friedman, M.; Stutta, M. J.; Gatzy J. T.; Boucher, R. C. Am. Rev. Respir. Dis. 1986,134, 938-943. 7. Brown, R. A.; Schanker, L. S. Drug Metub. Dispa. 1983, 11, 355-360. 8. Burch, S. G.; Erbland, M. E.; Gann, L. P.; Hiller, F. C. Am. Rev. Respir. Dis. 1989,140, 955-957. 9. Sepkovic, D. W.; Colosimo, S. G.; Axelrad, C. M.; Adams, J. D.; Haley, N. J. Am. J . Public Health 1986, 76, 1343-1344. 10. Schanker, L. S.; Mitchell, E. W.; Brown, R. A. Drug Metub. Dispos.1986,14,79-88. 11. Tmovec, T.; Durisova, M.; Bezek, S.; Kalla , Z.; Navarova, J.; Toncikova, 0.;Kettner, M.; Faltue, F.; Erichreb, M. Drug Metub. Dispos. 1984,12,641-644. 12. Brazzell, R. K.; Smith, R. B.; Kostenbauder, H. B. J . Phurm. Sci. i 9 8 2 , 7 i , 1268-1274. 13. Hemberger, J. A.; Schanker, L. S. Am. J . Physiol. 1978, 234, C191-Cl97. 14. Enna, S.J.; Schanker, L. S. Am. J . Physwl. 1972, 223, 12271231. 15. Enna, S. J.; Schanker, L. S. Am. J . Physiol. 1972,222,409-414. 16. Schanker, L. S.; Mitchell, E. W.; Brown, R. A. Phurmacology 1986,32, 176-180. 17. Dahl, A. R.; Felicetti, S.A.; Muggenburg, B. A. Fund. Appl. Toxicol. 1983,3, 293-297. 18. Feyerabend, C.; Rueeell, M. A. H. J.Pharm. Phurmacol. 1979,31, 73-76. 19. Jacob, P.; Wilson, M.; Benowitz, N. L. J . Chromtogr. 1981,222, 61-70. 20. Ebert, R. V.; McNabb, M. E.; McCusker, K. T.; Snow, S.L. JAMA 1983,250,2840-2842. 21. Brown, R. D.; Manna, J. F. J . Phurm. Sci. 1978,67,1687-1691. 22. Wagner, J. G. J . Phurmacokinet. Biophrm. 1976,4,443-467. 23. Glantz, S. A. In Primer o Biostatistics, 2nd ed.;Glantz, S.A.,Ed.; McGraw-Hill: New Yorl, 1987; pp 30-63. 24. Glantz, S. A. In PrimerofBiostatistics,2nded.; Glantz, S.A., Ed.; McGraw-Hill: New York, 1987; pp 64-100.
Acknowledgments We thank Cindy Ha e for secretarial assistance and Cristal McQuary and Cynthia f k b s , D.V.M., for technical assistance. We also thank Richard V. Ebert, M.D., for helpful advice and assistance with the nicotine assa and Paula Anderson, M.D., and Bill J. Gurley, Ph.D., for review of t i e manuscript.
