A Novel Dosing Method for Drug Administration to the Airways of the Isolated Perfused Rat Lung PETER
R. BYRON"AND RALPHw. NIVEN
Received August 7, 1987, from the Coll 8 of Pharmacy, University of Kentucky, Lexin ton KY 40536-0082. Accepted for publication Present addressegepartment of Pharmacy and Pharmaceutics,%irghia Commonwealth University, Richmond, VA 23298. February 25, 1988. Abstract 0A novel method is described for the reproducible administration of known liquid quantities to the peripheral airways of the isolated perfused rat lung. The basis of the technique was to use a 25-pL metered dose of fluorocarbon propellant to expel liquid (as a coarse spray) from an intratracheal dosing cartridge into the airways, while simultaneously inflating the lungs with a fixed volume of gas. The methodology is illustrated by administration of 100-pL volumes of aqueous disodium fluorescein solutions to a series of lung preparations. The reproducibility and regional distribution of dosing were determined by dissection, homogenization,and fluorimetric assay. Even though the dye was distributed nonuniformly between the lung lobes, in a series of preparations, 65.9 2 4.8% of the recovered dose was still deposited in the lung periphery, the site from which absorption is believed to occur. The method will enable the study of airway-to-perfusatetransfer kinetics for compounds administered in a variety of different liquid formulations.
Isolated lung system^^,^ have been used for some time in pharmacological^ and physiological investigatiorm6 Little emphasis has been placed on their capacity to give pharmaceutical information.2.4 Provided formulations can be delivered reproducibly to the airways, it should be feasible to use these systems to evaluate drug absorption and disposition in the lung (as functions of formulation), as well as to correlate in vivo7 with in vitro data. Dosing to the lung, both in vivo and in vitro, is normally performed by intratracheal injections-11 or by some means of aerosol inhalation.lO The former method suffers from an inability to reproducibly dose the deep lung because of nonuniform distribution of dose,lo difficulty in standardizing the rate of administration when delivered from a syringe, and the problem of widely varying degrees of expectoration of the dose. This final point, which occurs because of random bronchial and tracheal plugging by solution boli (followed by lung deflation), is a major obstacle to reproducible delivery by syringe. On numerous occasions, 100-pLintratracheal doses have been administered by syringe to horizontally positioned lungs prepared as described in this paper. The percentage of the dose reaching the lung lobes can vary from 40 to 90%. Aerosol delivery is more expensive, technically complicated,12 and very difficult to q~antify.~JS Many isolated lung systems employ vertical positioning of the lungs by suspending them from the trachea, which may then suffer from gravitational stretching. This nonphysiologic positioning, which may interfere with drug deposition and clearance, was avoided in this study. The development of a n inexpensive, simple, isolated perfused rat lung (IPRL) preparation was pursued with the objective of attaining validated and reproducible drug delivery to naturally positioned lungs.
Experimental Section Lung Preparation-Adult, male Wistar rats, weighing 250350 g, were anesthetized intraperitoneally with sodium pentobarbitone OO22-3549/88/08oO-0693$01.oO/O 0 1988, American Pharmaceutical Association
(60 mg/kg, 50 mg/mL; Nembutal, Abbott Laboratories, North Chicago, IL). The neck region was opened and a stainless steel cannula (length = 30 mm; 1.94 mm id; 2.29 mm od) was inserted 14 mm inside the trachea and tied in place. After opening the thorax at the diaphragm, the lungs were ventilated via the cannula (33 respirationdmin, 2-mL tidal volume). The rib cage was quickly severed on both sides and flapped backwards. Sodium heparin (0.1 mL; 1000 unitslml; Elkins-Sinn, Inc., Cherry Hill, NJ) was injected directly into the lumen of the right ventricle, and one loose ligature was placed around the pulmonary artery (PA) and aorta. Bubble-free perfusate at 37°C (Krebs-Henseleit buffer with 4% w/v bovine serum albumin [K4])was introduced through a blunt, 16G stainless steel-tipped cannula (length = 19 mm) that was inserted into the PA (but not to the point of the arterial bifurcation which hinders blood clearance) via an incision in the upper right ventricle. The cannula was clipped in place and the left ventricle and atrium were severed immediately to enable blood and perfusate to be pumped freely (15 mllmin; Masterflex peristaltic pump, type 7520-00, Cole-Parmer Instrument Company, Chicago, IL). Positive pressure ventilation was halted while the lungs remained partially inflated. The PA ligature was tightened and the clip was removed. The lungs were excised, washed with 37 "C K4, and a steel rod (length = 11 cm, diameter = 1.65mm) was slipped through the esophagus. The lungs were suspended horizontally in a jacketed glass thorax (volume = 1000 mL, 10.5 cm id) at 37°C (Figure l), with the tracheal exit cannula (length = 78 mm, 2.37 mm id, 2.82 mm od) fitting snugly around the tracheal cannula; this joint was sealed with a short piece of silicone tubing (Figure 2). No bends existed from the trachea, via the connections, to the external atmosphere. The whole procedure took no more than 8 min. Residual blood components were washed from the preparation for a further 5-10 min, after which the thorax was sealed and K4 was recirculated from a reservoir of 200 mL (Figure 1).
C
B
D
+
A-
J
F
Figure 1-Schematic ofthe isolatedperfusedlung preparation. Key: (A) carbogen supply; (6)peristaltic pump; (C) bubble trap; (D) syringe (20 mL); (E) water-jacketedglass thorax; (F) air removal trap; (G)waterjacketed perfusate reservoir; (H) automatic temperature probe; (I)pH electrode; (J) K4 perfusate. Journal of Pharmaceutical Sciences / 693 Vol. 77, No. 8, August 1988
Quantities of dye were determined in each of six separate tissue regions (Table I; the nontissue components, tracheal cannula, tracheal exit cannula, and residual fluid of dissection were also assayed), and their sum was compared with the administered dose. Because the lungs consist of a series of lobes, relative lobar concentrations (RLC) were determined, in fashion similar to the method of Raabe et a1.16 from:
I
II A
81
RLC C
D
i
F
n
Ill
=
(% of administered dose in lobe) (weight of individual lobe/total weight of lobes) x 100
U
Results and Discussion
Figure 2-The three stages in dosing the airways: (I) loading the cartridge;(11) insertion into the tracheal cannulae;(111) sealing the system and administering the dose. Key: (A) stainless steel dosing cartridge;(8) silastic tubing; (C) tracheal exit cannula (in artifical thorax, Figure 7); (D) tracheal cannula; (E) rat trachea; (F) metered dose inhaler; (G) silastic tubing connections sealing over the tracheal exit cannula; (H) dosing cartridge A, within tracheal cannulae, C and D, after expulsion of dose.
Dosing the Airways-The means of administering known fluid doses to the IPRL is shown schematically in Figure 2. The basis of the method was to simultaneously expel a coarse liquid spray into the airways and inflate the lungs with a fixed volume of gas. A dosing cartridge, A (Figure 2; length = 105 mm, 0.89 mm id, 1.64 mm od), was connected to a 60-mm length of silicone tubing, B1 (2 mm id, Silastic, Dow Corning, Midland, MI), filled with 0.1 mL of solution, and weighed (weighing was repeated after dose administration). The cartridge was slipped through the tracheal exit cannula, C, and tracheal cannula, D, until its tip just projected into the rat trachea, E (Figure 2, Stage 11). Part of the flexible tubing, B1, was pushed 3 mm over the outside of the exit cannula, C, so that it formed a gas tight seal on tubes A and C (Figure 2, Stage 111). A glass (Wheaton Glass, Mays Landing, NJ) metered-dose inhaler (MDI), F, fitted with a 25-& per actuation, inverted metered-dose valve (Valois DF 10, BLM Packaging, Inc., Greenwich, CT) and actuator (Valois 251/407/75 mm) was connected to the liquid-free end of the dosing cartridge. A propellant blend containing 40%(w/w) fluorocarbon 11 and 60% (w/w) fluorocarbon 12 (Dymel 11 and 12, Dupont, Wilmington, DE), with a vapor pressure of 43 psig, was employed for all the experiments reported here. The MDI was actuated so that 25 pL of propellant vaporized, thereby expelling the liquid in the dosing cartridge deep into the airways and inflating the lungs reproducibly. (The total vapor volume from this propellant blend was 6.1 mL. Reduced lung inflation could easily be accomplished if necessary by employing an additional vapor reservoir between the MDI and dosing cartridge.) One second after depressing the valve, the dosing cartridge was removed completely, allowing the lungs to deflate through the tracheal exit cannula, C. Dose Distribution and Reproducibility-Sixteen separate lungs were prepared as described above. A dose of 0.1 mL of disodium fluorescein (Fisher Scientific, Fair Lawn, NJ) was administered to each lung preparation, and perfusion was stopped immediately. The lungs were removed and dissected prior to fluorimetric assay.14
Dosing the Airway-A reproducible, 89.4 f 5.0%of the contents of each dosing cartridge was expelled as the "administered dose" into the IPRL. The average recovery of the administered dose was 94.6 5 2.9%.There was no evidence of significant adsorptive or binding losses and, at these doses, dilution and centrifugation of tissue homogenates obviate a need to allow for fluorescence quenching effects. However, over 25% (25.7 f 5.4%) of the dose leaving the dosing cartridge (Figure 2) was deposited in the nontissue components (tracheal cannulae) of the preparation, showing that some 25 pL of each 100 pL dose was refluxed along the tracheal cannulae as the cartridge was being withdrawn and the lungs were deflating. It is important to understand that solutions are not expelled from this dosing cartridge as fine droplet aerosols. Actuation with the cartridge tip outside of the lungs produced an apparently log-normal droplet size distribution, with a count median diameter of 350 pm and a geometric standard deviation of 2.0. With the tip inside the trachea, liquid subdivision into droplets must, of course, be constrained further. Table I gives the mean percentage of the recovered dose deposited in each of the lung regions. Anatomically, the rat lung consists of a single left and four right lobes. The dissection performed in this study separated the trachea and major bronchi from each lobe. Thus, in terms of airway distribution, each lobe contained minor conducting airways and alveoli. The majority of the dose was found in the left and the right inferior lobes. As indicated by their percentage weights (percent of total lobe weight), they constitute the largest gas exchange components of the rat lung. With the IPRL in a horizontal (Figure l), and therefore natural position, the dose distribution shown in Table I compares favorably with that presented by other workers for solution and suspension administration to live anima1s.l0J6 Dose Distribution and Reproducibility-Although the percentage of each dose reaching the lungs was reproducible (interlung coefficient of variation = 7.3%),intralung variations in dye distribution were pronounced. In Figure 3, where the data is presented as relative lobar concentrations, it is clear that fluorescein distribution was influenced by more than just the variation in lobe mass. A value of RLC = 1 (eq 1)would be expected if the distribution was uniform. Brain et 81.16 and Pritchard et 81.10 found a similar uneven distribution for intratracheally administered suspensions in vivo.
Table I-Percentage of Recovered Dose Delivered to Different Lung Components' PCb RS RMd
Yo Dose recovered" Yo Total lung weight'
(1)
11.7 f 3.9 6.7 f 0.6
5.3 f 2.6 6.6 f 0.6
3.1 2 1.9 7.9 ? 1.4
RI"
L'
Rg
22.1 f 6.3 14.4 2 1.5
23.7 +- 7.0 15.7 I 1.8
7.0 f 1.9 48.8 -1- 3.0
a Data presented is the mean f standard deviation. Right postcaval or median lobe. =Right superior or apical lobe. dRight middle lobe. " Right inferioror diaphrammatic lobe. 'Left lung. Remaining tissue components (fluorescein located primarily in trachea and major bronchi). " A total of 65.9 2 4.8% of the recovered dose was deposited in the lung lobes (PC + RS + RM + RI + L). 'Homogenized tissue dried at 100 "C for 24 h; (dry tissue weighvtotal lung weight) x 100.
694 / Journal of Pharmaceutical Sciences Vol. 77, No. 8, August 1988
syringe. It is this reliability, for drug delivery to the lobes, which is of most importance pharmaceutically. If this preparation is to be useful for screening different dosage forms for determination of, for example, relative rates of drug absorption, then reliable dosing must be accomplished prior to observing transfer. While differences in pulmonary epithelial permeability probably exist between the conducting airways and the alveolar regions,lg there is no evidence for a signifi cant variation in permeability between different lung l0bes.4-7 Furthermore, because of the incomplete bronchial circulation in these preparations, absorption into the perfusate is believed to occur only from the peripheral airways,4 which are predominantly in the lobes. Consequently, the method described here is capable of administering a range of different liquid formulations in known doses to the IPRL and can be used to facilitate investigation of the effects of different drugs and excipients on drug deposition and transfer in the lung. With some additional validation, this technique can be taken one step further by eliminating the dosing cartridge and directly administering drugs from a MDI.
References and Notes RM
pc
L
Flgure 3-Mean
relative lobar concentrations in the different lung regions (eq 1). Error bars are standard deviations. Significant differences occurred between (PC,Rl, L) and (RM, RS), with p < 0.05, as performed by Duncan's new multiple range test. Key: (PC)post cavat lobe; (RS) right superior lobe; (RM) right middle lobe; (RI) right inferior lobe; (L) left lung.
1. Niemeier, R.W. Environ. Health Pers 1984,56, 35-41. 2. Mehendale, H. M.; Angevine, L. S.; O k i y a , Y. Toxicology 1981, 21, 1-36. 3. Levey, S.; Gast, R. J. A pl. Physiol. 1966,21, 313316. 4. Byron, P. R.; Roberts, S. R.; Clark, A. R. J.Phurm. Sci. 1986, 75, 168-171. 5. Niemeier, R. W.; Bingham, E. Life Sci.1972, IlfPt. ZZ), 807-820. 6. Fisher, A. B.; Dodia, C.; Linask, J. Exp. Lung Res. 1980,1,13-
Pf
21.
Dubaybo and Thetl7 were able to obtain a uniform distribution of Evans blue dye in the right lung when administration was restricted to the right bronchus. To achieve this, the position of the rat, the dose volume, and the rate of administration required careful control.l7 By employing fluorocarbon blends with different vapor pressures and the methods described here, it is possible to exercise some degree of control over the rate of liquid expulsion and lung inflation. In these investigations, the rates were held relatively constant, as defined by the evaporation rate of the propellants. However, it is known that preferential inflation of different lobes occurs on in vivo inhalation, and regional differences in gas trapping within excized lungs probably requires different transpulmonary pressures to ventilate them.'* Thus, despite careful IPRL positioning and the use of the MDI, the variation shown in Figure 3 seemed inevitable. Even so, because the dose was metered, the percentage of each dose reaching the lobes of the IPRL (67.3 2 5.1%)could be predicted with far more certainty using the method described here than was the case with either aerosol administration4 (where deposition may have been more uniform) or with instillation from a
7. Ci&k, A. R.; B on, P. R. J. Phurm. Sci. 1985, 74, 939-942. 8. Kettle, E. H.; &ton, R. Lancet 1932,1, 1190-1192. 9. Cember. H.: Hatch. T. F.: Watson, J. A.: Grucci. T. B. Arch. Znd. Hve. Occ. Med. 1954.10; 124-129. '
T. T. Arch. Znt. Med. 1973, 13. Brown,.R. A.; Schanker, L. S. Drug Metub. Disp. 1983,11,355360.
14. Clark, A. R.; Byron, P. R.; Groom, C. V. J. Pharm. Phurmucol. 1981,33, 39P. 15. Raabe, 0. G.; Yeh, H. C.; Newton, G. J.;Phalen, R. F. InZnhuZed Particles, ZV; Walton, W. N., Ed.; Pergamon: U.K., 1977; pp 3nn az.
16. Brain, J. D.; Knudson, D. E.; Sorokin, S. P.; Davies, M. A. Enuimn. Res. 1976,11, 13-33. 17. Dubaybo, B. A.; met, L. A. J.Appl. Physiol. 1985,59,266-268. 18. Morgan, J. J.; Franz, D. N.; h a z e r , D. G. Resp. Physio2. 1984, 55,309316. 19. Ei€ros, R. M.; Mason, G. R. Am. Rev. Resp. Dis.1983,127, S59S65.
Acknowledgments This work was supported in part by the University of Kentucky Medical Center Research Fund.
Journal of Pharmaceutical Sciences / 695 Vol. 77,No. 8,August 1988