THE ANATOMICAL RECORD 196:313-321 (1980) Morphology of the Guinea Pig Respiratory Tract JAY P. SCHREIDER AND JOHN 0. HUTCHENS Department of Pharmacological and Physiological Sciences, University ofChicago, Chicago, Illinois 60637 ABSTRACT A morphologic description of the airways of the guinea pig was developed from measurements of casts of the lungs and nasal cavity and from measurements of frozen sections of the lungs. The lengths, diameters, branching pattern, and numbers of elements of the respiratory tract formed the basis for a representative model of the system. The branching pattern is irregular to the pulmonary region but regularly dichotomous thereafter. The nasopharyngeal-tracheobronchial region contributes 2.64 cm3of the total respiratory volume of 21.62 cm". The alveoli contribute 16.31 cm:' of the 18.98 cm' pulmonary region. The nasal region consists of convoluted and irregular airways with a functional volume of 0.48 cm3. Small rodents, including the guinea pig, are commonly used in inhalation toxicology research. The toxicity of an inhaled aerosol depends partially on the extent and loci of deposition in the respiratory tract. The disposition depends on particle characteristics (size, density, and shape), on physical characteristics of the respiratory system, and on the kinetics of gas flow within the system. Differences in deposition characteristics between humans and experimental animals will influence extrapolation of toxicity results obtained with small animals to humans. This study was performed to contribute to a n understanding of these factors. Findeisen ('35) derived equations describing particle deposition in the lungs by impaction, sedimentation, and diffusion. These equations, along with a model of the human respiratory tract, were used to calculate the deposition characteristics for man. That simple model assumed regular dichotomous branching of the airways. Landahl ('50, '63) derived a different set of equations governing particle deposition. Using these equations, along with a modified form of Findeisen's model, Landahl recalculated particle deposition characteristics for the human respiratory tract. Weibel ('63) also assumed dichotomous branching when developing a detailed model of the human respiratory system. Krahl ('64) and Davies ('61) presented models in which dichotomous branching was not assumed for the whole lung and the presence of side branching was considered. The structure of the respiratory systems and particle deposition characteristics have been established for only a few experimental animals. Hutchens and Sharp (unpublished data) developed a schematized model of the respiratory tract of the rat. Based on this model, particle deposition characteristics were calculated for the rat (Granito, '71). Raabe et al. ('71) presented measurements of the tracheobronchial portions of the respiratory tract of two rats, two beagle dogs, two humans, and one hamster. Kliment et a1 ('721,from latex casts of guinea pig lungs, measured lengths, diameters, and branching angles of the elements. Based on those measurements, a model of the respiratory tract was derived and the deposition characteristics were calculated. However, their results differed significantly from the results in this study and with the work of other authors (Tenney and Remmers, '63;Forrest and Weibel, '75). This study was part of a comprehensive effort a t the University of Chicago to establish the aerosol deposition characteristics of a number of laboratory animals commonly used in toxicology studies. Through the use of epoxy casts of the respiratory system and examination of the alveolar regions of fast frozen sections of the lungs, the morphology of guinea pig airways is reexamined. Based on these measurements, a 000-3276W80/1963-0313$01.700 1980 ALAN R. LISS, INC. Jay P. Schreider's present address is Radiobiology Laboratory, University of California, Davis, CA 95616, (916) 752-6918. Received May 15, 1979; accepted August 20, 1979. 313 314 JAY P. SCHREIDER AND JOHN 0. HUTCHENS model of the respiratory tract is presented. The anatomical nomenclature used is taken from Cooper and Schiller (’75). MATERIALS AND METHODS Guinea pigs of an unspecified strain, weighing about 600 g, were killed by a n intraperitoneal injection of 300 mgkg of sodium pentobarbital (USP “injection,” 5% solution). The animals died within five minutes of injection and did not exhibit any obvious respiratory trauma. Reflex bronchial constriction, a major problem in preparing guinea pig lung casts, was prevented by exposing the guinea pigs to an aerosol of epinephrine, generated by blowing oxygen through a modified DeVilbis type nebulizer containing an epinephrine solution (USP “injection”1mg/cc),from one minute preceding pentobarbital injection until respiration ceased. The effectiveness of the epinephrine pretreatment in preventing bronchial constric- tion was demonstrated by the observed uniformity of the resulting casts. Casts made from the epinephrine-treated animals showed no obvious differencesfrom the few satisfactory casts or portions of casts made from untreated animals. After death of the animal, the trachea was exposed and cannulated. The cannulation tube was connected to a n air pressure head of 15 cm of water. The chest cavity was opened (lungs still under pressure) and the lungs and heart removed by severing the mediastinal pleura, the esophagus, and the blood vessels leading to the heart (with the exception of the pulmonary vessels). The pressure of 15 cm water caused the lungs to just fill the opened (but not distorted) chest cavity and the edges of the lungs to overlap the heart. The lungs, still under pressure, were immersed in a saline solution (leaky specimens were discarded). Lung casts (Fig. 1) were prepared by with- Fig. 1. Undissected epoxy cast of guinea pig lung. Dorsal view. GUINEA PIG RESPIRATORY MORPHOLOGY drawing a volume of air through a syringe attached to the clamped tracheal cannula and then injecting a n equal volume of casting medium over a two minute period. The casting compound was prepared by mixing a two-part epoxy resin (No. 6201, High Strength Plastics, Chicago, IL), which was put under vacuum for 1-3 minutes to remove air bubbles. For some preparations, Lucite beads about 200 pm in diameter (separated by sieving from Lucite 40 Bead Polymer, E. I. Du Pont De Nemours and Co., Polychemicals Department, Wilmington, DE) were mixed 1:2 with the resin. The beads were used to prevent the resin from filling many of the smaller airways. After casting, the lungs were hung vertically, supported by the tracheal cannula, in a moist chamber overnight. The tissue then was digested in a bath of hot 2 N NaOH. The casts were washed in water and allowed to dry. A few lung casts were prepared by allowing epoxy resin t o flow into evacuated lungs supported by an artificial thorax. The artificial thorax was formed from W o agar poured into a paper cup mold containing the inflated lung specimen. The agar hardened around the buoyant lung, and only the clamped tracheal cannula, extending through a paraffin-sealed hole in the bottom of the cup, was exposed. The lung was then evacuated through the tracheal cannula with a water aspirator, a syringe containing epoxy resin was attached to the cannula, and resin was allowed to flow into the lung and harden overnight. After removal from the agar, the lung tissue was digested in the usual way. Measurements were made on the larger branches where distension might be expected to occur during injection under pressure. Casts prepared in this way were identical to those prepared by the injection method, indicating that the method (and injection pressures) did not cause distended or stretched branches. Lung casts were dissected under a binocular dissecting microscope. Beaded casts were used for studying the larger airways because the beads prevented filling of many of the smaller airways and thus simplified the dissection. Branches 200 microns and smaller that were filled with resin were removed, leaving a cast such as that pictured in Figure 2. The dissected fragments were saved for study of the smaller airways. When unbeaded resin was used, casting of the smaller airways was more frequent. Alveolar bundles (Fig. 3) were isolated from such casts. Several dozen casts were made. The best casts were dissected. One beaded cast was 315 completely dissected and measurements made on all lobes. A number of casts were partially dissected. Measurements made on the dissected portions served to clarify some measurements made on the completely dissected cast. Lung elements were measured with a linear micrometer occular on a dissecting microscope. The measurements were accurate to within i 50 microns under x 10 magnification, i 16 microns under x 30 magnification (used for most measurements), and ? 8 microns under x 60 magnification (used for pulmonary region measurements). All ducts through three orders of branching were measured, but measurements for the remaining orders were based on a random sampling of at least ten elements of each order. Lengths and diameters of ducts were recorded. Duct diameters were measured a t the midpoint. The size and number of ducts branching off a given element also were recorded. The largest diameter ducts branching from an element were designated as the next smaller order. When regular dichotomous branching began, the branching orders became obvious. The branching angle was defined as the angle between the axes of the parent duct and the daughter branch. A branch was designated as a side branch if the axis of the major daughter did not appreciably change in direction from the axis of the parent duct. A branch was designated as an end branch if the axes of both daughters changed direction from the axis of the parent duct. The angles were estimated through the dissecting microscope; however, orientation problems made exact measurement difficult. In addition, the branching angles of the larger branches may have been somewhat distorted by hanging the lungs by the trachea during curing. Therefore, an overall estimate was made of the angle for a side branch and for a n end branch. Although they are extensions of the primary bronchi, the left and right posterior bronchi were treated as branches (branching angle of 0")of the first order bronchi and were grouped with the other second order bronchi. Therefore, the lengths of the first order bronchi were measured from the tracheal bifurcation to the point where the first side branch was given off in the posterior lobes. This resulted in seemingly short lengths for the first order bronchi, but the model was easier to treat. Measurements of the smaller elements were made on many isolated fragments of the unbeaded casts. Measurement of the length of the alveolated ducts was complicated by orienta- 316 JAY P. SCHREIDER AND JOHN 0. HUTCHENS Fig. 2. Dissected cast of guinea pig lung. Dorsal view. Fig. 3. Fragment of alveolar bundle. GUINEA PIG RESPIRATORY MORPHOLOGY 317 which supply the cranial, middle, and accessory lobes, and then terminates with the right caudal lobe. The left first order bronchus (0.20 cm diameter) supplies branches to the left cranial and the left middle lobes, and then terminates in the left caudal lobe. The lobes are classified on the basis of their air supply. A section of the lung is considered to be a lobe if it is supplied by a second order bronchus. This treatment is unambiguous except in the case of the right cranial and middle lobes. Although the second order bronchi supplying these lobes are separate, there are tissue connections between the two lobes. The pictures of a nasal cast (Fig. 4a, b) show that the right and left passages are separate for about 25 mm. There is a bend of approximately 40"at the external nares. The sinus portion of the cast forms several dead pockets. The volume of the entire nasal cavity is 0.883 cm", while the volume of the nasal cavity minus the sinus or ethmoid area is 0.483 cm'. The measurements are summarized in a schematization of the respiratory tract presented in Tables 1and 2. Table 1illustrates the branching pattern of the ducts from the second order bronchi through the alveolated ducts. Smaller airways have fewer side branches. The branching pattern is irregular to the level of the 0.014 cm ducts. Three 0.009 cm ducts branch from each of the 0.014 cm ducts. The branching pattern is regularly dichotomous thereafter. The branching angle is 15" for the tracheal bifurcation. For the rest of the system, the angles are about 30" for an end branch and 60" for a side branch. As a duct gives off side branches, it decreases in diameter. From the branching pattern and the anatomical measurements, a model of the lung (Table 2) is generated in which a duct is placed in a given order according to its diameter and not its origin. Duct diameters did not occur as a continuum of sizes, but fell into fairly well defined ranges. Therefore, the diameters were recorded as grouped data with the midpoint of the group being listed in the tables. Because of their small diameter, the right cranial and left middle bronchi were grouped with the third order bronchi. With the exceptions of the fourth order alveolated duct and the second order bronchi, the volume of each element decreases with each order of branching. The volume of an order or region is a function of the number and volume of the elements in that region. RESULTS The alveoli are closely packed on all alveolThe guinea pig lung is divided into seven lobes. The right first order bronchus (0.23 cm ated elements, suggesting the absence of respihrnnchinles. - -._- - - ~ ~ -- ~ - Because the surface area of diameter) branches t o three secondary bronchi, rat,nrv tion illusions. Since the alveoli were closepacked on the surfaces of the alveolated ducts, the lengths of the airways were calculated by multiplying the number of alveoli along the lengths of the ducts by the alveolar diameter. In some cases, it was possible to count the total number of alveoli on a duct. The alveolar number was also estimated by calculating the number of spheres of a given radius that could be packed on the surface of a duct. There was good agreement between the two methods of estimation. Because the alveolar casting often was incomplete, alveolar dimensions were also estimated from frozen sections of lungs. Lungs were removed under pressure in the same manner used in preparing the casts, and quickly frozen in liquid nitrogen. Such fast freezing did not collapse the lungs because of the immediate formation of a frozen shell. Thick sections were taken from the frozen tissue and examined in a cryostat under a dissecting microscope. Because the alveoli were sectioned at random distances from their centers, the largest diameters found were accepted as the diameters of the alveoli. This method assumes that the alveoli are uniform in size in the frozen specimen. Examination of the alveolar bundle fragments supported this assumption. For the preparation of nasal casts (Fig. 41, the anterior end of the trachea was cannulated and epoxy resin injected towards the cranial region. After the mouth had filled, the jaws were tied shut. The head was suspended vertically with the nose pointing up, and resin was injected until it began to emerge from the nares. After the resin hardened, loose tissue was stripped from the head, and the remaining tissue was digested in hot 2 N NaOH. Pieces of bone adhering to the cast were digested with 1N HCL. The length of the nasopharynx on the nasal cast was measured from the choanae to the pharynx, using a micrometer occular and a dissecting microscope. The shape of the duct varied along its length, so a vernier caliper was used to measure twenty different diameters along the length. Acrylic molds (reverse casts) of the nasal cavity were prepared (Hutchens and Schreider, manuscript in preparation) and the volume of the nasal region measured from the amount of water needed to fill one of the molds. - J 318 JAY P. SCHREIDER AND JOHN 0. HUTCHENS A B Fig. 4. Epoxy cast of nasal cavity. (A) side view, (B) dorsal view. the alveolated ducts is constant, the number of alveoli per duct remains about t h e same throughout the pulmonary region. Total respiratory volume is 21.62 cm”. The dead air space (nasopharyngeal-tracheobronchial region) has a volume of 2.64 cm”. The pulmonary region has a volume of 18.98 cm:’, the majority of which is contributed by the alveoli (16.31 cm”). Examinations of the epoxy casts and the frozen sections indicated that the alveoli have the shape of a % sphere or a polyhedron (minus one side occupied by the alveolar opening) with a n average diameter of 80 microns. As the alveoli inflate, the shape may more closely approach that of a sphere with a narrow-necked opening to the duct. All alveoli are assumed to be in the same state of inflation. DISCUSSION Anatomists classify the airways of the lungs according to the amount of smooth muscle and cartilage contained in the walls and according to the types of cells lining the walls. In the classification presented here, histological make-up of the elements is not considered, and 3 R.C. = Right Caudal. R.Cr. R.C. bronchus (0.20) R.Cr. bronchus (0.11) R.M. bronchus (0.13) R.A. bronchus (0.15) L.C. bronchus (0.15) L.Cr. bronchus (0.15) L.M. bronchus (0.11) 0.10 0.065 0.050 0.040 0.030 0.023 0.014 0.009 0.007 0.006 0.006 Parent duct* (diameter, cm) = 3.0 1.0 1.0 1.0 4.0 0.10 = 2.6 1.7 3.0 3.0 7.0 Right Middle. R.A. 6.0 2.0 3.0 3.0 5.0 6.0 (no side branches) 3.0 6.0 Right Cranial. R.M. -. = Right Accessory. L.C. 1.0 3.0 3.0 2.6 2.4 1.0 1.1 3.0 4.0 4.0 4.0 4.0 4.0 1.9 1.3 3.0 3.0 = .- of = 0.10 0.10 0.10 0.10 0.10 0.065 0.065 0.065 0.050 0.040 0.030 0.023 0.014 0.009 0.007 0.006 0.006 0.005 0.005 0.065 0.10 0.065 0.065 0.065 0.065 0.050 0.050 0.040 0.030 0.023 0.014 0.014 0.009 0.007 0.006 0.006 0.009 Average diameter of ducts branching off end of duct of origin (cm) guinea pigs Left Cranial. L.M. = Left Middle. 2.0 2.0 2.4 1.0 0.014 Left Caudal. L.Cr. 2.0 1.6 1.0 Average no. ducts of given diameter branching off side of duct of origin diameter of branching duct (cm) 0.065 0.050 0.040 0.030 0.023 TABLE 1 . Branching pattern of ducts in the respiratory tract e 0 0 $ 8z z @ cn M 8a 3> 320 JAY P. SCHREIDER AND JOHN 0. HUTCHENS TABLE 2. Schematic representation of respiratory tract Region Diameter of duct or alveolus (cm) Cross-sectional area (crnl) of duct Nasal cavity Nasopharynx Trachea 1st order bronchi 2nd order bronchi 3rd order bronchi 4th order bronchi 5th order bronchi 6th order bronchi 7th order bronchi 8th order bronchi 9th order bronchi 1st alveolated duct 2nd alveolated duct 3rd alveolated duct 4th alveolated duct 5th alveolated duct Alveoli 0.32 0.30 0.22 0.16 0.10 0.065 0.050 0.040 0.030 0.023 0.014 0.009 0.007 0.006 0.006 0.005 0.008 7.9 x 10-2 7.1 x lo-' 3.7 x 10-2 2.0 x 10-2 7.9 x lo+ 3.3 x lo-:' 2.0 x lo-:' 1.3 x lo-:' 7.1 x 4.0 x 1.5 x x x x x x 6.4 3.9 2.8 2.8 2.0 the ducts are classified only according to their diameters and the presence or absence of alveoli on their walls. During nasal deposition studies (Hutchens and Schreider, manuscript in preparation), no material was ever recovered in the ethmoid area. Because the ethmoid area forms dead pockets, any material entering it should be deposited there. The conclusion is that very little or no tidal air enters the ethmoid area and that the ethmoid area is not part of the functional volume of the respiratory system. In this model of the lungs, all members of the same size are treated as members of the same order with identical paths from the trachea through the various branches and branching angles to that order. Examination of the lung cast shows that the anatomical paths to the various pulmonary regions may be quite different. However, the scheme presented is based on average values and should be a realistic model of the respiratory system. Tenny and Remmers ('63) related lung volume to body weight, and alveolar diameter and surface area to metabolic rate for several laboratory animals. Computations based on those formulas yield values of 25.3 cm' for total lung volume, 76.8 pm for the alveolar diameter, and 0.79 m2for total alveolar surface area. Forrest and Weibel ('75) estimated the pulmonary diffusion capacity of guinea pigs. From their graphs, extrapolations can be made to yield values for a 600 g guinea pig of 21.5 cm" total - lo-* 10 10-5 10-5 lo-' 10-5 of guinea pig Length (em) of duct 2.5 4.5 1.2 1.9 0.85 0.53 0.26 0.19 0.10 0.066 0.045 0.024 0.024 0.024 0.024 0.032 vol. of duct or alveolus (cm0 0.48 0.20 0.32 0.042 0.037 6.6 x lo-' 1.8 x 5.1 x 1.9 x 10-4 7.1 x 2.6 x lo-' 7.0 x 1.6 x 9.2 x 10-7 6.8 x lo-' 6.8 x 10 6.3 x lo-' 2.0 x 10-7 no. of ducts or alveoli 1 1 1 2 5 17 86 312 1.0 x 103 3.0 x 10' 9.5 x 10' 4.2 x 104 1.3 x lo5 2.5 x 105 5.0 x 1oj 1.0 x 108 2.0 x 10" 8.2 x 107 lung volume, 15 cm3alveolar air space volume, and 1.3m ' alveolar surface area. Data from this study show values of 21.6 cm'j total lung volume, 16.3cm" total alveolar volume, and 1.1m2 alveolar surface area (80 pm alveolar diameter). Kliment et al. ('72) presented a model of the respiratory tract of the guinea pig based on measurements of latex corrosion casts, but their values (extrapolated to a 600 g animal) for number of alveoli (6.8 x lo6),alveolar diameter (92 pm), calculated alveolar surface area assuming spherical alveoli (0.18 m2), and total lung volume (9.11 cm:')are quite different than the values of this paper, or values reported by Tenney and Remmers ('631, and Forrest and Weibel('79, and are not consistent with established relationships correlating alveolar surface area and lung volume with respiratory requirements and body weight. Kliment et al. ('72) also list fewer orders, fewer members of each order, and smaller dimensions of the elements of each order than were found in this investigation. Incomplete casting could have resulted in low numbers of alveoli and distal airways, leading to a low lung volume. In addition, the treatment of the nasal cavity as a straight cylinder is not appropriate for the convoluted and irregular shape of the nasal casts (Fig. 4). The model presented in this paper forms the basis for calculations of the aerosol deposition characteristics of the guinea pig using the di- 32 1 GUINEA PIG RESPIRATORY MORPHOLOGY TABLE 2lcont.) Schematic representation of respiratory tract of guinea pig. Region VOl. of region (cm") Cumulative vol. of ducts (cm') Nasal cavity Nasopharynx Trachea 1st order bronchi 2nd order bronchi 3rd order bronchi 4th order bronchi 5th order bronchi 6th order bronchi 7th order bronchi 8th order bronchi 9th order bronchi 1st alveolated duct 2nd alveolated duct 3rd alveolated duct 4th alveolated duct 5th alveolated duct Alveoli 0.48 0.20 0.32 0.084 0.19 0.11 0.15 0.16 0.19 0.21 0.25 0.29 0.19 0.23 0.34 0.68 1.24 16.31 0.48 0.68 1.00 1.09 1.27 1.39 1.54 1.69 1.89 2.10 2.35 2.64 2.83 3.06 3.40 4.07 5.31 21.62 ameters, lengths, and numbers of the ducts of the respiratory tract; the branching angles and branching pattern of the ducts; the size, shape, and number of the alveoli; the extent of alveolation of the ducts; and the physics of aerosols and airflow (Schreider and Hutchens, '79). These calculations are used to predict regions of the respiratory airways in which particles of a given size or distribution may deposit in inhalation toxicology studies. The results of these calculations are generally in line with the results of exposure studies on man and various experimental animals (Schreider and Hutchens, '79). Alveoli/ duct Alveoli/ region Alveolar vol. (cm% region 23 21 19 19 23 2.88 x 106 5.41 x lo6 9.46 x lo6 1.89 x loG 4.55 x 10" 0.57 1.08 1.88 3.76 9.05 Davies, C.N. (1961) A formalized anatomy of the human respiratory tract. In: Inhaled Particles and Vapors. Proc. Int. Symp., Br. Occupational Hygiene SOC.,Oxford, March 29-April 1, 1960. Pergamon Press, London, pp. 82-91. Findeisen, W. (1935) Uber das absetzen kleiner, in der luft suspender teilchen in der menschlichen lunge bei der atmung. Arch. Ges. Physiol., 236t367-379. ,Forrest, J.,and E. Weibel(1975) Morphometric estimation of pulmonary diffusion capacity. VII. The normal guinea pig. Lung. Respir. Physiol., 24: 191-204. Granito, S. (1971) Calculated Retention of Aerosol Particles in the Rat Lung. MS. dissertation, Department of Physiology, University of Chicago. Kliment, V., J. Lihich, and V. Kandersova (1972) Geometry of guinea pig respiratory tract and application of Landahl's model ofdeposition of aerosol particles. J. Hyg. Epidemiol. Microbiol. Immunol., 16: 107-115. Krahl, V. (1964) Anatomy of the mammalian lung. In: Handbook of Physiology: Respirations. Am. Physiol. SOC., Washington, D.C., vol. 1, pp. 213-284. Landahl, H. (1950) On the removal of airborne droplets by ACKNOWLEDGMENTS the human respiratory tract: I. The lung. Bull. Math. Biophys., 1243-56. The research for this study was supported by H. (1963) Particle removal by the respiratory sysPublic Health Service grant PHS R01- Landahl, tem. Note on the removal of airborne particulates by the ES00855 from the National Institutes of human respiratory tract with particular reference to the Health. role of discussion. Bull. Math. Biophys., 25:29-39. JPS. received personal support from Na- Raabe, O., H. Yeh, and G. Schum (1971) Tracheobronchial geometry; human, dog, rat, hamster-A compilation of tional Institutes of Health National Research selected data from the project respiratory tract deposition Service Award PHS 5-T32 GM-07151from the models. U S . Govt. Printing Office, Washington, D.C. National Institute of General Medical Services. Schreider, J., and J. Hutchens (1979)Particle deposition in the guinea pig respiratory tract. J. Aerosol Sci., (In press). Tenney, S., and J. Remmers (1963) Comparative quantitative morphology of the mammalian lung: Diffusion area. LITERATURE CITED Nature, 197:54-56. Cooper, G., and A. Schiller, (1975) Anatomy of the Guinea Weibel, E. (1963) Morphometry of the Human Lung. Academic Press, New York. Pig. Harvard Univ. Press, Cambridge.