Digital image tiles: a method for the processing of large sections

Segmentation of large areas of light microscopic slides into N by N fields, and each of these fields into M digital image tiles, allows the scanning, storage and digital processing of large images. Any of the original N2 fields or composites of M adjacent tiles can be recalled to the video display for analysis. Developed procedures for use on a microscope equipped with a precision scanning stage allow registration of the image coordinates (X-Y) for any original or composite field and the alignment of one of these fields along the depth (Z) axis by means of external, machined fiducial marks in serial sections. To facilitate work whenever unavoidable, we have incorporated methods for digital image panning and zooming (changes of magnification) and discuss their use and implications.

A morphometric study of the carotid body in chronically hypoxic rats

We performed morphometric studies of carotid body in acutely and chronically hypoxic rats (inspired PO2 = 70 Torr, at sea level). Acute exposure was for the duration of about 10 min, and chronic exposure lasted for 28 days. We confirmed that the total volume of the organ increased by severalfold. At the light-microscopy level we found an enlargement of the volume density of the blood sinuses from 14 to 31% due to chronic hypoxia. The morphometric hematocrit increased from 39 to 70% paralleling changes in the conventionally measured venous hematocrit. These data do not show any specific plasma skimming in the carotid body blood vessels. With the electron microscope we found that the mean average volume of type I cells increased from 320 micron3 in controls to 1,120 micron3 in the chronically hypoxic rats without hyperplasia, whereas type II cells had increased in number without alteration in size. Qualitative observations revealed that the normal appearance of clusters of ovoid type I cells interspersed by capillaries had been transformed into a pattern of individual cells forming plates between expanded blood vessels with a large increase of contact area between the cells and vessels. Type II cells appeared to have proliferated without changes in individual size to cover the enlarged periphery of type I cells. The observed structural changes in the carotid body parenchyma and vasculature appear to be physiologically adaptive and provide further support for the idea that various elements in the organ are particularly sensitive to hypoxia.

Methods in laboratory investigation. Digital reversal and enhancement of negatives in video-based interactive morphometry

We describe a system for the digital reversal of negatives in interactive morphometry. Minimum requirements, in addition to a high-resolution black and white video camera and monitor, are a readily available digital video frame store and a computer system as simple as a microprocessor-based personal computer. In a video-based analysis system, the addition of a digital image memory device and computer obviates the time and expense of photographic processing of the large number of negatives to positive prints required for an appropriate sampling size in morphometry. Additionally, the enhancement of digital images to recover details in the negative is feasible. This study represents a part of a continuing effort to develop an efficient, interactive morphometry system that encourages the use of a large number of random samples by simplifying processing.

Reconstruction experience with myocutaneous and osteomyocutaneous skin flaps in oncological surgery of the head and neck

Our experience in the field of oncological surgery of the head and neck enables us to present 26 cases operated upon, analysing the complications and the results, and suggesting immediate reconstruction using myocutaneous and osteomyocutaneous skin flaps due to their advantages as shown. We also present the different criteria considered in the choice of each of the three different flaps which our department uses.

Immunotherapy of chronic active viral hepatitis B with propionibacterium granulosum

12 HBsAg-positive patients with chronic active hepatitis B were treated monthly with cell walls (5 or 10 mg) of Propionibacterium granulosum KP-45 (PG) intravenously administered for a period of 6-10 months. The clinical state was evaluated according to an arbitrary 6-grade scale based on serum biochemical tests, HBV markers and the morphology of liver biopsy samples. Intravenous administration of PG was well tolerated, and no serious side effects were observed. All patients exhibited complete or partial normalization of biochemical parameters and HB5AG and DNA-polymerase levels were decreasing after 6 months of therapy. Prior to treatment, HBeAG was detected in 9 patients, in 4 of whom seroconversion occurred with appearance of anti-HBeAg. Excellent results were seen in 3 patients and beneficial results in a further 3 patients; signs of improvement were seen in 5 patients, and only one patient did not react in a clearly positive way. In 11 of these patients, before and during treatment, isolated peripheral blood lymphocytes were investigated for their ability to show rosette E formation, their reactivity to phytomitogens (PHA, ConA), and their suppressor T lymphocyte activity. In five patients interferon levels were investigated.

The use of touch-sensitive screen in interactive morphometry

We propose the use of a touch-sensitive screen as a tool to facilitate acquisition of data both in point counting and morphometric planimetry. It obviates the use of a 'correspondence image' required with many planimetric devices in which the means for data acquisition and display are separate. The touch-sensitive screen offers operational advantages over the light pen. The characteristics of this new graphical input device, its assembly and basic operational considerations are introduced and discussed.

The connective tissue of the rat lung: electron immunohistochemical studies

The ultrastructural distribution of specific connective-tissue components in the normal rat lung was studied by electron immunohistochemistry. Three of these components were localized: type I collagen, fibronectin and laminin. Type I collagen was present not only in major airways and vascular structures, but also in alveolar septa. Laminin was found in all basement membranes, and only in basement membranes, demonstrating once more that this glycoprotein is an intrinsic component of the basement membrane. Fibronectin was found free in the interstitium and on the surfaces of collagen fibers. The basement membranes of bronchial, glandular and endothelial cells of large vessels lacked fibronectin; however, capillary endothelial and occasionally epithelial alveolar basement membranes contained some fibronectin in an irregular, spotty distribution. This localization suggests that in the lung, as in other tissues, fibronectin is not an intrinsic component of the basement membrane, but rather a stromal and plasma protein. Only basement membranes in the alveolar parenchyma contained "trapped" plasma fibronectin.

Early cell alterations induced by histamine and epinephrine in rabbit lung

This study was concerned with the early effects on lung cells of infusion of 18 micrograms/kg histamine to the rabbit. For comparison purposes, other rabbits received a bolus of 0.4 micrograms/kg epinephrine. After administration, the lungs were immediately fixed by vascular perfusion of osmium tetroxide. In the group that received histamine, we observed selective alterations in the type I epithelial cells of the alveolar wall, consisting of localized thickening, increased number of cytoplasmic ribosomes, and homogeneity and increased electron density with reduction of the number of plasmalemmal vesicles. These alterations were quantitated; the vesicular load (no. of vesicles/cm2) was computed separately for each cell front of epithelial and endothelial cells. Epinephrine induced only edema in epithelial and endothelial cells, resulting in high electron lucency, thickening, and irregular cell profiles. We interpret the alterations seen in histamine-treated animals as reflecting uptake of histamine by the alveolar epithelium leading to an increased level of metabolic activity. Histamine induced no interstitial edema and no immediate reduction of capillary volume. The cellular edema related to epinephrine, a substance which is not taken up or metabolized by the lung, represented a form of cell injury.

Morphology and function of liver after intravenous and intraportal injections of the immunomodulating agent Propionibacterium granulosum

An immunomodulator, cell walls from Propionibacterium granulosum strain KP-45 (PG), has been applied in rabbits to test its influence on morphology and function of liver cells. For investigation of distribution in internal organs, the PG preparation was 51Cr-labelled and injected i.v. into mice. Functional liver tests were performed in rabbits injected intravenously or intraportally with PG. Morphologic studies (light and electron microscopy) of liver sections were performed in mice and rats injected intravenously with PG, 1-60 d prior to liver examination. Intravenous or intraportal application of PG did not cause detectable clinical symptoms during the whole observation period (1-60 d). During the first 3-7 d, elevation of serum globulins and transaminases (SGOT, SGPT) were noted with reappearance of normal values after 14-20 d. 131I-labelled Bengal Rose injected into PG-treated rabbits was trapped mostly in the spleen, and to a smaller extent in liver, where radioactivity was significantly lower than in saline-injected control animals. 51Cr-labelled PG injected i.v. into mice were after 60-90 min localized preferentially in lungs and liver, where the radioactivity was much higher than in the spleen. Morphological studies revealed that 6 h following i.v. injection of PG, there appeared partial clearing of the hepatocyte cytoplasm and lowering of their basophilic content with enhancement of pinocytic activity of the hepatocytes. This was accompanied by enlargement of RES cells in liver with enhanced intracellular digestion, higher number of lysosomes and appearance of phagocytic vacuoles.(ABSTRACT TRUNCATED AT 250 WORDS)

Inhibition of Lewis lung carcinoma in mice by local microwave hyperthermia combined with immunomodulating Propionibacterium granulosum KP-45

C-57 BL/6 mice with Lewis lung carcinoma were treated 2 weeks after tumor implantation with local microwave hyperthermia (2450 MHz, tumor temperature 43.5 degrees C, 30 min) and/or intratumoral or intraperitoneal injection of 1 mg cell walls of Propionibacterium granulosum KP-45. Tumor growth, survival time of the animals, and the delayed skin hypersensitivity to oxazolone were followed up, as well as the 3H-thymidine uptake of tumorous tissues and the number of lung metastases. The combined treatment of microwave hyperthermia with immunomodulating P. granulosum KP-45 resulted in significantly stronger inhibition of tumor growth than with each of these methods alone. The number of lung metastases could be significantly lowered, and the skin reactivity to oxazolone remained enhanced during the whole observation period (over 70 days after tumor implantation). The implications of the test observation are discussed.

Alveolar epithelial lesions induced by angiotensin in rabbit lungs

We slowly administered a bolus of 10 micrograms/kg of angiotensin II in saline to anesthetized rabbits. Their lungs are fixed by vascular perfusion of fixatives 2-3 minutes after the end of the infusion. Electron-microscopic examination of the lung parenchyma did not reveal signs of hemodynamic edema, but several epithelial alterations were observed that could be interpreted as the result of incorporation of plasmalemmal vesicles into the plasma membrane: these included undulating membrane profiles, transcellular channels formed by coalescing vesicles, and/or deep infoldings of the cell membrane. These appear to have the capability of causing demarcations in the cellular cytoplasm of Type I epithelial cells, which result in cell fractionation and localized destruction of the squamous alveolar epithelium, which leads to denudation of the basement membrane from the air side. These epithelial lesions were often associated with intracapillary platelet-fibrin aggregates, and cell destruction was most apparent in the presence of substantial amounts of fibrin. Morphologic signs of damage of the capillary endothelial cells existed but were less pronounced; in particular, there was no endothelial denudation under the platelet aggregates.

Isolation of lamellar bodies from rat granular pneumocytes in primary culture

A lamellar body fraction was isolated from rat alveolar granular pneumocytes in primary culture by upward flotation on a discontinuous sucrose gradient and compared with a similar fraction isolated from lung homogenates. Lamellar bodies from granular pneumocytes were free of detectable contamination with either succinate dehydrogenase or NADPH-cytochrome c reductase. There was an enrichment of acid phosphatase activity, which, based on distribution of enzyme activity on the gradient, did not appear to be a contamination from other fractions. The lamellar body fraction of granular pneumocytes yielded approx. 1 microgram protein/10(6) cells with a phospholipid-to-protein ratio (mg/mg) of 9.6 +/- 0.4 (n = 7). Composition with respect to total phospholipids was 71.0% phosphatidylcholine (disaturated phosphatidylcholine, 45.2%), 8.4% phosphatidylglycerol and 12.8% phosphatidylethanolamine. Palmitic acid comprised 66% of the fatty acids in phosphatidylcholine and 34% of those in phosphatidylglycerol. The lamellar body fraction from granular pneumocytes was similar to that from whole lung with respect to phospholipid-to-protein ratio and phospholipid composition and showed only minor differences in fatty acid composition. Ultrastructurally, lamellar bodies showed generally intact limiting membranes and lamellated structure. Lamellar bodies from granular pneumocytes showed occasional multinucleated whorls which were not seen in those isolated from lung homogenates. This study describes a method for preparing a homogeneous fraction of intact lamellar bodies from small amounts of material (6 X 10(7) granular pneumocytes). The yield on a per cell basis was higher when compared with a similar preparation from whole lung, although overall yield is small, due to loss of cells during the cell isolation procedure. This preparation may be useful to evaluate the role of lamellar bodies in the synthesis and secretion of lung surfactant by isolated granular pneumocytes.

Ethanol, hepatocellular organelles, and microtubules. A morphometric study in vivo and in vitro

The effects of chronic ethanol intoxication on hepatic microtubules are controversial. We, therefore, designed a morphometric study to examine ethanol-induced alterations in microtubules in vivo and in vitro. Using a computer-assisted point-counting system we determined the volume, surface, and length densities of hepatic microtubules, in addition to those of other hepatic organelles. Consistent with previous qualitative descriptions, chronic ethanol feeding caused increases in the volume and surface densities of mitochondria and microsomes. These changes account, in part, for the increased protein content of hepatocytes after chronic ethanol consumption. However, there was no effect on the volume, surface, or length densities of hepatic microtubules, nor on their calculated radii. The structure of isolated bovine brain microtubules after polymerization in vitro was not changed by ethanol or its metabolites, acetaldehyde and acetate. These data indicate that neither chronic ethanol intoxication nor the presence of ethanol, or its metabolites, affects the number or structure of microtubules.

Number and distribution of plasmalemmal vesicles in the lung

Vesicles in the lung, particularly in capillary endothelial and squamous alveolar epithelial cells, are membrane differentiations that form a system of impressive aggregate size. Strong evidence from electron microscopic tracer studies indicates that they have the potential of acting as a transendothelial transport mechanism, but their precise level of activity and the significance of vesicular transport as a pathway different from hydraulic permeability have not been clarified. It has been proposed that vesicles may form either permanent or transient transcellular channels. The relative significance of vesicles as passage sites compared to the leaky intercellular endothelial junctions awaits clarification. Quantitation of individual cell size and number of vesicles reveals that in both endothelial and epithelial cells more than 70% of the total plasma membrane is located in vesicles. Coated pits and vesicles are readily identifiable in all cells of the lung parenchyma, although they are much less frequent than noncoated vesicles. These are important because they have been identified in cell biology as sites of receptor-mediated endocytosis and therefore offer the basis for the presence of a system of selective permeability and/or uptake of substances by the lung, which, so far, has not been explored.

Secretion of Weibel-Palade bodies observed in extra-alveolar vessels of rabbit lung

We fixed rabbit lungs by perfusion of osmium into the pulmonary artery and examined in light and transmission electron microscopy a large number of extra-alveolar vessels with a diameter of 0.1-0.25 mm, with emphasis on Weibel-Palade bodies (endothelial specific granules). Weibel-Palade bodies are organelles specific to endothelial cells. Their function is unknown, but they are useful markers for identification of endothelial cells in culture. We were able to observe release of the content of these bodies into the vascular lumen; this indicates that they are secretory.

A 2.5-year follow-up of local immunotherapy of advanced stomach and intestinal adenocarcinoma with Propionibacterium granulosum

Intratumoral injections of 10 mg cells walls from Propionibacterium granulosum strain KP-45 (PG) were applied in 14 patients with advanced, metastatic stomach (five cases) and colorectal (nine cases) adenocarcinoma. Each patient had his own "twin" control. All patients received no other anticancer treatment, except analgetics and/or palliative surgery. Treatment with PG resulted in partial regression of tumors accompanied by improvement of the clinical state of these patients as well as the reappearance of normal values in blood count biochemical parameters. In each pair of twin cases, survival of the PG-treated patient was longer than the untreated control. The mean survival of PG-treated patients was 23.5 months (4 of 14 patients being still alive after 2.5 years follow-up), while all control patients died with a mean survival period of 6.4 months. The difference between these two patient groups of about 17 months is significant (p greater than 0.01).

[Local immunomodulation in advanced Crohn disease with a cell-wall preparation of Propionibacterium granulosum KP-45]

Four patients aged 30-34 years with Crohn's disease in and advanced stage, having lasted 2 or 3 years, were treated by several injections of 10mg lyophilized cell wall material of P. granulosum strain KP-45 locally into the lesion. In all 4 cases a preceding conventional therapy with Salazopyrine and Prednisone or surgery had been unsuccessful. This local immunomodulatory treatment led to decisive improvement of the disease in all 4 cases, in 3 cases inflammatory symptoms disappeared completely. For this reason a local immunomodulation with P. granulosum KP-45 is recommended as an additional therapeutic procedure in patients who are otherwise resistant to therapy.

Early tissue damage in ethchlorvynol-induced alveolar edema in rabbit lung

The purpose of this study was to examine the morphologic changes that occur in the early stages of intraalveolar edema. Anesthetized rabbits were intravenously administered a bolus of 40 mg/kg of ethchlorvynol, a mild hypnotic known to induce respiratory distress syndrome in humans and laboratory animals when given intravenously. After 15 min their lungs were fixed for transmission electron microscopy. Examination revealed variable amounts of irregularly distributed intraalveolar edema with erythrocytes and fibrin strands that coexisted with modest or nonexisting interstitial edema suggesting that primary hemorrhagic alveolar flooding had taken place. Most alveolar epithelial and endothelial cells appeared normal except in localized areas. In lungs fixed by vascular perfusion, in which normal capillaries were flushed, obstructions were noticed in alveolar corner capillaries. These areas were identified by light microscopy and selectively sectioned for electron microscopy. They contained intravascular cell-fibrin aggregates consisting of plugs of degranulated platelets, fibrin, erythrocytes, and leukocytes. Endothelial and epithelial cells in the vicinity of the plugs showed variable degrees of injury. In places the damage was so severe that vascular and alveolar spaces were separated only by the basal lamina. Our work shows the previously unnoticed existence of capillary microemboli or microthrombi in a well-known experimental model of respiratory distress syndrome and suggests extravasation of blood elements through discrete sites of cell injury associated with the fibrin-platelet aggregates rather than diffuse increase of permeability as cause of early alveolar flooding.

Alveolar wall relations

We have presented a highly dynamic view of the alveolar septum and its main enclosed structure, the dense capillary network. The septal or perimicrovascular interstitium is the space between alveolar epithelial sheets after exclusion of the capillary network. It contains cells, fibers, and a viscous matrix. Capillaries form a very complex network, which closely follows the geometry of the terminal airways and participates in functional adaptations of the wall, particularly septal pleating. The level of filling and configuration of different capillaries ranging from collapse to full distension are variable, depending on factors such as transmural balance of forces but also on tissular configuration. Alveolar flooding of any cause will produce an immediate change of capillary configuration and volume.

Immunostimulation by systemic administration of Propionibacterium granulosum in patients with primary or secondary lung tumours

The effect of a cell suspension of Propionibacterium granulosum strain KP-45 given intravenously to patients with advanced primary or secondary lung carcinoma, and simultaneously treated with cytostatic drugs, was estimated. The control group consisted of patients treated with cytostatics only. The clinical state of patients treated with the KP-45 preparation and with this immunostimulant plus cytostatics, was compared by evaluation of tumour size by X-ray examination, peripheral blood picture, and biochemical and immunological parameters. Treatment with P. granulosum KP-45 did not cause any serious adverse reaction or complication. No evident inhibitory effect on primary or secondary lung tumour development was observed. However, injections of the KP-45 preparation in cytostatics-treated patients resulted in decreased adverse effect of these drugs on the haemopoietic system and absence of bacterial infections, in contrast to patients treated with cytostatics only.

Morphometric study of rat lung cells. I. Numerical and dimensional characteristics of parenchymal cell population

Four normal rat lungs were positive-pressure inflated (7 mm Hg downslope) and perfusion-fixed for electron microscopy. Morphometric analysis yielded estimates for compartmental volume densities in alveolar septal lung tissue excluding capillaries: type I cells, 0.126; type II cells, 0.097; endothelium, 0.264; interstitial cells, 0.358; interstitial space, 0.116; macrophages, 0.039. Numerically, the largest cell population of the parenchymal tissue was the endothelial cells (43%) followed by interstitial (31.8%), type II (14.5%), and type I (7.5%) epithelial cells; 3.2% were macrophages. Individual type I cells possessed the largest cell volume (915 micron3) as well as th largest luminal surface (4,518 micron3), covering 97.5% of the alveolar surface. Type II cells (volume, 366 micron3; surface, 62 micron2) line only 2.5% of the alveolus. Total endothelial surface is 16% larger than the alveolar surface, although individual endothelial cells are much smaller (volume, 336 micron3; luminal surface, 946 micron 2) than type I cells. Macrophages and interstitial cells (cell volume, 665 and 615 micron3, respectively) are of an intermediate size. By restricting the reference frame to alveolar septal tissue, surface and volume estimates agree well in most cases with instillation-fixed lungs. Caution is nevertheless indicated in interpreting such comparisons because differing fixation/inflation techniques generate unequal osmotic and oncotic pressure gradients.

Distribution of vesicles in cells of air-blood barrier in the rabbit

Vesicular transport has been proposed as a pressure-independent shuttle mechanism for transport of macromolecules across thin cells. The vesicular populations of cells found in the gas-exchanging parenchyma were quantitated by morphometric means. We found that the numerical densities (number of vesicles/unit of volume) were comparable in endothelial (131/micrometer 3) and epithelial (145/microm3) type I cells but much smaller for interstitial cells. The vesicular loads or numbers of vesicles/unit of surface area were computed separately for each cell front (luminal and abluminal) and found to be similar for all surfaces, ranging from 150 to 196/micrometer 2 except for the interstitial front of epithelial cells where the load was higher (230/micrometer 2). Although this represents a very high transport potential, little can be said about vesicular effective performance because of insufficient knowledge of vesicular biology and biophysics. A frequent observation in the capillary network is attenuations of endothelial cells, which are thinner than the vesicular diameter and devoid of vesicles.

Local immunotherapy of stomach and intestinal carcinoma by Propionibacterium granulosum

The effect of cell suspension of Propionibacterium granulosum strain KP-45 injected intratumorally was estimated in patients with advanced malignant neoplastic disease of the upper and lower part of the digestive tract. These patients received no other anticancer treatment, and results of therapy with P. granulosum were compared with a control group of nontreated patients. In the course of immunotherapy, general clinical state, dynamics of tumor growth as judged by endoscopy and morphologic, biochemical and immunologic parameters of the peripheral blood, were evaluated. Treatment with KP-45 preparation usually resulted in partial regression of tumor growth, accompanied by improvement of the clinical state of these patients, as well as reappearance of normal values of blood picture and biochemical parameters. Enhancement of cellular immune response, evaluated by increased number of peripheral blood T lymphocytes and increased skin reactivity to PHA and PPD, was observed. No serious side effects or complications of local immunotherapy were noted. No patients who had immunotherapy suffered from bacterial infection; this was not the case in the control group.

Morphometry of pinocytotic vesicles in the capillary endothelium of rabbit lungs using automated equipment

Pinocytotic vesicles are one of several anatomic factors relevant to the permeability for macromolecules of capillary endothelia. We quantitified the number and distribution of pinocytotic vesicles present in the endothelium of pulmonary capillaries. Several similar studies have been reported in the past for endothelia of systemic capillaries, but the technical difficulties involved in this difficult application of morphometry never have been discussed. In the present work, an on-line microprocessor was used to assist with the processing of the point-counting data. The average figures from nine rabbit lungs for parameters thought to be relevant to characterize the population of vesicles are: thickness of the endothelium (TAUen) 0.192 microns; number of vesicles per volume unit of cell (NVen), 131 vesicles/microns3; vesicular load of each cell front (N/S)b, 196 vesicle/microns2, and (N/S)i, 181 vesicles/microns2. The number of vesicles per volume unit (NVen) is smaller than that which has been described for systemic capillaries.

Fusion of pinocytotic vesicles in liquid filled lungs: a mechanism of cellular damage

Lungs of anesthetized rabbits were filled unilaterally with macromolecule-containing solutions (Dextran 40 and albumin) for 20 min. which generally resulted in fluid resorption from the flooded lung. Electron microscopic examination revealed the following: (1) formation of chains of vesicles, similar or identical to pinocytotic vesicles, which start from depressions or fluid-immersed cellular surfaces in macrophages, or endothelial and squamous epithelial cells; and (2) formation of intracellular blisters as a result of the apparent fusion of numerous pinocytotic vesicles. Both observations seem to be augmented forms of normal phenomena. The hypothesis is advanced that the chains of vesicles are related to the increased amount of undulation of the cell surface and to cellular invaginations that follow relaxation of tensions at the surface of cells immersed in fluid. The blisters could be the result of the increase of bidirectional flow of vesicles across the squamous cells. This brings an increased risk of collision because of the absence of lysosomes in the thin cytoplasmic extensions of endothelial and squamous epithelial cells. It is concluded that, even in very early stages, flooding per se produced some form of cellular alterations with a potential for leading to more overt lung pathology such as cell death and desquamation which would greatly increase tissue permeability.

Alveolar volume-surface area relation in air- and saline-filled lungs fixed by vascular perfusion

The influence of volume changes and interfacial forces on the geometry of peripheral air spaces was studied in excised rabbit lungs inflated with either air or saline and fixed by vascular perfusion at four points of the deflation limb of the pressure-volume curve corresponding to 100, 80, 60, and 40% of the total lung capacity (TLC). In air-filled lungs pleating and folding of alveolar septa were observed, especially in alveolar corners. However, the alveolar surfaces were smooth, except at low lung volumes where some surface crumpling occurred. In saline-filled lungs pleats were absent; the alveolar surface was irregular at all inflation levels due to undulating walls and bulging capillaries. Morphometry indicated that at all alveolar volumes (VA) the surface areas (SA) were larger in saline- than air-filled lungs. No simple mathematical function was found to characterize the relation between SA and VA over the entire volume range studied. Within the range of normal breaths (80 to 40% TLC) the best fit for n in the function SA = k.VnA was 0.58 for saline-filled lungs (r = 0.93) and 0.33 for air-filled lungs (r = 0.68), suggesting different and complex deflation patterns.

Morphologic aspects of alveolar microcirculation

Direct observations of the flow direction and connections between arteries and veins in the mammal lung are difficult. When we divide the lung into smaller units like acini or segments we can observe a central supply of the unit with arterial blood that has venous drainage at different points of the periphery. Consideration of the situation prevailing at birth strongly suggests a preferential flow direction through paths located in primary septa at the bottom of alveoli. Capillaries of the secondary septa placed between alveoli open to the same duct represent collaterals of the mainstream flow filled only if pressure conditions permit. Another significant feature is the presence of pleated alveolar septa. While capillaries inside the interalveolar wall mostly appear flat or collapsed, the capillaries of the pleated alveolar corners are always wide open. Often they show openings into a small venule placed inside the pleated area, which strongly suggests that the pleated areas contain the venous side of the capillaries.