Scanning Microscopy

Volume 4 Number 4 Article 7

9-19-1990

Characteristics of Granulation Tissue which Promote Hypertrophic Scarring

C. Ward Kischer The University of Arizona

Jana Pindur The University of Arizona

Peggy Krasovitch The University of Arizona

Eric Kischer The University of Arizona

Follow this and additional works at: https://digitalcommons.usu.edu/microscopy

Part of the Biology Commons

Recommended Citation Kischer, C. Ward; Pindur, Jana; Krasovitch, Peggy; and Kischer, Eric (1990) "Characteristics of Granulation Tissue which Promote Hypertrophic Scarring," Scanning Microscopy: Vol. 4 : No. 4 , Article 7. Available at: https://digitalcommons.usu.edu/microscopy/vol4/iss4/7

This Article is brought to you for free and open access by the Western Dairy Center at DigitalCommons@USU. It has been accepted for inclusion in Scanning Microscopy by an authorized administrator of DigitalCommons@USU. For more information, please contact [email protected]. Scanning Microscopy, Vol. 4, No. 4, 1990 (Pages 877-888) 0891-7035/90$3.00+.00 Scanning Microscopy International, Chicago (AMF O'Hare), IL 60666 USA

CHARACTERISTICS OF GRANULATION TISSUE WHICH PROMOTE HYPERTROPHIC SCARRING

C. Ward Kischer*, Jana Pindur, Peggy Krasovitch and Eric Kischer

Department of Anatomy, The University of Arizona, College of Medicine Tucson, Arizona 85724

(Received for publication June 22, 1990, and in revised form September 19, 1990)

Abstract Introduction

Hyper trophic and keloids are Hypertrophic scars are peculiar to humans characterized by nodules of that (Kischer et al., 1982a) and occur as sequelae originate in granulation tissue arising from to deep injury of the body surface. A related full thickness or deep 2° injuries to the lesion is the keloid, which, unlike the skin. Fifty-six granulation tissues of varying hypertrophic , tends to outgrow the ages post-injury were examined morphologically boundaries of the original injury (Peacock for evidences of how the nodules and, thus, the et al., 1970). Both types of lesions may be scar form. New microvessels grow in ascension prompted by the surgeon's knife, a thermal burn towards the free surface in a milieu of or equivalent-type injury. They may undergo inf lammatory cells and . Collagen self-limiting resolution (not so prevalent in deposition increases with time from the base of the case of the keloid). The time for the the wound to the free surface and begins to self-limiting process to occur may range from a concentrate between lateral branching of the few months to ma ny years and cannot be new microvessels. Computer derived serial predicted. These fibrotic lesions are hard, reconstructions of hypertrophic scar nodules e levated, erythemic and contracting. indicate they are of varying shape and size Every hypertrophic scar and keloid contains probably due to fusion of adjacent structural units of collagen called nodules, microvascular collagen masses between lateral which are never present in fully mature scars branches. This is accomplished by the gradual (Kischer and Brody, 1981). Therefore, these but persistent degeneration of microvascular anatomical units actually define the scar and endothel.i.a and pericyte s. Fifty-six pieces of keloid. How do these nodules develop? From granulation tissue taken from 5 cases of where and from what do they arise? varying age post-injury were implanted into An earlier study strongly suggested the nude mice. Several proceeded to develop scar origin of nodules to be in granulation tissue and some of those developed nodules. The (Linares et al., 1973), but did not specify how latter developed only when the zero-time they might be formed. The nodules have long implant contained lateral microvascular axes parallel to one another and, branches. Hypertrophic scars and keloids are a predominantly, the free surface of the skin. product of granulation tissue elements, the They are rarely aligned perpendicular to the most important of which are primed active free surface. They also are oriented parallel fibroblasts and excessive microvascular to the lines of flexion across joints. Nodules regeneration, including lateral branching, are composed of unidirectionally aligned which subsequently degenerates, in part, collagen (parallel to the long axis of the promoting nodule formation and remodeling. nodule), which shows virtually no undulation. Interstitial space is not observed by electron microscopy within the nodule (Kischer, 1974). The fibroblasts are similarly oriented and the nodules are avascular. However, the nodules KEY WORDS: granulation, hypertrophic scar, are circumscribed with a network of keloid, light microscopy, electron microscopy, microvessels (Kischer and Brody, 1981). tissue culture, implantation, nude mice, Is it a .foregone conclusion that a deep , platelet derived . inJury will always result in a hypertrophic scar (in some cases a keloid)? If nodules *Address for Correspondence: form, the answer is yes. C. Ward Kischer, Ph.D., Department of Anatomy We need to know how the nodule is formed, The University of Arizona College of Medicine and why. This knowledge might provide the Tucson, Arizona 85724 means by which development of the nodule could Phone No. (602) 626-6090 be prevented, which might then obviate the

877 C.W. Kischer, et al .

development of the hypertrophic scar. and coated with 300A0 of gold in a Polaron Therefore, the present study was designed Sputter Coater, Model #5100, using argon gas. to investigate the following: 1) the organiza­ Examination of the tissues was conducted in an tion of granulation tissues over time after ETEC Autoscan using 20 kV. injury with particular attention to micro­ vascular distribution, collagen deposition and Implants cellular orientation; 2) to determine if there might be some developmental relationship Pieces of full thickness granulation between the above (#1) information and the tissues were implanted into subcutaneous nodules; 3) to determine if pieces of pockets over the scapular areas of nude mice. granulation tiss ue implanted into the nude Five cases of granulation tissues, three, mouse will proceed to develop nodules and/or 1 month or younger, and two, 4 months scar. post-injury, provided 56 i mplants into 22 mice. The implants were carried from 6 to 160 days. Materials and Methods All procedures from acquisition of the tissue to implantation were carried out under sterile conditions. Fifty-six different human granulation Nude mice (nu/nu), which are also athymic tissues, each from full thickness wounds, vary­ (Krueger and Briggaman, 1982) and do not reject ing in age post-injury from less than 1 month foreign tissue implants, were obtained from to 4 years, obtained by biopsy or surgical Harlan Sprague Dawley, Inc., Indianapolis, IN, scrapings prior to grafting, have been analyzed and from Charles River, Inc., Charles River, by light and electron microscopy. In terms of MA. All mice used were males and were kept in post-injury age, 13 were 1 month or less, 8 were presterilized microisolator units with filter 2 months, 4 were 3 months, and 6 were 4 months bonnets on top and lined up in front of laminar post-injury. Twenty-two were 5 months or older flow units. All handling of these mice was and 3 were of unknown age. done in a laminar flow hood under sterile Pieces of these tissues were processed for conditions. The mice were fed with sterilized all subsequent microscopic analyses by initial Purina Mouse Chow and given sterilized water to fixing in Karnovsky's fixative (Karnovsky, drink. They were housed one mouse to a unit. 1965). Granulation tissues from full thickness wounds were obtained under sterile conditions, Light Microscopy (LM) kept in sterile containers over ice, and transported to the animal laboratory for For light microscopic analysis, the tissues surgical implantation into the mice. The mice were dehydrated in graded ethanols then through were anesthetized with sodium pentobarbital xylene and embedded in paraffin. Subsequent using approximately 60-70 mg/kg of body tissue sections of 10 pm thickness were stained weight. Before the tissue pieces were by Hematoxylin and Eosin or by Masson' s tri­ implanted, smaller representative pieces were chrome method. General morphology and the taken for microscopic study. Those pieces for distribution and location of collagen for each microscopic studies were taken in full sample of tissue was recorded. Transverse thickness, placed in Karnovsky's fixative, and sections were mostly studied, although a few later processed for light and electron tissues were examined that were cut parallel to microscopy. the free surface. The implantation procedure has been previously described (Kischer et al. 1989a). Electron Microscopy (EM) The suprascapular areas on the back were incised 1 cm in length and the incision Transmission EM (TEM): Smaller pieces of slightly widened with tips of fine scissors to the same tissues were processed from the initial form a subcutaneous pouch. The tissue piece, Karnovsky's fixative through dehydration in standardized at approximately 5 x 8 x 5 mm in graded ethanols, to absolute alcohol to size, were inserted into this pouch; two pieces propylene oxide, and embedded in Epox 812 (Ladd were implanted in each mouse, one on each side, Research Co.). Thin sections were cut by although some mice received 3 or 4 implants. diamond knives and stained with lead citrate In these cases the pouch was widened to and examined in a Philips 300 electron accommodate 2 pieces but which would not abut microscope. one another. The incisions were closed using Scanning EM (SEM): Tissues to be studied 11 mm wound clips. No dressing was applied by SEM were transferred from Karnovsky's although the incised area with wound clips fixative to graded ethanols, 50% to absolute attached was sprayed with 95% ethanol. alcohol, then placed in the chamber of a When the tissues were harvested, the mice Tousimis Samdri-790 Critical Point Dryer. The were again anesthetized with sodium pento­ intermediate fluid was alcohol and the barbital, swabbed aseptically with alcohol, the transitional fluid was liquid CO2, Tissue back skin again incised and the implants blocks were mounted for cross-sectional study removed according to the schedule in TABLE I:

878 Granulation Tissues and Hypertrophic Scars

TABLE I decrease in number, while fibroblasts Schedule of Granulation Tissue Implants increase. Staining for collagen becomes Into Nude Mice visible. By approximately 4 to 5 months old these characteristics are prominent. Lateral Days as Implants II of Implants branching of the microvessels occurs with 0 - 19 days 9 variation between 2 and 3 months post-injury. 20 - 39 days 6 This branching appears at the same time that 40 - 59 days 4 collagen begins to accrue, and seems to arise 60 - 79 6 near the deepest layer first (Fig. 3). The 80 - 99 10* microvascular branches include many pericytes 100 129 7* (Fig. 4) which differ in appearance from the 130 - 159 6 interstitial fibroblasts (Fig. 5). 160 O** In still older granulations, the collagen * - 2 implants not found accumulates to a dense amount and is ** - 4 implants not found circumscribed by sections of microvessels forming whorls and circular profiles of Nodule Structure -like cells (Fig. 6). The older granulations demonstrate The s true ture and conformation of the degenerate remains of microvessels (Fig. 7). collagen nodule is integral to the formation of Within the circular (pre-nodule) configurations the hypertrophic scar. Knowing its form might microvessels are seldom observed. The older provide a clue as to how it develops. nodular-type appear avascular (Fig. 8). Therefore, serial sections of selected hypertrophic scars were made showing Scanning Electron Microscopy cross-sections through nodules. Photographs were made of the same "window frame" of each When viewed by scanning electron section, and sometimes up to 100 sections were microscopy, the upper half to one-third of made in succession. The outlines of the early granulation tissue in cross section often nodules in each window frame were drawn on the shows a granular appearance and is composed of computer screen, then each frame was aligned small cells and erythrocytes (Fig. 9). As the with the one before it. Upon completing the granulation tissue matures a scaffolding of aligning process through the computer software, fine filamentous material appears obscuring the next step was to enter ]-dimensional discrete structures and cells (Fig. 10). A coordinates specifying the location of the graded occurrence of collagen from the basal viewing position. The software then calculated area to the surface is observed with large the new screen pixel locations for a view of well-defined collagen fascicles identified in the sectional stack, from a specific angle and the former area (Fig. 11). Often, these areas elevation. With the complete viewing suggest small nodules. calculations computed, the next step was to Implants of granulation tissues removed present and edit each window or layer. Editing after 40 days show progressive increases of is defined here as simply the omission of whole collagen and demonstrate a marked resemblance independent parts from specified layers in to the nodular appearance in hypertrophic scars order to isolate a clear view of a specific (compare figure 11 with figure 12). An implant area. Thus, two adjacent nodule outlines or of 69 days (Fig. 13) is compared with normal profiles could be the sole focus of a given dermis at the same magnification (Fig. 14) to series. The outlines were rotated through show the dramatic increase in collagen and its 360° in order to view all sides. organization into nodules during implantation.

Results Transmission Electron Microscopy

Light Microscopy The early granulation tissues of one to two months demonstrate scattered collagen, a Evaluation of cross-sect.ions through full background of amorphous material, presence of thickness granulation tissues demonstrates a fibrin within the interstitium and walls of the progressive gradient differentiation from basal new vessels, and portions of cells ( some area to the surface over time. degenerating) (Fig. 15). The microvessels, Granulations of 1 month age demonstrate a however, are usually patent and sometimes show loose network of fine filamentous and granular diapedesis (Fig. 16). The endothelial cells material, interstitial fibrin with a display a rich array of rough endoplasmic predominant cellular content of white cells and reticulum (RER) as do the pericytes (Fig. 17). . Occasional fibroblasts are In the older granulation tissues, many of observed. The microvessels are oriented the microvessels are occluded by an excessive predominately in ascension towards the free number of endothelial cells and are surrounded surface (Fig. 1). The surface of the tissues by multiple layers of active appearing contains a fibrin clot, and patent ends of pericytes (Fig. 18). are the microvessels terminate directly into that layer predominant interstitial cell type (Fig. 19). (Fig. 2). Little collagen is observed in the These cells can only be identified by upper layer. transmission electron microscopy by their With increasing age white cells tend to arrays of micro filaments and cytoplasmic

879 C.W. Kischer, et al.

Figure 1. Microvessels growing towards the Figure 2. Newly regenerating microvessels in free surface of granulation tissue one month a granulation tissue one month post-injury. post-injury. Interstitial cells are white Some microvessels terminate in the fibrin layer blood cells and fibroblasts. H & E stain. (F). H & Estain. Bar= lOOpm. Bar = lOOµm.

Figure 3. Basal area of chronic granulation Figure 4. Microvessels in upper one-third of tissue 6 months post-injury. Whorl area (W) two month old granulation tissue. Note circumscribed by microvessel < ➔). pericytes around vessels < ➔). H&E stain. Interstitium contains considerable collagen. Bar= 100)1111. Trichrome stain. Bar= lOOpm. ------·------densities. showed nodule formation. Sixteen others of the Often in the middle and lower third of the 33 implants demonstrated active fibroblasts and granulation tissue the microvessels contain increased collagen. Implants prior to 40 days degenerating pericytes and endothelial cells did not show sufficient changes to indicate (Figs. 20 and 21). This degeneration is not scarring except one implant at 22 days and one accompanied by the presence of macrophages or at 25 days. Those implants demonstrating inflammatory cells. nodules came from granulations of 6 months and 20 months post-injury. The appearance of Implants coll.agen and nodular structure is a function of time. Up to approximately 40 days there is a Ten of 33 implants of 40 days or more gradual increase in collagen and fibroblasts

880 Granulation Tissues and Hypertrophic Scars

Figure 5 . Thick plastic section of occluded microvessel with surrounding pericytes (P) in granulation tissue of 1 month post-injury. Note fibroblasts nearby ( ➔). Toluidine Blue. Bar= 10pm.

. ,., .,.. 'r ., , . " . ' r ,,r .,. . I t, ~ --- "" Figure 6. Whorl of collagen (W) in basal l. ? ,,. - ~ '">. - .. ?' t ...~ "" granulation area of 5 months post-injury. Note . . . .. ~ peripheral microvessel ( ➔ ). Trichrome ? • ' ' .. stain • Bar = l00pm. ,,, - i '• r w ,.,.-· , .,. \ ' .. ~ 7

Figure 7. Area of granulation tissue 4 months p:+injury showing degenerating microvessels ( ). H&E stain. Bar= l0qpm.

and a corresponding diminution of fibrin and inflammatory cells. Many of the early granulation implants show areas of necrosis. Beyond 40 days, the magnitude of and formation of nodules is considerably greater. By 40 days, virtually all fibrin and inflammatory cells had disappeared. Thereafter, collagen increases and discrete compartments or nodules of collagen become prominent and essentially walled off by microvessels (compare Figs. 22 and 23; also, see Figs. 12 and 13). Eight implants were not found upon attempts of recovery. All 8 were 90 days or longer post-implantation.

Nodule Structure

The graphic representations produced by the Figure 8. Nodular-like areas (N) of collagen computer represent an unpatterned structure of in granulation 5 months post-injury. Trichrome different nodules. Within a given set of stain. Bar= l00pm.

881 C.W. Kisc her, e t al.

Figure 9. SEM of one month old granulation Figure 10. SEM of three month old granula:ion tissue. Granular texture with many small cells tissue. Scaffold of fine filaments cove~ing and erythrocytes. Bar= 10pm. underlying structures, such as collagen bundles, cells and portions of micro vessels. Bar = 10pm.

Figure 11. SEM of lower area of granulation in Figure 12. SEM of granulation tissue implanted f igure 10. Collagen is more prominent and 50 days showing nodule (N). Bar= 10pm. suggests formation of nodule (N). Bar = 10pm.

Figure 14. SEM of normal dermis i n cross Figure 13. SEM of granulation tissue implanted section. Compare with Figure 13. Bar= l0pn. 69 days shows scar development and beginning nodule formation (N). Bar= 10pm.

882 Granulation Tissues and Hypertrophic Scars

. .... f::...) • . . : . \ •}.):.f ·• ·•~:t;t . i

\

\

Figure 16. TEM of patent microvessel 17um in diameter from granulation tissue 2 months post-injury. Note diapedesis (~) and Figure 15. TEM of upper area of granulation interstitial fibrin (F). Bar= ]pm. tissue 2 months post;;n➔ ry. Mostly fibrin (F), some collagen ( ) and portions of fibroblas t cells (C). Bar= 1pm. layers, one can observe independent nodules, nodules merging at a specific point, and others that merge and shortly thereafter separate again. The succession of pictures clearly presents the variable forms nodules may have through serial reconstruction and 360° rotation. Figure 24, followed through 43 serial sections (total depth = 430 pms) shows two adjacent nodules which continue separated for some length then fuse or merge. Figure 25 shows a small circularly-shaped nodule surrounded by a larger bizarre-shaped nodule. The two nodules did not fuse throughout 26 serial sections.

Discussion

The overall major objective of these studies of granulation tissue was to determine in some measure the dynamic events which might lead to the development of the hypertrophic scar, especially, the collagen nodules. Young granulation tissue in the case of a deep skin wound is characterized by a massive regeneration of new microvasculature growing from the vascular remnants predominantly at the base and perhaps some from the sides of the wound. They grow beneath and into a fibrin clot, and predominantly in a direction towards the free surface. Accompanying pericytes increase markedly and display vast arrays of rough endoplasmic reticulum, presumably Figure 1 7. TEM of microvessel in granulation synthesizing either collagen or extracellular tissue 1 month post-injury. Lumen (outlined matrix, or both. This is especially true in - - -) occluded by platelets and white cells. the area of microvascular lateral branching in RER in Endothelial cells and pericytes (*). the mid- and lower presumptive reticularis Bar = ]pm. areas. Amidst the vast arrays of new

883 C.W. Kischer, et al.

Figure 18. TEM of occluded microvessel in 3 Figure 19. TEM of myofibroblasts (M) in month old granulation tissue. Both endothelial advanced granulation tissue of 1 month (E) and pericyte (P) cells contain substantial post-injury. Bar= 1pm. amounts of RER. Note degenerating fibroblast (D). Bar = 1pm.

Figure 20. TEM of occluded microvessel in Figure 21. TEM of occluded microvessel in 6 granulation tissue 6 months post-injury. Note month old granulat ion tissue. Endothelial degenerating pericytes (D). Bar= 1)-llll. cells and pericytes are degenerating (D). Bar = 1pm.

884 Granulation Tissues and Hypertrophic Scars

Figure 22. Granulation tissue three months of Figure :u. Edge of ':10 day implant of age, showing surface-directed microvessels in granulation tissue from same case as in Figure milieu of white cells, macrophages and 22. Note collagen mass and nodules (N). fibroblasts. Trichrome stain. Bar = lOOpm. Trichrome stain. Bar= lOOpm.

24 25

Figure 24. Computer depiction of reconstructed Figure 25. Computer depiction of reconstructed nodules from several sections of hypertrophic nodules from serial sections of hypertrophic scar. Through 43 sections nodule A remains scar. Through 26 sections nodule A remains separate from nodule B until they fuse at separated from nodule B. designated point( ➔).

the present study, we searched for evidence microvessels are white cells, fibrin, and that the pronounced increase in pericytes about occasional fibroblasts. the new microvessels, sometimes an apparent 5 From where do these first few fibroblasts layers, might be streaming off to form the new originate? This study was not designed to population of fibroblasts. We could not answer that question. But, it is important to confirm this possibility from our study. consider from the standpoint of other However, Beranek (1989) has proposed that some observations made in the present study. pericytes and/or some endothelial cells may be Heretofore, several theories have been put transforming into fibroblasts from the early forth including their origin from blood-borne growing tips of new capillary buds. It appears monocytes (Gillman and Wright, 1966), pericytes we did not examine granulations of a young (MacDonald, 1959; Rudolph et al., 1977; Kischer enough age to find this evidence. et al . , 1982a,b) or from peripheral resting Although we are unable to confirm this fibrocytes (Dunphy, 1963; Hadfield, 1963). In possibility from our observations, the idea is

885 C.W. Kische r, et al. appealing. It would not take much of a Granulations of an earlier age, presumably contribution (perhaps from growing micro­ before lateral vascular branching, would vascular tips) to produce a new actively probably not form nodules. It is of note that proliferating population of fibroblasts. This Estrem et al. (1987) were able to show the seems more plausible than a blood-borne stem development of keloid mass from a transplant of cell or inward migration from the periphery of cultured keloidal fibroblasts into the nude the wound. mouse. However, this mass does not show What we see throughout the granulation nodules. tissues with advancing age are increasing The collagen forming the nodules is most examples of degenerating pericytes and likely produced in copious amounts because of endothelial cells, mainly found in the lateral the fact that the vast majority of microvessels branching areas of older granulations. There in the older granulations are occluded by an is no corresponding accompaniment by increase of their own endothelial cells. This macrophages. phenomenon was first observed in the hyper­ Our studies indicate that hypertrophic trophic scar (Kischer et al. 1971), later scarring (nodule formation) shows a gradient reported for granulation tissue (Kischer, 1979) from the basal area to the surface. As and confirmed statistically in another study microvascular branching occurs groups of new (Kischer et al. 1982b). From another study of fibroblasts, which could have originated by microvascular occlusion, we predicted that exclusion or transformation from the early hypertrophic scars should be hypoxic (Kischer growing tips of new microvessels, become et al., 1975) because occlusion should obviously somewhat compartmentalized between the lateral produce blood stasis and, therefore, lower vascular branches. Thus, these areas in planar oxygen tension (P02). Later, hypoxia was profile between two lateral branches of a stem confirmed in hypertrophic scars through direct microvessel, oriented perpendicular towards the microprobe by Sloan et al. (1978). Occlusion free surface, probably are the "whorls" (the is sustained in the hypertrophic scar and forerunners of the nodules), which we reported keloid (Kischer et al., 1982b). As long as in an earlier study of granulation tissue occlusion persists the scar will persist. (Linares et al., 1973). Because the fibroblasts are facultative The increased numbers of active pericytes anaerobes, their ability to survive low P02 in young granulations are probably synthesizing may be greater than the case of pericytes and and excreting an interstitial matrix which endothelial cells and, thus, may live longer. would necessarily be aligned according to the Resolution and maturation of the lesion must growing alignment of the vascular branches. await the degeneration of a critical number of Thus, fibroblasts '"trapped" between these fibroblasts which would then reduce the collagen branches would proliferate in a corresponding synthesis/degradation balance in favor of the alignment accounting for unidirectional latter. alignment of cells and collagen within the nodule. Acknowledgments Contrary to an earlier study (Kischer and Brody, 1981), our computer generated model of This research was supported, in part, by the nodules suggests a variable shape, which NIH research grant #5R01GM34928. more nearly conforms to the events occurring during the early branching process of the new References microvessels and during the later events when some microvessels are degenerating and Beranek JT (1989). Ingrowth of hyperplastic disappearing. Loss of the latter would remove capillary sprouts into fibrin clots: further the interdiction between adjacent whorls evidence in favor of the angiogenic hypothesis (nodules) and allow for fusion. Unregulated of repair and fibrosis. Meds. Hypotheses 28, loss of the microvessels could account for the 271-273. variation and unusual shapes of some of the Dunphy JE (1963). The fibroblast a nodules. ubiquitous alley for the surgeon. New Eng. J. The fact that many of our implants proceed Med. 268, 1367-1377. to scarring ( where nodules form it would be Estrem SA, Domayer M, Bardach J, Cram AE hypertrophic scarring) provides evidence that (1987). Implantation of human keloid into the template for production of the scar is, athymic mice. Laryngoscope 97, 1214-1218. indeed, within the granulation tissue, but most Gillman T, Wright LJ TI966). Autoradio­ likely, after some lateral microvascular graphic evidence suggesting in vivo trans­ branching has occurred. We have demonstrated formation of some blood mononuclears in repair that vascular anastomosis occurs in and fibrosis. Nature (Lond.) 209, 1086-1090. hypertrophic scar implants by 16 days (Kischer Hadfield G (1963). The tissue of origin of et al., 1989b). It appears in this present the fibroblasts of granulation tissue. Brit. implant study that it also occurs here because J. Surg. 50, 870-881. of the later compartmentalization of the Karnovsky MJ (1965). A formaldehyde­ collagen bundles (nodules) by microvessels at a glutaraldehyde fixative of high osmolality for later implant stage. Thus, at least two of the use in electron microscopy. J. Cell Biol. E._, elements of the template necessary for 137a-138a. production of the scar appear to be branched Kischer CW (1974) . Collagen and dermal microvessels and stimulated fibroblasts. patterns in the hypertrophic scar. Anat. Rec.

886 Granulation Tissues and Hypertrophic Scars

179, 137-146. forming a capsule around the -- Kischer CW (1979). Fine structure of implant. However, in parallel studies of granulation tissues from deep injury. J. hypertrophic scar implants (Kischer et al. Invest. Dematol. 72, 147-152. Anat. Rec. 225:189, 1989), we found mouse cells Kischer CW, Brody GS (1981). Structure of do not invade the implant essentially beyond the collagen nodule. Scan. Electr. Microsc. the capsule. Therefore, the implant retains III, 371-376. its original, pre-implantation, morphology -- Kischer CW, Linares HA, Dobrkovsky M, throughout the period of implantation (Kischer Larson DL (1971). Electron microscopy of the et al. J. Trauma 29:672-677, 1989). There is hypertrophic scar. In 29th Am. Proc. Electr. no comparison possible between the implant mode Microsc. Soc. Amer. Claitor's Puhl. Div., Baton and development of a hypertrophic scar because Rouge, pp. 302-303. the scar has already developed and is not Kischer CW, Shetlar MR, Shetlar CL (1975). altered morphologically as an implant. Alteration of hypertrophic scars induced by mechanical pressure. Arch. Dermatol. 111, C.J. Doillon: How do you know if the early 60-64. wounds (1 month or less) will behave to either Kischer CW, Shetlar MR, Chvapil M (1982a). a hypertrophic scar tissue/keloid or a mature Hypertrophic scars and keloids: A review and scar tissue? Are the patients followed at long new concept concerning their origin. Scan. term? Electr. Microsc. IV, 1699-1713. Authors: All the parameters involved in the Kischer CW, Theis AC, Chvapil M (1982b). development of a hypertrophic scar are not Perivascular myofibroblasts and microvascular known. That is one reason why we embarked on occlusion in hypertrophic scars and keloids. this current study. However, all of the Human Path. 13, 819-824. granulation tissues we studied were from full Kischer CW, Pindur, J, Shetlar MR, Shetlar thickness injuries. Under that condition all CL (1989a). Implants of hypertrophic scars and will proceed to produce a hypertrophic scar. keloids into the nude (athymic) mouse: If there is a doubt about this it would be viability and morphology. J. Trauma 29, found in cases of chronic granulation tissue 672-677. and in aged patients. The term "mature" scar Kischer CW, Sheridan D, Pindur J ( 1989b). has often been used synonymously with "normal" Use of nude (athymic) mice for the study of scar. This implies that this kind of scar hypertrophic scars and keloids: vascular never goes through a stage of hypertrophy. We continuity between mouse and implants. Anat. do not believe this occurs. The period of Rec. 225, 189-196, hypertrophy in those cases may be of Krueger GG, Briggaman RA (1982). The nude sufficiently short duration so as to escape the mouse in the biology and pathology of skin. eye of the clinician. Most of these patients Exp. Clin. Res. 2, 301-322. were children and, indeed, were followed Linares HA,- Kischer CW, Dobrkovsky M, clinically for some time and either fitted for Larson DL (1973). On the origin of the pressure wraps or had their subsequent scars hypertrophic scar. J. Trauma 13, 70-75. reconstructed. MacDonald RA (1959). Origin of fibroblasts in experimental wounds: C.J. Doillon: How do you know the nodules autoradiographic studies using tritiated formed in nude mice are from human origin? thymidine. Surgery 46, 376-382. Authors: We believe the implanted granulation Peacock EE Jr~ Madden JW, Trier WC tissues formed nodules within (and by) the (1970). Biologic basis for the treatment of donor tissue, not originating from the mouse. keloids and hypertrophic scars. Southern Med. In parallel studies of implants from J. 63, 755-759. hypertrophic scars we used a Hoechst stain Rudolph R, Guber S, Suzuki M, Woodward M procedure that clearly distinguishes mouse (1977). The life cycle of the . cells from human cells (Kischer et al. 1989 Surg. Gyn. Obst. 145, 389-394. Anat. Rec. 225:189-196). Mouse cells may be Sloan DF, Brown RD, Wells CH, Hilton JG incorporated--minimally into the capsule (1978). Tissue gasses in human hypertrophic surrounding the implant, but they do not invade burn scars. Plast. Reconstr. Surg. 61, 431-436. the implant itself. Mouse endothelial cells anastomose with peripheral endothelial cells Discussion with Reviewers from donor microvessels and establish vascular continuity by about the 16th day J.P. Waterhouse: The granulation tissue post-implantation. fragments grafted into a nude mouse are placed into a non-identical immunologically impaired A.J. Wasserman: In addition to the branched host, and are then subject to a host response microvasculature, what role do other short of graft rejection. This will probably macromolecules such as fibrinogen, heparin, include the growth of fibrous tissue around it and transglutaminases play, if as a reaction. Under what circumstances can any, in nodule formation? What matrix morphological appearances in this situation be macromolecules are responsible for the compared to those in the development of a "stimulated fibroblasts" which you claim are at hypertrophic scar in a single injured least one "of the elements of the template individual? necessary for production of the scar"? Authors: The mouse indeed participates in Authors: This question goes beyond anything

887 C.W. Kischer, et al. yet known about nodule formation and any answer (Peacock et al. 1981. Ann. Surg. 193:592-597) would be nothing more than idle speculation. that treatment with BAPN in a caseof excised Fibroblasts may be stimulated by one or more of keloids restricted the recurrent lesions to the a host of factors and molecules in the milieu status of hypertrophk scars. No morphological of granulation tissue, including PDGF, as shown studies were done on the recurring lesions. by in-vitro studies (Kischer & Pindur, 1990. Cytotech. 1_, 238-238). B. Persky: How might one attempt to control the regeneration of microvasculature in order A.J. Wasserman: You have speculated quite a to prevent/reduce hypertropic scar or keloid bit about the genesis of the early granulation formation? tissue fibroblasts. Would you care to try and Authors: That is a difficult question to identify the macromolecules responsible for answer. In some cases, as soon as autografts inducing the differentiation of the pericytes "'take"' the grafted area is wrapped with light or endothelial cells into fibroblasts. pressure. This has some effect on reducing the Authors: No one knows anything about specific amount of host-donor vascular anastomoses, and inductive macromolecules. The reference by may reduce the incidence or severity of a Beranek (1989) is intriguing and could be subsequent hypertrophic scar. However, grounds for a concentrated study, however. clinically, I am not aware of any treatment to prevent microvascular regeneration in A.J. Wasserman: This question refers to Figure granulation tissue. This would be contrary to 10. Depicted in the micrograph is 3 month old good surgical procedure because grafts "'take" granulation tissue which contains fine on a well supplied vascular bed. Grafting as filaments covering collagen bundles, early as possible is desirable but not always microvessels and cells. The fine filaments possible. Early grafting would have the look like collagen in the formative stages, indirect effect of reducing the overall typical of regenerating tissue. Please regeneration of the microvessels. Other than speculate as to what you think this filamentous the above, not enough is known as yet, to material is. manipulate the granulation tissue appropriately. Authors: I speculate that this material may be 1) proteoglycans; 2) young collagen fibrils of B. Persky: Can hypertropic scarring be small diameter, about 200 to 400 angstroms D; selectively induced in order to enhance the 3) a combination of both. interface between the skin and artificial prosthetic implants? A.J. Wasserman: What are your feelings a bout Authors: This is a clinical question about the use of lathyrogens to regulate fibrotic which I have little knowledge. I would say, tissue development and perhaps nodule formation? however, that fibrosis occurs rather quickly Authors: Lathyrogens, such as BAPN around prostheses. If one wants more scarring, (betaaminoproprionitrile), have been used the key would be to promote more granulation experimentally, and, in some restricted cases, tissue by invasive injury combined with new clinically in the attempt to control collagen microvascular regeneration. There are a number production (i.e., fibrosis). BAPN apparently of factors which promote increased collagen acts by inhibiting the cross-linking enzyme synthesis; however, I am unaware of any lysyl oxidase. To my knowledge it is not beneficial clinical or therapeutic use of these approved for clinical treatment by the FDA. factors at this time. There have been claims in the literature

888