Nanomedicine, Volume IIA: Biocompatibility

© 2003 Robert A. Freitas Jr. All Rights Reserved.

Robert A. Freitas Jr., Nanomedicine, Volume IIA: Biocompatibility, Landes Bioscience, Georgetown, TX, 2003 Foreign Body Granulomatous Reaction

As a general principle [1841], the human body reacts to insoluble foreign bodies placed within it either by extruding them (if they can be moved and an external wall is close at hand) or by walling them off by exactly the same process as wound granuloma formation* (Chapter 24). Willert [3157] has pointed out that while small amounts of indigestible particles can be stored locally or transported away through the lymphatic drainage (Section, large quantities of particles can overwhelm the normal process and produce (1) a histiocytic granulation tissue with accompanying fibrosis, which results from attempts to encapsulate and isolate the reaction, and (2) progressive tissue loss through necrosis and attempts at remodeling [234], a phenomenon sometimes called “small particle disease” [234, 2669, 3157] or “nano-pathology” [5638].

* This process is relatively slow, with mostly neutrophils arriving during the first 6-24 hours, replaced by monocytes after 24-48 hours [1841].

As an example, in one experiment up to 7.5 mg/kg of glass fibers (~0.2 cm3/70 kg) instilled peritoneally in rats were taken up by peritoneal organs in 1-2 days. But at higher doses, the excess foreign material formed clumps of fibers (nodules) that were either free in the peritoneal cavity or loosely bound to peritoneal organs. The nodules displayed classic foreign body reactions with an associated granulomatous inflammatory response [3158]. The granulomatous reaction of the body to Teflon particles has been exploited therapeutically (Section, and granulomatous foreign body reaction has been reported for a diverse range of materials including colloidal carbon [3159], cholesterol [3160, 3161], collagen [3162], cotton and other surgical textiles [3163-3165], fish bone [3166], gallstone [3167], glass [3168], graphite [2513, 2514], hair [6166-6169], mercury [3169], metal particles [3170, 5824], plastics [3171-3173], silica [3174-3177], silver needle [3178], sutures [3179, 3180], swabs [3181], talc [3182], thorns [3183-3185], and wood [3186-3188]. The possibility of nanorobotic foreign-body carcinoma is discussed in Section 15.2.8.

Studies of silica-induced fibrosis [3134] suggest that activation of macrophages by foreign materials [3189, 3190] is a prerequisite for release of chemotactic factors (which summon other phagocytes to the site) and cytokines [3191]. The chemical activity of phagocytosable particles does not seem to be primarily responsible for their cellular stimulatory effect [234]. Activated macrophages that encounter foreign particles larger than a few microns in size can multiply by mitosis or be stimulated to merge with other macrophages [2857, 2858] to form a relatively sessile multinuclear foreign body giant cell (FBGC) [2668, 3192]. Reaching up to 80 microns in diameter, the FBGC cell can more aggressively phagocytize larger particles than individual neutrophils, eosinophils, or macrophages alone can attack. For example, inhaled short inorganic fibers (<5 microns) are phagocytized by alveolar macrophages, but long fibers (>10 microns) are phagocytized by FBGCs in rats, hamsters and guinea pigs [3193]. Also, a macroscopic ocular lens implanted in mice produced multi-macrophage FBGC aggregates [3194]. Fibroblasts then surround the FBGC aggregate and form a fibrous wall around the object. Encapsulation or marsupialization (see below) could well be the fate of an immobilized medical nanorobot that is unable to avoid triggering phagocyte activation (Section

As a general principle, granulomas are proximately mediated by the local release of interleukins such as IL-1beta [4650] and other interleukins [4651, 4652], and by proinflammatory C-C cytokines such as monocyte chemotactic proteins MCP-1 and MCP-2 [4650] and other cytokines [4653-4656]. These cytokines help to recruit new leukocytic cells to the site. Granulomas display characteristic cytokine profiles with coordinated expression that is under cytokine-mediated regulation [4656]. Medical nanorobots may be equipped with molecular sorting rotors to absorb some or all of these recruitment or key mediating cytokines [4657], thus reducing their local concentration to near-background levels and effectively short-circuiting the granuloma-formation process. The effects will be similar to the results in knockout mice lacking critical chemokine receptors whose ability to form granulomas is thereby artificially impaired [4658-4660]. A less elegant alternative would be to release anti-interleukin antibodies which have been shown to partially abrogate pulmonary granuloma formation and to inhibit leukocyte recruitment in mice in vivo [4661], or to release a receptor antagonist for IL-1 [2157]. Other granuloma inhibition strategies might also be pursued [4662-4664, 5367-5379]. In the case of long-term nanorobot missions or augmentations (Chapter 30), a key design issue will be whether granuloma inhibition can be achieved locally without blocking the function systemically, or alternatively, how to replace the lost function served by granuloma formation, using artificial means, once the natural means have been permanently systemically suppressed.

As Peacock [1841] pithily describes the process of granuloma formation: “In granulomatous reactions, the macrophage is usually found immediately adjacent to the inciting material or it may actually have phagocytosed it. Fibroblasts move into the area and surround the cluster of macrophages. Collagen is laid down, eventually enclosing the lesion in a dense fibrous capsule. These hard spheres of fibrous tissue constitute the granuloma.”

Granulomas can be comprised of macrophages (foreign body reaction), epithelioid cells (immune granulomas of sarcoidosis, tuberculosis), or skin macrophages or Langerhans’ cells (histiocytosis X) [3195] that have ingested foreign material but cannot digest or exocytose it. Activated macrophages and tissue monocytes release cytokines such as angiogenic growth factors that induce the invasion of capillaries into the granulation tissue [234]. Large amounts of mucopolysaccharides and collagen are synthesized and formed into a scaffold for cellular reconstruction and remodeling in this tissue, a process called fibroplasia [234, 3196]. Similar processes (granuloma formation) in latex-bead- and ink-particle-stuffed fibroblasts that become trapped in the connective tissue of the dermis are responsible for the long-term persistence of tattoos [778]. By one month after implantation, the granuloma has become a relatively acellular [3197] fibrous capsule that is maintained by the presence of the implant [234]. If the foreign body is then removed, the capsule may collapse into a residual scar or be completely remodeled [234]. The potential for encapsulation may apply to isolated particles such as medical nanorobots, aggregates of such particles such as communicyte (Section 7.3.2) or navicyte (Section 8.3.3) arrays, or to the outer surfaces of macroscale implants such as artificial nanorobotic organs. Trapped nanorobots can still communicate chemically with the external environment even in the absence of transgranulomatous mechanical penetration [6140] by nanorobot appendages, e.g., via simple chemical diffusion (Section 3.2) of small molecules through granuloma walls [6141-6144].

Many factors can influence the thickness of the fibrous capsule. Chemically active materials such as corrosible metals or leachable polymers will mediate formation of a capsule whose thickness is directly proportional to the rate of release of these constituents [234, 3201]. Besides concentration, the chemical nature of the released materials or surface composition may be cytotoxic, inhibitory, or neutral [3199-3201]. For example, pure titanium may elicit a minimal fibrous encapsulation under some conditions, whereas stainless steel implants can induce a thick fibrous layer up to 2 mm deep [3201]. In experiments with rats [3200], polyethylene implants coated with RGD (a tripeptide) or poly-L-lysine had thicker capsule formation than RGE-coated implants. Active medical nanorobots should be able to control these emissions and surface characteristics. Even disabled devices, if constructed of diamondoid materials and physically intact, should remain chemically inert and not corrode (Section or leach (Section 15.3.7).

Besides chemical inertness, mechanical factors are also important in mediating capsule formation [234]. For example, an absolutely smooth surface discourages extensive fibrosis, although a slight roughness of the surface (even microscopic irregularities), particularly if the roughness is ordered as in linear scratches or in the weave of a fabric, leads to increased fibrous reaction [1841]. Formation of a fibrous capsule around a nanorobotic organ implant will be markedly aided if very fine lines are etched on the surface of the implant because fibroblasts show directional movement on an oriented substrate by a process called “contact guidance” (Section

Capsules also become thicker with increased relative motion between implant and tissue [234, 3201-3207], sometimes [1841] but not always [3202] as a result of mild injury to adjacent tissue. Variation in the distribution of strain between implant and tissues can alter the spatial pattern of fibrous tissue thickness surrounding the implant [3211]. In extreme cases, a painful fluid-filled bursa mimicking a synovial capsule may form around the implant [234]. One study [3207] found that cylindrical implants inserted into dog femur bone and laterally oscillated in vivo for 8 hours per day produced stable bone ingrowth up to 20 microns of oscillation, but not at 40-150 microns of oscillation, which produced excess fibrous ingrowth. When implanting a material, caution should be taken that the implanted material has roughly the same mechanical properties as the surrounding tissue [3207-3210]. A significant mismatch (Section, such as a difference in Young’s modulus between the implant and the surrounding tissue, could induce the formation of a relatively thick capsule. This factor may be most relevant to such nanomedical systems as in vivo tethers (Sections, fiber networks (Section 7.3.3), pressure ulcer resistant nanorobotic garments (Section, vasculomobile nanoaggregates (Section 15.5.3), transdermal portals (Chapter 19), and nanoorgans (Chapter 14) employing external mechanical effectors.

Implant shape [5728] may also affect the thickness of the fibrous capsule [234, 3212-3215]. Capsules will become thicker over edges and sharp changes in surface features. For example, the capsule surrounding a rectangular slab of reactive material will be dogbone or club-shaped, a phenomenon called “clubbing” [3212]. All else equal, cylindrical implants form stronger soft-tissue attachments than flat rectangular implants [3214]. Implants with features offering a reduced solid angle to surrounding tissue reduces the accessibility of that nearby tissue to microbe-killing neutrophils if it ever becomes infected [234]. Similarly, implants having corners with the most acute angles produce higher inflammatory response in the absence of infection – in one study [3216], otherwise similar implants having a triangular shape showed the highest enzyme activity and cellular response, pentagon shapes showed less, and circular rods showed the least activity. The existence of “dead spaces” – volumes filled with cell-free fluid rather than tissue – is a special geometric hazard because this fluid can act as an in vivo culture medium for bacteria [234]. Thus in nanorobotic organs, adherence of soft tissue to the implant will usually be desirable in order to eliminate these fluid-filled cavities, thus helping to decrease the risk of infections. Other characteristics of implants that may lead to surface infection and the spread of biofilms have already been reviewed in Section

Electrical currents, such as those emanating from an implanted stimulatory electrode, can also produce capsules [3217-3219] whose thickness is sometimes related to current density [234], although sometimes there is little or no fibrous reaction [3220]. Electrodes can release corrosion products while also mediating changes in local pH and pO2, so effects due to direct electrical (faradic) and indirect electrochemical (electrodic) stimulation can easily be confused [234].

Fibrous tissue capsule thickness is also influenced by implantation location within the body. In one series of experiments with rats [3198-3200], intraperitoneally-placed implants had a more extensive fibrous and vascular tissue formation and more numerous associated inflammatory cells than subcutaneously-placed implants. In another experiment with dogs [3218], intramuscular-placed electrodes produced thicker fibrous capsules than epimysial-placed electrodes.

Finally, thickness tends to increase with duration of the implant in the body. For instance, the external fibrous capsule surrounding one implant reached 4-5 mm in thickness, 19 years post-implantation [3221].

Black [234] notes that implant “resolution” – a final state after which no further progressive biological changes occur – can have four possible outcomes:

(1) Resorption. A resorbable implant [3222] eventually resolves to a collapsed scar, or in the case of bone, may entirely disappear. Most medical nanorobots probably will not be resorbable. However, if nanorobots retain mobility after encapsulation they could later migrate away from the site, producing a similar outcome.

(2) Extrusion. The local host response to an implant in contact with epithelial tissue will be the formation of a pocket or pouch continuous with the adjacent epithelial membrane, a process called “marsupialization” [3223, 3224] due to the structural similarity to a kangaroo’s pouch. In the case of the external epithelium (skin), marsupialization results in the extrusion of the implant from the host unless the implant is anchored in the deep connective tissue or other deep tissue [3223]. Nanorobot control and mobility systems should prevent this outcome, unless it is desired.

(3) Integration. In a very few cases such as the implantation of pure titanium in bone [3225], a close, possibly adhesive, approximation of nearly normal host tissue to the implant is possible without an intervening granulomatous capsule, although inflammatory cells may persist in small numbers. With proper surface engineering (Section 15.2.2), good tissue integration is a very real possibility for medical nanorobots or nanoorgans.

(4) Encapsulation. This is the most common response to, for example, implant wear particles [2668-2670], carbon particles [902], or cosmetic microimplant particles [3171-3173]: formulation of granulation tissue with a fibrotic capsule surrounding the foreign body. If an implant is placed in a location where bone may form (e.g., within a medullary space) and does not achieve osseointegration, then the fibrotic capsule may become mineralized in which case the granuloma is called a “sequestrum.” Small particles also can elicit the release of cytokines that stimulate large phagocytic cells called osteoclasts to resorb bone [3226]. Such particles may even inhibit bone formation by osteoblasts [254], resulting in overall bone loss and a loosening of the implant at the implant-bone interface, possibly with some local tissue necrosis. Selective absorption or emission of appropriate factors (e.g., cytokines to stimulate osteogenesis [5618-5620] or revascularization via angiogenesis [5621-5623], or bone morphogenetic proteins [5624]) by nanomedical implants could reduce or eliminate these negative effects. If encapsulation is inevitable, nanorobots can be designed to accommodate this natural reaction. For example, an encapsulated nanorobot could extend sensor-tipped telescoping stalks through the capsule, enabling collection of sensor data outside the capsule’s outer wall. Once sufficient readings have been taken, the sensor stalks could be retracted back into the nanorobot without further disturbing the capsule.

“Whether each of these resolution outcomes represents success or failure of the implant depends on the circumstances [and] the desired consequences of the insertion of the implant,” observes Black [234]. “This is the basic idea of biocompatibility: biological performance in a specific application that is judged suitable to that situation.”


Last updated on 30 April 2004