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

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

15.3.3.4 Carbon Fiber Composites

Although carbon fiber electrodes are in common use as biosensors [4847-4851] and related applications [4852-4854], carbon fiber composites such as carbon fiber-reinforced carbon (CFRC) materials have a mixed record where biocompatibility is concerned [4855]. For example, in one experiment [829] a vascular prosthesis made of pyrolytic carbon fibers was implanted on the infrarenal aorta of growing pigs, then was removed up to 120 days after surgery. The carbon grafts showed thromboresistance of the inner surface at the time of implantation, development of a thin neointima with good viability, rapid and complete endothelialization of the flow surface, and solid anchorage to perigraft tissues. But another in vivo study [830] of carbon-carbon composites in contact with blood showed the accumulation of platelets on exposed surface material having any surface morphology, although platelet concentration in blood remained constant. Bulk structure of composites influences the adhesion mechanism of entrapped platelets (e.g., active adhesion vs. mechanical adhesion).

Carbon fiber patches inserted as prostheses into the dorsal lumbar fascia of rabbits initially had poor mechanical properties but developed good connective tissue response after several weeks [831]. In another study [832], carbon fabric with 35- to 50-micron diameter fibers infiltrated with low temperature pyrolytic carbon produced a tightly woven porous structure with maximal pore size >200 microns; 30 days after percutaneous implantation in a calf, epidermal downgrowth was minimal. Later, a thin fibrous capsule surrounded the implant and mature connective tissue with accompanying blood vessels filled the pores of the fine trabecularized carbon layer, allowing a biocompatible connection between an artificial internal organ system and the external environment [832].

Carbon fiber has often been tested as a prosthesis for ligaments because filamentous carbon is a known fibrogenic material, inducing the formation of replacement collagen [833, 934, 935]. In one early experiment that gave good results [834], the biological reaction of tissues to carbon fiber ligament prostheses was examined in sheep knees. Connective tissue and bone grew into the prosthesis under physiological conditions at the insertion points in cancellous bone, and there was tissue ingrowth around the carbon fiber ligaments intra- and extra-articularly. Carbon fibers were reported to be a very good scaffolding and a permanent prosthesis for ligament replacement.

But other experiments gave poorer results. One study [835] found that carbon fiber used to reconstruct anterior cruciate ligament in the knee did not bond to bone nor did it induce the formation of new ligament. There was only a very minor fibroblastic response despite the presence of numerous particles of carbon fiber scattered throughout the knees [835]. In yet another study [837], part of the patella ligament in rabbits was resected and replaced by carbon fiber implants. After residence times ranging from 1 week to 1 year, the carbon implants along with surrounding tissues and iliac lymph nodes were removed and examined by light- and transmission electron microscopy to determine whether the carbon fiber implant would be removed by phagocytosis and substituted by new ligament or some other adequate repair tissue. In this study, there was no indication of successful removal of carbon fibers by phagocytosis and the implant was surrounded by dense connective tissue like a scar. No vital dense or regular connective tissue was found in deep layers of the implant, even after 3 and 12 months, and no replacement of the carbon fiber implant by new ligament or tendon. A persistent foreign body reaction was observed, leading the authors to conclude that “it is very doubtful [that] good results with ligament and tendon replacement by carbon fiber implants can be expected in patients.”

Similar results were obtained in humans. For example, carbon fiber ligaments implanted in human patients has evidenced a considerable foreign-body reaction to the carbon fibers and insufficient metaplasia of the allogenic material to connective tissue [836]. In another study [838], more than a year after carbon fiber was used to reconstruct the lateral collateral ligament of a human knee, histological study suggested that the implant was unlikely ever to acquire the structure of a natural ligament. This was true even though the implant was biologically compatible and was deemed biomechanically sufficient as long as the entire tow of carbon fibers was preserved [838].

Carbon fiber has been tested as a bone prosthesis with better outcomes, failing only rarely [839]. For instance, long-term middle-ear implantation of carbon-carbon prostheses in guinea pigs produced no significant tissue destruction or inflammation, no digestion or erosion of the implants, and no passage of carbon particles into the reticuloendothelial system [840]. Carbon fiber-reinforced carbon implanted in holes drilled in rat femurs developed a thin layer of fibrous tissue bridging the gap between bone and implant for a period up to 8 weeks, but by 10 weeks bone was observed adjacent to the implant, giving firm fixation [841]. Another study [842] found that carbon-carbon cloth sandwich provided good osseocompatibility. The fiber-like surface texture gave a degree of bony attachment of greater strength than for titanium during 4-40 weeks post implantation [842]. In yet another experiment [843], CFRC material with 30-micron pores was implanted as femoral transverse diaphyseal pins in rats. By 45 weeks, most specimens showed direct implant-bone contact over most of the interface although the interface was chemically abrupt with no cross-diffusion of ionic species. Implant pores were partially filled with tissue including fresh bone organized de novo deep within.

What about carbon fiber and carbon composite particulates? In inhalation experiments, no fibrosis, local reactive pulmonary inflammation, or other significant effects were observed in rats exposed to 7-micron thick, 20- to 60-micron long carbon fibers for 30 hours per week during a 16-week experiment at an average chamber concentration of 20 mg/m³, although the inhaled particles were phagocytosed by alveolar macrophages [224]. Rats showed no fibrosis or other ill effects from inhaling 20 mg/m³ (25 million fibers/m³) of carbon fibers measuring 3.5 microns in diameter and 10-60 microns in length for 30 hours per week during a 16-week experiment [765]. Nonfibrous dust particles from pounded carbon fiber, inhaled by guinea pigs, were phagocytosed. Carbon fibers longer than 5 microns were still extracellular after 27 weeks and were uncoated; no pathological effects were observed [223, 844]. In another experiment [2584], a 5-day exposure to respirable carbon fibers at 50-120 gm/m³ produced dose-dependent transient inflammatory responses in rat lungs, but no significant difference in the morphology or in vitro phagocytic capacities of macrophages were observed. Medical examination of carbon fiber production workers has revealed no adverse effects on the lungs [225]. However, one Russian animal study found slight pulmonary fibrosis and respiratory tract irritation from carbon fiber dust [226] and a Japanese study found morphological changes in rat lungs due to some kinds of carbon fibers [762]. One other study of several aerosolized carbon composites found some that showed little toxicity, but found others that were consistently toxic for alveolar macrophages and caused significant accumulations of airway cells and neutrophils in rat lungs [845].

Carbon fiber particles can elicit a cellular response. In one experiment [846], wear particles produced from Versigraft carbon when added to rabbit synovial cell culture induced significantly elevated collagenase and gelatinase enzyme production. 1 mg/kg particles injected into rabbit knees accumulated in the periarticular synovial folds and induced strong macrophage infiltration in the synovium [846]. In another study [784], carbon fiber-reinforced carbon particles of up to 20 microns in diameter were phagocytosed when presented to in vitro cultures of murine macrophages. Larger particles were not phagocytosed but became surrounded by aggregations of macrophages, some of which migrated onto the particle surfaces [784]. Cells presented with a large excess of particles became rounded and detached from the substrate, and some underwent lysis [784]. In yet another experiment [847], wear particles from carbon prosthetic materials were cultured with rabbit synovial fibroblasts. Internalized particles induced collagenase, but even carbon particles that remained extracellular provoked considerable collagenase synthesis. Synovicytes that contained no particles nevertheless produced collagenase when co-cultured with cells that did contain particles. This indicates that carbon fiber particle phagocytosis, besides inducing collagenase, also provokes the release of cell-activating factors which then activate other cells in the culture [847].

The soft tissue response to long carbon fibers and carbon fiber microparticles is said to be excellent [848, 902]. In general, carbon fibers are integrated by the organism without causing significant foreign body reaction (inflammation), with normal tissue growth around (and encasement of) the individual fibers [848]. There is, however, progressive rupturing of pure carbon fiber implants. The resulting carbon fiber microparticles are absorbed either by macrophages or by foreign body giant cells and are distributed throughout the body via the lymphatic system [848]. In one animal study [826], particles from carbon fiber reinforced carbon of sizes 11 microns and 30 microns implanted into the triceps surae muscle of Wistar rats produced no muscle tissue necrosis or exudative reaction during the acute phase (~1 week). During the chronic phase (up to 52 weeks), the 11-micron particles induced only a modest inflammatory infiltration of fibroblasts and phagocytes while the 30-micron particles induced a much larger infiltration of fibroblasts, macrophages, and giant cells. The study by Helbing et al [902] found excellent tissue tolerance of <~8-micron particles of carbon-fiber reinforced carbon, in rats. In humans, the soft tissue response to carbon fiber was studied histologically one and a half years after being used to reconstruct the lateral collateral ligament of the human knee [838]. A remarkably consistent pattern was seen in the induced ligament. The basic pattern was a composite unit, consisting of a core of carbon fiber enveloped in a concentric manner by coherent layers of fibroblasts and collagen fibers. The new structure seemed to have been induced by continuous irritation caused by the physical structure of the carbon fibers.

Bones and joints seem to tolerate carbon fiber particles rather well. In one experiment [849], carbon fiber fragments with a diameter of 7 microns and a length between 20-100 microns were injected in the medullary canal (in long bones) of 16 rabbits, and evaluated after periods of 2 and 12 weeks. There was phagocytosis of small carbon fiber fragments by macrophages, but only a minimal foreign body reaction to the intramedullary carbon fiber fragments. A small amount of fibrosis was observed around some carbon fibers along with a small amount of new bone formation with inclusion of carbon. Only a few carbon fragments were transported to the parenchymal organs, with no foreign body reaction [849]. In a similar study in humans [835], carbon fiber wear particles scattered throughout the knees stimulated only a very minor fibroblastic response.

Last updated on 30 April 2004