PubMed1998
Although the histology of capsular tissue is well described in the literature, most studies in humans do not correlate histologic findings with implant age (number of years an implant was in place before sampling). As such, questions regarding the long-term histology in humans remain. The microanatomy of 93 human periprosthetic capsular tissues surrounding 22 textured and 71 smooth silicone-surfaced prostheses was studied. The implants were divided into two groups according to the time between implantation and capsulectomy: between 0 and 5 years or more than 5 years. Hematoxylin and eosin and Masson trichrome-stained sections were analyzed by light microscopy, with and without polarization. Eighteen of the textured implants contained silicone gel and four contained saline. Sixty of the smooth implants contained silicone gel, eight contained saline, and in three, the filler type was not known. For the majority of patients, surgery was performed for augmentation mammaplasty, and the implants were removed because of capsular contracture. The following histologic features were assessed: synovial-like metaplasia, villous hyperplasia, density of the collagenous capsule, alignment of collagen fibers within the capsule, and the presence of foreign material and of a foreign body reaction. The following trends were observed. In smooth implants, increasing implant duration was associated with a decrease in the presence of synovial-like metaplasia (p = 0.003) and villous hyperplasia; there was no significant difference in the presence of a dense collagenous capsule, the orientation of collagen fibers, or the presence of a foreign body reaction. An increase was observed in the presence of foreign material (p = 0.01). In textured implants, increasing implant duration was associated with a decrease in the presence of synovial-like metaplasia, villous hyperplasia (p = 0.003), dense collagenous architecture, and parallel orientation of collagen fibers (p = 0.017). An increase in the presence of a foreign body type reaction and foreign material (p = 0.024) was observed. In comparing textured and smooth-surfaced implants, synovial-like metaplasia was observed more often in the textured group, both at 0 to 5 years (p = 0.01) and at greater than 5 years (p < 0.01). Textured implants more often had villous hyperplasia at 0 to 5 years (p = 0.03) but not beyond 5 years. Smooth implants more often had a dense collagenous capsule than textured implants after 5 years. No significant difference was seen in the orientation of collagen fibers in capsules around smooth and textured implants at 0 to 5 years. After 5 years, the incidence of capsules with collagen fibers arranged parallel to the implant surface was significantly greater in the smooth group than in the textured group (p = 0.01). The presence of a foreign body type reaction was seen more often in the textured group between 0 and 5 years (p = 0.01) and at greater than 5 years (p < 0.01), and the presence of foreign material was more often seen in the textured group between 0 and 5 years (p = 0.06) and at greater than 5 years (p < 0.01). In summary, the cytologic changes around implants seem to be dynamic in nature, and implantation duration and shell type play a significant role. Synovial-like metaplasia, villous hyperplasia, and foreign material were more often observed in the textured group within the 0 to 5 year interval. Beyond 5 years, synovial-like metaplasia, a foreign body type reaction, and foreign material were more often observed in the textured group. Differences in the density of collagenous capsules were not significant at any time point, and collagen fibers oriented parallel to the implant surface were more often observed in the smooth group after 5 years. The significance of these findings awaits further investigation.