Seven RCTs (480 implants) were included in the meta-analysis of efficacy. The evaluation of complications was based on 532 patients. Reasons for the discrepancy between the numbers analysed for efficacy and safety were not reported.
One study was assigned a Jadad score of 2 for quality; all other scores ranged from 3 to 5 out of 5 points.
There was a statistically significant reduction in the rate of capsular contraction associated with textured compared with smooth breast implants (OR 0.19, 95% CI: 0.07, 0.52). Significant statistical heterogeneity was found (p=0.006). Four of the seven studies reported no significant difference between treatments, while three studies reported a statistically significant reduction in capsular contraction with textured compared with smooth breast implants.
Subgroup analyses showed a significant reduction in capsular contracture with textured versus smooth breast implants for both saline or silicone implants, for implants with either Siltex or Biocell surfaces, for studies with longer duration of follow-up (more than 3 years) and for subglandular placement of the implant. There was no significant difference between textured and smooth implants for submuscular implants (based on one study that the authors considered might have been underpowered; 104 implants).
Some studies reported complications without stating which treatment group the patients had been allocated to; as such, these events could not be included in the meta-analysis.
There was no significant difference between textured and smooth breast implants in the number of complications (4 of 271 patients versus 5 of 261 patients), reoperation for complications (3 out of 271 versus 5 out of 261), infection (3 out of 271 versus 1 out of 261), or reoperation for infection (2 out of 271 versus 1 out of 261).