Research Article: PCR-hybridization after sonication improves diagnosis of implant-related infection

Date Published: June 4, 2012

Publisher: Informa Healthcare

Author(s): Jaime Esteban, Noelia Alonso-Rodriguez, Gema del-Prado, Alberto Ortiz-Pérez, Diana Molina-Manso, Jose Cordero-Ampuero, Enrique Sandoval, Ricardo Fernández-Roblas, Enrique Gómez-Barrena.


We wanted to improve the diagnosis of implant-related infection using molecular biological techniques after sonication.

We studied 258 retrieved implant components (185 prosthetic implants and 73 osteosynthesis implants) from 126 patients. 47 patients had a clinical diagnosis of infection (108 components) and 79 patients did not (150 components). The fluids from sonication of retrieved implants were tested in culture and were also analyzed using a modified commercial PCR kit for detection of Gram-positive and Gram-negative bacteria (GenoType BC; Hain Lifescience) after extraction of the DNA.

38 of 47 patients with a clinical diagnosis of infection were also diagnosed as being infected using culture and/or PCR (35 by culture alone). Also, 24 patients of the 79 cases with no clinical diagnosis of infection were identified microbiologically as being infected (4 by culture, 16 by PCR, and 4 by both culture and PCR). Comparing culture and PCR, positive culture results were obtained in 28 of the 79 patients and positive PCR results were obtained in 35. There were 21 discordant results in patients who were originally clinically diagnosed as being infected and 28 discordant results in patients who had no clinical diagnosis of infection.

For prosthetic joint infections and relative to culture, molecular detection can increase (by one tenth) the number of patients diagnosed as having an infection. Positive results from patients who have no clinical diagnosis of infection must be interpreted carefully.

Partial Text

The mean age of the patients was 66 (23–96) years; two-thirds were women. As expected, the mean age was lower for patients with osteosyntheses than for patients with prostheses (56 vs. 73 years, p < 0.001, Kruskal-Wallis test). All cases of infection were late (i.e. infections that appeared 3 or more weeks after surgery) except 2 cases of S. aureus, which were acute prosthetic joint infections. Most reports on the use of molecular biological tools to detect bacteria in periprosthetic tissues have been based on amplification of a broad-range bacterial target, such as 16S rRNA sequences, and subsequent identification of the organisms detected by sequencing of the fragment. More recently, the use of specific primers to detect specific pathogens has also been evaluated (Piper et al. 2009). The main problem with these techniques is the lack of standardization, so their use in clinical practice is problematic.   Source: