Search tips
Search criteria 


Logo of intorthopspringer.comThis journalToc AlertsOpen ChoiceSubmit Online
Int Orthop. 2009 June; 33(3): 829–833.
Published online 2008 March 8. doi:  10.1007/s00264-008-0534-y
PMCID: PMC2903123

Language: | French

What is the significance of a positive Propionibacterium acnes culture around a joint replacement?


The purpose of this study was to show the significance of a positive Propionibacterium acnes sample around a joint replacement. Records from the microbiology laboratory data over a 3-year period were reviewed to identify patients with prosthetic joints from whom Propionibacterium acnes was isolated at least once. The medical records of all those patients were retrieved and the demographic, clinical, microbiological and haematological data were collected and examined. The preoperative values of erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) were recorded. Fifty patients underwent a routine revision of a joint arthroplasty; six patients had a joint aspiration. Only one patient had further revision surgery for infection. The preoperative values of ESR and CRP were very variable. The presence of a positive sample around a joint arthroplasty is of uncertain significance. Further studies are needed in order to establish uniform criteria for the diagnosis of infection caused by Propionibacterium acnes.


Le propos de cette étude était de montrer l’importance de la présence autour d’une prothèse articulaire de propionibactério acnes. L’étude des données d’un laboratoire de microbiologie sur une période de trois ans a permis d’identifier les patients porteurs de prothèses articulaires et pour lesquels la propionibactério acnes a été isolée au moins une fois. Les données médicales de ces patients mais aussi les données cliniques, microbiologiques, hématologiques ont été collectées et examinées de même que les valeurs pré-opératoires de la CRP et de l’ESR. 50 patients ont été revus, six avaient nécessité une aspiration articulaire et un une révision chirurgicale pour infection. Les valeurs préopératoires de l’ESR et de la CRP sont variables. La présence autour d’une articulation de propionibactério acnes a une signification très incertaine. Il sera nécessaire de pratiquer des études ultérieures de façon à établir les critères du diagnostic d’une infection secondaire à propionibactério acnes.


Propionibacterium acnes (P. acnes) is an anaerobic diphtheroid which is well known to inhabit the sebaceous glands and is the most abundant organism isolated on routine skin cultures, far outnumbering aerobic bacteria [7]. It is traditionally considered non-pathogenic [6]. The true incidence of anaerobic skeletal infections remains in some dispute. In a combined study from the Mayo Clinic and the Hospital for Special Surgery on periprosthetic hip infections, the incidence of anaerobic bacteria was 12% whereas the most common organisms isolated were Staphylococcus epidermidis and Staphylococcus aureus with an incidence of 54% [4]. A review of 34 papers on infected hip arthroplasty from 1977 to 1999 showed that propionibacteria accounted for 1.7% of the infections [3].

Our aim was to investigate the significance of a positive P. acnes culture around a joint arthroplasty and their correlation to the inflammatory markers erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP).

Materials and methods

Records from the clinical microbiology laboratory data between May 2001 and May 2004 were reviewed to identify patients with prosthetic joints from whom P. acnes was isolated at least once. The following criteria were required for inclusion of a patient in the study: a presence of a joint arthroplasty, at least one isolation of P. acnes in a culture of a surgical specimen (joint fluid, para-articular fluid, perioperative irrigation-drainage fluid) or joint fluid obtained from a transcutaneous puncture.

We then retrieved the medical records of all those patients and the demographic, clinical, microbiological and haematological data were collected together with the therapeutic options chosen. There was no history of previous periprosthetic infection. The preoperative values of the ESR and CRP were recorded. For the purposes of this study, an ESR rate of 30 mm/h or higher and a CRP level of 10 mg/l or higher were considered to be suggestive of infection and were deemed a positive result.


Between May 2001 and May 2004, 56 patients with prosthetic joint replacements had a positive sample of P. acnes. Fifty patients underwent a routine revision of a joint arthroplasty and six patients had a joint aspiration only. There were 35 males and 21 females. The average age was 66.9 years (range: 18–91). There were 30 hip joint replacements, 21 knee joint replacements, 4 endoprosthetic replacements of the femur and 1 shoulder joint replacement (Table 1).

Table 1
Demographic factors, clinical and microbiological characteristics, and outcome of patients with positive samples for P. acnes around a joint replacement

Pain and/or loosening of the prosthesis were the main reasons for revision arthroplasty in all 50 patients. The treatment methods included surgical intervention and antibiotic treatment. Thirty-five patients did not receive any antibiotics, 12 patients received antibiotics (either through the enteral or parenteral route or a combination of both) for a total time of 6 weeks and in 9 patients the records were inadequate regarding antibiotic treatment. The number of total culture samples taken and sent off for microbiological analysis varied markedly (from one to eight samples). In 40 patients P. acnes bacteria were isolated only whereas in the rest of the 16 patients a mixture of both P. acnes and other organisms were isolated. These included 12 coagulase-negative staphylococci, 2 bacilli, 1 Staphylococcus aureus and 1 enterococcus pathogens. Only one patient underwent further revision for infection. The average follow-up period was 20.5 months (1–48).

In the group where only P. acnes were isolated there was a variable CRP and ESR result (Table 2). In 19 patients both ESR and CRP were normal, in 7 patients they were both elevated, whereas for the rest of the patients either ESR or CRP was elevated.

Table 2
Preoperative values of ESR and CRP

In the group where there was a mixture of P. acnes and other isolates there was also a variable result (Table 2). In nine patients both CRP and ESR were normal, in only one patient they were both elevated, whereas in the rest of the patients either ESR or CRP was elevated.


P. acnes is a low-virulence organism, part of normal skin and dental microbiota, often causing delayed clinical infections with a long interval between the inoculation of the bacteria and the onset of symptoms. Tunney et al. [13] in their paper have shown that P. acnes was isolated as the single infecting organism in 46% of their samples taken from hip replacements and a further 16% of the samples were infected by a combination of P. acnes and an aerobic Gram-positive organism. They then concluded that infection of retrieved hip prostheses is currently underestimated. It is likely that patients who had routine revision surgery for ‘aseptic’ loosening of prostheses probably harboured a low-grade infection. The purpose of our study was to evaluate the significance of joint arthroplasties colonised with P. acnes. Only one patient in our study underwent further revision for periprosthetic infection. P. acnes was the sole causative organism and the patient was also given a 6-week course of antibiotics following the routine revision.

It is interesting to note from our study that all the P. acnes isolates were sensitive to penicillin. None of the rest of the patients underwent revision surgery despite the positive P. acnes samples. We could also argue that in patients where a mixture of organisms was isolated, P. acnes could be regarded as a contaminant but it is very difficult to justify it.

Intraoperative cultures are considered the ‘gold standard’ for the diagnosis of infected arthroplasties. The area that appears to be most inflamed should be sampled and at least three tissue samples should be sent for culture to improve the yield and to decrease the likelihood of a false-negative culture. Atkins et al. [1] in their prospective study of infections in arthroplasty showed that at least three positive samples were necessary to predict infections, regardless of the pathogen involved. In our series only two patients had three or more positive samples. In one of the patients P. acnes was the sole organism whereas in the other patient coagulase-negative staphylococcus was also isolated. None of these patients underwent further revision surgery. Lutz et al. [8] in their study included a group of patients with joint arthroplasty infected with P. acnes. In that group the patients had signs of either prosthetic dysfunction or pseudoarthrosis without the signs of sepsis. The average number of positive samples in this group was one. Of their 17 patients, 12 were treated exclusively by surgery and without antibiotic treatment; only 1 patient failed treatment. They concluded that making a suitable decision in this group of patients is particularly difficult, as the lack of signs of sepsis does not exclude infection. In our series none of the patients had signs of sepsis preoperatively and whether they had a low-grade infection is difficult to say.

In another study by Marculescu et al. [9] two patients who underwent revision arthroplasty and had at least two positive samples for P. acnes obtained at the time of revision surgery received no antimicrobial treatment; both had a favourable outcome without a second removal of the prosthesis. In our study three patients had at least two positive intraoperative samples for P. acnes solely and they all had a favourable outcome.

The significance of positive intraoperative cultures taken during revision surgery in a case that was not thought to be infected following preoperative investigation remains controversial. Tsukuyama et al. [12] reported their experience of 31 of 275 patients who underwent revision arthroplasties in whom intraoperative cultures were positive but infection was not suspected preoperatively. All were treated with a 6-week course of intravenous antibiotics. During their study period 5 of those 31 patients had recurrent infections. It would seem that in at least some cases, positive cultures represent previously undiagnosed infection and require treatment.

On the contrary, in another study by Byrne et al. [2], the authors examined 80 patients who underwent hip and knee cemented arthroplasty in order to determine the incidence of perioperative contamination in these patients. Contaminated samples were identified in 18 patients (22.5%) with Staphylococcus epidermidis being the most common contaminating organism involved. Despite the high incidence of intraoperative contamination, this was not reflected by a similar rate of post-operative infection as none of the patients developed clinical evidence of deep prosthetic infection at follow-up.

Although it has been recommended that all cases of positive intraoperative cultures should be treated with antibiotics [5, 12], no prospective study has looked at the results of treatment with antibiotics as compared with no additional antibiotic treatment in addition to the revision surgery and perioperative antibiotics.

The most commonly used preoperative investigations for a suspected infected joint arthroplasty are laboratory screening tests, such as measurements of the ESR and the level of CRP.

An elevated CRP may be more indicative of infection than a rise in the ESR, but is still non-specific. The measurement of both CRP levels and ESR is marginally more accurate [10].

A prospective analysis of Spangehl et al. [11] has shown that when both the measurement of ESR and CRP values reveal negative findings, the probability of infection is 0.00 (0 of 95; 95% confidence interval: 0.00–0.04); when both tests are positive, the probability of infection is 0.83 (20 of 24; 95% confidence interval: 0.62–0.95). They did not define though the causative organism in their infected cases. CRP concentration seems to be normal in some infections with bacteria of low virulence such as propionibacteria, whereas high values are generally seen in infections with Staphylococcus aureus and β-streptococci [10]. In the study by Lutz et al. [8] in their group of patients who had signs of sepsis and at least one positive intraoperative sample for P. acnes, the ESR was elevated in 96% and the CRP was elevated in 93% of the patients. Unfortunately, in the rest of the patients with no signs of sepsis they did not perform a preoperative measurement of the inflammatory markers. In our study the preoperative values of ESR and CRP were very variable and difficult to interpret. For example, of seven patients colonised with P. acnes and both ESR and CRP elevated, only one patient underwent further revision surgery whereas the rest of the patients did not have further revision surgery.

There are no studies in the medical literature to show the preoperative ESR and CRP values in colonised or infected joint arthroplasties with P. acnes.

From our study we have drawn some messages. The presence of a positive P. acnes sample around a prosthetic joint is of uncertain significance. We can confidently say that a positive sample of P. acnes is not an indication of severe infection. We were unable to explain the elevated preoperative values of ESR and CRP. All P. acnes isolates were sensitive to penicillin and whether these patients should have a course of antibiotics is questionable. We do support the value of multiple samples, which could lead to less false-positive results. It is important to follow up the patients at the long term as a low-grade infection can declare itself in the future. Uniform criteria have not been established for the diagnosis of infection of prosthetic joints caused by P. acnes and further studies are needed.


1. Atkins BL, Athanasou N, Deeks JJ, OSIRIS Collaborative Study Group et al. Prospective evaluation of criteria for microbiological diagnosis of prosthetic joint infection at revision arthroplasty. J Clin Microbiol. 1998;36:2932–2939. [PMC free article] [PubMed]
2. Byrne AM, Morris S, McCarthy T, et al. Outcome following deep wound contamination in cemented arthroplasty. Int Orthop. 2007;31:27–31. doi: 10.1007/s00264-006-0121-z. [PMC free article] [PubMed] [Cross Ref]
3. Dunlop DJ, Masri BA, Garbuz DS, Greidanus NV, Duncan CP. The infected total hip arthroplasty. In: Sinha RK, editor. Hip replacement. Current trends and controversies. New York: Marcel Dekker; 2002. pp. 299–350.
4. Garvin KL, Hanssen AD. Current concepts review. Infection after total hip arthroplasty. Past, present, and future. J Bone Joint Surg Am. 1995;77:1576–1588. [PubMed]
5. Graziani AL, Hines JM, Morgan AS, MacGregor RR, Esterhai JL., Jr . Infecting organisms and antibiotics. In: Steinberg ME, Garino JP, editors. Revision total hip arthroplasty. Philadelphia: Lippincott, Williams & Wilkins; 1999. pp. 407–417.
6. Hillier SL, Moncla BJ. Peptostreptococcus, Propionibacterium, Eubacterium and other non-spore-forming anaerobic gram-positive bacteria. In: Murray R, Baron EJ, Pfaller MA, Tenover F, Yolken RH, editors. Manual of clinical microbiology. 6. Washington, D.C.: ASM; 1999. pp. 587–602.
7. Launder WJ, Hungerford DS. Late infection of total hip arthroplasty with Propionibacterium acnes. Clin Orthop. 1981;157:170–177. [PubMed]
8. Lutz MF, Berthelot P, Fresard A, Cazorla C, Carricajo A, et al. Arthroplastic and osteosynthetic infections due to Propionibacterium acnes: a retrospective study of 52 cases, 1995–2002. Eur J Clin Microbiol Infect Dis. 2005;24:739–744. doi: 10.1007/s10096-005-0040-8. [PubMed] [Cross Ref]
9. Marculescu CE, Berbari EF, Hanssen AD, Steckelberg JM, Osmon DR. Prosthetic joint infection diagnosed postoperatively by intraoperative culture. Clin Orthop Relat Res. 2005;439:38–42. doi: 10.1097/01.blo.0000183091.83509.d8. [PubMed] [Cross Ref]
10. Shih LY, Wu JJ, Yang DJ. Erythrocyte sedimentation rate and C-reactive protein values in patients with total hip arthroplasty. Clin Orthop. 1987;225:238–246. [PubMed]
11. Spanghel M, Masri B, et al. Prospective analysis of preoperative and intraoperative investigations for the diagnosis of infection at the sites of two hundred and two revision total hip arthroplasties. J Bone Joint Surg Am. 1999;81(5):672–683. [PubMed]
12. Tsukayama DT, Strada R, Gustilo RB. Infection after total hip arthroplasty. A study of the treatment of one hundred and six infections. J Bone Joint Surg Am. 1996;78:512–523. [PubMed]
13. Tunney MM, Patrick S, Gorman SP, Nixon JR, et al. Improved detection of infection in hip replacements. A currently underestimated problem. J Bone Joint Surg Br. 1998;80:568–572. doi: 10.1302/0301-620X.80B4.8473. [PubMed] [Cross Ref]

Articles from International Orthopaedics are provided here courtesy of Springer-Verlag