Treatment of prosthetic joint infections due to Propionibacterium

Figures & data

Table 1. Patient characteristics of 60 patients with Propionibacterium-associated PJI, presented according to their postoperative antimicrobial treatment

Characteristic	Rifampicin (n = 39)	No rifampicin (n = 21)	Total group (n = 60)	p-value
Location of joint, n				1.0
Knee	15	9	24
Hip	12	6	18
Shoulder	12	6	18
Type of arthroplasty, n
Primary/revision	31/8	15/6	46/14
Age at PJI diagnosis^a	69 (40–78)	69 (47–80)	69 (40–80)	0.5
Sex, female/male	22/17	7/14	29/31	0.09
BMI, kg/m^2^a	28 (21–50)	28 (22–35)	28 (21–50)	0.2
Medical history of PJI, n	2	2	4	0.6
Missing data	6	0	6
Clinical presentation, n
Missing data	1	0	1
Joint pain	31	18	49	1.0
Stiffness	15	10	25	0.5
Tumor	9	7	16	0.4
Instability	7	8	15	0.1
Rubor	5	1	6	0.4
Sinus tract	2	2	4	0.6
Calor	2	1	3	1.0
Fever	1	1	2	1.0
Wound leakage	1	1	2	1.0
Laboratory diagnostics
ESR > 30 mm/h	12	6	18	0.5
Missing ESR data	5	3	8
CRP > 10 mg/L	10	6	16	0.6
Missing CRP data	8	7	15
Leucocytes, × 10^9/L^a	7.9 (5.2–15)	7.5 (4.9–13)	7.7 (4.9–15)	0.6
Missing data, n	13	10	23
Microbiological diagnostics
No. of tissue cultures^a	7 (6–9)	6 (5–10)	6 (5–10)	0.3
No. of positive cultures^a	3 (2–9)	3 (2–7)	3 (2–9)	0.8
Mono/polymicrobial	28/11	19/2	47/13	0.1
PJI classification^b				0.7
Early postoperative	3	1	4
Late chronic	9	4	13
Acute hematogenous	2	0	2
Positive intraoperative cultures	25	16	41

PJI: periprosthetic joint infection; BMI: body mass index; ESR: erythrocyte sedimentation rate; CRP: C-reactive protein.

a Values are median (range)

b According to Tsukayama et al. (1996): early postoperative infection (< 1 month after index surgery), late chronic infection (> 1 month after index surgery), acute hematogenous infection (antecedent bacteremia with acute onset of symptoms in affected joint with the prosthesis), or infection diagnosed from positive intraoperative cultures (≥ 2 positive cultures of the same specimen obtained at the time of revision operation).

Table 2. Surgical treatment and postoperative antimicrobial regimen in 60 patients with Propionibacterium-associated PJI, presented according to their postoperative antimicrobial treatment

Characteristic	Rifampicin (n = 39)	No rifampicin (n = 21)	Total group (n = 60)	p-value
Surgical treatment				0.5
Debridement and prosthesis retention	5	1	6
1-stage revision (partial revision)	25 (5)	16 (5)	41 (10)
2-stage revision	9	4	13
Antibiotic treatment (daily doses)
Clindamycin 600 mg x 3 and rifampicin 450 mg x 2	33	–	33
Teicoplanin 400 mg x 1 i.v. and rifampicin 450 mg x 2	6	–	6
Clindamycin 600 mg x 3	–	16	16
Amoxicillin 500 mg x 4	–	1	1
Ciprofloxacin 750 mg x 2 and clindamycin 600 mg x 3	–	1	1
Doxycycline 200 mg x 1	–	1	1
Linezolid 600 mg x 2	–	1	1
Teicoplanin 400 mg x 1 i.v.	–	1	1

i.v.: intravenously.

Table 3. Number and types of failures in 60 patients with Propionibacterium-associated PJI, presented according to their postoperative antimicrobial treatment

Characteristic	Rifampicin (n = 39)	No rifampicin (n = 21)	Total group (n = 60)	p-value
Failures
1-year follow-up	2/39	2/21	4/60	0.7
2-year follow-up	4/23	3/13	7/36	0.6
Survival, median (range), months	19 (0.1–49)	23 (0.2–47)	21 (0.1–49)	0.9
Type of failure
Relapse^a	2	2	4	0.4
Reinfection^b	2	1	3	0.5

a Relapse: defined as positive cultures growing the same microorganism as the initial intra­operative samples.

b Reinfection: defined as a new infection with a pathogen other than that in the initial intra­operative samples.

Table 4. Overview of the patient characteristics of 7 failed cases treated for Propionibacterium-associated PJI

Case	Age	Sex	Location of joint	Preoperative diagnosis	Micro-organism(s)	PJI classification^a	Antibiotic treatment	Change in antibiotic treatment	Surgery	Duration of survival (months)	Type of failure
1	51	M	Shoulder	Suspected infection	P. acnes	Late chronic	Clindamycin	No	2-stage revision	0.2	Relapse
2	72	F	Shoulder	Dysfunction	P. acnes	PIOC	Clindamycin + rifampicin	No	1-stage revision (total revision)	19	Reinfection
3	56	M	Knee	Dysfunction	P. acnes	PIOC	Clindamycin + rifampicin	No	1-stage revision (total revision)	0.1	Reinfection
4	40	F	Hip	Aseptic loosening	P. acnes	PIOC	Clindamycin + rifampicin	No	1-stage revision (partial revision)	7	Relapse
5	78	M	Shoulder	Suspected infection	P. acnes	Late chronic	Clindamycin + rifampicin	No	2-stage revision	14	Relapse
6	50	M	Hip	Suspected infection	P. acnes + Morganelli morganii	Early postop.	Clindamycin + ciprofloxacin	Yes^b	Debridement and prosthesis retention	0.1	Reinfection
7	56	M	Shoulder	Dysfunction	P. acnes	PIOC	Clindamycin	No	1-stage revision (total revision)	23	Relapse

a According to Tsukayama et al. (1996), see Table 1. PIOC: Positive intraoperative cultures

b Ciprofloxacin dose was reduced because of adverse effects.

Figure 1. Kaplan-Meier survival curve of 60 patients treated for Propionibacterium-associated PJI. The cumulative success rate was 93% (95% CI: 83–97) and 86% (95% CI: 71–93) after 1 year and 2 years, respectively. The small vertical spikes represent the censored data.

Figure 2. Comparison of Kaplan-Meier survival curves of patients treated for Propionibacterium-associated PJI with and without rifampicin combination therapy. A cumulative success rate of 90% (95% CI: 67–98) and 82% (95% CI: 53–94) was found in patients treated without rifampicin after 1 year and 2 years, respectively. A cumulative success rate of 95% (95% CI: 81–99) after 1 year and 88% (95% CI: 69–95) after 2 year was reached in patients treated with rifampicin. Overall comparison of the cumulative success rates revealed a p-value of 0.7 (log-rank test). The small vertical spikes represent the censored data.

Tsukayama D T, Estrada R, Gustilo R B. Infection after total hip arthroplasty. A study of the treatment of one hundred and six infections. J Bone Joint Surg Am 1996; 78 (4): 512-23.

 PubMed Web of Science ®Google Scholar