Peptoniphilus olsenii in a polymicrobial surgical site infection following open reduction and internal fixation of a calcaneal fracture: Case report

Abstract

Surgical treatment of calcaneal fractures is associated with a high incidence of postoperative wound infections, ranging from 7% to 25%. Their treatment is a real challenge in orthopedic surgery. Herein, we present a rare case of a polymicrobial surgical site infection involving the anaerobic species Peptoniphilus olsenii in a 59-year-old male. The infection was treated in multiple stages by aggressive surgical debridement together with filling the formed calcaneal cavity with bioactive glass, silver-impregnated dressing, vacuum-assisted closure, skin grafting, and targeted antibiotic therapy. Finally, we briefly discuss our treatment algorithm according to literature data.

Introduction

Displaced intraarticular calcaneal fractures are often treated surgically with the aim of restoring normal anatomy and function [1]. Since deep infections of the calcaneus present a treatment challenge, considerable attention is needed to reduce the incidence of deep infection after calcaneal plate fixation [2]. Postoperative wound infection rates vary from 7% to 25% [3]. Poeze et al. (2008) reported that the median deep infection rate was 5.1% [4]. However, only scarce data exists about the causative pathogens of surgical site infections [3], [5], [6].

Herein, we present a rare case of polymicrobial postoperative calcaneal infection with isolation of a relatively recently recognized bacterial species – Peptoniphilus olsenii. We also briefly discuss our treatment algorithm according to literature data.

Case

Presentation

A 59-year-old Caucasian male construction worker presented to our institution with a painful dehisced surgical wound in his left foot. Three months earlier in another hospital, he had had open reduction and internal fixation of a calcaneal fracture sustained after a fall from a ladder. The wound had never healed and gradually became more and more painful with increasing purulent discharge and putrid smell. The patient was a heavy smoker and he had gout with no recent exacerbations.

Clinically, the patient presented in a wheelchair. His left heel was very painful, with an L-shaped surgical incision on its lateral aspect. The longitudinal limb of the incision had healed but the horizontal limb was dehisced, with necrotic areas, purulent discharge, and visible orthopedic hardware (Fig. 1a). Preoperative imaging evaluation included radiographs and computed tomography (Fig. 1a-f). He had no leukocytosis, his C-reactive protein (CRP) was 9.3 mg/L at presentation and it fluctuated towards normal during the treatment period. Antibacterial treatment was initiated with meropenem 1 g every 8 h and vancomycin 1 g every 12 h intravenously. As a first-stage procedure, surgical debridement was performed (Fig. 1 g). The bony defect was filled with 45S5 bioactive glass granules and covered with a silver-impregnated sheet dressing in order to prevent premature closure of the granulating soft tissues. The wound was covered with a vacuum-assisted closure (VAC) system (Fig. 1 h). Intraoperative samples from the curetted necrotic bone tissue were processed essentially according to Rieber et al [7], with 10 days of enrichment in a modified semifluid thioglycollate broth (LT) containing beef liver extract and a mixture of horse serum and hemin (SIFIN, Berlin, Germany) [8]. Subcultures were grown on Chocolate and Columbia agar (Diachim, Bulgaria) at 36°C in 5.5% CO₂. Anaerobic subcultures were performed on Fastidious Anaerobe Agar (Liofilchem, Italy) and Columbia CNA Agar with Crystal Violet (CNA-CV, Diachim, Bulgaria) incubated at 36°C for 48 h. Microbiological sampling revealed a polymicrobial infection comprising Proteus mirabilis, Escherichia coli, Enterococcus faecalis, and Peptoniphilus olsenii, which were identified by MALDI Biotyper (Bruker, Germany) in accordance with the manufacturer's instructions. Interestingly, we were only able to isolate Peptoniphilus olsenii on CNA-CV, a medium that effectively inhibits the growth of Enterobacterales and Staphylococci, which would otherwise mask the slow and weak growth of anaerobes. Antimicrobial susceptibility testing was performed using either the MIC determination method (Micronaut-S, Bruker, Germany) or the disk diffusion method (Oxoid, UK). The results were interpreted according to The European Committee on Antimicrobial Susceptibility Testing - EUCAST v15 with the exception of P. olsenii whose susceptibility was interpreted according to The Clinical and Laboratory Standards Institute (CLSI) M100 35ed for anaerobes (Table 1). Interestingly, we evidenced two unusual microbiological findings: E. coli resistant to colistin and E. faecalis resistant to ampicillin. This might be the result from previous antibiotic treatment, however there was no pertinent medical documentation available. Overall P. olsenii was susceptible to all tested antimicrobials except for clindamycin consistent with previous reports [8]. Antibiotic therapy was continued as initiated. At week one, a second debridement was performed. The wound was found to be almost completely covered by fresh granulation tissue with no visible signs of active infection (Fig. 1i). The silver-impregnated dressing was removed and the granulation tissue was closed over the bony cavity with a single suture. A VAC system was reapplied until week two when the wound bed was found to be completely covered by granulation tissue so the skin defect was covered with a free full-thickness skin graft. After hospital discharge at week four and 28 consecutive days on meropenem and vancomycin, deescalating antibacterial therapy was continued with amoxicillin/clavulanic acid 1 g every 12 h orally and ciprofloxacin 500 mg every 12 h orally until the follow-up visit at week seven. After removal of the compression dressing, the skin graft was found to be viable in the central portion of the wound right over the former bone defect but necrotic around the periphery without signs of active infection. Wound cultures were taken at the bedside and enzymatic debridement was initiated. The same antibiotic treatment was continued during the eighth week until extended-sptectrum beta-lactamase (ESBL) producing Klebsiella pneumoniae and Pseudomonas aeruginosa were identified (Table 1). Treatment was reverted to meropenem 1 g every eight hours intravenously until week 10 when two consecutive sterile wound cultures were obtained. By week 12, i.e., after four weeks of enzymatic debridement and antibacterial therapy, the wound was almost completely epithelialized. The patient was free of symptoms and was instructed about dressing changes. After 10 months, there were neither clinical nor radiographical signs of infection (Fig. 1j-l).

Fig. 1.

a: At presentation, the horizontal limb of the surgical approach was dehisced, the wound bed was covered in necrotic tissue that reached straight to the hardware, and the wound edges were swollen and partially macerated. b, c: Lateral and axial radiographs of the calcaneus at presentation show an ill-defined osteolytic area in the posterior part of the calcaneus; Sudeck’s bone dystrophy was visible, especially in the head of the fifth metatarsal; faulty fracture fixation, which is beyond the scope of this paper, was also evident. d, e, f: Sagittal, coronal, and transverse CT slices of the calcaneus demonstrating the extent of bone destruction. Sudeck’s type of patchy osteoporosis was also more evident that on the radiographs. g: After initial debridement, which required additional partial incision of the longitudinal arm of the original scar, and filling the bony cavity with bioactive glass granules. h: After initial application of the VAC system. i: After one week of VAC treatment, the wound bed was almost completely covered in granulation tissue and despite some skin maceration the wound edges were not undermined; there was no additional skin loss; the granulation tissues overlying the area of bone trepanation were sutured at this stage. j, k: Lateral and axial radiographs with the glass granules in place. l: Postoperative photograph 10 months after surgery.

Table 1.

Antimicrobial Susceptibility Results.

	E. coli*	E. faecalis*	P. mirabilis*	P. olsenii*	K. pneumoniae**	P. aeruginosa**
	MIC [mg/l]	DD [mm]	MIC [mg/l]	MIC [mg/l]	DD [mm]	DD [mm]
Benzylpenicillin	-	-	-	< 0.063(S)	-	-
Ampicillin	6 mm (R)	7 mm (R)	13 mm (R)	0.125 (S)	-	-
Piperacillin	> 32 (R)	ND	≤ 4 (S)	-	ND	ND
Piperacillin/ Tazobactam	≤ 1 (S)	ND	≤ 1 (S)	< 1 (S)	18 mm (R)	22 mm (I)
Amoxicillin/ Clavulanic acid	6 mm (R)	6 mm (R)	24 mm (S)	≤ 0.5 (S)	10 mm (R)	-
Ampicillin/ Sulbactam	17 mm (S)	-	21 mm (S)	-	9 mm (R)	-
Cefuroxime	22 mm (I)	-	17 mm (R)	-	6 mm (R)	-
Ceftazidime	≤ 0.25 (S)	-	≤ 0.25 (S)	-	11 mm (R)	10 mm (R)
Ceftazidime/ Avibactam	≤ 1(S)	-	≤ 1 (S)	-	22 mm (S)	30 mm (S)
Cefepime	≤ 1(S)	-	≤ 1 (S)	-	16 mm (R)	26 mm (I)
Ceftolozane/ Tazobactam	≤ 1(S)	-	≤ 1 (S)	-	ND	ND
Amikacin	≤ 4 (S)	-	≤ 4 (S)	-	18 mm (S)	21 mm (S)
Gentamicin	2 (S)	15 WT	2 (S)	-	7 mm (R)	-
Tobramycin	1 (S)	-	2 (S)	-	9 mm (R)	ND
Ertapenem	-	-	-	0.125 (S)	ND	-
Imipenem	≤ 1 (S)	ND	2 (I)	0.5 (S)	25 mm (S)	28 mm (I)
Meropenem	≤ 0.125 (S)	-	≤ 0.125 (S)	0.5 (S)	28 mm (S)	30 mm (S)
Ciprofloxacin	≤ 0.06 (S)	20 mm (S)	≤ 0.06 (S)	-	6 mm (R)	29 mm (S)
Levofloxacin	≤ 0.125 (S)	22 mm (S)	≤ 0.125 (S)	-	8 mm (R)	30 mm (I)
Moxifloxacin	25 mm (S)	-	-	0.5 (S)	6 mm (R)	-
Teicoplanin	-	19 mm (S)	-	-	-	-
Vancomycin	-	16 mm (S)	-	≤ 2 WT	-	-
Clindamycin	-	-	-	> 8 (R)	-	-
Linezolid	-	22 mm (S)	-	-	-	-
Doxycycline	-	-	-	0.5 (S)	-	-
Tigecycline	21 mm (S)	25 mm (S)	-	< 1 NA	0.5 (S)	-
Colistin	> 8 (R)	-	-	-	0.5 (S)	1 (S)
Aztreonam	≤ 1 (S)	-	≤ 1 (S)	-	8 mm (R)	26 mm (I)
Fosfomycin	> 128 (R)	-	> 128 (R)	-	ND	-
Metronidazole	-	-	-	0.25 (S)	-	-
Trimethoprim/ Sulfamethoxazole	> 8 (R)	ND	≤ 1 (S)	-	6 mm (R)	-

“*”= isolate recovered intraoperativelly at initial visit

“**”= isolate recovered from wound at the follow-up

“S” = susceptible; “I” = susceptible, increased exposure; “R” = resistant

mm = inhibition zone values in millimeters obtained from the disk diffusion method

“-“ = intrinsic resistance or inappropriate for therapy

“ND” = not tested

“NA” = not applicable

“WT” = wild type

During the whole treatment, weight-bearing was allowed as tolerated and foot elevation was encouraged. The importance of cessation of smoking had to be repeatedly emphasized with little effect.

Discussion

Postoperative infection in traumatology is a significant factor for morbidity; only limited data regarding bacterial speciation and antibiotic resistance exist [9]. The postoperative wound infection of the calcaneus is established in up to 25% of cases [2], [3]. Folk et al [10]. analyzed early wound complications in 190 surgically treated calcaneal fractures. They reported a wound complication rate of 25%, with 21% of cases needing additional surgical interventions. Smoking, diabetes mellitus, and open fractures were identified as significant predictors of wound complications.

Backes et al [3]. in their study established that the postoperative infections are most commonly caused by Enterobacteriaceae, S. aureus, and Streptococci. Thus, the recommended antibiotic prophylaxis and treatment after calcaneal fractures should take these pathogens in mind. If postoperative wound infection is suspected, empirical antibiotic therapy should be started; if needed, corrections of the therapy can be made after bacterial identification. Obligate fastidious anaerobic bacteria represent a significant, yet frequently underestimated, burden in post-operative wound infections, complicating clinical management due to inherent challenges in their detection and identification, especially in polymicrobial cultures [11].

Peptoniphilus species belong to the group of Gram-positive anaerobic cocci (GPAC) and were reclassified in 2001 by Ezaki et al [12]. Although GPAC account for approximately 25–30% of anaerobic infections, progress in understanding their clinical relevance has been limited [8]. Advances in 16S ribosomal RNA sequencing have significantly improved GPAC taxonomy, leading to the first identification of Peptoniphilus olsenii in clinical samples in 2007 [13], [14].

Peptoniphilus species are most commonly associated with chronic infections of the skin and soft tissues but have also been found in bloodstream infections [15], [16]. Peptoniphilus olsenii was first identified in patients with dry gangrene related to peripheral vascular disease, diabetic foot ulcers, and toe infections [14]. Yamamoto et al [8]. also detected Peptoniphilus olsenii and Gleimia europaea in a case of subcutaneous abscess and septic shock in a patient with COVID-19. Peptoniphilus species are notoriously difficult to characterize using conventional microbiological methods owing to their slow growth (48–72 h) and fastidious anaerobic nature. Even following successful isolation, accurate identification frequently requires advanced techniques like MALDI-TOF MS and DNA sequencing [16]. Ultimately, these phenotypic limitations drive diagnostic delays, hinder the delivery of targeted antimicrobial therapy, and contribute to the systemic underreporting of these pathogens. In this study, we show that Peptoniphilus olsenii could be isolated after appropriate enrichment and subculturing on CNA-CV selective media even in the presence of fast growing Enterobacterales and Enterococcus spp.

As part of the initial evaluation for suspected osteomyelitis, laboratory investigations such as erythrocyte sedimentation rate (ESR) and CRP need to be evaluated [10]. CRP is accepted as a reliable indicator of osteomyelitis in the corresponding clinical settings. Its serum levels rise after 6 h and peak at around 48 h; CRP returns to normal within about one week after appropriate treatment [10]. Although ESR is a sensitive marker, its specificity is limited. Its levels increase in the setting of fractures and other inflammatory conditions and reach peak values at 3–5 days after infection and need up to three weeks to return to normal after adequate treatment [10]. In our case the diagnosis was clear based on the clinical findings, so CRP was assessed at baseline for monitoring inflammatory activity during treatment rather than as a diagnostic tool, and ESR was not assessed because its slow reactivity would not provide a useful overview of the treatment process.

When infection is established, meticulous surgical debridement is needed to remove all necrotic and infected tissues; finally, a substantial bone defect could result. Most commonly, the defect is filled with antibiotic-loaded polymethyl methacrylate (PMMA) cement, as pre-shaped beads or as various custom-made antibiotic-loaded spacers. However, this requires a second surgical procedure for spacer removal, followed by definitive fracture fixation and bone grafting [17]. Calcium sulfate packing is also a treatment option that is used in calcaneal osteomyelitis with favorable clinical outcomes [18]. In contrast to PMMA, it can deliver a broader spectrum of antibiotics and is fully biodegradable, thus no secondary surgery is needed [19]. Another option is bioactive glass, the first of which – 45S5 (45% SiO₂, 24.5% Na₂O, 24.5% CaO, and 6% P₂O₅) was synthesized in 1969, and the one most commonly used is S53P4 (53% SiO₂, 23% Na₂O, 20% CaO, and 4% P₂O₅). Bioactive glasses are a group of synthetic materials based on silicon dioxide; they have valuable mechanical properties and the possibility of bone integration [20]. This compound has osteoinductive properties; moreover, it increases local pH levels and osmotic pressure which creates an unfavorable environment for both aerobic and anaerobic bacteria [17], [20]. The 45S5, and especially the S53P4, also demonstrates antibiofilm activity against various pathogens, effectively reducing biofilms produced by S. aureus, S. epidermidis, Acinetobacter baumannii, and K. pneumoniae [20]. This advantage of bioactive glass eliminates the need for topical antibiotics [17]. The osteoinductive properties arise from the formation of a silicon-rich calcium phosphate layer on the glass surface which subsequently crystallizes into hydroxyapatite and stimulates new bone formation [17]. In our case, several surgical debridements were performed after implant removal. A significant bone defect was formed after removal of necrotic and infected tissue. We filled the defect with bioactive glass for two reasons: firstly, it has osteoinductive activity and secondly, it increases the local pH and osmotic pressure, which leads to its antibacterial effect. We adopted a more aggressive surgical approach due to the deep infection.

Conclusion

Calcaneal infections are among the most difficult to manage and eradicate. Careful clinical assessment with close postoperative surveillance is essential. Successful treatment requires accurate microbiological diagnosis, aggressive surgical debridement, adequate soft-tissue coverage, and antibiotic therapy. Herein, we present a multibacterial deep infection caused by Peptoniphilus olsenii, Proteus mirabilis, Escherichia coli, Enterococcus faecalis, Klebsiella pneumoniae, and Pseudomonas aeruginosa. To our knowledge, this is the first case of P. olsenii reported in orthopedic and traumatology literature. In summary, in advanced infections, a multidisciplinary team is essential, with active partnership from all team members.

Author statement

The authors certify that they have obtained the consent of the patient to publish this report. The patient understands that his identity will not be revealed through the figures presented.

CRediT authorship contribution statement

Rumyana Hristova: Methodology, Investigation, Formal analysis, Data curation. Ivelina Zhivkova: Project administration, Investigation, Formal analysis, Data curation. Ivan N. Ivanov: Writing – review & editing, Writing – original draft, Validation, Supervision. Georgiev Georgi: Writing – review & editing, Writing – original draft, Validation, Supervision, Investigation, Formal analysis, Data curation, Conceptualization. Iva N. Dimitrova: Writing – review & editing, Visualization, Supervision, Methodology. Svetoslav A. Slavchev: Writing – review & editing, Writing – original draft, Methodology.

Funding

There was no specific funding for this study.

Declaration of Competing Interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:Georgi P. Georgiev reports was provided by Medical University-Sofia Faculty of Medicine. Georgi P. Georgiev reports a relationship with Medical University-Sofia Faculty of Medicine that includes: employment. Georgi P. Georgiev has patent No patent pending to No. No conflict of interest If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.