Prosthetic Knee Joint Infection with Haemophilus influenzae 1 Year After Total Knee Arthroplasty: A Case Report and Literature Review

Introduction

Periprosthetic joint infection (PJI) imposes a major burden on healthcare systems [1]. Total hip and total knee arthroplasty are 2 of the most common surgical procedures performed worldwide [2]. Around 1–2% of patients who undergo primary total joint arthroplasty of the knee or hip joint are affected by PJI [2]. This type of infection, which is considered a serious complication of joint arthroplasty, can lead to further revision or excision arthroplasty, prolonged hospitalization, a long course of antibiotics, and poor functional outcomes [3].

PJIs are caused by a spectrum of microorganisms and can be influenced by a many host, environmental, and technical factors throughout the continuum of care [1]. These infections are most commonly caused by staphylococci species, including Staphylococcus aureus and the coagulase-negative staphylococci, especially Staphylococcus epidermidis [4]. Many other organisms, including streptococci, enterococci, Enterobacterales, Pseudomonas aeruginosa, and anaerobic bacteria (eg, Cutibacterium acnes), can be involved [4]. PJI due to H. influenzae is rare [5]. To the best of our knowledge, only 8 cases of PJI caused by H. influenzae have been reported in the literature [3,5–11].

Here, we present a successfully treated case of a 59-year-old woman presenting with prosthetic knee joint infection caused by H. influenzae 1 year after total knee arthroplasty. This is the first case of H. influenzae causing PJI reported in a country in the Middle East.

Case Report

A 59-year-old woman presented with a 1-day history of acute right knee pain. Prior to this, she reported mild right knee pain but maintained independent ambulation. The current episode was characterized by severe pain, subjective fever, and inability to bear weight. Notably, she had undergone a right total knee arthroplasty (TKA) 1 year prior to the presentation as management of osteoarthritis. She also reported a history of recurrent upper-respiratory tract infections, with the most recent episode occurring 3 months prior. She had no remarkable past medical history and had received all the required vaccines according to the national immunization program.

Physical examination of the right knee revealed significant swelling, tenderness, and warmth around the joint. Tenderness was elicited with minimal passive flexion, and full extension was limited. Erythema was notably absent.

Initial laboratory investigations demonstrated elevated inflammatory markers: erythrocyte sedimentation rate (ESR) 60 mm/hour (≤20 mm/hour), C-reactive protein (CRP) 129 mg/L (≤9 mg/L), and white blood cell count (WBC) of 9.89×109/L (3.4–9.6×109/L) with a high neutrophils percentage of 91.50% (40–75%).

A plain radiograph of the right knee showed the implant in situ, with no evidence of periprosthetic radiolucency or fracture. Subsequently, she underwent a triple-phase bone scintigraphy, which revealed increased uptake at the periprosthetic region in all 3 phases of scanning. Right knee computed tomography (CT) imaging demonstrated 2 distinct fluid collections surrounding the knee joint, measuring 1.2 cm in maximal thickness. The collections were surrounded by soft-tissue swelling and edema, without evidence of a gas pocket. We also found diffuse reduction in muscle bulk, with fatty infiltration of the scanned muscles, and focal sclerosis at the lateral tibial condyle (Figure 1).

Aspiration of the right knee joint was performed under aseptic conditions, yielding approximately 5 mL of minimally bloody fluid. The aspirated synovial fluid was sent for analysis. The initial plan was to admit her for intravenous amoxicillin-clavulanate (1.2 g, every 8 hours). Knee aspiration culture yielded scanty growth of Haemophilus influenzae. Based on the culture result, amoxicillin-clavulanate was given only for 2 days, then was switched to intravenous ceftriaxone (2 g, once daily). Peripheral blood culture collected on hospital day 1 was sterile after 7 days of incubation. On hospital day 3, right knee joint fluid and peripheral blood were collected and cultured. The peripheral blood culture grew Haemophilus influenzae and Staphylococcus epidermidis. Vancomycin (1 g, every 12 hours) was added for 2 days and then stopped per the infectious disease specialist and medical microbiologist’s recommendation, since Staphylococcus epidermidis is a common skin colonizer and was recovered from only 1 set of samples. The right knee joint fluid culture showed a heavy growth of mucoid H. influenzae.

The recovered H. influenzae was identified using matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) technology (Bruker Biotyper sirius®). Antibiotic susceptibility testing was performed using the disk diffusion method in accordance with the latest Clinical and Laboratory Standards Institute (CLSI) guideline (M100, 2025). The beta-lactamase test (Cefinase disc) was negative, and the isolate was susceptible to ampicillin, cefuroxime, ceftriaxone, ciprofloxacin, and trimethoprim-sulfamethoxazole. The isolate was sent to the public health laboratory for typing and was identified as a non-type B.

She underwent a first-stage revision of the right TKA. Intraoperative findings included loosening of the femoral and tibial components with frank pus seen in the knee joint surrounding the implant, mostly in the anteromedial and posteromedial region of the knee deep to the pes anserinus (Figure 2). Extensive debridement was done to the soft tissue surrounding the implant, most notably the medial compartment, as the biggest pus pockets were noticed. Irrigation of the knee joint was also performed, and an antibiotic-impregnated cement spacer was implanted.

Multiple intraoperative specimens were collected and submitted for analysis. The synovial fluid and 3 tissue biopsies from the suprapatellar region, femoral notch region, and tibial region were sterile by culture. The synovial fluid white blood cell count was 9453 cells/μL, mostly neutrophils. Only the synovial fluid that was inoculated in the anerobic blood culture bottle tested positive after 4 days of incubation. Nonfermentative gram-negative bacteria were isolated from this vial, which failed to be identified by MALDI-TOF and were linked mostly to environmental origin. H. influenzae was detected from this positive anerobic vial using the multiplex polymerase chain reaction (PCR) system (BioFire FilmArray blood culture identification 2 panel, bioMerieux). However, H. influenzae was not isolated by culture from aerobic and anerobic blood culture vials. Histopathological examination of the submitted tissue biopsy revealed synovium exhibiting congestion, synovial hyperplasia, giant cells, and a dense infiltrate of acute inflammatory cells. Neutrophils were abundant, exceeding 60 per high-power field (HPF) (Figures 3, 4). No microorganisms, necrosis, or granulomas were observed. The final diagnosis of synovial tissue with acute inflammation was made.

The patient was allowed to fully weight bear as tolerated the next morning after the operation with the assistance of the physiotherapy team. Subsequently, blood cultures collected from various sites were sterile. Ceftriaxone was given for a total of 30 days, and then the patient was discharged in good condition after 35 days of hospital stay on cefuroxime orally (500 mg, every 12 hours) for 14 days.

The patient was kept in an antibiotic-free window for 6 weeks. A second-stage revision debridement was done. A knee aspiration and 3 tissue cultures were collected intraoperatively, and all were sterile. There was no evidence of clinical, radiological, and microbiological treatment failure. She will be followed up in the clinic to ensure there is no relapse.

Discussion

PJI can present in a wide range of clinical features, from clear signs of infection to more indolent symptoms, such as joint dysfunction or pain [12]. The most sensitive clinical findings in PJI are pain and reduced range of movement [13]. Moreover, the only fully specific clinical finding of PJI is a sinus tract communicating with the joint or exposed prosthesis [12]. Intraoperative finding of purulent fluid around the prosthesis is a subjective assessment by the surgeon because of difficulty in differentiation between pus and other turbid fluids present in other conditions, such as adverse local tissue reaction and crystal arthropathy [14]. Therefore, the presence of pus can suggest infection; however, it is not a confirmatory sign of PJI [12].

The diagnosis of PJI can be difficult and utilizes many different approaches. The currently available diagnostic methods have problems with accuracy and interpretation of results, and there is no reference standard definition of PJI [12,15]. A project developed by the European Bone and Joint Infection Society (EBJIS) and supported by the Musculoskeletal Infection Society (MSIS) and the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) Study Group for Implant-Associated Infections (ESGIAI) evolved a 3-level approach for the diagnosis of PJI [12]. This approach provides practical guidance and allows consistency in definition among orthopedic surgeons. In our presented case, the histopathological finding showed more than 5 neutrophils in more than 5 high-power fields, which confirmed the diagnosis of PJI using the EBJIS approach. The presence of acute inflammatory cells in tissue was found to be highly specific for the diagnosis of PJI [16–18]. Experts recommend collecting at least 3 deep samples, especially from the bone-implant interface membrane, synovium/pseudocapsule, or abnormal tissue for better evaluation of the inflammatory cells [19]. The American Academy of Orthopaedic Surgeons (AAOS) and Infectious Diseases Society of America (IDSA) strongly support the use of histopathology and preoperative serum (ESR, CRP, interleukin-6) in the diagnosis of PJI [1].

The microbiological diagnosis of PJI is ideally made pre-operatively, but if this is not possible, this should be pursued during revision or resection arthroplasty [4]. For ideal recovery of the pathogen, the IDSA and the American Society for Microbiology (ASM) recommend 3–4 separate tissue samples be submitted for aerobic and anaerobic culture, sonication of explanted prostheses, semi-quantitative aerobic and anaerobic culture of the resultant sonicate fluid, and inoculation of synovial fluid and tissues into blood culture bottles [4]. They also consider 2 or more intraoperative cultures or a combination of preoperative aspiration and intraoperative cultures yielding the same organism to be definitive evidence of PJI [4]. Applying these criteria to our case, H. influenzae was found to be the definitive causative agent for the right prosthetic knee joint infection. The main microbiological challenge in the diagnosis of PJI is culture-negative cases [12]. Molecular techniques, which have gained popularity in orthopedic surgery, such as the 16S ribosomal RNA gene sequencing or the multiplex PCR panel, may be considered if cultures are negative [4,12]. However, the lack of clear interpretative criteria for molecular methods cannot completely replace conventional culture workup [12,20]. Moreover, a positive result by a molecular assay does not indicate the presence of viable organisms [20], as occurred in the present case. Therefore, caution is needed when amplification-based molecular assays are used for diagnosis.

H. influenzae is a gram-negative coccobacillus that is part of the normal flora of the mucosal surfaces of the upper-respiratory tract of many healthy individuals [21]. The presence of a concomitant or preceding viral infection can predispose previously healthy carriers to infection by invasion of the bacteria, inflamed or damaged respiratory mucosa, and entry to the bloodstream [21]. PJI is usually acquired from contamination at the time of arthroplasty implantation but can occur due to extension from adjacent sites or subsequent hematogenous seeding [4]. In our patient, it is likely that the PJI developed because of hematological spread from the respiratory tract due to a previous history of recurrent upper-respiratory tract infection. In 1980, Borenstein and Simon described the first case of PJI caused by H. influenzae [7]. The literature on patients with PJI caused by H. influenzae is summarized in Table 1.

The polysaccharide capsule is the main virulence factor of H. influenzae [21]. Strains of H. influenzae can produce 1 of 6 distinct capsular polysaccharides (a to f) or may be nonencapsulated (non-typeable H. influenzae) [21]. Nonencapsulated strains produce smaller, buff-colored colonies, while encapsulated strains often have a mucoid appearance [21]. Invasive infections caused by H. influenzae, such as epiglottitis, orbital cellulitis, meningitis, and bacteremia, are usually caused by the capsular strains [21]. Since the introduction of the H. influenzae type b conjugate antigen vaccine in the 1980s, the distribution of serotypes causing invasive infection has shifted [21]. Khan et al highlighted the importance of vaccination against H. influenzae in patients with prosthetic joints [3]. The role of preventive vaccination in our patient is unclear; she received the H. influenzae type b vaccine as part of the national immunization program. The World Health Organization reported variation in the incidence of each serotype [22]. Recent reports suggest most invasive disease is attributed to non-typeable H. influenzae and H. influenzae serotype f [21]. The strain in the present case study was mucoid in appearance; however, only serotype b was tested. For this reason, it might be instructive to know the capsular serotypes of H. influenzae strains recovered from patients with invasive disease.

The management of PJI necessitates the need for surgical intervention and prolonged courses of antimicrobial therapy [23,24]. Because there have been only a few cases reported of H. influenzae causing PJI, a unique guideline for antimicrobial therapy is difficult to formulate. Two cases of PJI caused by H. influenzae with beta-lactamase-positive and amoxicillin–clavulanic acid-resistant (BLPACR) previously reported [9,10]. In the absence of reliable evidence, antibiotics should be selected based on susceptibility testing, beta-lactamase test, individual risk factors of the patient, and multidisciplinary support of institutional experts. The case described in this report was beta-lactamase-negative; therefore, cephalosporins were used successfully to eradicate the infection after a prolonged course of antibiotic therapy.

Conclusions

The diagnosis and management of periprosthetic joint infection (PJI) are difficult, especially if the cause is a rare pathogen such as Haemophilus influenzae. The case described in this report with the literature review shows the pathogenic potential of H. influenzae as a rare cause of PJI. An essential component of the care of patients with a rare PJI is a strong collaboration between all involved medical specialists in a multidisciplinary approach.