Active septic arthritis of the hip in adults: what’s new in the treatment? A systematic review

Search strategy and criteria

For this systematic review we searched the electronic databases PubMed, Scopus, and Embase using the search terms: “hip” or “native hip” and “septic arthritis” or “coxitis” (Table 1). The search was limited to articles published in the past 20 years because few case reports were published before 2000 and most case series or cohort studies have appeared since then. Articles were reviewed according to the Preferred Reported Items for Systematic Reviews and Meta-Analyses Statement for Individual Patient Data (PRISMA-IPD). ⁹

The research questions, inclusion and exclusion criteria were decided a priori. The inclusion criteria included human studies published in English and studies on the treatment of bacterial septic arthritis of the native hip. Exclusion criteria were: septic arthritis in children; infection after hip arthroplasty; avascular necrosis of the femoral head; tubercular or fungal infection; rheumatic disease; septic arthritis involving other joints (knee, shoulder, sternoclavicular, wrist, pubic symphysis); tumours; hip osteoarthritis or impingement; hip dysplasia; not a systematic review; studies on animals/cadavers; epidemiologic studies; aseptic disease.

Study screening

After excluding the duplicates, two reviewers (FD, LM) screened the title, abstract, and full text of each study. Disagreements were solved by discussion after full-text evaluation. The reference lists of the studies were manually searched for other publications that may have eluded the initial search.

Search results

A total of 3926 articles were retrieved for preliminary evaluation (Fig. 1). After the duplicates were eliminated, 3081 original articles remained. Two authors (FD, LM) reviewed the titles: 2939 studies were excluded because they did not meet the inclusion criteria and 19 because they were published in a language other than English. After the abstracts were reviewed, 102 studies were excluded because they did not meet the inclusion criteria. Of the remaining 21 articles, ten were excluded after evaluation of the full text. Eight articles were added to 11 eligible studies from their references, for an end total of, 19 studies in the final analysis and quality assessment. All 19 studies reported data on treatment of bacterial septic arthritis in the adult native hip (first research question); on success rates and outcomes after surgery (second research question); 17 reported data on antibiotic regimen and its duration (third research question).

View Full Size

Study flowchart.

Note. THA, total hip arthroplasty; AVNF, AVascular Necrosis of Femoral head

Citation: EFORT Open Reviews 6, 3; 10.1302/2058-5241.6.200082

• Download Figure
• Download figure as PowerPoint slide

Assessment of study quality

Two reviewers (FD, LM) independently evaluated each study for quality. For this systematic review, the Methodological Index for Non-randomized Studies (MINORS) questionnaire was used. ¹⁰ The final analysis included retrospective case-control studies (n = 3), retrospective case series (n = 14), a systematic review and a prospective study (n = 1 each) (Table 2). The mean MINORS score for retrospective case series was 11/16 (range, 7–13); the MINORS score for the case-control studies was 17/24 (range, 15–21). The inclusion of only one prospective study was the main study quality deficiency.

Data collection

Data from each study included number of patients, sex, age, follow-up, pathogen, treatment, success rate, outcomes, and complications (Table 3). The studies were grouped and compared by treatment: arthroscopy, two-stage open treatment, one-stage open treatment (Tables 4, 5 and 6). Complications associated with treatment were assessed; the range is reported.

Treatment options

Treatment options for bacterial arthritis of the native hip can be grouped into three categories:

1. Arthroscopic treatment (Table 4): hip arthroscopy is performed in standard supine position with a three-portal (anterior, anterolateral, posterolateral) procedure. ^{13,15,18,19,25} Yamamoto et al describe the use of three different arthroscopes, each from a single portal. ¹⁹ After gentle distraction, a 70° arthroscope and then a 30° arthroscope is inserted. ^13,15,18 Lee et al suggest mixing 1 mL 0.1% epinephrine solution to the saline fluid to obtain a more precise view. ¹⁵ There is no a defined regimen for irrigation volume: Nusem et al ¹⁸ use 6–8 L of saline solution, while other authors reported using more than 20 L. ^13,19,25 One or two suction drainages are usually positioned and then removed 2–3 days later. ^15,18,25 Kim et al describe drain removal after 5–14 days. ¹³ Tolerable weight-bearing with crutches and unrestricted range of motion exercise are allowed immediately. ^13,15,18,25 Yamamoto et al permit weight-bearing after three weeks. ¹⁹

2. Two-stage treatment (Table 5): the first stage entails debridement of the infected tissues, resection arthroplasty of the proximal femur followed by implantation of an antibiotic-loaded cement spacer or cement beads. The second stage involves implantation of a total hip replacement. ^{4,6,12,16,21–24,27} Anagnostakos and Kelm suggest high-pressure pulsatile lavage with 20 L of antiseptic solution (chlorhexidine) and Ringer solution. ^6,26 The femoral neck should be cut as proximal as possible to preserve femoral bone. ^23,26

   The use of antibiotic-loaded hip spacers obtains a high local concentration of antibiotic, while maintaining leg length and function in the interim period, thus making subsequent hip replacement easier and more functional. ^{12,16,21–23} Antibiotics in spacers or beads should be based on culture result; the most commonly used are gentamicin and vancomycin. ^{4,6,16,21–24,26,27} Use of a spacer cup is reported in the case of affected acetabulum. ⁶ Cement spacers should be positioned without further opening the femoral canal, thus preventing the spread of infection, and fixed with antibiotic-loaded cement, according to ‘glove’ or ‘press-fit’ methods. ^4,22,26 Cho et al describe an innovative technique in which four cancellous screws or Steinmann pins are inserted into the remaining proximal femoral neck parallel to the long axis of the femur. Antibiotic-loaded cement is then fixed to the screws and moulded to create the shape of the femoral head. ²³

   After first-stage surgery, patients can walk with crutches and toe-touch weight. ^6,26 The range of motion of the hip is permitted up to 60° of flexion. ^6,26 The second stage is performed only when satisfactory infection control has been achieved: wound healing and normal erythrocyte sedimentation rate/C-reactive protein level measured at three consecutive follow-up visits indicate eradication of infection. ^{4,6,12,16,21,24} Additional aspiration for cell count and culture can be performed to confirm the absence of septic arthritis. ^{4,16,21,24,27} Anagnostakos et al compare negative findings of intraoperative frozen tissue sections before THA can be performed. ⁶

3. Open surgical procedures (Table 6): surgical debridement aims to improve patient survival and prevention of radical excisional tissue removal; the surgical approaches varied on surgeon’s preferences by paying particular attention to the vascular supply of the femoral head. Routine open debridement includes removal of all accessible synovial tissue, capsulectomy, and drain placement. New approaches are described by Ohtsuru et al and Kelm et al. ^14,17 Ohtsuru et al describe placement of a pedicled vastus lateralis muscle flap after infection eradication. ¹⁴ An anterolateral thigh skin paddle flap is created and a pedicled vastus lateralis muscle flap is detached. The flap is rotated 180° and placed into the incision of the approach below the subcutaneous fat. The muscle flap is placed to fill the dead space in the hip and a suction tube is inserted. Hip motion is permitted two weeks after surgery and full weight-bearing after three months.

   Kelm et al report the use of vacuum-assisted closure (VAC) therapy with polyvinyl alcohol sponges. ¹⁷ Polyvinyl is preferred because it provokes less pain and can be left in situ for longer than polyurethane. Postoperatively, a continuous subatmospheric pressure of 200 mm Hg is applied via a vacuum therapy system. Sponges are removed after infection eradication, defined as clinical healing, normalization of laboratory parameters, and inspection of the drained fluid.

Outcomes and success rate

Successful treatment of bacterial arthritis is usually defined as eradication of the infection without reinfections or infective complications, and with the restoration of hip joint function. Outcomes and success rates depend on treatment choice:

1. Arthroscopic treatment (Table 4): all the studies report a 100% success rate. ^{13,15,18,19,25} Only Lee et al report a success rate of 88.8% after the first arthroscopy and complete success after second-look arthroscopy. ¹⁵ Most studies report no arthroscopy-related complications. ^{13,15,18,20,25} Kim describes a case of hip osteoarthritis and heterotopic ossification in one patient. ¹³ Yamamoto et al describe damage and detachment of cartilaginous fragments in all the cases treated. ¹⁹ Finally, Kim et al mention a potential risk of abdominal compartment syndrome due to arthroscopy fluid extravasation. ¹³

2. Two-stage treatment (Table 5): the two-stage technique is associated with positive outcomes and very low failure rates, reinfection or persistent infection. Most studies define therapeutic success as no infection recurrence and infection eradication. ^{4,6,12,16,21–24,27} The success rate is reported from 73.7% to 92.8% after spacer implantation and from 80.0% to 100.0% after THA replacement. ^{4,6,12,21–24,26,27} Harris Hip Score improves from 11.5–58.8 before to 61.6–71.0 after surgery with spacers, and to 93.3–93.6 after THA. ^16,22,23

   The most common local complications are recurrent infection (range, 9.8–10.7%), drainage sinus following THA implantation (44% in one study), periprosthetic fracture (10.7% in one study), aseptic loosening (7.1% in one study), broken stem (3.5% in one study), heterotopic ossification (12–17%). ^{4,6,12,16,21,27} Spacer-specific complications are rare: Anagnostakos et al report moulded spacer fractures in three patients and moulded spacer dislocation in two. Papanna et al report spacer dislocation in one case. When recurrent dislocations occur, the authors suggest a Girdlestone resection arthroplasty. ^6,27

   Reported general complications are cerebellar haemorrhage (3.5%). ²¹ Kao et al report a 13.5% mortality rate in patients with septic hip arthritis. ¹² The rate rises to 18% after the first stage. The authors report that patients who undergo repeat surgical debridement have favourable clinical results.

3. Open surgical procedures (Table 6): the two studies report a 100% success rate. ^14,17 No complications are observed during VAC removal, while one case of intramuscular abscess of a muscle flap is described. ^14,17 Ohtsuru et al report no need for additional surgery and no recurrence in patients who received musculocutaneous flap transposition; the hip range of motion is 0–90° and the Harris Hip Score 77. ¹⁴ VAC therapy is reported as successful for the eradication of septic hip arthritis, without recurrence. ¹⁷

Antibiotic therapy

Antibiotic treatment is mandatory to eradicate hip infection in association with surgical treatment. All the studies describe antibiotic therapy initiated according to culture results and bacterial sensitivity: pending the culture results, empirical antibiotic therapy is recommended. ^{5,13,15–19,23–27} If cultures are negative, vancomycin, teicoplanin, levofloxacin, clindamycin, gentamicin or first-generation cephalosporine can be administered intravenously. ^{4,17,23,24,26} In most studies the duration of antibiotic therapy is from three to six weeks and is generally successful in infection eradication. ^{13,15–18,21–25,26,27} Intravenous antibiotics are immediately initiated after surgery and oral antibiotics are introduced after the first 3–4 weeks, then continued for further 2–3 weeks. ^{16,18,23,25–27} Discontinuation of intravenous administration depends on clinical or laboratory evidence of resolution of ongoing infection. ^16,22,26 Oral antibiotics are given until C-reactive protein levels return to normal. ^24,26 Nusem et al report the use of an elastomeric continuous infusion device for intravenous antibiotic therapy after discharge from the hospital. ¹⁸ Yamamoto describes a restricted antibiotic protocol, with daily oral administration of cefazolin or cefdinir for one week: no reinfection is reported. ¹⁹