Appropriate initial treatment of an infection depends on the extent of the infection, the presence of draining sinuses, the virulence of the organism, when the infection becomes apparent, whether the implants are loose, and the patient’s general medical condition. Although the initial treatment of deep infection after total hip arthroplasty is typically surgical, the decision of whether to remove or retain the components may largely be guided by the chronicity of the infection.
Tsukayama classified periprosthetic infections into four categories:
A careful history and physical examination are crucial in making the diagnosis of total hip infection. Although the diagnosis of early postoperative infection or acute hematogenous infection is often not difficult, late chronic infections can be challenging to distinguish from other causes of pain in a patient with a previous total hip arthroplasty. Early or late acute infections may be characterized by pain, fever, wound drainage, or erythema. Pain unrelieved by a seemingly well-functioning arthroplasty may be a clue to a chronic infection. A history of excessive wound drainage after the initial arthroplasty, multiple episodes of wound erythema, and prolonged antibiotic treatment by the operating surgeon also are worrisome. Physical examination focuses on the presence or absence of painful hip range of motion, swelling, erythema, sinus formation, or fluctuance.
Often radiographs of the affected hip are normal or at best may be indistinguishable from aseptic loosening of the prosthesis. Progressive radiolucencies or periosteal reaction occasionally may be seen and is indicative of possible infection. Pseudobursae seen on arthrography of the hip have been described as diagnostic of deep infection though not very routinely done.
Laboratory evaluation includes ESR and CRP. White blood cell count is rarely elevated in late chronic infection and is not a sensitive screening tool. ESR greater than 30 mm/h and CRP greater than 10 mg/L have been shown to be reasonably sensitive and specific for the diagnosis of infection. In noninfected patients, the ESR may take 1 year to return to normal, whereas the CRP should normalize within 3 weeks after hip replacement. A single-center study showed that the serum interleukin-6 level is very sensitive and specific for periprosthetic infection.
Hip aspiration is warranted if either the ESR or the CRP is elevated, or if the index of suspicion for infection is high despite normal ESR and CRP values. Aspiration should not be undertaken until at least 2 weeks after discontinuation of antibiotic therapy. This is done in an outpatient setting with the patient under local anesthesia, and usually, in our practice, the aspiration is performed by a musculoskeletal radiologist. Fluoroscopy is necessary for accurate insertion of the needle. The aspiration is done with the same attention to sterile technique as a surgical procedure, with a full surgical scrub and preparation. Skin flora may be introduced into the cultures and confuse the results, or, worse, they may be introduced into the joint. An 18-gauge spinal needle is inserted from anterior at a point just lateral to the femoral artery along a line from the symphysis pubis to the anterior superior iliac spine . As an alternative, the needle is inserted laterally, just superior to the greater trochanter. The tip of the needle must enter the joint and must be seen and felt to come in contact with the metal of the neck of the femoral component. Gentle rotation of the extremity helps bring fluid toward the needle if none is easily withdrawn after entering the joint. Gram stain, aerobic and anaerobic cultures, and cell count with differential are obtained on the aspirant. According to Salvati et al., if the WBC count is greater than 25,000 leukocytes/mL with greater than 25% polymorphonuclear leukocytes, infection should be suspected.
Nuclear medicine studies may be obtained if the diagnosis of infection is not confirmed by hip aspiration, and the index of suspicion remains high. The indium-111–labeled white blood cell scan seems to be more reliable than previously studied methods. In a prospective study of musculoskeletal infections, Merkel and Brown found indium-111 scans to be accurate in predicting infection in 88% of cases compared with 62% accuracy for differential technetium and gallium scans. Scher et al. found 54% positive and 95% negative predictive values for indium-111 scanning and suggested a limited role for these studies in which the absence of infection may be confirmed. Similar results were noted for the combination of technetium-99m sulfur colloid and indium-111 leukocyte scanning with 66% sensitivity, 89% negative predictive value, and 90% accuracy.
|2.||Débridement and irrigation of the hip with component retention|
|3.||Débridement and irrigation of the hip with component removal|
|4.||One-stage or two-stage reimplantation of total hip arthroplasty|
Management choices are made based on the chronicity of the infection, the virulence of the offending organism, the status of the wound and surrounding soft tissues, and the physiologic status of the patient.