Answer: Treatment of prosthetic joint infections (PJIs), such as total hip replacements, is challenging. Before we decide on which surgical approach to take, there are two key questions to ask about the patient's history: 1) How long has it been since total hip replacement surgery? and 2) How long have the symptoms of infection been present?

While infections that develop soon after implantation may be successfully treated without device removal, late-onset infections are likely to require device removal for cure.^1,2 The Infectious Diseases Society of America (IDSA) notes that early-onset MRSA-associated PJIs that occur <2 months after surgery, or acute hematogenous infections with a stable implant and short duration (≤3 weeks) of symptoms may be treated with surgical debridement and device retention. That approach is accompanied by IV antibiotic therapy, as well as adjunctive oral rifampin therapy for 6 weeks followed by 3-6 months of oral antibiotic therapy. (Note that current IDSA guidelines recommend 2 weeks of parenteral antibiotic therapy.) Oral antibiotic therapy may include a variety of different antibiotics that are active against MRSA, such as trimethoprim sulfamethoxazole (TMP-SMZ), a tetracycline, clindamycin, or a fluoroquinolone.³ Cure rates with this approach are not universal, but have been reported to be over 70%.^4,5 Device removal (plus debridement) is recommended, whenever feasible, for patients with unstable implants or >3 weeks of symptoms.

In terms of surgical therapy, randomized clinical trials of approaches are lacking, so expert recommendations are typically based on observational studies. It appears that the ideal treatment is a one-step or two-step exchange (Figure 1). A one-stage (ie, direct exchange) procedure requires removal of the infected joint, debridement of the infected site, and reimplantation of a new prosthesis during a single surgery. The published success rates can be over 85%.^2,5 However, for virulent organisms such as MRSA, the cure rates may be lower.⁵

Most experts recommend a two-step exchange procedure for all PJIs, including those caused by MRSA, as published cure rates are typically about 90%.^2,6 The two-stage exchange procedure involves removal of the infected prosthesis, debridement, and placement of an antimicrobial-impregnated spacer. This surgery is followed by 4-6 weeks of intravenous antibiotics active against the infecting organism(s). Reimplantation of a new prosthesis is performed following completion of antibiotic therapy. Some experts recommend that, before the new prosthesis is implanted, follow-up cultures of the infected sites are done for patients with virulent organisms like MRSA, although not all studies have shown this approach is associated with improved cure.⁷

Back to top

Answer: First of all, patients with MRSA bacteremia must not only be treated for their MRSA bacteremia, but must have source control as well. Common sources for MRSA bacteremia include IV catheters, pneumonia, skin and skin structure infections (SSSIs), endocarditis, and bone and joint infections, although other less common sources are possible. Some of these infections such as endocarditis and bone and joint infections require 4-6 weeks of therapy.³

The IDSA guidelines emphasize that all patients with MRSA bacteremia should undergo echocardiography, preferably using transesophageal echocardiography as opposed to transthoracic echocardiography,³ as the proportion of patients with MRSA bacteremia that have endocarditis is not insignificant—13% in some case series, including 21% of community-associated MRSA bacteremia.⁸ However, recently published literature shows that patients who are at low risk of infective endocarditis may not routinely require transesophageal echocardiography.⁹

Treatment of bacteremia with short-course therapy (2 weeks) may be feasible in select patients. Experts recommend that patients eligible for short-course therapy should fit several criteria, including removal of catheters that are suspected sources of infection, performance of a transesophageal echocardiogram, no implanted prosthesis (eg, arthroplasty, prosthetic valve), negative follow-up blood cultures 2-4 days after the initial culture, defervescence within 72 hours of starting effective antistaphylococcal therapy, and no signs or symptoms of metastatic staphylococcal infection (Table 1).¹⁰

However, episodes of MRSA bacteremia that don’t fit these criteria need prolonged therapy, typically at least 4-6 weeks, depending on the type of infection such as osteomyelitis or endocarditis. Some have tried to push short-course therapy to even shorter courses than 2 weeks. However, observational studies have found that treatment of bacteremia with courses for <2 weeks is associated with high rates of clinical failure,¹⁰ so even the most uncomplicated and rapidly responding of MRSA bacteremias should be treated for at least 2 weeks.

While intravenous therapy was the standard of care for MRSA bacteremia for many decades, more recently there has been interest in “step-down” therapy with oral antibiotics, such as linezolid. Oral therapy has many advantages, including earlier hospital discharge, avoidance of home IV therapy and the risks associated with IV catheters (such as infection and thrombosis), and lower cost compared with IV antibiotics.

In a pooled analysis of randomized studies, linezolid was associated with similar clinical outcomes as IV vancomycin for the treatment of selected MRSA infections (such as pneumonia and complicated skin and soft tissue infections) complicated by bacteremia.¹¹ Many of these patients were transitioned from IV to oral linezolid after 7 days of therapy. In this study, cure rates were similar between those treated with linezolid and those treated with vancomycin. In another study, linezolid was used for salvage treatment of persistent Staphylococcus aureus bacteremia. Compared with vancomycin, linezolid was associated with a high success rate (75% vs 17%, P = .006) and lower Staphylococcus aureus–related mortality (13% vs 53%).¹² The above findings suggest that oral step-down therapy with linezolid is acceptable for patients who lack the risk factors for poor outcomes described in Table 1.

Of note, a 2007 letter from the FDA noted that there was increased mortality in a clinical trial of linezolid versus vancomycin for the treatment of catheter-related bacteremia in patients who were treated with linezolid. However, the higher mortality in the linezolid group was seen in subgroups with gram-negative infections, mixed gram-negative and gram-positive infections, or those who had no identified pathogen. Mortality differences were not seen in patients with gram-positive infections, such as MRSA. Thus, if MRSA bacteremia is associated with a catheter-associated infection and there is evidence of concomitant gram-negative infections, the use of linezolid alone would not be sufficient, as expected, given its lack of gram-negative activity. Use of linezolid in catheter-associated infections in which the pathogen cannot be identified may also be a suboptimal choice. For other MRSA bacteremias, linezolid may offer the advantages outlined above.

Back to top

Answer: MRSA pneumonia is a very serious and often deadly infection. Vancomycin was the traditional treatment of choice because there were no adequately studied or available alternatives until the past decade.

Despite its activity against MRSA, vancomycin has never been considered an optimal treatment for pneumonia. The concentration of vancomycin in the epithelial lining fluid (ELF), a site in which drug concentration is believed to be important for pneumonia treatment, is relatively low, typically 25%-33% of serum concentrations.¹³ This suggests that vancomycin concentration at the target organ of infection may be suboptimal. Other antibiotics have higher concentrations in the ELF. The best studied is probably linezolid, which has ELF concentrations one to four times higher than plasma concentrations.¹³ However, some studies of ELF concentrations were performed in healthy volunteers, and pharmacokinetics in this population may differ from that in ill patients.

Regardless of ELF concentrations, clinical efficacy from randomized trials is more relevant for patient care than pharmacokinetic studies. A recently published clinical trial demonstrated that linezolid had higher clinical cure rates compared with vancomycin in the treatment of hospital-acquired and healthcare-associated MRSA pneumonia (58% vs 47%, P = .04).¹⁴ However, in this study, treatment was initiated with either drug, and treatment was started in the first 48 hours of antibiotic therapy, so it did not represent a salvage therapy situation, as in the case described above. Additionally, the study did not include patients with community-associated MRSA pneumonia. Thus, the role of linezolid as salvage therapy for patients failing vancomycin for MRSA pneumonia or as primary treatment for community-associated MRSA pneumonia is unclear, although one would expect its efficacy to be as good as or possibly better than vancomycin. 

While TMP-SMZ appears to be reasonable therapy for SSSIs caused by MRSA in observational studies, its use for serious Staphylococcus aureus infections has been studied and is not promising. In a randomized clinical trial of hospitalized patients with Staphylococcus aureus infection, over half of whom had MRSA, cure rates were lower with TMP-SMZ than with vancomycin (86% vs 98%, P = .01).¹⁵ Thus, it appears to be a suboptimal choice for salvage therapy for critically ill patients.

Longer-acting tetracyclines such as minocycline and doxycycline appear to be reasonable therapy for SSSIs caused by MRSA, but data on their efficacy for the treatment of MRSA pneumonia are lacking. Thus, it’s difficult to recommend their use for MRSA pneumonia or patients failing vancomycin unless other treatment options are contraindicated.

Telavancin has been studied for the treatment of hospital-acquired pneumonia in two randomized clinical trials. Compared with vancomycin, telavancin was found to be non-inferior.¹⁶ Thus, telavancin could be considered in this setting, although there is no clear advantage over vancomycin. It is unclear how this drug compares with linezolid for pneumonia, as they have not been compared in randomized clinical trials.

Ceftaroline is approved for the treatment of community-acquired pneumonia and is active against almost all MRSA strains or all MRSA isolates in clinical trials. It is not specifically approved for the treatment of community-acquired MRSA pneumonia, due to the small number of patients with this infection enrolled in registry trials. However, it is approved for the treatment of skin infections caused by MRSA. It has been studied as salvage therapy for difficult-to-treat MRSA infections such as endocarditis and, in a small case series, was associated with infection resolution in five of six patients.¹⁷ However, the drug has not been systematically studied as salvage therapy for MRSA, but based on these studies, it appears promising and could be considered as salvage therapy, especially if the patient cannot take linezolid. Clearly, however, more robust studies are needed in terms of its role for the treatment of MRSA pneumonia.

In summary, linezolid is the best-studied option for patients with MRSA pneumonia that is refractory to vancomycin. Based on current studies, linezolid would probably be the treatment of choice as salvage therapy. Ceftaroline and telavancin could also be considered, especially if there was evidence of in vitro resistance to linezolid in the infecting isolate or the patient was unable to take linezolid because of drug interactions or contraindications.

Back to top