CORR Insights^®: Late Reinfection May Recur More Than 5 Years After Reimplantation of THA and TKA: Analysis of Pathogen Factors

Where Are We Now?

For more than three decades, the two-stage protocol has been a standard—and in many settings, the preferred—treatment for periprosthetic joint infections (PJIs). Although studies examining two-stage protocol generally present acceptable-to-excellent results, most of these studies describe short-term followup of less than 5 years [3, 6, 7, 10-12, 16, 18]. Indeed, we only have scarce knowledge about the long-term future of these patients and of their implants.

However, researchers are now examining PJI treatment with one-stage exchange and/or Débridement, Antibiotics, Irrigation, Retention (DAIR). And because these studies are showing good results [2, 5, 8, 14, 17], more surgeons are changing their everyday practices and treating at least some of their patients with single-stage approaches of one kind or another.

Still, in most parts of the world, two-stage revisions for PJI remain the most-common approach. And where the two-stage approach is concerned, there are large variations in treatment protocols, even with respect to what would appear to be relatively fundamental elements of the approach, such as antibiotics (length of treatment, route of administration, combinations), spacers (whether or not to use them, whether they should be static or articulating, and whether prefabricated ones are preferred to spacers that are handcrafted in the operating room), time to reimplantation, use of bone graft, and use of cemented versus cementless implants at the time of definitive reconstruction.

The present study by Garvin and colleagues [4] therefore is an important step in the right direction. Not only do the authors analyze the long-term results of the two-stage protocol [1, 9, 15], but they also examine some factors related to late reinfection. Additionally, the study focuses on antibiotic resistance and on the important issue of positive cultures at the time of reimplantation.

Where Do We Need To Go?

Although the current study is a major step forward, important gaps in our knowledge remain with respect to two-stage revision for PJI. For example, future studies will need to teach us more about spacers: When to use them, what kind(s) to use, and how long they should remain in place [10]. We need to learn more about what the best routes, durations, and types of antibiotics to use, as well as how to use them in combination most effectively [3, 6, 13]. Additionally, we need to know if and when it is safe to use bone grafts, and when metallic supplements might be preferred to treat bone defects that are so common in these arthroplasties. And, of course, the cemented versus cementless implant question remains largely unanswered in this setting.

We also need to recognize that two-stage revision is not the only approach to PJI; variations on the single-stage approach (including, perhaps, DAIR) may have a role to play. What are the long-term results of one-stage and DAIR protocols, and will these long-term results be superior or inferior to those obtained with the two-stage approach? What is the long-term mortality of these patients when compared with age-matched or comorbidity-matched controls? In general, the patients with PJI are older, and so I wonder whether patients treated with single-stage approaches may achieve better functional results because they have gone through less surgery than our patients treated with two-stage revisions; future studies should evaluate this. Since many single-stage PJI revisions are performed with cemented implants (in order to use antibiotic-impregnated cement at the time of the revision), one wonders how the durability of this approach might compare with two-stage revision for PJI, where cementless implants are more of an option.

Ideally, we would answer these questions using robust data, rather than what we now have, which generally is limited to anecdotes and case series.

How Do We Get There?

In order to answer the questions above, researchers should tackle the topic in three ways. First, researchers should conduct a contemporary review of previously published studies on PJI treatment. Researchers could limit analysis to patients treated longer than 10 years clinically, radiographically, and serologically. This contemporary review would supply enough data for patient mortality and its causes and function level. In this quite simple way, we could obtain plenty of information about long-term results with the different modalities of treatment used years ago.

Next, researchers should once again take advantage of previously published data. Many of those old series inadequately described global infectious results; only a short number of them tried to find differences among subgroups. But if we apply the appropriate statistical analysis to long-term results, we could potentially find differences between patient characteristics and his or her conditions. A more sophisticated degree of study could include data from similar or equivalent series, obtaining high numbers of patients, which would increase the statistical power of results. A systematic review or metanalysis would be most desirable, but unlikely considering researchers would be working with a retrospective series.

Finally, a prospective, multicenter series of patients managed with a randomly assigned therapeutic protocol would be ideal. The difficulty arises not only from the coordination of several centers and including enough patients with PJIs, but, also with the amount of time it would take to obtain important data. Perhaps an analysis of large databases from insurance companies or national registries could help.