All Conditions & Treatments

Prevention of Infections in Total Joint Replacements (Arthroplasty)

Interviews with Experts

Paget: I am pleased today to present Dr. Barry Brause, an infectious disease specialist -- a specialty in the general area of internal medicine. Over the past 20 years, he has developed a particular interest in prosthetic joint infections, although he has a general approach to infections in this hospital and in the setting of musculoskeletal diseases.

Dr. Brause, we do approximately 2,000 to 3,000 joint replacements here out of our 15,000 surgeries a year. What percentage of people might develop an infection in that joint?

Brause: The infection rate is less than 1% for the prosthetic joint infections implanted at this hospital, but that just covers the initial three months post-operatively. There is a larger risk that goes forward from there--an infection of the joint by blood stream sources and from infections elsewhere in the body. It is generally thought that the infection rate varies between 1% and 4% over the course of the life of the prosthesis.

Paget: Has that infection rate changed since John Charnley developed the hip replacement in England in 1960?

Brause: Yes, the infection rate actually was as high as 14% back when the prosthesis was first introduced, as John Charnley devised it. And then, as it was performed by a number of other surgeons, they all had to go through a learning process. As the learning curved progressed, the infection rate came down. A number of preventive techniques were employed that helped that infection rate go down to the low rate it is at now.

Paget: So let's talk about those points. A person comes to a doctor and has a knee or a hip pain, and eventually is diagnosed as having either osteoarthritis or rheumatoid arthritis, and it's deemed that a joint replacement is needed. What should that set up both pre-operatively, just post-operatively, and then in the long term, as far as a process to prevent these infections, because they are so horrible when they occur?

Brause: I think the first thing is that patients have to become knowledgeable about the fact that they are having a foreign body introduced permanently into their body. As a foreign body, it cannot defend itself against infection the way the rest of the tissue in the body can. If other tissues in the body are exposed to bacteria, those tissues are likely to be able to defend themselves. But when those same bacteria are allowed access to tissues that are right next to the metal prosthesis, then there is an enhanced capacity to establish an infection, because our bodies cannot clear the germs from those locations when there is a foreign body in the midst of the tissues. So I think they need to know that that is the problem. Prostheses, as wonderful as they are in terms of restoring function, do represent a substantial increase in risk for infection at that site.

Then there are a number of issues that should be brought up to the patient prior to the implantation. There are three areas that are particularly prominent that all patients should be thinking about along with their physicians.

The first is that if they have any skin conditions, they should be under very good control. If they have a chronic dermatitis, like psoriasis, it should be very well controlled prior to the implantation. And then they should be on a regimen, and a continuing cure protocol with the dermatologist, to keep it under control, because the biggest defense we have against infection is an intact skin. When they don't have an intact skin, when they have any skin condition that would allow a hole to develop in the skin surface that represents a risk of infection, that can seed the blood stream and can wind up infecting the prosthesis.

The second area is the teeth and the gums, an area that just cannot be neglected. It needs constant maintenance to be sure that the ordinary bacteremia, the ordinary number of bacteria into the blood stream associated with chewing and brushing teeth, is minimal. If patients get worried about going to see the dentist because they have been told that seeing the dentist represents a risk for their prosthesis, which it may, some people respond by not going to the dentist, and that is the wrong approach.

So first they need to be encouraged to make sure that they are up to date with their dentist and that there is no unfinished business in their mouths. They should avoid the idea that "Well, I'll take care of this after my joint is replaced", because it is very important to have it taken care of before the joint is replaced. We should talk later about what should be done once the prosthesis is in to reduce the risk of those dental visits, because dental cleaning is a significant cause of bacteremia and there are situations where the data has been fairly persuasive that prophylactic antibiotics are of use.

The third area would be urologic issues. In men who have an enlarging prostate and urinary problems, where they are already having bladder dysfunction, many of them will say, "Well, I'll have my prostate taken care of after my prosthesis is put in", and that is really the wrong way. If the prostate is already at a point where it needs to be operated upon, then that surgery should not be delayed until the prosthesis is put in.

It is preferable to have it done before the prosthesis is put in, so that the enlarged prostate, which represents a risk of a urinary tract infection, will no longer be such a risk. And the risk of a urinary tract infection will be largely reduced. So when the prosthesis is in place, it won't be vulnerable to the potential bacteremia that such a urinary tract infection could represent, particularly in the perioperative period. Post-operatively, it is very common for these patients to require bladder catheterization, which can be a particular issue if your prostate is very large and you don't have good bladder function.

In women, it is a different anatomic situation, but it is the same functional result, in that women who are predisposed to urinary tract infections need to bring that under control. We may not be able to eliminate the cause of their predisposition, but we should be able to go through risk reduction in coordination with urologists and other specialists in the area. If they need to have a bladder resuspension, that shouldn't be put off if that is indicated. That would be a fairly rare intervention. I think, more commonly, acidification of the urine, drugs like Methenamine Mandelate (Mandelamine) and other types of procedures that are going to be specific for the individual should be employed. The patient should be fairly knowledgeable about them. But taking advantage of the opportunity that the prosthesis represents, the physician can motivate them to keep that type of problem at the lowest risk possible.

Paget: So we have addressed the issue of what a patient should do prior to even coming in for the surgery. What has science and a hospital like Hospital for Special Surgery done to decrease the incidence of infection while they are in the operating room and in the hospital?

Brause: As I mentioned before, the biggest improvement in infection rates occurred as the techniques of surgery improved. The second biggest improvement is associated with using perioperative antibiotic therapy. That has been shown to substantially reduce the risk of having a prosthetic joint infection. I think that, universally, people are using prophylactic antibiotic therapy for about 24 hours, at the most 48 hours, post-operatively, starting immediately before the implant is inserted. I think that continues to be an effective deterrent and risk reducer with regard to perioperative infection rates.

In addition, some prostheses are cemented in place, and some surgeons may choose to use antibiotics in the cement, which may represent an additional advance in terms of risk reduction. It is not clear that that actually represents a risk reduction in humans. So that is something that should be decided individually by the surgeon, related to the specific type of prosthesis that is put into which individual. That is an option.

The third area that has been of some benefit is control of the air flow in the operating room, which is a very expensive investment by the hospital, to create a system in which the flow is all directed in a fashion to reduce the risk of infection. This is a very expensive outlay, and it does give a very small decrease in the infection rate. It is the type of decrease in infection rate that is so small that most general hospitals probably would not make that investment, but hospitals that specialize in putting in prostheses might have made that investment, as we have here at Hospital for Special Surgery.

Paget: And post-operatively, if there are wound infections or urinary tract infections, one needs to be aggressive in treating those and calming them down. So if somebody has had joint replacement, is home and kind of getting back to their lives again, what other methods of infection prevention should be carried out?

Brause: The first thing that I tell patients is that they have to become more adept and responsive to their first aid needs. Commonly, when we are doing something and we cut ourselves, we finish the task then we say, "I'll take care of that cut later." What happens later is that you forget about the cut because it doesn't hurt anymore. So it's not until the next day, when its festering, it's red and it gives you a little pain that you start to do something about it. By that time, it may be too late to protect you from a potential risk that it represents to your prosthesis. So if you get some injury while you are doing something, you have to stop what you are doing and take care of that injured skin area with some proper first aid -- using antiseptics and some Band-Aids -- and then go back to finish the task. It is your attitude toward risk that could represent a problem for your prosthesis.

In the same way, if you get symptoms that you think are consistent with urinary tract infection, talk to your doctor about it. In the past, you may have been able to drop your urine off, have it cultured and wait to see what the result was. Then, if the culture was positive, you would be treated for it. If you have a prosthetic joint and you have symptoms that are at all compatible with urinary tract infection, you should be treated for the urinary tract infection first and wait for the culture to come back. You can stop the therapy if it turns out that you don't have an infection. The risk that that additional 24 or 48 hours represents while you are waiting for the urine culture to come back may be a critical period of time. If the infection advances, it may become bacteremia during that period, and that may be something that could have been avoided if antibiotics had been started earlier.

Paget: So close collaboration and partnership with the physician is obviously important. Now patients will commonly have procedures. They'll have a dental procedure. They may need a cystoscopy or a colonoscopy. Is there a specific type of prophylaxis that they should think about and anticipate the need for in that case?

Brause: Yes, this has been in the past a very controversial area. I think some aspects of it are still subject to controversy. But the area that has received fairly universal agreement is that patients who are undergoing dental procedures that have an invasive nature to them -- that is a procedure that is associated with gum bleeding for instance -- should receive prophylactic antibiotics prior to undergoing those procedures. The data that is based on is largely the ideas that are behind prevention of endocarditis. And some data shows that the patients, for about two years after the implant is put in, have an increased incidence of dental bacteria causing their prosthetic joint infection.

It is thought that they are more prone to dental-origin infections at their prosthesis because of the enriched blood supply through the prosthesis during that two-year period of time, when the bone and the surrounding tissues are still healing. Statistics show that the dental bacteremia probably represents a risk all the time, but it is only enhanced vascularly during that two years. That's what the statistics show. It is on that basis that some parties who are not convinced that prophylactic antibiotics would be of use, have become convinced. Now there is a consensus that prophylactic antibiotics should be used starting about an hour before the invasive dental procedure. Just one cluster of pills is taken at that one time without any follow-up antibiotics, unless the dentist finds something that he thinks the patient should receive longer course of therapy for.

Paget: Are there specific types of antibiotics that you might mention, so the patients can talk to their physicians about them?

Brause: Well, for patients who are not allergic to penicillin, I think the preferred agent is amoxicillin, and it's usually as 2 grams of amoxicillin, which would be four 500 mg tablets taken one hour prior to the dental procedure. There are alternative drugs. Cephalexin (brand named Keflex) has been an agreed upon alternative to amoxicillin. I think we have to bear in mind that cephalexin does not cover anaerobes as well as amoxicillin does.

An alternative agent, for those who are allergic to penicillins and cephalosporins, would be clindamycin (Cleocin), which covers most of the strep in the mouth and anaerobes very well. That also should be taken one hour prior to the dental procedure and the dose is 600 mg, and that usually comes as 150 mg and 300 mg tablets and capsules.

Paget: What about other procedures like colonoscopy or cystoscopy -- do they need to be prophylaxed as well?

Brause: We don't know the real answer for that. It is subject to debate and probably it is going to be very difficult to get good numbers that would tell us that it has to be done. I think those decisions should be made on an individual basis between the patient and the person doing the procedure in consultation with a rheumatologist and even an orthopaedic surgeon, to make sure that the proper risk reduction is employed. And there should be a balance to that approach, so that patients should not receive antibiotics unless it would be in their best interest.

Paget: Say a patient has a procedure and is doing well and then there is a concern that there might be a joint infection in the prosthesis. How such patients present? How would they know that that is happening, so they can communicate with their physician?

Brause: The only really common symptom associated with prosthetic joint infection is pain. Somewhere between 90% and 95% of patients with prosthetic joints present with pain, but most of the people who have painful prostheses are not infected. Probably one in four or one in five of patients who have a chronically painful prosthesis is infected. So a painful prosthesis is one symptom that the patient would go back to the orthopaedist or the rheumatologist to be evaluated for. And one of the differential diagnoses to be entertained is whether that pain is due to infection.

The other types of presentations are things like fever, redness, swelling, and drainage from the incision. Those are all present in about 25% to 35% of cases, because generally, prosthetic joint infections present in one of two fashions. The most common type is where they present with pain which slowly progresses over a period of many months until it is so bad that the patient no longer can tolerate it.

Paget: So they will have a pain-free period between their surgery and when that new pain happens and it develops slowly?

Brause: That is exactly right. Or, they will never have the normal progression to painless ambulation that we expect. If they don't experience the benefits of the prosthesis, they may have a very early infection. But most of them have been pain-free, with everything going well, and then there is a disruption. Something goes wrong and nobody knows what that is, and one of the possibilities is that it could be an infection.

Paget: So they are erring on the side of safety. They call their doctor. How do you make the diagnosis?

Brause: The most direct way is to put a needle into the joint, using radiographic guidance when it is a hip prosthesis, because it is very difficult to actually be confident that the needle is placed in the joint space when you have a total hip replacement in place. So we always really try to do that radiographically. We do a little bit of an arthrogram, put the needle in place, and then put some dye through the needle to confirm that the needle actually is in the joint space. Then we make an aspiration and use that material for culture and for gram stain to determine whether an infection is present.

If it is a knee prosthesis, that is also a very good way of going, but because it is more superficially located, a number of orthopaedic surgeons and rheumatologists feel sufficiently comfortable with the anatomy to do an aspiration without using an imaging technique. The other types of diagnostic approaches, using imaging techniques such as centigraphy, iridium scans, gallium scans, and technetium scans, are really fraught with too much non-specificity to be helpful. There are too many false positives and too many false negatives to really carry the burden of making a diagnosis about whether prosthetic joint infection is present or absent.

Paget: So how are these prosthetic joint infections treated?

Brause: There are three basic approaches. The classic approach is one that was actually developed here at Hospital for Special Surgery. Although this approach has the highest success rate, it is also the one that is the hardest to perform -- the prosthesis is removed. The patient remains without a prosthesis in place for six weeks while they receive antibiotics that are specific for the germ that is causing their infection. At the end of the antibiotic therapy, they receive a nice new implant at that site. That takes six weeks plus the rehab at the other end after the prosthesis is put back in. So the patient really is out of their normal functioning stride for about two months.

Another approach is to take the prosthesis out and to put another prosthesis in immediately, during that same operative procedure. That is an approach that was developed in Europe, and it has a substantially lower success rate than a two-stage procedure. But it is used in those situations where the patient, for other reasons, usually other medical reasons, cannot have a period of time with the prosthesis out. When the prosthesis is out, the patient is not normally ambulatory, and if they can't handle that period of relative bed rest or walking around with crutches and a walker, then it may be in their best interest to have what is called an 'exchange procedure' which is that one operation.

Paget: And they get the six weeks of antibiotics as well on top of that?

Brause: That is the protocol. We use that protocol. In Europe, they actually do not do that, and that may be one of the reasons why their success rate was lower. But even when you use the antibiotics after the prosthesis has been exchanged, it should be expected that the persistent infection rate will be substantial because you never really had the opportunity to treat the patient when the prosthesis was absent. As we said at the beginning of this talk, the main predisposition to the infection is the presence of the foreign body. If you don't get a chance to treat the germ in the absence of the foreign body, your success rate is going to be substantially lower.

Paget: And what is the third scenario?

Brause: The third scenario is reserved for people who cannot undergo surgical removal of the prosthesis, for medical or orthopaedic reasons, and where the prosthesis is not loose (so we still have a functional prosthesis), where the organism causing the infection is not causing the patient to be systemically ill, where the organism is exquisitely sensitive to all agents, and where the patient can tolerate such antibiotics.

When those criteria are fulfilled, then we will approach the patient with an idea to suppress the infection. That is where we acknowledge that we will not be able to eradicate the infection, but instead we try to control the infection to such degree that the patient will never know that there is an infection going on.

That requires, often times, a surgical procedure at the beginning to drain out the fluids around the prosthesis, although there are cases where that is not included. That is followed by intravenous antibiotic therapy, usually to get a head start on the infection, but there are also cases where that has not been employed.

What is employed is protracting all antibiotic therapy, designed specifically on the basis of the germ that is causing the infection, using antibiotics that are well absorbed through the gastrointestinal tract, so that we can get good tissue levels and maintain high-dose therapy, with very slow tapering of the dosage. The patient stays on antibiotic therapy for the life of the prosthesis, and this may not be for the patient's whole life, but it will be until the prosthesis needs to be replaced.

It may well be that the prosthesis will survive for the patient's normal life span, but then it could loosen because of mechanical reasons. When the prosthesis needs to be replaced, we should take advantage of that opportunity to eradicate the infection completely.

Paget: We have come so far in 42 years. Are we going to come even farther in the next 42 years and, if so, what will that entail?

Brause: I'm confident that we are going to make improvements, but I have no idea. I think that the innovations that are presently being worked on, not just thought about, are innovations that go beyond antibiotic therapy. These are trying to devise techniques by which we make the prosthesis capable of repelling the organism. So the organism becomes incapable of attaching itself to the surface of the prosthesis. Those techniques are now in the laboratory, and we hope that they are going to be useful in the future.

Paget: And obviously the responsibility is mine as a rheumatologist to prevent the joint problem in the first place, and finally, the issue of antibiotic resistance. We have just heard that some streps are now resistant to erythromycin, one of our mainstays. What can patients and their physicians do to try to avoid changing the flora of the world so that these bugs become our enemies and we can't get rid of them?

Brause: That is a very good topic to discuss at every forum, because I think it is something that we need to address as part of our society as a whole. I think that this is a good opportunity to make a pitch for not using antibiotics unnecessarily. That is in the individual that we are talking about, who has a prosthesis, but it is also in society in general, because it is by the overuse of antibiotics in general that we induce resistance. It is not just the use of antibiotics in patients. It's the use of antibiotics in cattle and other food producing livestock. We need to encourage our government and representatives to encourage the people in that industry to reduce their use of antibiotics in livestock, because that is generating a whole host of resistant organisms.

In addition, more recently, patients have taken to using antibacterial soaps, which are not a good idea to use on a chronic basis. If there is an issue, or if their dermatologist, rheumatologist or somebody has said that for a short period of time using an antibacterial soap would be an advantage to them, that is fine. But I think if they use it chronically, then they will be altering the organisms on their skin surface, and they will be colonized with resistant organisms. So if they ever get an infection from germs on the skin surface, it will be harder to treat.

Dr. Brause was interviewed by Dr. Stephen A. Paget, Physician-in-Chief, Hospital for Special Surgery


Stephen A. Paget, MD, FACP, FACR
Physician-in-Chief Emeritus, Hospital for Special Surgery
Stephen A. Paget Rheumatology Leadership Chair
Barry D. Brause, MD

Attending Physician, Hospital for Special Surgery
Professor of Clinical Medicine, Weill Cornell Medical College

    Other resources

    Success Stories


    In-person and virtual appointments

    Related Content