Diagnosing And Treating Osteomyelitis

Dr. Armstrong: Let’s talk about more of our diagnostic modalities. What about some of the things that have been talked about as gold standards in diagnosing deep infections such as osteomyelitis? What about probing to bone for instance? Is it an accurate means of assessing osteomyelitis? Is it worthwhile? Should we be doing it in all of our patients?
Dr. Lipsky: One study described using this technique for diagnosing osteomyelitis. Because the positive predictive value of the test was 89 percent and it’s cheap and easy to do, people have widely adopted it as a means of diagnosing osteomyelitis, without going any further.²
Clinicians need to understand that there are some caveats to interpreting the results of that study. It took place at one center and the probing was done by only one or two people. Secondly, all of those patients were enrolled in a study of antimicrobial therapy that was looking at people with limb-threatening infection. Because of that, the pre-test probability of osteomyelitis in that group of patients was 66 percent. Therefore, the positive predictive value of any test would be pretty good and probably overstated in this population.

The next problem is that the test is often not performed as described by practitioners. I often see people say they’re probing to bone using the back of a wooden swab. The feel of bone under a wooden swab is not the same as it is under a metal probe.
The final issue with this diagnostic modality is to ask about the inter- and intra-observer variability of the test. If the four of us were to do this test blinded on a patient, would we all agree that it was positive or negative? Even if I was to do the test on the same patient more than once, would I agree with myself?

What Studies Reveal About The Probe To Bone Test
With these questions in mind, we put the probe to bone method to test in a study that we recently completed, comparing linezolid against an aminopenicillin beta-lactamase inhibitor combination.³
That study was conducted at multiple centers on 370 patients using the technique, as defined in the original paper, with a metal 14-gauge eye probe. The prevalence of osteomyelitis in our patients was only a third of that in the original paper. We hope to report our results soon. In the meantime, I view the probe to bone as a useful test, but not predictive enough to stop doing other diagnostic tests if the rest of the clinical picture is not highly suggestive of osteomyelitis.

Addressing Other Misconceptions About The Test
Dr. Armstrong: So it’s possible that the more high-risk patients you are seeing, the greater the likelihood that this probe to bone test will in fact be positive and potentially useful. However, perhaps for many clinicians who are in private practice and who are not seeing these osteomyelitic patients on a daily basis being referred to them, the pre-test probability is lower. Therefore, there’s a lower likelihood that this probe to bone test will be a definitive diagnostic test for osteomyelitis.
We sometimes bandy about the notion that these probes are a “poor man’s bone scan.” I think that’s probably a bit injudicious. How then should we go forward in diagnosing bone infections complicating soft tissue infections? If we can’t probe to bone, are we going to get a bone scan or a Ceretec scan or an Indium 111 or an MRI?
Dr. Joseph: People always come up to me and say they had this case in which they could touch the bone or see the bone. Therefore, they said they had to treat it for osteomyelitis. I ask them exactly what they were using to probe. But I always liked the old podiatry packer-spatula. It works wonderfully. It works as well as the eye probe.
Dr. Lipsky: But to the extent that the test was defined with a certain technique, you can’t really say that you can use a non-validated technique and had a positive probe to bone. It may well be that there are other devices that are more effective, but until we’ve tested them, it’s difficult to say …
Dr. Joseph: Right. It just goes to exactly what you were talking about earlier. We don’t have any of these questions answered. Does it make a difference if you use a 14-gauge eye probe vs. a packer-spatula? I agree with you that the wood feel is going to blunt it.

When You Should Pursue Other Diagnostic Tests
Dr. Joseph: It’s not only the patient population, but the severity of the infection (you have to consider). We see patients come in and we debride a lesion or an ulceration. OK, there’s a little exposed bone there. We can freely see it, we can freely probe it, but there’s no other sign of infection and the bone is solid when you probe it. Does that mean there’s osteomyelitis there?
I think with some of the data that Ben has accrued, we may not automatically assume that which we were doing before. If there are no clinical signs of infection and it just so happens that this infection is deep enough that bone is exposed, it doesn’t now mean there is osteomyelitis.
We’re all aware of the MRI studies, the Ceretec studies, the labeled white cell studies and the bone scan work that’s been done. I think most people would agree MRIs have probably become the most sensitive and specific test with the highest predictive value, but that comes at quite a cost. It’s also very operator-dependent.
I don’t have a good answer on a definitive diagnosis for osteomyelitis other than a bone culture taken properly. There’s also been an incredible overuse of the old technetium bone scan. Everyone gets those and they’re always positive for lots of reasons other than osteomyelitis.

What Potential Impact Will The Test Have On Treatment?
Dr. Armstrong: Andrew, how do you diagnose osteomyelitis? This is obviously a problem you see very frequently. Your pre-test probability for osteomyelitis is probably higher than most, but you also see community patients as well as referrals.

Dr. Boulton: As Ben has said, it may not be much better than tossing a coin doing a probe to bone test, which we relied on, perhaps falsely, in the past. You have to ask the question if you’re going to do a test: is it going to alter your management if you find the result? If you’re going to do a lot of tests and you’re still not going to know the result, then you have spent a lot of money and it’s not going to alter your management.
Also, before answering the question directly, you have to ask if we have a definitive treatment for osteomyelitis. If you do have treatment X which is going to eradicate at 99 percent, then the importance of the test becomes greater.
It seems to me at the moment that we’re as much confused about the diagnosis of osteomyelitis as we are the management. Increasingly, physicians such as myself, podiatrists and perhaps even some surgeons are relying more on medical treatment for localized osteomyelitis in the diabetic foot. Traditionally, in most of the older textbooks, osteomyelitis equals surgery.
Now if you’ve got a well perfused foot with localized osteomyelitis in a digit and you’ve got the possibility of a medical treatment and you can then afford to observe it for a while, then the importance of a test of diagnosis is less marked. Of course, the cost of antibiotics is a consideration.
(Getting back to your question), I think we really don’t know. What we tend to do is still rely upon clinical suspicion and repeated radiographs which are cheap. We do use Indium white cell scans and I still think they are probably the most sensitive as proven in the literature. We have used MRI scans but not very frequently.

Is There Any Evidence On Antibiotics For Osteomyelitis?
Dr. Armstrong: Are you telling me that we should be treating bone infections in the distal forefoot just with prolonged antibiotics alone?
Dr. Boulton: I’m not telling you that, but I’m suggesting that I have used that in the past with success. There are a number of new techniques. We’ve used a combination of antibiotics in the well perfused foot. We use that plus local debridement surgery to remove the infected bone. We’ve also used antibiotic beads, which are unproven but interesting.⁴
Dr. Joseph: I was recently asked to give a lecture called “Antibiotics In Osteomyelitis” and I did a Medline search on the subject. There was a very interesting paper by Stengel in Lancet Infectious Diseases just last year.⁵ They looked at all of the published and unpublished control trials from 1966 to 2000 (all bone and joint infections). Basically, his conclusion was and this is a direct quote: “There exists little high quality evidence on antibiotic therapy for osteomyelitis.”

To me, that was amazing. If you combine that with a quote from Carl Norden, MD, a number of years even before that when he says something to the effect of: “Vital questions such as which antibiotic to use and the duration of therapy have yet to be answered and no answers are forthcoming.”
The data’s not there. There have been studies looking at everything from surgical to just medical, all small little trials. You’ve got some data, from primarily the orthopedic surgeons it seems, that come right out and say you need to do ablative surgery. Then there were some studies that say it’s not necessary. According to Venkatesan in a retrospective study in Diabetic Medicine in 1997, 77 percent of patients with osteomyelitis resolved completely with oral antibiotics alone and he had a six-year follow-up.⁶
I think the bottom line is we don’t have good answers to these questions. Every case is individual. Some we’ll treat medically and some we’ll treat surgically.

Other Pertinent Tips On Detecting Osteomyelitis
Dr. Lipsky: Warren has raised an important point. Certainly, the fact that you can put a probe down and touch bone or even the fact that you can see bone does not necessarily mean that it’s infected. Carl Norden’s studies found that it was very difficult to infect rabbit bone.⁷ You had to induce necrosis of the bone and inject enormous numbers of organisms in order to cause that infection. So the bone is quite resistant to infection if it’s normal bone and reasonably well perfused, which may not always be the case in diabetic patients.
I’d like to return to a point Andrew made about the value of a plain radiograph. Most of us have these available whereas we may not have ready access to scans and MRIs. If you’re not sure if a patient has osteomyelitis, you get a baseline foot X-ray. If you see no changes suggesting osteomyelitis, you can treat that patient as if he or she has a soft tissue infection for a couple of weeks and then repeat the X-ray. If the X-ray is still negative, then your patient probably does not have osteomyelitis. If, on the other hand, the patient now develops bony changes, he or she probably does have osteomyelitis. It is a very inexpensive way to make the diagnosis.

When The Degree Of Infection Comes Into Play
Dr. Armstrong: I think we sometimes forget that most of these problems, especially in terms of bone infections, are not so acute that we have to act immediately. One of my former mentors, William Todd, DPM, used to always tell me there are two kinds of infections, those that you have to treat immediately and those that can wait.
For many of these infections that tend to be brewing, I think we can afford to take the cautious road and put the person on an antibiotic for a period of time, wait and see how these patients do.
Dr. Lipsky: I’d like to take that one step further. You can sometimes wait to put the patient on antibiotics if you think you may need to do a biopsy. If the plain X-ray shows a foot abnormality that might be either a Charcot neuroarthropathy or an infection, and the patient is systemically well, you can actually hold the antibiotics and do the bone biopsy to culture the bone.
Dr. Armstrong: Even if the patient has local signs of infection? The patient comes into your general medicine clinic and has a wound that probes down to bone. You fear osteomyelitis but you say maybe that’s Charcot in the midfoot. This is a whole other can of worms. Are you going to wait two or three days even with local signs of infection?
Dr. Lipsky: There are two issues to address in that question. If the patient has a serious soft tissue infection, and certainly, if he or she has any signs of systemic infection (fever, chills, leukocytosis) or even metabolic instability, then you certainly would start them on antimicrobial therapy immediately and not worry about the diagnostic test.
On the other hand, we see many more patients who come in with chronic, indolent problems. The ulcer has been present for a long time. The X-ray is taken and there’s some abnormality in the bone. In that situation, you can wait to start antimicrobial therapy.
The other issue is how quickly you can mobilize whomever it is in your hospital who does the bone biopsy. If they can do it quickly (i.e. in a day or two), it’s well worth waiting before starting antimicrobial therapy. Remember that treating for osteomyelitis requires many weeks of therapy.

What About The Value Of The Bone Biopsy?
I just want to make another comment about the bone biopsy. Warren has raised the issue of whether it’s the culture or the histology that’s most useful diagnostically. This is yet another question that has not been answered definitively. Some would argue that because so many of these patients are put on antimicrobials, a negative culture but positive histology probably means they have osteomyelitis. On the other hand, many would say the histology can be abnormal because of an overlying inflammatory response of the soft tissue and therefore you have to count on a bone culture.
Dr. Joseph: Bone pathology is not very specific. Unless you have a really fine bone pathologist at your hospital or center, most pathologists aren’t used to looking at bone, any inflammation in the bone. Any necrosis in the bone at all, even a post-surgical change, is going to be read out as osteomyelitis. We all know that when you’re doing a procedure and they ask you for a diagnosis on the form, as soon as it says ‘Rule out osteomyelitis,’ they’re going to make you happy and say it’s osteomyelitis.
Let’s say that patient comes in and he or she has been on antibiotics like you suggested. Ideally, how long does he or she need to be off antibiotics before the patient should be cultured?
Dr. Lipsky: It’s a great question that I have discussed with others, including Carl Norden, who is the dean of investigators in the field of osteomyelitis. While no one knows the answer, what we can say is there probably (needs to be) at least three drug half-lives to reduce the antibiotic down to a sub-therapeutic level. Probably, the longer you can go without antibiotics to some degree, the better. Our general rule has been to take people off for at least two or three days prior to doing a bone biopsy, if it’s safe to do that.
We’ve had relatively few cases among the 200 or so bone biopsies that we’ve done over the past 10 years, where somebody who we highly suspected had osteomyelitis, and had histological osteomyelitis, had an unexpectedly negative bone culture.
However, as you know, osteomyelitis can be a focal disease and you can miss it with a bone biopsy. So we encourage our interventional radiologists, who do our bone biopsies, to take at least two or possibly three cores of bone, to make sure we haven’t just missed the organisms.

References:

References

2. Grayson ML, G.W. Gibbons, et. al. “Probing to bone in infected pedal ulcers: a clinical sign of underlying osteomyelitis in diabetic patients.” JAMA (1995) 273(9): 721-723.
3. Lipsky BA, Armstrong DG, Acin F, et. al. Treating diabetic foot infections with linezolid versus aminopenicillins: a randomized international multicenter trial. Abstract 189, Infectious Diseases Society of America, October 25, 2002, Chicago, IL.
4. Armstrong DG, Findlow AH, Oyibo SO, Boulton AJM. The use of absorbable antibiotic-impregnated calcium sulphate pellets in the management of diabetic foot infections. Diabetic Med (2001); 18: 942-943.
5. Stengel D, Bauwens K, Sehouli J, Ekkernkamp A, Porzsolt F. Systematic review and meta-analysis of antibiotic therapy for bone and joint infections. Lancet Infect Dis (2001 Oct); 1(3):175-88.
6. Venkatesan P, Lawn S, Macfarlane RM, Fletcher EM, Finch RG, Jeffcoate WJ. Conservative management of osteomyelitis in the feet of diabetic patients. Diabet Med 1997 Jun;14(6):487-90.
7. Norden CW. “Lessons learned from animal models of osteomyelitis.” Rev Infect Dis (1988);10:103-9.