Why is it that we don’t have antibiograms for anaerobes? I had a patient recently discharged on oral moxifloxacin after perforated appendicitis. She required emergency surgery and was discharged before the final culture results were available. The cultures grew 4+ streptococcus anginosus, 4+ gram positive bacilli and 4+ gram negative bacilli, both of which were identified as anaerobes, including Bacteroides fragilis (as Dr. Partlow predicted), but sensitivities were not performed. I was reading that resistance of bacteroides to fluoroquinolones is quite high, however, we don’t have any local data to guide therapy. In addition, moxifloxacin has been approved and studied for the treatment of complex intra-abdominal infections and is frequently used for this indication, which seems counter-intuitive based on its anaerobic coverage. Could you comment whether or not this treatment is appropriate?
First of all, although anaerobes are common pathogens from virtually any anatomic site, we don’t have an antibiogram as recovery is highly variable. To begin with, growing anaerobes in the laboratory requires that enough specimen is collected so that a separate anaerobic culture can be performed in addition to regular culture under aerobic conditions. Growing these organisms is very time consuming and involves biochemical testing under strict anaerobic conditions which are nutritionally and atmospherically demanding, and in some cases, gas liquid chromatograpy and molecular typing before definitive identification can be made. As in the case of your patient, many infections involving anaerobes are polymicrobial, which makes sensitivity testing very tedious, needing to separate multiple different bacteria. On the other end, false negatives are also common due to inadequate culture techniques. These procedures are costly and often are not practical for many clinical microbiology laboratories. Because of this, the CLSI recommends performing anaerobic cultures for individual cases only if the patient is clinically worsening despite presumably adequate therapy, there is difficulty selecting empiric treatment for reasons such as allergies or to check for activity if a prolonged treatment course is going to be required.
Practically speaking however, the presence of anaerobes is highly predictable based on the type of infection you are dealing with, making testing somewhat redundant. In such cases, anaerobes are simply treated empirically. These infections include many head and neck infections such as brain and peritonsilar abscesses, perimandibular space infections and dental abscesses. In respiratory infections, aspiration pneumonia and empyemas often have anaerobic involvement, and to a lesser extent, HAP. Intra-abdominal infections, particularly abscesses, appendicitis and peritonitis usually have an anaerobic component, as do pelvic infections such as pelvic inflammatory disease. Finally, for SSTI, we suspect anaerobes in diabetic foot infections, type 1 necrotizing fasciitis, bite wounds and infections of the “boxer short area”. Basically if it’s putrid smelling, involves your bum or an animal or if it’s an abscess in your belly, anaerobes are game.
In addition to these clinical scenarios, anaerobes can be implicated based on the gram stain. If organisms are seen (gram positive bacilli or cocci, and GNB), and they are not identified further, this usually means these organisms are anaerobes, and the lab will report them as simply that in the final culture. Gram positive rod-shaped anaerobes includes colostridium and actinomyces, anaerobic GPCs are usually peptostreptococcus,. GNB tend to be fusobacterium, bacteroides, prevotella and porphyromonas. In some cases, the lab can then do some basic identification tests and report what type of an anaerobe they are seeing, such as with actinomyces and bacteroides, but they don’t routinely perform sensitivities.
For selecting empiric treatment, we tend to divide the anaerobes as “above the waist”, i.e. oral, and “below the waist”. Oral anaerobes, such as those implicated in respiratory infections or dental infections tend to include the Bacteroides oralis group, the pigmented Prevotella (melaninogenicus group), Porphyromonas, Fusobacterium and Peptostreptococcus. For these, we can use clindamycin, moxifloxacin, doxy and even, in some cases good old penicillin. In below the waist situations, especially intra-abdominal infections, Bacteroides, including fragilis and different types of Colostridium predominate, and these agents don’t offer reliable coverage. Here, you want to use metronidazole, beta-lactam/beta-lactamase like piptazo or a carbapenem.
To address your question about moxifloxacin; yes, it has been studied in the setting of complicated intra-abdominal infections. There was one study specifically, done in 2006, that compared it to piptazo and showed that they two treatments are equivalent. However, bacteroides spp are becoming more and more resistant, even when compared to 10 years ago when this study was done. In fact, in one surveillance report, the rate of resistance went from 5% to 38% from 1997-2004 in the US. Overall, the current resistance rates to moxi in Canada is about 25%. In addition, in this particular study, bacteroides was only isolated in about 25% of the patients, or 40 patients in each arm, so a relatively small proportion of patients were affected by the organism where resistance is seen the most today.
In general, based on its broad spectrum of activity, ease of use and older clinical data for IA infections, moxifloxacin is not a terrible choice. However, with emerging resistance in mind, and the availability of other choices, something like Clavulin or a metronidazole containing regimen is probably better. I would not chose moxi in a patient who was critically ill either, or someone with a high chance of complications. We should do an RCT on 7N to that regard, no?