Transcription
[Music] hello again. This is the fifth lecture in this course on antibiotics. And today's topic is antibiotics for gram-negative infections.
The learning objectives are: first, to know the general spectrum of activity for antibiotics against gram-negative organisms. Second, to be able to compare the spectrum of activity and major side effects of the antibiotics used for highly resistant gram-negative infections, such as those by pseudomonas, Acinetobacter, and Stenotrophomonas. And last, to be able to select an appropriate antibiotic for infections with gram-negative organisms.
As with the last lecture, I'll start with a general antibiotic susceptibility table. Less available information in references and the literature means that the the assignment of pluses and minuses with this lecture will necessarily be more arbitrary than the last. But I'll do my best to have these signs represent a combination of typical in vitro activity and clinical usage. And I'll just cover the most clinically relevant gram-negatives.
The majority of these are enteric gram-negative rods, although there is much overlap with the clinical diseases they cause. These enteric gram-negatives tend to form groups with similar antibiotic resistance patterns. For example, E. coli, Klebsiella, and Proteus typically share a similar resistance profile, as do Acinetobacter and Citrobacter. Another important gram-negative bacteria is H. flu, as is the genus Neisseria, to which belongs the species Neisseria gonorrhoeae and Neisseria meningitidis.
As the first-generation cephalosporins have poor gram-negative coverage, I'll start just with the second generation, which cover the E. coli, Klebsiella, Proteus group very well, along with H. flu. The third-generation cephalosporins add good Neisseria coverage and some activity against the Acinetobacter, Citrobacter group. And the fourth-generation cephalosporins, predominantly Cefepime, shows excellent activity against all of these.
Combinations of ampicillin and beta-lactamase inhibitors, as in Augmentin and Unasyn, have excellent coverage of everything listed, but the Enterobacteriaceae, Acinetobacter, Citrobacter group. And the anti-pseudomonal penicillin beta-lactamase inhibitor combo, Zosyn, has great coverage of routine gram-negatives that is equal or near equal to Cefepime.
Looking at a couple more of the antibiotic classes: Carbapenems have great coverage of everything as well. Here is Ertapenem and Meropenem are interesting in that they have excellent activity against most things with the exception of rising resistance, specifically in E. coli, that has been seen in the last five years or so in the United States. Also, resistance to quinolones among Neisseria is highly specific to geography, with some countries measuring a 10% resistance rate and others a 100% resistance rate. Aminoglycosides cover all of these well, except Neisseria.
Most of the species listed so far don't usually pose a significant treatment challenge, provided adequate source control and host immune function. However, there are a collection of gram-negatives that are classified as highly resistant. That term is variably defined, more so with the gram-negatives than with the gram-positives, as we saw in the last lecture, based on how they impact the clinical care of a patient. I personally consider the highly resistant gram-negatives to be: first, Pseudomonas; second, ESBL. ESBL is an interesting term that many doctors use grammatically as if it were a specific bacterial species in itself, um, but ESBL actually stands for Extended Spectrum Beta-Lactamase. And the term is usually used interchangeably with ESBL-producing bacteria. ESBL-producing bacteria are always aerobic gram-negatives, most commonly strains of E. coli and Klebsiella.
What exactly is ESBL? Well, typical beta-lactamases are bacterial enzymes that confer resistance to penicillins and narrow-spectrum cephalosporins by cleaving the beta-lactam ring of those drugs. ESBLs increase resistance to also include the anti-pseudomonal penicillins like Zosyn, the third and fourth-generation cephalosporins, and Ertapenem. The result of this is that it greatly limits the number of antibiotics that are available to fight these infections.
Also on the list of highly resistant gram-negatives are Acinetobacter and the tongue twister Stenotrophomonas. Both of which are usually just seen in either ventilator-associated or hospital-acquired pneumonia. All four of these share some common risk factors: prolonged or recent hospitalization, recent antibiotic use, mechanical ventilation, residence in a long-term care facility, and indwelling catheters, lines, and tubes. In addition, each bacteria type has been found to have some additional specific risk factors as well.
Species-specific risk factors for Pseudomonas include immunodeficiency, cystic fibrosis, bronchiectasis, and severe cutaneous burns. Risk factors for ESBL are emergency abdominal surgery. Acinetobacter shows significant seasonal and geographic variation, being most common in summer and in warm and humid climates. Exposure to war or natural disasters has also been found to be a risk factor. And risk factors for Stenotrophomonas include immunodeficiency and cystic fibrosis.
Antibiotics to consider when treating one of these highly resistant gram-negative infections include the third-generation cephalosporin Cefepime, Ceftazidime, Piperacillin and Tazobactam, carbapenems excluding Ertapenem, Aztreonam, aminoglycosides, and Colistin.
Let's take a look at these in some more detail. Cefepime inhibits cell wall synthesis. It is great against Pseudomonas with some activity against Acinetobacter and unreliable activity against Stenotrophomonas. Notable adverse reactions include the possibility that it lowers seizure threshold. It's the only third-generation cephalosporin with anti-pseudomonal activity. Ceftazidime, a fourth-generation cephalosporin, also inhibits cell synthesis. It's active against Pseudomonas and a little against Acinetobacter. Ceftazidime also lowers seizure threshold and interestingly can cause a positive test without triggering hemolysis.
Pip-Tazo, or Zosyn, has excellent Pseudomonas coverage, has Acinetobacter and Stenotrophomonas coverage, and occasionally, but not usually, covers ESBL. Remember that the anti-pseudomonal dosing of Zosyn is higher than regular dosing. The carbapenems, specifically Imipenem and Meropenem, have excellent Pseudomonas and ESBL coverage and some Acinetobacter. The activity of Ertapenem against Pseudomonas and ESBL is less robust and should be avoided when these pathogens are suspected. Aztreonam only really covers Pseudomonas among these four, although it's usually reserved for patients with a severe beta-lactam allergy, as there is generally no cross-reactivity. It should not be used in patients who have had a severe allergic reaction to Ceftazidime, as the two share a side chain and thus a theoretical risk of cross-reactivity. As Aztreonam has no activity against gram-positives or Acinetobacter, the aminoglycosides, which inhibit the 30S ribosomal subunit, have great activity against Pseudomonas with some activity against Acinetobacter. The aminoglycosides have significant adverse effects, most prominently nephro- and ototoxicity, which necessitates close monitoring of serum drug levels. The aminoglycosides also have poor lung penetration, though Tobramycin specifically is available in an aerosolized form that is useful in treating Acinetobacter infections in patients with cystic fibrosis.
Colistin, also known as Polymyxin E, is an interesting antibiotic. It works by damaging the cell membrane and shows at least some activity against all four of these difficult-to-treat pathogens. Despite most clinicians assuming it's a new antibiotic developed to treat highly resistant strains of Pseudomonas, it actually was discovered in 1947 and used significantly in the 1960s and 70s, at which point safety concerns led physicians to abandon its use in favor of newer options. However, the evolution of antibiotic resistance has demanded a return of attention to it. And in retrospect, concerns over Colistin toxicity may no longer be quite as great. We understand more about its pharmacokinetics, understand appropriate dosing, are using different and more purified forms of the drug, and can monitor more closely for side effects. Today, toxicity of Colistin may actually be less than that of aminoglycosides, the use of which initially supplanted Colistin because they were felt to be safer. For now, however, Colistin remains the agent of last choice for the most highly resistant gram-negative infections.
There are a few more gram-negative antibiotic pearls of which you should be familiar. As you may recall from lecture three on the classification of antibiotics, Ciprofloxacin and Levofloxacin show generally good activity against Pseudomonas. Also, the combination of Amikacin and Sulbactam, which is marketed under the name Unasyn in the US, usually has good activity against Acinetobacter and may be a reasonable, more narrow-spectrum alternative to the above. Finally, Trimethoprim-Sulfamethoxazole has excellent coverage of Stenotrophomonas and is generally the antibiotic of choice when no contraindications exist, such as renal failure.
I'm going to end this lecture with a discussion of a common question that arises during treatment of a possible Pseudomonas infection: Should an empiric antibiotic regimen chosen for possible Pseudomonas infection contain not one, but two antibiotics active against this bacteria? While I can't speak about the rest of the world, this is fairly common practice in the United States, most prominent when treating a suspected Pseudomonas pneumonia. The exact combination used is usually an anti-pseudomonal beta-lactam plus either an anti-pseudomonal quinolone or an aminoglycoside. The theoretical benefits of this approach include an increased chance that the suspected pathogen is covered by at least one antibiotic prior to the micro lab's reporting of sensitivities, prevention of resistance emerging during therapy, and exploitation of synergy that may be observed in vitro when two agents are used simultaneously.
Despite these theoretical benefits, the clinical data supporting this approach is relatively weak and conflicting. A 2003 Cochrane review concluded that there were worse outcomes when an aminoglycoside is added to a beta-lactam for empiric treatment in cancer patients with neutropenic fever. A 2006 Cochrane review concluded there was no benefit to adding an aminoglycoside to a beta-lactam for empiric treatment in sepsis, but did find increased nephrotoxicity with double coverage. And a 2008 meta-analysis of 41 randomized control trials evaluating empiric antibiotic regimens for suspected ventilator-associated pneumonia found that overall monotherapy is not inferior to combination therapy in the empiric treatment of these infections. However, the methodologic quality of the trials included were noted by the authors to be low. And other experts in the field have criticized the meta-analysis itself for reaching an overly broad conclusion when the included trials generally did not specifically include the patients who were at highest risk of having multi-drug resistant organisms.
In addition to the evidence, the most recent joint guidelines from the American Thoracic Society and Infectious Disease Society of America recommended the following regimen for management of hospital-acquired, ventilated-associated, and healthcare-associated pneumonias: an anti-pseudomonal beta-lactam plus either an anti-pseudomonal quinolone or an aminoglycoside, plus either Vancomycin or Linezolid if MRSA risk factors are present. Of course, that last qualification is kind of silly, since whatever features of a patient's history place them into one of these pneumonia subtypes to begin with would almost always also be a risk factor for MRSA. It's also important to note that the level of evidence behind this recommended regimen was designated by the authors as Level 3, which they defined as evidence based on case studies, expert opinions, and antibiotic susceptibility data without clinical observations. In other words, it was poor quality evidence.
So, overall, what's the verdict? Should we empirically double cover for Pseudomonas? I think we don't actually know. Since I work at an academic institution where I am encouraged to provide some degree of autonomy to the medical residents, I usually go along with whatever they propose when it comes to this issue. However, if I were working alone, my a general approach would be to double cover only for those patients who have been in the hospital for a week or longer, who are chronically ventilator dependent, who had previously been infected with a highly resistant strain of Pseudomonas, or whom had cystic fibrosis. I'm not saying that this is the correct approach per se, because I don't think we actually know, as I just said. Uh, it's just what I would do in the absence of more definitive data.
That concludes this lecture on antibiotics for gram-negative infections. The next lecture will be a similar discussion on antibiotics for anaerobic infections.
[Music]
[Music]