0025-7125/06/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.mcna.2006.07.007 medical.theclinics.com strain of S pneumoniae, and the antibiotic achieves therapeutic concentrations in serum/tissue in excess of the intermediate breakpoint, then such organisms should be considered "sensitive" rather than "resistant." Therefore, in organisms that are in the resistant range, resistance may be viewed as being moderate or high. Moderately resistant strains have MICs that are achievable with the commonly used doses of antibiotics against the organism. Highly resistant strains may require antibiotics from a different class [7-10].
Resistance is not a generalized phenomenon but is limited to relatively few organisms. For example, among streptococci, resistance to penicillin is of concern with S pneumoniae but not with others (ie, group A, group B, group C, nonenterococcal group D, and group G streptococci). Acquired resistance may be the result of widespread dissemination of a resistant clone that has occurred because of point mutation. In surveys of antibiotic resistance, the practitioner should be careful to differentiate between an increase in acquired resistance secondary to antibiotic use and an increase in apparent antibiotic resistance secondary to the dissemination of a resistant clone [3,7,8,10].
It is a popular misconception that antibiotic resistance is related to the volume or duration of antimicrobial use and that acquired resistance is inevitable. Although resistance has increased over the past several decades, the problem of acquired antibiotic resistance is not generally related to antibiotic overuse in terms of volume or antibiotic tonnage—that is, there has been no appreciable increase in S pneumoniae resistance to doxycycline after decades of extensive worldwide use [11-13]. Doxycycline also illustrates another point. Despite concern about rare strains of S pneumoniae highly resistant to doxycycline, doxycycline retains its activity against the other common respiratory pathogens, namely Hemophilus influenzae and Morax-ella catarrhalis, as well as the atypical pulmonary pathogens causing respiratory tract infections [14,15]. Doxycycline has retained its usefulness against the zoonotic pathogens, which have also been implicated in bioterrorist attacks. It retains its activity against all these pathogens without the development of appreciable resistance [16,17].
The antibiotic resistance that occurs with most organisms and antibiotics is acquired resistance. Some antibiotics, such as macrolides, encounter both intrinsic and acquired resistance when used against certain organisms (ie, S pneumoniae) [9,11,12]. Approximately 30% of strains of S pneumoniae are naturally or intrinsically resistant to macrolides. Intrinsic resistance is not a function of volume of use, but acquired macrolide resistance is a function of volume of use. Therefore, extensive macrolide use results in resistance that is additive. That is, the combined effect of a background of natural resistance superimposed on the component of acquired resistance related to use results in very high levels of macrolide-resistant S pneumoniae (MRSP) (ie, approximately 40%) [18-23].
For reasons that are unclear, resistance is largely limited to certain members of various antibiotic classes and is usually relegated to one or two organisms. For example, trimethoprim sulfamethoxazole (TMP-SMX) use is associated with an increase in penicillin-resistant S pneumoniae (PRSP) as well as in MRSP. However, the activity of TMP-SMX against strains of methicillin-sensitive S aureus (MSSA) remains good. With conventional tetracycline, there has been an increase in resistance in MSSA, but other members of the class (ie, doxycycline and minocycline) maintain their anti-MSSA activity [12,13]. At the present time, doxycycline is even active against community-acquired MRSA (CA-MRSA), and minocycline is highly active against strains of hospital-acquired MRSA (HA-MRSA) .
In vitro antibiotic susceptibility and in vivo effectiveness
For the clinician, there are interpretation problems with antimicrobial susceptibility reports. With certain antibiotic organism combinations, a discrepancy may exist between in vitro and in vivo activity. Certain antibiotics may be reported as "sensitive" in vitro to certain organisms, but the drugs are, in fact, ineffective in vivo in clinical experience. All streptococci are reported as being sensitive in vitro to aminoglycosides, but in fact they are all intrinsically resistant to aminoglycosides when used alone. When an aminoglycoside is combined with penicillin, for example, the combination has antienterococcal activity against most strains of E faeca-lis, namely vancomycin-sensitive enterococci (VSE). Discrepancy between in vivo and in vitro sensitivity testing is also a practical problem in treating HA-MRSA. Strains of HA-MRSA are often reported as being "sensitive'' in vitro to a wide variety of antibiotics, which usually does not correlate with in vivo effectiveness. In treating MRSA infections, it is prudent to disregard susceptibility testing and select a drug for empiric therapy with proven clinical effectiveness against HA-MRSA (ie, vancomycin, quinupristin-dalfopristin, minocycline, linezolid, daptomycin, tigecycline) [20,23,24].
Interestingly, strains of CA-MRSA resemble MSSA in susceptibility testing as well as in their clinical expression. CA-MRSA strains without the Panton-Valentine leukocidin gene cause the same spectrum of disease as does MSSA. Strains of CA-MSSA (SCCmec IV) are sensitive to antibiotics that are not effective in treating HA-MRSA (ie, clindamycin, TMP-SMX, doxycycline). Clinicians should know when to rely on in vitro susceptibility testing to guide therapy and when to rely on clinical experience [20,25].
Prolonged or extensive use of an antibiotic does not, per se, result in an increase in resistance. The example of doxycycline has been given in regard to respiratory pathogens, but the same may be said of first-, second-, and third-generation cephalosporins in regard to S pneumoniae. With prerespir-atory quinolones, S pneumoniae resistance was a problem. But, after decades of intensive worldwide use of ceftriaxone, there has been no increase in resistance in S pneumoniae [11-13]. Among enterococci, ampicillin remains highly active. The same is true for vancomycin, considering its volume of use. The increase in VRE has resulted in but was not caused by widespread vancomycin usage. Enterococci are part of the normal fecal flora. Approximately 95% of fecal enterococci are of the E faecalis variety. Because virtually all strains of E faecalis are sensitive to vancomycin, E faecalis is essentially synonymous with VSE. When vancomycin is given, the VSE component of the fecal flora is diminished. The naturally or intrinsically resistant E faecium strains (ie, VRE) then become the predominant part of the fecal flora. As a result, VRE have become more prevalent but not more resistant per se [5,6]. This is a good example of how increased antibiotic use may have an indirect effect on antibiotic resistance, not by increasing acquired resistance but by permitting the emergence of an uncommon organism in the fecal flora, which is given a selective advantage as vancomycin increases its numbers. For this and other reasons, VRE has emerged as a highly resistant gram-positive pathogen without becoming more resistant to vancomycin [26-29].
Because antibiotic resistance is complex, clinicians should become familiar with the nuances of resistance terminology and the factors that affect resistance. Clinicians should also be familiar with differentiation between colonization and infection, because colonization is ordinarily not treated, whereas infection is. Furthermore, clinicians should be familiar with the differences between vitro sensitivity testing and in vivo effectiveness, so that effective empiric therapy may be selected when treatment of resistant organisms is necessary [9,11-13,20,30].
Because S pneumoniae is the most important bacterial pathogen in upper and lower respiratory tract infections, there is concern about increased antibiotic resistance in S pneumoniae. ''Penicillin resistance," particularly of the intermediate variety, has limited clinical significance . As mentioned earlier, when b-lactams are given in the usual dose appropriate for the anatomic location of the respiratory tract infection due to S pneumoniae, these strains are, in fact, responsive and clinically susceptible to b-lactam antibiotics. Overuse of b-lactam antibiotics per se has not resulted in an increase in S pneumoniae resistance [9,11-13]. An increase in the MICs (ie, ''MIC drift") has occurred with some b-lactams, but the achievable blood/tissue levels of b-lactams in non-central nervous system infections are sufficiently high that these strains are easily eliminated [7,8,10].
Because b-lactam antibiotics obey time-dependent kinetics, any concentration in excess of the MIC will eliminate the organism if the concentration is maintained for about three quarters of the dosing interval. Pharmacokinetic and pharmacodynamic information readily explains this phenomenon. The breakpoints for penicillin for S pneumoniae are as follows: less than 1 mg/mL is sensitive, between 1-2 mg/mL is intermediate, and 2 mg/mL or greater is reported. With intermediate or even resistant strains, with MICs of 6-10 mg/ mL, empiric therapy with ceftriaxone, for example, provides serum levels far in excess of 6 mg/mL (ie, after a 1-g intravenous dose, achievable serum levels are approximately 150 mg/mL). Even when one uses oral antimicrobial therapy for infections except those of the central nervous system (CNS), concentrations in sinus fluid, middle ear fluid, bronchial fluid, and lung parenchyma are readily achieved with usual full doses of orally administered b-lactams. The only therapeutic concern regarding PRSP involves the treatment of CNS infections. Even in CNS infections, as long as therapeutic concentrations are achieved in the ce-rebrospinal fluid, PRSP meningitis is not a major therapeutic problem [20,31,32].
Acquired resistance to "high-resistance-potential" antibiotics is related to volume of antibiotic use. Overuse of TMP-SMX and macrolides for the treatment of respiratory tract infections has led to problems with PRSP and MRSP worldwide. TMP-SMX predisposes to PRSP and MRSP as well as to multidrug-resistant S pneumoniae (MDRSP). The use of TMP-SMX should be avoided when possible in respiratory tract infections. TMP-SMX does not predispose to resistance in other organisms and continues to be useful in nonpneumococcal infections (ie, aerobic gram-negative bacillary infections, MSSA). Overuse of macrolides in particular has resulted in the emergence of MDRSP because of the macrolide-induced PRSP and MRSP. Like TMP-SMX, macrolides should be avoided if possible in the treatment of respiratory tract infections [9-12,18-23]. Other antibiotics with the appropriate spectrum and a ''low resistance potential" should be used preferentially. "Macrolide-sparing" antibiotics include doxy-cycline, telithromycin, and respiratory quinolones [11,13,20]. Telithromycin and the respiratory quinolones are effective against PRSP, MRSP, and MDRSP [33-39]. From an antibiotic resistance perspective, it is preferable to use antibiotics with a ''low resistance potential,'' such as doxycycline, te-lithromycin, and respiratory quinolones, to treat respiratory tract infections. Using these agents preferentially both treats and prevents further pneumococcal resistance (Table 1) [7-9,11-13].
VSE and VRE are part of the normal fecal flora. The relative proportions of VSE and VRE may be modified by antimicrobial therapy or prolonged hospitalization. As part of the normal fecal flora, enterococci commonly
Clinical comparison of Streptococcus pneumoniae antibiotics
Usual adult dose Side effects
PRSP/MRSP Amoxicillin 1 g (PO) q 8 h Drug fever/rash
Doxycycline 100-200 mg Nausea/vomiting
Ceftriaxone 1 g (IV) q 24 h Non-C difficile diarrhea
MDRSP Levofloxacin 750 mg Nausea/vomiting/diarrhea
Moxifloxacin 400 mg (IV/PO) q 24 h
Nausea/vomiting/diarrhea T QTc
Excellent bioavailability (90%) Low resistance potential Excellent bioavailability (93%) IV/PO formulation Active against typical and atypical respiratory pathogens Low resistance potential After 1 g (IV) serum concentrations ~ 150 |xg/mL exceed MICs of even highly resistant strains (MICs 6-10 ng/mL) May be given IM (with lidocaine) Low resistance potential IV/PO formulation Excellent bioavailability (99%) Active against typical and atypical respiratory pathogens Low resistance potential IV/PO formulation Excellent bioavailability (90%) Active against typical and atypical respiratory pathogens
No IV formulation Inactive against atypical respiratory pathogens None
No PO formulation Inactive against atypical respiratory pathogens
Gatifloxacin 400 mg (IV/PO) q 24 h
Gemifloxacin 320 mg (PO) q 24 h
Telithromycin 800 mg (PO) q 24 h
Tigecycline 100 mg (IV) x 1 dose, then 50 mg (IV) q 24 h Ertapenem
Nausea/vomiting/diarrhea Î QTc
Nausea/vomiting/diarrhea Î QTc
Nausea/vomiting/diarrhea Î QTc
Nausea/vomiting/diarrhea Î QTc
Low resistance potential IV/PO formulation Excellent bioavailability (96%) Active against typical and atypical respiratory pathogens Low resistance potential Good bioavailability (71%) Active against typical and atypical respiratory pathogens Low resistance potential Good bioavailability (57%) Active against typical and atypical respiratory pathogens Low resistance potential
Low resistance potential Once daily dosing May be given IM (with lidocaine) Low resistance potential
Avoid in diabetics
No IV formulation
No IV formulation
No PO formulation
No PO formulation Inactive against atypical pathogens No PO formulation
Abbreviations: IM, intramuscular; IV, intravenous; PO, oral.
Adapted from Cunha BA. Antibiotic essentials. 5th edition. Royal Oak (MI): Physicians' Press; 2006.
colonize the perirectal area as well as the skin of the lower extremities. En-terococci are not highly invasive organisms, and colonization represents the majority of strains isolated in the hospital setting. In general, infections, not colonization, should be treated with antimicrobial therapy. Because of the proximity to the rectum, enterococci commonly colonize wounds of the abdomen, lower extremities, and urine. Enterococci may occasionally contaminate the blood culture contaminants in blood samples drawn from the antecubital fossa. In patients whose skin below the waist is colonized with enterococci and who are turning in bed, it is easy to understand how the antecubital fossa could be colonized with enterococci and contaminate blood cultures during venipuncture [5,6].
Enterococci should be viewed as permissive pathogens, capable of causing infection in selected situations. Enterococci are common single pathogens in biliary tract and urinary tract infections. They are uncommon causes of intravenous-line infection. Enterococci also cause bacterial endocarditis. They are the most common organisms associated with bacterial endocarditis with the focus of infection below the waist (ie, the gastrointestinal or the genitourinary tract). Enterococci are not important pulmonary or neuropathogens. They may be pathogens in complicated skin and soft tissue infections below the waist. Enterococci alone are incapable of intra-abdom-inal infection. In intra-abdominal infections (ie, between the urinary bladder and the gallbladder), enterococci are permissive pathogens, that is, they cause infection only in concert with another pathogen [5,6,40].
The spectrum of enterococcal infections is the same for VSE and VRE. The only difference between VSE and VRE is related to the therapeutic approach . Virtually all strains of VSE remain susceptible to ampicillin and antipseudomonal penicillins, that is, ticarcillin, azlocillin, mezlocillin, piper-acillin, and vancomycin. An example of the difference between in vitro susceptibility and in vivo activity is that of penicillin with regard to enterococci . Enterococci are usually reported as being sensitive to penicillin, but penicillin monotherapy is ineffective in treating enterococcal infections. Penicillin combined with an aminoglycoside, such as gentamicin, is active against VSE because of synergy. Clinicians should be aware that, with the exception of the third-generation cephalosporins, cefoperazone, cephalosporins have no anti-VSE activity. Although the MIC90 of cefoperazone to VSE is 32 mg/mL and is considerably higher than with ampicillin, serum concentrations of less than 32 mg/mL are readily achieved after 2 g of cefoperazone (intravenous) with resultant serum levels of approximately 240 mg/mL . Quinupristin/dalfopristin is active against VRE but inactive against E faeca-lis, that is, VSE [42-47]. Quinolones have variable anti-VSE activity and should be used in VSE infections with a demonstrated susceptibility to qui-nolones [5,6].
Strains of VRE are by definition vancomycin resistant and are also resistant to the usual antibiotics that are active against VSE. Fortunately, in vitro susceptibility testing is reliable in VRE, by contrast with VSE or MRSA, and susceptibility results may be used to guide antibiotic selection for VRE infections [22-24]. As with VSE, VRE infection, not colonization, should be treated with antibiotics. Antibiotics with proved effectiveness against VRE include quinupristin/dalfopristin, doxycycline, chloramphenicol, linezolid, daptomycin, and tigecycline [5,6,20,40,41,48-63]. Anti-VRE drugs are also effective against VSE with the exception of quinupristin/dalfopristin, which is only effective against VRE. Nitrofurantoin is the preferred oral antibiotic for treating VRE cystitis or catheter-associated bacteriuria . Oral agents available to treat serious systemic VRE infections include doxycycline, chloramphenicol, and linezolid. High-dose daptomycin should be used for VRE infections (ie, 12 mg/kg intravenously [IV] every 24 hours [with normal renal function]), because the MIC90 of VRE is more than double the MIC90 for MRSA (ie, 6 mg/kg [IV] every 24 hours) [61-63].
Staphylococci are part of the normal skin flora. The nares is the primary site of colonization for staphylococci (versus the feces for VSE and VRE). Because staphylococcal colonization of the skin is so common, MSSA/ MRSA as well as Coagulase Negative Staphylococci (CoNS) are common blood culture contaminants: the skin organisms are commonly introduced during venipuncture. Colonization is common in respiratory secretions of in-tubated patients, on broad-spectrum antimicrobial therapy, in nonpurulent surgical wounds, draining body fluids (chest tube or abdominal drainage), and urine in catheterized or instrumented patients. As with enterococci, staphylococcal colonization should not be treated; treatment should be reserved for infection. The spectrum of infection with MSSA resembles that of MRSA with some important differences. Staphylococci are the most common cause of suppurative wound infections, are important causes of IV-line infection, and are now the most common cause of acute bacterial endocarditis (ABE) (Table 2) [20,64].
Antibiotics traditionally effective against MSSA include antistaphylococ-cal penicillins, such as oxacillin, nafcillin, first-, second-, and third-generation cephalosporins (excluding ceftazidime), TMP-SMX, clindamycin, doxycycline, gentamicin, vancomycin, piperacillin/tazobactam, and cefe-pime. The preferred antibiotics to use for MSSA infections include antista-phylococcal penicillins, antistaphylococcal cephalosporins, and cefepime . The preferred oral agents commonly used to treat MSSA infections include TMP-SMX, clindamycin, and oral first-generation cephalosporins. Oral antipseudomonal penicillins are erratically absorbed and are frequently associated with gastrointestinal upset. If a b-lactam is selected, first-generation cephalosporins are preferred as oral anti-MSSA therapy .
Susceptibility testing with MRSA presents interpretation difficulties because there is often a discrepancy between in vitro susceptibility and in
Clinical comparison of enterococcal (vancomycin-resistant enterococci) and staphylococcal (methicillin-sensitive Staphylococcus aureus/methicillin-resistant Staphylococcus aureus) antibiotics
Antistaphylococcal agent Usual adult dosea Side effects Advantages Disadvantages
Complicated skin/soft tissue infections 4 mg/kg (IV) q 24 h
Bacteremias 6 mg/kg (IV) q 24 h (q 24 h dosing CrCl O 30 mL/min; q 48 h dosing CrCl <30 mL/min) 600 mg (IV/PO) q 12 h
Painful myalgias (rare)
Skin discoloration with prolonged use
Most effective/rapidly bactericidal MSSA/MRSA antibiotic No [ VRE prevalence Also useful for VISA/VRSA Once daily dosing
PO formulation (excellent bioavailability) Excellent CNS penetrations (70%)
Also useful for VISA/VRSA Proven effectiveness
PO formulation (excellent bioavailability) Excellent CNS penetration
No PO formulation Potential cytochrome P450 interactions
Potential cytochrome P450
interactions Serotonin syndrome
No PO formulation Q 8 h dosing Ineffective against VSE Poor CNS penetration (< 10%) Limited VRE experience
Neutropenia "Red Man'' syndrome
Proven effectiveness No nephrotoxicity (vancomycin levels unnecessary; dose ~ CrCl) Poor CNS penetration (15%)
No side effects No [ VRE prevalence No potential cytochrome P450 interactions
[ MICs/resistance (VISA/ VRSA)
No PO formulation for MRSA/ VSE
[ VRE prevalence Slow resolution of MSSA/
MRSA infections [ Therapeutic failures Total cost of IV vancomycin more expensive than PO linezolid No PO formulation
Abbreviations: CrCl, creatinine clearance; PO, oral; VISA, vancomycin-intermediate S aureus; VRSA, vancomycin-resistant S aureus. a Normal renal/hepatic function.
Adapted from Cunha BA. Antibiotic essentials. 5th edition. Royal Oak (MI): Physicians' Press; 2006.
vivo effectiveness [23,24]. Many antibiotics appear to be sensitive against strains of HA-MRSA by in vitro susceptibility testing but are inconsistently reliable clinically, and other agents should be used. Antibiotics with demonstrated MRSA activity include vancomycin, quinupristin-dalfopristin, minocycline, linezolid, daptomycin, and tigecycline [64-90]. The only oral agents available to treat MRSA infections are minocycline and linezolid [20,58].
Community-acquired methicillin-resistant Staphylococcus aureus and hospital-acquired methicillin-resistant Staphylococcus aureus
Recently, there has been an increase in reports of CA-MRSA. For decades, MRSA was primarily a hospital organism (ie, HA-MRSA), but over the years, strains also originated from the community that were related genetically to hospital strains. Currently, CA-MRSA implies not only origin in the community but also a different genetic strain of MRSA. CA-MRSA differs from HA-MRSA in its spectrum and antibiotic susceptibility.
CA-MRSA strains are of the SCCmecIV type. CA-MRSA strains may carry the Panton-Valentine leukocidin (PVL) gene. PVL is a potent toxin which causes tissue necrosis. PVL-positive CA-MRSA strains are highly virulent and are the cause of severe pyodermas/necrotizing fasciitis as well as necrotizing community-acquired pneumonia (CAP). PVL-negative CA-MRSA strains are less virulent and resemble clinically the spectrum of infection caused by MSSA. Patients presenting with otherwise unexplained, unusually severe necrotizing fasciitis/pyomyositis should be suspected of having PVL-positive CA-MRSA until proven otherwise. To date, virtually all of the cases of severe/necrotizing MRSA CAP have been due to PVL-positive CA-MRSA following viral influenza. MSSA/MRSA is a rare cause of CAP. MSSA/MRSA CAP is rare even in diabetics commonly colonized with staphylococci. Essentially, all staphylococcal CAP occurs in the postviral influenza setting. Extremely severe necrotic MRSA CAP may occur after viral influenza in patients infected with PVL-positive CA-MRSA strains (Table 3) [91-105].
Strains of HA-MRSA are usually pen-resistant and are susceptible and responsive to only a small number of antibiotics, such as vancomycin, minocycline, quinupristin-dalfopristin, linezolid, daptomycin, and tigecy-cline. Antibiotics with known effectiveness against HA-MRSA are also effective against CA-MRSA. CA-MRSA strains are pauci-resistant to antistaphylococcal antibiotics and are sensitive to some of the agents active against MSSA strains. For this reason, CA-MRSA strains may be treated with clindamycin, doxycycline, or TMP-SMX [20,104,105]. However, serious or fulminant infections due to CA-MRSA PVL-positive strains should be treated with antibiotics known to be effective against HA-MRSA.
Comparative clinical features of community-acquired methicillin-resistant Staphylococcus aureus and hospital-acquired methicillin-resistant Staphylococcus aureus infections
MRSA clinical and laboratory features CA-MRSA
Was this article helpful?