INTRODUCTION — Streptococcus pneumoniae (pneumococcus), the most common cause of bacterial respiratory tract infections in children and adults, was susceptible to virtually all antibiotics used in treating such infections until outbreaks of infection due to antibiotic-resistant pneumococci were recognized in South Africa in the late 1970s [1,2]. Although the responsible organisms were called penicillin-resistant pneumococci (PRP), they had acquired genetic material that encoded resistance both to penicillin and to other commonly used antibiotics.
In the ensuing decades, resistance of pneumococci to several clinically relevant classes of antibiotics has evolved from an ominous medical curiosity to a worldwide health problem.
Macrolides, azalides, and lincosamides are related drugs that inhibit protein synthesis at the same site in the bacterial ribosome and are generally active against the same microorganisms. Macrolides and azalides are (or, at least, used to be) generally active against S. pneumoniae, Haemophilus influenzae, Moraxella catarrhalis, Legionella species, Chlamydia pneumoniae, and Mycoplasma pneumoniae. Clindamycin, the only lincosamide currently in use, is effective against most pneumococci but is not active against such pathogens as H. influenzae and M. catarrhalis.
The mechanisms of action and resistance of pneumococci to macrolides, azalides, and lincosamides, as well as clinical data on the outcome of therapy with these drugs for otitis, sinusitis, acute exacerbations of chronic bronchitis, pneumonia, and meningitis, will be reviewed here. Resistance to the other classes of drugs is discussed separately. (See "Resistance of Streptococcus pneumoniae to beta-lactam antibiotics" and "Resistance of Streptococcus pneumoniae to the fluoroquinolones, doxycycline, and trimethoprim-sulfamethoxazole".)
MACROLIDES AND AZALIDES — The first of the macrolides, erythromycin, is poorly tolerated. After oral administration, erythromycin causes gastrointestinal distress and is not reliably absorbed; it also frequently causes thrombophlebitis after intravenous administration. As a result, erythromycin has been largely replaced by clarithromycin, a newer macrolide, and azithromycin, an azalide. It is important to note that most pneumococci that are resistant to erythromycin are also resistant to the newer macrolides and the azalides.
Mechanisms of action and resistance — Macrolides and azalides insert into a pocket of the 23S subunit of the 50S ribosome specifically by attaching at domain V of the peptidyl transferase loop, thereby blocking protein assembly. In doing so, these drugs exert a bactericidal effect on S. pneumoniae [1,2].
Acquisition of genetic material, designated ermB or mefA, encodes for resistance to macrolides and azalides [3,4]:
●ermB encodes methylation of a base in domain V of the 23S ribosomal ribonucleic acid (rRNA) (A2058) that alters the site of attachment such that the macrolide no longer recognizes and binds to the ribosome. This lack of binding creates high-level resistance (>64 mcg/mL); thus, increasing the dose with a resulting increase in the concentration of the drug has little effect.
●mefA encodes a pump by which the organism expels macrolides. This resistance is at a lower level (usually <16 mcg/mL); high antibiotic concentrations might be expected to overcome the pump, forcing enough antibiotic into the bacterium to exert an antibacterial effect. Clarithromycin and azithromycin are more active against pneumococci than is erythromycin, and it has been thought that they might be effective in vivo against mefA-containing strains. However, the level of resistance in these strains has been rising [5], and it seems increasingly likely that the resistance observed in vitro will be clinically meaningful.
Some resistant isolates carry both ermB and mefA. In a small percentage of isolates, other mutations induce resistance, for example, by causing other base substitutions in domain V or by altering protein sequences within or adjacent to the macrolide binding site, especially involving ribosomal proteins L4 and L22 [6,7].
Prevalence of resistance — In 1998, only 18 percent of pneumococcal isolates were resistant to macrolides [8]. Between 1998 and 2011, there was a steady increase in the rate of resistance such that, in 2011, 25 to 45 percent of pneumococci in the United States were resistant to macrolides. By 2019, resistance rates were on average about 40 percent in the United States [9-12].
For reasons that may be related to cultural differences within the health care professions relating to antibiotic use, there is substantial regional variation [13,14]. Resistance is attributable to mefA and ermB in approximately equal numbers of isolates, with about 25 percent of them carrying both elements [14].
Although the introduction of the pneumococcal conjugate vaccine into the vaccine schedule for children in the United States in 2000 was anticipated to reduce resistance by leading to the elimination of resistant nasal carriage strains, this appears not to have been the case, since "replacement" strains that have emerged also have a high rate of resistance [15].
A higher proportion of pneumococci in Europe are macrolide resistant and, in the majority of phenotypically resistant isolates, ermB is responsible [16]. Rates of resistance are lower in Canada than in the United States and higher in the Far East than in Europe [17-19]. The rates of erythromycin resistance in Asia were 55 percent overall, 92 percent in Vietnam, and 70 percent in Japan [18,19]. In the study from Japan, half of the isolates were highly resistant (minimum inhibitory concentration [MIC] ≥16 mcg/mL), and resistance was most frequently identified among children younger than two years of age [19]. Similarly, high rates of resistance are now reported in Eastern Europe as well [20].
A prospective cohort study of several thousand patients with invasive pneumococcal infection identified apparent risk factors for the acquisition of macrolide-resistant pneumococcal strains [21]. These include previous use of azithromycin (odds ratio [OR] 9.9), clarithromycin (OR 3.9), penicillin (OR 1.8), and trimethoprim-sulfamethoxazole (OR 2.1).
In another study, following mass azithromycin administration (a single dose of 20 mg/kg every three months for one year) for trachoma in children in Ethiopia, the diversity of pneumococcal sequence types detected by multilocus sequence typing of nasopharyngeal swab specimens decreased significantly, and resistant clones present before mass azithromycin administration increased in frequency (from 5 percent before treatment to 15 percent after treatment) [22]. Higher rates of pneumococcal resistance to macrolides were also found following mass administration of azithromycin in Niger [23,24]. These studies support the hypothesis that antibiotic selection pressure results in clonal expansion of existing resistant strains among pneumococci colonizing the nasopharynx.
Resistance and the outcome of therapy
Otitis media — Careful studies of otitis media have established a close relationship between the susceptibility of the infecting pneumococcus and the response to macrolides [25-28]. One review summarized an extensive clinical experience by stating that azithromycin cured nearly 95 percent of cases of otitis media when the organisms were susceptible to the drug but only 20 percent when they were not [27]. In one study, for example, azithromycin (10 mg/kg on the first day and 5 mg/kg for four additional days) cured 23 of 25 children who had pneumococcal otitis media due to a fully susceptible organism (MIC <0.5 mcg/mL) compared with three of eight children whose infecting organism was only susceptible to >2 mcg/mL [25]. A systematic review and meta-analysis of randomized controlled studies that compared azithromycin with amoxicillin-clavulanate found no difference in the outcome of treatment with either drug regimen [29]. Studies of treatment for otitis media are highly dependent upon (1) the methods used to make the clinical diagnosis, (2) the methods used to establish the bacteriologic diagnosis, and (3) the methods used to define cure, and response to a placebo has been documented in careful prospective studies. As the proportion of cases of otitis due to pneumococci declines and the proportion due to Haemophilus increases, one might expect better outcomes with macrolide therapy, but empiric treatment with a macrolide for otitis media is not recommended, although it could be considered in patients who cannot take beta-lactam antibiotics.
Sinusitis — In some studies, azithromycin (500 mg daily for 3 to 5 days) or clarithromycin (500 mg twice daily for 10 to 14 days) effectively treated patients with acute bacterial sinusitis [30,31]. A study from Croatia where, at the time, 25 percent of pneumococci were erythromycin resistant [32] showed a 95 percent cure rate by azithromycin [33]. These results might support the notion of a discrepancy between in vitro and in vivo resistance to this class of drugs, but a limitation of such studies is that most cases of sinusitis eventually resolve spontaneously, many cases of sinusitis are of viral etiology, and the outcome of any study is determined by the severity of disease in the patients who are included. Macrolides are not recommended for empiric therapy of bacterial sinusitis. This is discussed in detail separately. (See "Uncomplicated acute sinusitis and rhinosinusitis in adults: Treatment".)
Pneumonia — Most of the literature on macrolide resistance in pneumonia deals with groups of patients who are lumped together under the diagnosis of "community-acquired pneumonia" (CAP). CAP is a heterogenous syndrome that can be caused by a variety of pathogens in addition to pneumococcus and, for a substantial proportion of cases, the causative pathogen is not known. Treatment regimens frequently include a macrolide plus a second antibiotic. When the use of macrolide monotherapy has been reported, the absence of an identified etiologic agent renders it impossible to interpret the results [28]. In evaluating reports of responses to therapy, it is essential to note the years during which patients were studied and the prevailing rate and level of resistance at those times, which is clearly a "moving target" in recent years. Thus, determining the relationship among drug levels, inhibitory concentrations, and clinical outcomes is very difficult.
While the Infectious Diseases Society of America and American Thoracic Society guidelines for the treatment of CAP conditionally recommend either azithromycin or clarithromycin for CAP treatment in selected otherwise healthy outpatients in areas where <25 percent of pneumococcal isolates are macrolide resistant (ie, MIC ≥16 mcg/mL), such areas are uncommon, and clinicians typically do not know local resistance rates. Therefore, we do not routinely recommend empiric macrolide monotherapy. S. pneumoniae is the most potentially dangerous pathogen and the clinician should not miss this pathogen when selecting an empiric regimen. (See "Treatment of community-acquired pneumonia in adults in the outpatient setting", section on 'Empiric antibiotic treatment'.)
Macrolide monotherapy is not recommended for the treatment of hospitalized patients with CAP, but macrolides and azalides are commonly used in the United States as part of a combination regimen. (See "Treatment of community-acquired pneumonia in adults who require hospitalization".)
LINCOSAMIDES — Clindamycin is the only lincosamide in current use.
Mechanisms of action and resistance — Although structurally different from the macrolides, clindamycin acts at the same site in the ribosome, and its activity is blocked if the site is methylated; thus, pneumococci that have the ermB mutation are also resistant to clindamycin. However, this drug is not excluded from bacterial cells by the efflux pump, and mefA does not convey resistance.
Prevalence of resistance — Approximately 5 to 10 percent of pneumococci in the United States are resistant to clindamycin with regional variation, as noted above [34].
Treatment and response — Based upon relatively limited clinical data, clindamycin provides effective therapy for pneumococcal pneumonia due to susceptible isolates [35], but it is not recommended for pneumonia of uncertain etiology because of the lack of efficacy against other common pathogens such as H. influenzae, M. catarrhalis, Legionella spp, and M. pneumoniae.
SUMMARY
●Rising rates of macrolide resistance – The macrolides were important agents for treating otitis media, sinusitis, and outpatients with community-acquired pneumonia (CAP) in the past, in large measure because of their excellent activity against pneumococci. However, increases in the proportion of pneumococci that are resistant to macrolides have reduced the utility of macrolides for these conditions. (See 'Macrolides and azalides' above.)
●Efficacy of clindamycin – Clindamycin is effective against pneumococci that have mefA resistance but not against those that have ermB; the lack of activity against other respiratory pathogens (eg, Haemophilus influenzae) limits the utility of this drug in otitis media, acute bacterial sinusitis, or CAP unless pneumococci are known to be responsible. (See 'Lincosamides' above.)
Do you want to add Medilib to your home screen?