Resistance of Streptococcus pneumoniae to the fluoroquinolones, doxycycline, and trimethoprim-sulfamethoxazole

INTRODUCTION — Streptococcus pneumoniae (pneumococcus) is among the most commonly identified bacterial causes of upper and lower respiratory tract infections including pneumonia, otitis media, acute rhinosinusitis, and acute exacerbations of chronic obstruction pulmonary disease (COPD). However, rates of pneumococcal infections are overall declining, largely due to widespread use of pneumococcal conjugate vaccination in children and adults [1,2].

Penicillin (or amoxicillin because of more reliable absorption after oral administration and a much longer half-life) remains the drug of choice for treating susceptible pneumococcal infection. In this section, alternative drugs will be discussed.

The fluoroquinolones (often called quinolones) have been used widely to treat adults with these conditions. In 2016, the US Food and Drug Administration stated that the serious adverse effects associated with fluoroquinolones generally outweigh the benefits for patients with acute rhinosinusitis (as well as acute bronchitis and uncomplicated urinary tract infections) who have other treatment options [3]. This safety alert is likely to reduce the indiscriminate use of the fluoroquinolones, thereby preserving the susceptibility of pneumococci. (See "Fluoroquinolones", section on 'Benefits and risks of use'.)

Doxycycline is recommended as an alternative agent for empiric treatment of pneumonia in adult outpatients [4], acute exacerbation of COPD [5], and acute bacterial rhinosinusitis [6] and as an alternative agent for patients with penicillin allergy.

Trimethoprim-sulfamethoxazole (TMP-SMX) was commonly used to treat these conditions from the mid-1970s to the mid-1990s, but this combination drug has largely fallen out of favor because of the high rate of pneumococcal resistance.

Pneumococci were uniformly susceptible to all antibiotics used to treat respiratory tract bacterial infections until outbreaks of infection due to antibiotic-resistant pneumococci occurred in South Africa in the late 1970s [7,8]. Although the responsible organisms were called penicillin-resistant pneumococci, they had acquired genetic material that encoded broad resistance both to penicillin and to other commonly used antibiotics. In the ensuing decades, pneumococcal resistance has arisen in a number of clinically relevant classes of antibiotics.

The mechanisms of resistance and the prevalence of resistance to the fluoroquinolones, doxycycline, and TMP-SMX are 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 macrolides, azalides, and lincosamides".)

FLUOROQUINOLONES — The fluoroquinolones consist of a family of related compounds, including ciprofloxacin, gemifloxacin, levofloxacin, moxifloxacin, and ofloxacin. Use of these drugs to treat respiratory infection represented a major therapeutic advance. At the time of their introduction, these drugs were uniformly active against S. pneumoniae as well as Haemophilus, Moraxella, Legionella, Mycoplasma, and Chlamydia spp. Except for very low rates of resistance of pneumococci (not exceeding 1 to 2 percent of isolates in the United States) and rare reports of resistance among Haemophilus, they remain so to the present. (See "Fluoroquinolones".)

The term "respiratory fluoroquinolone" has been applied to levofloxacin, moxifloxacin, and gemifloxacin based on their very low minimum inhibitory concentrations (MICs) for S. pneumoniae and other respiratory pathogens and achievable drug levels. They exhibit increased potency relative to ciprofloxacin and ofloxacin against S. pneumoniae [9-11]. Although there are remarkably few reports of treatment failures with ciprofloxacin in the United States, this drug is not regarded as a "respiratory fluoroquinolone" and, accordingly, is not recommended for treatment of respiratory infections by the Infectious Diseases Society of America or the American Thoracic Society [1].

Mechanisms of action and resistance — The fluoroquinolones bind topoisomerases, enzymes that govern the twisting and knotting of double-stranded deoxyribonucleic acid (DNA). The principal enzymes involved are gyrase and topoisomerase IV, more specifically, two subunits of gyrase, gyrA and gyrB, and two subunits of topoisomerase IV, parC and parE. By binding these sites, fluoroquinolones block enzymatic activity such that bacterial replication cannot take place. The specific site of action of a quinolone is determined by the avidity with which it binds each enzyme [12-14]. For example, ciprofloxacin binds exclusively with topoisomerase IV; levofloxacin binds more avidly with topoisomerase IV and also binds gyrase. Moxifloxacin reacts more strongly with gyrase. Gemifloxacin binds avidly with both.

Fluoroquinolone resistance in S. pneumoniae generally occurs when several mutations in gyrA or parC reduce binding of the drug to the site of activity [15,16]. Occasionally, mutations in parE are responsible. Resistance is stepwise; single mutations generally have little effect, but the more mutations that have occurred, the more resistant an organism is likely to be. In general, isolates with mutations in parC only are resistant to ciprofloxacin but not to the other quinolones, whereas isolates with mutations in gyrA only and isolates with mutations in both parC and gyrA tend to be resistant to all fluoroquinolones [15]. Isolates with more than one mutation are more likely to remain susceptible to moxifloxacin or gemifloxacin than to ciprofloxacin or levofloxacin [16]. Many pneumococcal isolates that are still regarded as susceptible have already acquired one mutation, but the fear that widespread resistance might result has not yet materialized.

Prevalence of resistance — Early surveillance studies showed that no more than 1 percent of S. pneumoniae isolates in the United States were resistant to levofloxacin, moxifloxacin, or gemifloxacin [17]. This also appears to be true for so-called replacement strains that have emerged since the widespread use of the 7-valent pneumococcal conjugate vaccine (PCV7), such as serotypes 19A, 35B, and 11A [18]. Overall rates of resistance to levofloxacin among nasopharyngeal isolates in Buffalo, New York remained stable from 2010 to 2016 but, after an initial decline, increased to ofloxacin during that same time period [19]. The increase is small and is probably of no clinical significance, but the trend was a steady increase in resistance to ofloxacin during the time studied.  

Widespread use of any of the fluoroquinolones increases the risk for emergence of resistance [20]. In a report from Canada, the prevalence increased from 1.5 percent in 1993 to 1994 to 2.9 percent in 1997 to 1998 [21], paralleling the use of fluoroquinolones; as the usage of these drugs has been curbed, the rate of resistance appears to have fallen to <2 percent [22,23]. There may be pockets of higher rates of resistance in settings in which there is excessive use of fluoroquinolones. A study in assisted living facilities, for example, found 6 percent of all colonizing pneumococci to exhibit resistance to fluoroquinolones [20]. A review summarized concepts about the relationships among proper fluoroquinolone use, inappropriate use, and pneumococcal resistance [24].

Higher rates of resistance have been noted in other countries. In Croatia [25] and Hong Kong [26], 4 to 13 percent of pneumococci were reported resistant to fluoroquinolones, and a study from Asia in 2009 and 2010 found up to 4 percent prevalence of resistance to newer fluoroquinolones [27]. Similarly, in Belgium, resistance of noninvasive isolates to levofloxacin increased to about 3 percent and high-level resistance to ciprofloxacin increased to 9 percent in 2013 [28]. Within countries that report increasing incidence of resistance, the proportion of resistant isolates is much higher among older subjects and patients with chronic lung disease [26]. In contrast, in countries where antimicrobial use is conservative, such as Germany, almost no resistance has been reported [21].

A prospective cohort study of 3339 patients with invasive pneumococcal infection between 1995 and 2002 identified apparent risk factors for the acquisition of fluoroquinolone-resistant pneumococcal strains [24,29]. These include:

●Previous fluoroquinolone use (odds ratio [OR] 12.1)

●Current residence in a nursing home (OR 12.9)

●Nosocomial acquisition of infection (OR 9.9)

DOXYCYCLINE AND MINOCYCLINE — The tetracyclines were discovered in the 1940s and, along with chloramphenicol, were the first broad-spectrum antibiotics. For a variety of reasons relating to efficacy against pneumococci, pharmacokinetics, toxicity, cost, and marketing, doxycycline has become the tetracycline of choice. However, in 2015, the cost of doxycycline increased 50-fold, from about 6 cents to more than $3.00 per pill, whereas minocycline remains at about 50 cents per pill. There appear to be no notable differences in the in vitro susceptibility of pneumococci to minocycline or doxycycline, with both drugs being more effective than tetracycline [30]. Resistance of pneumococci to doxycycline had exceeded 20 percent for several decades but, as of 2019, has fallen to 12.5 percent [31], probably reflecting lack of use in both pediatric and adult populations. These drugs are also effective against most isolates of H. influenzae, Moraxella catarrhalis, Legionella, Mycoplasma, and Chlamydia spp, explaining their potential usefulness as empiric therapy in community-acquired pneumonia (CAP) [4]. Glycylcyclines, such as tigecycline, regarded as advanced tetracyclines, have greater in vitro activity against pneumococci [32].

The literature is remarkable for a paucity of either in vitro or in vivo studies of the efficacy of doxycycline or minocycline in treating pneumococcal (or any other) pneumonia; this paucity appears to reflect the economics of the medical marketplace more than it reflects interest in or potential efficacy of the drug. Pharmaceutical manufacturers, who support comparative studies of antibiotics, usually choose to compare experimental compounds with drugs that are currently in vogue in order to position them to compete for market share. Some potentially excellent antibiotics are simply not studied, and it can be very difficult to obtain data on them. This principle applies to drugs such as doxycycline, minocycline, and trimethoprim-sulfamethoxazole (TMP-SMX) and includes studies in vitro or in vivo. Two investigator-initiated randomized controlled studies compared doxycycline with quinolones in CAP and found no difference in outcomes [33,34], but these studies were small, and only a fraction of patients had pneumococcal pneumonia. A meta-analysis supports the use of doxycycline in pneumonia of mild to moderate severity [35]. Omadacycline, a newly developed tetracycline, is effective in vitro against virtually all pneumococci [35] and could be considered for treatment.

Mechanisms of action and resistance — Tetracyclines inhibit protein synthesis by binding to the 30S ribosomal unit, specifically blocking the binding of aminoacyl transfer ribonucleic acid (RNA) to the mRNA-ribosome complex. Although regarded as bacteriostatic (which it is for Staphylococcus aureus), doxycycline is bactericidal for pneumococcus. Resistance is mediated by one or more proteins that bind the drug into vesicles that are then extruded from the bacterium [36].

Prevalence of resistance — At the Boston City Hospital in the late 1960s, pneumococci were uniformly susceptible to doxycycline, with a median minimum inhibitory concentration (MIC) of 0.39 mcg/mL (range 0.04 to 0.78 mcg/mL) [37]. Higher concentrations of tetracycline were required to inhibit pneumococci, but no isolates were resistant. Rare cases of infection due to tetracycline-resistant pneumococci were reported in the United States in the 1960s [38]. By that time, a small but measurable percentage of isolates were resistant in England [39] and Scandinavia [40]. By the mid-1970s, England had an overall rate of resistance of approximately 13 percent, with regional variation ranging from 2 to 32 percent [41].

By 2010, in the United States, approximately 25 percent of pneumococcal isolates were tetracycline resistant [17], with resistance to doxycycline or minocycline at about 20 percent. Tetracycline resistance was substantially greater among penicillin-resistant isolates. Studies from the United States and Canada have shown nearly results identical, with 87.5 percent of pneumococcal isolates being susceptible to doxycycline [23,31]. These findings validate the recommendation of the American Thoracic Society/Infectious Diseases Society of America guidelines for the empiric use of doxycycline (or minocycline, depending upon cost and availability) for empiric treatment of mild to moderate severity CAP. Susceptibility varies greatly in other countries, so a recommendation to use doxycycline cannot be universalized, and the use of amoxicillin or amoxicillin-clavulanate for empiric therapy may still be preferable in many parts of the world.

TRIMETHOPRIM-SULFAMETHOXAZOLE — Trimethoprim-sulfamethoxazole (TMP-SMX) was regarded as excellent therapy specifically for pneumococcal infection as well as more generally for otitis, sinusitis, and acute exacerbation of chronic bronchitis until the mid-1990s when it fell out of favor partially due to the increased prevalence of resistance of S. pneumoniae and partially to the development of alternative drugs, such as the newer macrolides and the fluoroquinolones.

Mechanisms of action and resistance — Trimethoprim and sulfamethoxazole block two separate steps in the synthesis of folic acid, a needed substrate for bacterial survival. Sulfamethoxazole displaces paraaminobenzoic acid, blocking biosynthesis of dihydrofolate [42]. Trimethoprim binds avidly to the active site of the bacterial enzyme that reduces dihydrofolate to tetrahydrofolate [43].

When TMP-SMX was released, it was touted as a drug to which bacteria would never become resistant because the mutation rate to each component is 1 in 10⁷ and a double mutation would only occur in 1 of 10¹⁴ organisms. This prediction proved to be wrong because mutant bacteria simply developed alternative bypass pathways for folate synthesis.

Activity and prevalence of resistance — Resistance of S. pneumoniae to TMP-SMX had clearly been evolving during the 1980s, but it first became prominent in 1986 when a cluster of penicillin-resistant isolates in Brooklyn was also found to be resistant to TMP-SMX [44]. Two years later, 11 percent of adult carriers and 30 percent of children in a daycare center in North Carolina who were carrying pneumococcus harbored TMP-SMX-resistant isolates [45]. Continued spread and selection of resistance was probably enhanced in children by the extensive use of this combination drug to treat otitis media [45] and in adults by the prophylactic use of TMP-SMX in acquired immunodeficiency syndrome (AIDS) patients [46].

Approximately one-third of all pneumococcal strains in the United States are resistant to TMP-SMX [18], ranging from 25 percent in the Northeast to 45 percent in the Southeast [47,48]. The prospective cohort study of 3339 patients with invasive pneumococcal infection cited above identified apparent risk factors for the acquisition of TMP-SMX-resistant pneumococcal strains [49]. These include previous use of TMP-SMX (odds ratio [OR] 4.7), azithromycin (OR 3.5), and penicillin (OR 1.7). In many other countries, rates of resistance are even higher, approaching 80 percent in some southeastern Asian countries.

With the widespread use of the seven-valent pneumococcal conjugate vaccine, rates of resistance initially declined because resistance was highest among vaccine strains. Resistance, however, has increased with the emergence of drug-resistant replacement strains. The most prevalent replacement strain (serotype 19A) shows an 83 percent rate of TMP-SMX resistance, but others, such as serotypes 35B and 7F, are uniformly susceptible [18]. This combination antibiotic is not recommended for treatment of respiratory infections that may be due to pneumococcus.

SUMMARY

●Common clinical manifestations of pneumococcal infection – Streptococcus pneumoniae (pneumococcus) is among the most commonly identified bacterial causes of pneumonia, acute rhinosinusitis, acute otitis media, and acute exacerbations of chronic obstructive pulmonary disease. (See 'Introduction' above.)

●Penicillin resistance – Pneumococci, which used to be broadly antibiotic susceptible, have acquired genetic material that encoded resistance to penicillin and other commonly used antibiotics. (See 'Introduction' above.)

●Fluoroquinolone resistance – In the United States, approximately 1 to 2 percent of S. pneumoniae isolates are resistant to levofloxacin, moxifloxacin, or gemifloxacin. Resistance rates are higher in United States nursing homes and in other areas of the world where fluoroquinolones are often used. (See 'Prevalence of resistance' above.)

●Tetracycline resistance – In the United States, approximately 20 percent of pneumococcal isolates are resistant to doxycycline and minocycline and a slightly higher percentage are resistant to tetracycline. This resistance rate is substantially higher among penicillin-resistant than among penicillin-susceptible strains. There is substantial variation worldwide, with the rate of tetracycline resistance ranging from 6 percent in Canada to 25 to 40 percent in many other countries. (See 'Prevalence of resistance' above.)

●Trimethoprim-sulfamethoxazole resistance – Approximately one-third of all pneumococcal strains in the United States are resistant to trimethoprim-sulfamethoxazole, with regional variation ranging from 25 percent in the Northeast to 45 percent in the Southeast. In many other countries, rates of resistance are higher, approaching 80 percent in some Southeast Asian countries. (See 'Activity and prevalence of resistance' above.)