|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Microbiology and Infectious Diseases (Dr. Weiss), Maisonneuve-Rosemont Hospital, University of Montreal, Montreal, QC, Canada; and Oscient Pharmaceuticals, Waltham, MA (Mr. Tillotson).
Correspondence to: Karl Weiss, MD, MSc, Department of Microbiology and Infectious Diseases, Maisonneuve-Rosemont Hospital, 5415 LAssomption, Montreal, QC, Canada H1T 2M4; e-mail: weisscan{at}aol.com
| Abstract |
|---|
|
|
|---|
Aim: The purpose of this review was to examine the current clinical evidence to support one option or the other by gathering all of the available published literature. We will review the existing controversies in terms of microbiology, immunology, and clinical outcomes comparing dual therapy (ie, with any combination of ß-lactams, macrolides, or fluoroquinolones) with monotherapy in the treatment of CAP.
Results: For the vast majority of patients with CAP (ie, outpatients and inpatients on medical wards), the type of antibiotic regimen prescribed does not have any significant impact. For patients with severe pneumonia, for which there is no accepted definition so far, the controversy remains alive. Mortality from pneumococcal pneumonia has been reduced over the last decades, but despite improved medical care, bacteremic pneumococcal pneumonia is still as lethal as ever, probably because of the aging population, the greater number of immunocompromised patients, and the number of patients with frequent comorbid conditions. Worldwide, the increasing rates of resistance of Streptococcus pneumoniae to antibiotics are also a serious concern, and the clinical implications are not always obvious. Although limited in number, the four studies showing the importance of adding a macrolide to a ß-lactam regimen for the treatment of bacteremic S pneumoniae pneumonia are retrospective and nonblinded, the findings are consistent, and they point to a trend that has to be explored more thoroughly. Studies published in the last few years suggest that combination therapy may be superior for bacteremic S pneumoniae pneumonia.
Conclusion: In the meantime, for practical purposes, patients hospitalized with a diagnosis of severe CAP may benefit from a dual antibiotic therapy combining a third-generation cephalosporin and a macrolide. For the majority of hospitalized patients with CAP who are not severely ill, fluoroquinolone monotherapy remains an approved, tested, and reliable option. Indeed, the time for more aggressive outpatient fluoroquinolone therapy may reduce the number of patients who are hospitalized with CAP. Independent prospective studies comparing combination therapy with standard monotherapy are urgently required for hospitalized patients with severe CAP.
Key Words: community-acquired pneumonia single vs dual antibiotic therapy
| Introduction |
|---|
|
|
|---|
Not surprisingly, various CAP guidelines17 have identified this issue as one of the most important issues in their recommendations. Any improvement in the management of bacteremic streptococcal pneumonia will significantly reduce the overall mortality rate of patients with CAP, hence the importance of addressing the dual-therapy issue. Dual therapy can be defined as any combination regimen involving a ß-lactam, a macrolide, or a fluoroquinolone; by contrast, monotherapy would be the use of a ß-lactam-based regimen or a fluoroquinolone agent alone.
| The In Vitro Paradox |
|---|
|
|
|---|
However, the "doomsday scenario," in which large numbers of patients do not respond to therapy while receiving these agents, has not occurred.9 There have been several anecdotal case reports and some retrospective studies published showing the relationship between in vitro resistance and in vivo clinical failures, but these remain the exception rather than the rule. Indeed, many primary care physicians report rarely seeing such clinical failures, and yet they are aware of growing antimicrobial resistance. In patients with pneumonia, only those with S pneumoniae with a minimum inhibitory concentration to penicillin of
2 mg/L may not respond to therapy in significantly greater numbers compared with those patients with susceptible strains.101112 Austrian and Gold5 showed many years ago that the overall mortality rate for patients with bacteremic pneumococcal pneumonia who were being treated with penicillin was 17% when resistance was not an issue. This suggests that other factors, likely immunologic or genetic, influence the outcome of this infectious condition. The drug-bug relationship alone cannot explain everything.
In Canada and the United States, macrolides remain the recommended first-line therapy for CAP for outpatients without comorbidities. For hospitalized patients, monotherapy with a respiratory fluoroquinolone, especially levofloxacin, has become a popular choice as an empiric therapy notably for hospitalized patients,17 although high-dose amoxicillin is still the suggested treatment in some European countries.13 Despite increasing macrolide resistance, significant clinical lack of response with these agents in previously healthy individuals is rarely seen and reported.
With regard to macrolides, efflux-mediated (or mef) resistance may rarely lead to clinical lack of response when dealing with infections contained in the respiratory tract,14 because the respiratory tissue concentrations of these antibiotic agents are tremendously higher than their serum concentrations. Tens of millions of macrolide prescriptions are written every year in North America, yet the number of clinical failures, even if underreported, is very low. Macrolide resistance has been associated with lack of response in cases of bacteremia, but this was foreseeable in selected high-risk patients with underlying medical conditions and remains rare.15
Presently in North America, S pneumoniae resistance to newer respiratory fluoroquinolones (ie, moxifloxacin and gemifloxacin) remains at the
0.5% level despite increasing use of this class of agent.16 However, the clinical failure of fluoroquinolones in respiratory tract infections has been summarized by Scheld17 and, although very rare and anecdotal, becomes a reality in our daily practice. Thus, this article and other publications15171819 have raised the issue of increasing clinically significant resistance to these classes of antibiotic agents.
It is true that the definition of clinical failure has also not been established, and mortality alone or the occurrence of severe life-threatening conditions such as meningitis while receiving therapy may only represent the tip of the iceberg. A lack of improvement in the condition of a CAP patient after 72 h of treatment may also be considered a clinical failure; however, these cases are not reported or seldom reported. We would suggest that the definition of clinical resistance, rather than strictly relying on National Committee for Clinical Laboratory Standards criteria, should be linked to the inability of an antibiotic agent to reach a concentration high enough at the site of infection to inhibit local bacterial replication and to the need for a second antibiotic prescription 72 h after starting therapy because of the deterioration or nonsignificant improvement of the initial medical condition.
| Does Antimicrobial Resistance Affect More Than Susceptibility? |
|---|
|
|
|---|
There is some controversy regarding the potential decreased virulence of invasive S pneumoniae strains. Consistently, large-scale, longitudinal in vitro studies2122 have shown lower levels of resistance to penicillin and macrolides among invasive strains compared with isolates from sputum or other nonblood infections. Although not proven, it has been advocated that antimicrobial agent resistance may be associated with a decreased fitness of the resistant organism or an increased energy cost for the bacteria, thus potentially rendering the organism less virulent because of the selection of life support processes over pathogenesis mechanisms.23
Surveillance programs can be criticized on many levels, including a bias toward respiratory tract isolates compared with blood or other systemic isolates. Additionally, the isolates are usually collected from patients presenting at a tertiary care center rather than the primary care setting. The patients in the tertiary or university clinics often have risk factors for antibiotic agent resistance, have chronic underlying diseases, have received numerous antibiotic treatments (eg, for COPD), or are immunocompromised patients (eg, due to HIV infection and chemotherapy). We may be overestimating resistance levels for first-line practicing physicians.
| Transatlantic Rift or Same Opinion Expressed Differently? |
|---|
|
|
|---|
Some data25262728 have suggested that combination therapy may be superior for the treatment of bacteremic S pneumoniae pneumonia. This group of patients represents only a small fraction of all of the CAP cases, but in terms of mortality it is the most important strata.
Additional support for combination therapy is the enhanced coverage of potentially resistant strains, the appropriate coverage of the organisms implicated in polymicrobial CAP, pneumococci, and atypical organisms, including both Chlamydia pneumoniae and Mycoplasma pneumoniae. Published data29 have suggested that a proportion of bacteremic S pneumoniae pneumonia patients have concomitant M pneumoniae or, rarely, Legionella sp infections. However, there is some transatlantic controversy between North American physicians, who support the importance of the atypical organisms, and European doctors, who are yet to be convinced of the benefit of the systematic use of a macrolide. In bacteremic streptococcal pneumonia, at least four retrospective clinical studies25262728 have shown a positive impact on mortality.
| Aiming Right |
|---|
|
|
|---|
The fact that combination therapy, as defined previously, may be superior to monotherapy in the treatment of patients with severe CAP may be explained by different factors, including a better coverage of atypical microorganisms and the fact that by acting at two different sites in the bacteria (cell wall for ß-lactams and the inhibition of protein synthesis for macrolides), this combination targets the microorganism at two different levels. Macrolides have been shown to have some very effective antiinflammatory properties. They are known to reduce the production of interleukin-8 and tumor necrosis factor-
, and erythromycin has been shown to reduce the adherence of S pneumoniae to respiratory epithelial cells.33 ß-Lactams and fluoroquinolones used as bactericidal agents are known to kill and destroy a large number of bacteria in a short amount of time, thus releasing numerous intracellular components responsible for perpetrating the inflammation process.
On the other side, macrolides inhibiting protein synthesis potentially decrease the production of virulence factors. This has been shown in patients with severe group A Streptococcus infections in whom clindamycin, acting as a protein synthesis inhibitor, may add a significant benefit to the global treatment.34
The importance of the immune system in cases of S pneumoniae bacteremia has been demonstrated in a pivotal 1964 study5 in which serum therapy alone significantly decreased the mortality rate compared with a cohort of control subjects with no serum therapy. Therefore, factors other than the simple concept of in vitro activity alone are playing key roles in fighting the infectious process. One study35 demonstrated the absence of in vitro synergy when combining treatment against S pneumoniae with erythromycin with penicillin or cefotaxime. Therefore, we cannot attribute the observed effect of adding a macrolide in patients with severe CAP to a synergistic action between these two antibiotic agent classes.
| What Is the Clinical Data Supporting Combination Therapy? |
|---|
|
|
|---|
In our institution, a similar retrospective study28 on the outcome of bacteremic S pneumoniae pneumonia in adults showed the same outcome. The mortality rate was significantly higher in the group receiving monotherapy (ß-lactam) compared with that in the dual-therapy group (ie, ß-lactam plus a macrolide) [11 of 42 patients [26%] vs 4 of 53 patients [7.5%], respectively; p = 0.02]. Patients were comparable in terms of disease severity as the initial pneumonia severity index scores were 113 and 114, respectively.36 Penicillin nonsusceptibility (ie, strains categorized as intermediate or resistant to penicillin) was not associated with increased mortality (1 of 10 patients [10%] vs 14 of 85 patients [16.4%], respectively; difference not significant) for sensitive strains.
Although all of these studies were nonrandomized retrospective studies, they confirmed the importance of initially adding a macrolide to a ß-lactam regimen for the treatment of bacteremic S pneumoniae pneumonia. However, these conclusions are limited to patients with severe pneumonia for whom the mortality risk is high. The value of these results cannot be extrapolated for the treatment of inpatients with moderate disease, for whom fluoroquinolones still remain an excellent therapy choice.
A few studies1237 demonstrated that resistance alone in bacteremic pneumococcal pneumonia does not have any significant impact on outcomes. Moroney et al37 showed that there was no relationship between the presence of resistant strains and the mortality rate or the requirement for ICU admission.
Almost all of the clinical studies comparing newer fluoroquinolones with the standard therapeutic regimen were designed to show noninferiority or bioequivalence so as to gain approval from registration agencies, and, therefore, high-risk patients (Fine IV or V categories) were usually excluded from these clinical trials. For example, an outpatient trial38 comparing two fluoroquinolones (gemifloxacin and trovafloxacin) showed response rates of 92.5% and 87.3%, respectively. However, this trial was not designed to evaluate fluoroquinolone efficacy in severe pneumonia.
The capital study published by Marrie et al39 showed a better outcome for treatment with fluoroquinolones, but this may have been explained by the study protocol itself. Implementing an active and highly intervention-oriented clinical pathway led to better resource utilization. One study40 has revealed an economic benefit for fluoroquinolone monotherapy over a combination treatment, but the vast majority of cases were not severe CAP. Studies comparing dual therapy with one of the newer respiratory fluoroquinolones with an enhanced activity against S pneumoniae (eg, moxifloxacin, levofloxacin, and gatifloxacin) in patients with severe CAP are not available yet. There is no clear definition of severe CAP so far, but the presence of pneumococcal bacteremia is certainly something that can be included in this definition. Therefore, severe CAP can be defined as the presence of S pneumoniae bacteremia, the necessity of ventilatory support, or patients with a pneumonia index score classifying them as being in Fine classes 4 or 5. In this scoring system, classes 4 and 5 were linked to a higher mortality rate.36 Although this system was not designed to evaluate severity, it can be used as a good indicator for this purpose. A new rule, designated as CURB-65 and using a 6-point score based on confusion, urea level, respiratory rate, BP, and age, has also been shown to be an excellent indicator for defining the mortality risk.41
| Future Challenges |
|---|
|
|
|---|
Better and earlier detection methods to rule out the presence of bacteremia at the initial presentation will certainly help. Rapid methods based on DNA detection of S pneumoniae in blood are not far away and will make life easier for clinicians.42 The arrival of a S pneumoniae urinary antigen test may be helpful in the future.43
Although most studies showing the importance of adding a macrolide to a ß-lactam regimen for bacteremic S pneumoniae pneumonia are retrospective and nonblinded, the findings are consistent and point to a trend that has to be explored more thoroughly. In the mean time, for practical purposes, patients who are hospitalized with a diagnosis of severe CAP may benefit from a dual antibiotic therapy combining a third-generation cephalosporin and a macrolide. For the majority of patients with CAP who are hospitalized and not severely ill, fluoroquinolone monotherapy remains an approved, tested, cost-effective, and reliable option. Indeed, the time for more aggressive outpatient fluoroquinolone therapy may reduce the number of patients who are hospitalized with CAP.44
Independent prospective studies comparing combination therapy with standard monotherapy are urgently required for severe hospitalized CAP patients. The definition of severe CAP has to be agreed on at an international level, and a three-armed controlled trial comparing a fluoroquinolone, a third-generation cephalosporin, plus a macrolide, and the same third-generation cephalosporin and fluoroquinolone will likely give the answer to this crucial debate. The main hurdle in conducting such a study is the difficulty of recruiting enough patients with bacteremic pneumococcal pneumonia.
| Footnotes |
|---|
Received for publication September 28, 2004. Accepted for publication January 20, 2005.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |