Antimicrobial Agents in Respiratory Illness: Beyond Bactericidal

This program was supported by an unrestricted educational grant from Abbott Laboratories.

Faculty disclosures for this summary are provided on page 24 of this publication.

Anti-Inflammatory Activities of Macrolide Antibiotics

Jun Tamaoki, MD, Associate Professor of Medicine, Pulmonary Division, Tokyo Women’s Medical University, Japan, reviewed his research on the anti-inflammatory effects of macrolide antibiotics.

“Patients with chronic inflammation of the airways have numerous neutrophils and eosinophils present in the mucosa, as well as epithelial damage and goblet cell metaplasia that can lead to airway hypersecretion,” Dr. Tamaoki said. “The increased airway mucous layer may be associated with mucociliary dysfunction and recurrent airway infection.”

Macrolides can inhibit chemotaxis of inflammatory cells and block cytokine production, reactive oxygen species production, and adhesion molecule expression, Dr. Tamaoki added. They are also known to reduce airway mucous secretion. “As a result,” he said, “treatment with macrolides may produce improvements in chronic airway inflammation and airway hypersecretion.”

While in vitro studies of the effect of macrolides on neutrophils have been contradictory, he noted, “many in vivo studies have shown that macrolides generally inhibit infiltration of neutrophils into the airways.”

Epithelial cell expression of ICAM 1 attracts neutrophils, and macrolides appear to reduce neutrophil accumulation by blocking expression of the gene.

Research has suggested that immune complex triggers alveolar macrophages to release nitric oxide (NO), and that chemicals generated from NO may cause lung inflammation and injury. “There-fore, we speculated that if macrolides inhibit NO generation, then the drug can protect against immune complex-induced lung inflammation,” Dr. Tamaoki explained. In vitro and in vivo experiments showed that erythromycin, clarithro-mycin and other macrolides reduce NO generation and expression of inducible nitric oxide synthase (iNOS) messenger RNA, and suggest this effect is due to inhibition of cytokine synthesis.

Dr. Tamaoki concluded by discussing the effects of macrolides on airway mucous secretion. Patients with diffuse panbronchiolitis (DPB) have progressive airway limitation and chronic inflammation, mainly in the respiratory bronchioles. Their airway mucosa contains many secretory granules, which suggests mucous hypersecretion.

Giving patients with DPB 400 mg of clarithromycin reduced sputum production without changing sputum flora, he reported. This reduction may be due to the anti-inflammatory effects of the drug, and macrolides may also act directly on mucus-producing cells in the airways. In vitro studies have suggested that erythromycin blocks respiratory glycoconjugate release from human airway explants, Dr. Tamaoki noted, and his own in vitro studies found that the drugs inhibited MUC-5 AC gene expression and thus mucous glycoprotein hypersecretion. “This effect,” he added, “likely at least partly involves inhibition of transcription factor I KAPPA B alpha.”

“Macrolide antibiotics have a variety of anti-inflammatory and immuno-modulatory activities and may thus be considered for the treatment of certain types of chronic airway inflammatory disease,” Dr. Tamaoki concluded.

The Use of Macrolides in Diffuse Panbronchiolitis and Cystic Fibrosis Lung Disease

Bruce Rubin, MEngr, MD, Professor of Pediatrics and Chief of Pulmonary Pediatrics, Wake Forest University, Winston-Salem, North Carolina, discussed his research on the mechanism of action of macrolides in lung disease.

“The non-antibiotic properties of macrolides are quite a few and have been known for a while,” Dr. Rubin noted. For example, he explained, earlier-generation erythromycin A derivatives were known to stimulate the motilin receptor, causing gastrointestinal side effects. Non-antibiotic macrolides are now being developed specifically to stimulate this receptor for use as prokinetic agents.

And troleandomycin, an erythro-mycin A derivative, has been shown to be steroid-sparing in people with steroid-dependent asthma. Patients on chronic oral prednisolone given troleandomycin, erythromycin or oleandomycin could reduce and sometimes eliminate their dosage of oral corticosteroids. “This was the first indication that these medications, the 14- and now the 15-member macrolide derivatives, have immuno-modulatory effects,” he said. Other research, he noted, has shown that these macrolides block angiogenesis and have anti-tumor effects.

Macrolides have been studied for asthma, sinusitis, bronchorrhea and bronchitis, while the largest studies have been of diffuse panbronchiolitis (DPB). The disease, he explained, is most often seen in Japan, Korea, and China, and is thought to have a genetic component. The disease was first described in the early 1960s.

“These patients have chronic cough, sputum, progressive dyspnea with airflow limitation, and early death,” Dr. Rubin said. Most patients present with symptoms in the second, third, and fourth decade of life, and most are non-smokers. Careful histories reveal that many had sinusitis and rhinitis dating back to early childhood, he noted. “It’s thought to be a lifelong disease, possibly starting in childhood and leading to very severe and progressive airway inflammation and infection, sputum, and chronic Pseudomonas tracheaobronchitis, with biofilm formation very common,” he continued. “This sounds very much like cystic fibrosis.”

Macrolides were first used to treat DPB by Miyasawa and Kudoh in 1982. There had been no effective treatment for DPB, Dr. Rubin explained. In the initial open-label trial, 60% of DPB patients on erythromycin showed improvement, compared to 15% on placebo. “In 1984 the five-year survival for DPB was 26%,” he added. “After the introduction of macrolides, the 10-year survival has increased to 94%, largely attributable to macrolide antibiotics,” Dr. Rubin said. “The Japanese Ministry of Health wanted to set up a randomized placebo-controlled trial but the effects were so dramatic that recruitment was almost impossible.”

Studies show that levels of IL-8, a strong prokinetic agent for neutrophils, fall dramatically in the serum and bronchial lavage fluid of DPB patients on erythromycin, Dr. Rubin said. There are also data that clarithromycin has a similar effect, and it’s likely that azithromycin does as well.

Dr. Rubin went on to discuss studies of macrolides for the therapy of cystic fibrosis (CF). Pilot studies have demonstrated improvement in CF patients given clarithromycin or azithro-mycin. Bush and colleagues performed a randomized double-blind placebo-controlled crossover trial of azithromycin in CF patients lasting 15 months (Lancet. 2002;360:978-84). FEV1, the primary outcome measure, increased 6% more on azithromycin vs. placebo, for a relative difference of 11%. While half of patients had an FEV1 improvement of 10% or more on azithromycin, some patients had no improvement and even some deterioration. Similar findings have been shown for clarithromycin.

Of interest is that patients homozygous for the most common CF defect, delta F 508, were four times more likely to have an improved FEV1 response with this macrolide. “Fortunately, this genotype accounts for about half of all patients with cystic fibrosis in Europe and North America,” Dr. Rubin noted. It was also found that patients on dornase alfa did not improve with azithro-mycin, which may be because the drug binds to the antibiotic.

A three-month placebo-controlled trial of azithromycin in 60 adults with severe but stable CF found the drug improved patients’ quality of life and reduced the rate of decline in pulmonary function, and also resulted in their needing fewer courses of IV antibiotics.

“The immunomodulatory effects of macrolides in CF and DPB are well established, with strong data for these effects in bronchiectasis and growing data for them in bronchitis and sinusitis,” Dr. Rubin said. “It’s possible that primary ciliary dyskinesia, chronic secretory otitis media and other chronic inflammatory diseases of the airways may benefit from these immunomodulatory effects,” he added.

Macrolides may also directly affect Pseudomonas and other bacteria by impairing their ability to form biofilms, Dr. Rubin noted. Pseudomonas can form biofilms throughout the body, he noted, for example in the lungs of CF patients, in the urinary tract of patients on catheters, and on the teeth, causing dental caries. According to a recent New England Journal of Medicine review, Dr. Rubin added, biofilms account for 60% of clinically important infections in humans.

Biofilms form when planktonic bacteria land on a surface and begin to produce colonies. “These sessile colonies build up, spread by twitching motility, and interact, through quorum sensing and then change their phenotype so that the colonies deep within will go into metabolic sleep, have lower oxygen requirement and are very well set up to develop resistance to antibiotics,” Dr. Rubin explained. Clarithromycin inhibits biofilm formation by affecting twitching motility by type IV fimbrae, which is necessary for adherence to epithelial surfaces and for virulence, he said.

Clarithromycin, as well as the newer ketolide ABT-773, “profoundly inhibit twitching motility,” he added. But mutants that need glutamate for metabolism are not sensitive to these drugs, Dr. Rubin said.

Macrolides for Acute Exacerbations of Chronic Bronchitis in COPD

Michael Niederman, MD, Professor of Medicine, SUNY at Stony Brook and Chairman of the Department of Medicine, Winthrop University Hospital, discussed the value of macrolides in treating acute exacerbations of chronic bronchitis.

He began by reviewing the classifications of these exacerbations. The three cardinal symptoms are increased dyspnea, increased sputum volume, and increased sputum purulence. Anthonisen and colleagues have classified exacerbations into Types 1 to 3. A type 1 exacerbation is one in which all three symptoms are present, while type 2 involves two symptoms and type 3 one symptom. Antibiotics are useful in type 1 or 2 exacerbations, which account for about 80% of all exacerbations, Dr. Niederman noted, and are not useful in type 3.

The three organisms important in the disease are non-typeable Haemo-philus influenza, pneumococcus and Moraxella, he added.

In pneumococcus infections, “drug resistance is common” in many populations, he added, including adults with chronic obstructive pulmonary disease (COPD).

There are two types of pneumococcal resistance to macrolides, Dr. Niederman noted: an efflux mechanism with a relatively low level of resistance and a ribosomal mechanism with a higher level of resistance. In the U.S., most macrolide resistance in pneumococci involves the efflux mechanism. “Although usage is driving resistance, recent studies have shown that there is very little change in the frequency of the ribosomal mechanism related to macrolide usage,” Dr. Niederman pointed out.

Drug resistance in Haemophilus influenza develops through a different mechanism, he continued: production of beta lactamase enzymes, seen in as many as 40% of these organisms. The problem is even worse in Moraxella catarrhalis, he added. One survey found that 95% of 700 isolates of the organism produced beta lactamase enzymes.

“There is not any question in my mind that antibiotics are useful in exacerbations,” Dr. Niederman said. While there have been no randomized placebo-controlled trials showing whether new broad-spectrum antibiotics are better, he argued that such aggressive therapy should be used in patients whose infections are most likely to recur. “In addition, if we have a chronically ill patient whose cost of failure of therapy would be great,” he said, “we might want to use a more potent agent preemptively in this population so that they are effectively treated because they can’t tolerate repeated episodes.”

Macrolides may be the best treatment in exacerbations for some populations, he added, because they are the most focused, while quinolones may be ideal for some patients, but too broad spectrum for others.

The patient with uncomplicated COPD who has less than four exacerbations a year and FEV1 greater than 50% would be a good candidate for a macrolide, he said. More complicated patients, who are older with worse lung function and more frequent exacerbations, likely should be treated with a quinolone or amoxicillin clavulanate.

Dr. Niederman emphasized the importance of maintaining heterogeneity in antibiotic use, in order to prevent the development of resistance. “I think that we have to consider what our alternatives are and try to select antibiotics appropriately for populations where we can use a variety of different ones in sequence and we can focus on not using broader therapy than is necessary,” he concluded.

Macrolides for the Treatment of COPD, Asthma, and Sinusitis

Mark Gotfried, MD, Professor of Medicine, University of Arizona, Phoenix and Medical Director, Kindred Hospital in Phoenix, discussed the clinical use
of macrolides in COPD, asthma, and sinusitis.

He reviewed the previous speakers’ points on the effects of macrolides in terms of inflammation. “Looking specifically in the respiratory tract, we note that the macrolides may attenuate the inflammatory response in the lungs by protecting ciliated epithelium against oxidative damage,” he said.

Dr. Gotfried cited one study comparing clarithromycin with prednisolone in a culture of nasal mucosa from patients with chronic rhinosinusitis, which found that clarithromycin reduced expression of pro-inflammatory cytokines. A study of chronic rhinosinusitis patients with nasal polyps found that three months of treatment with 14-member macrolides, chiefly clarithromycin, reduced IL-8 levels and nasal polyp size.

Regarding sinusitis, Dr. Gotfried continued, a study by MacLeod of 25 patients given 500 mg of clarithromycin twice daily for 14 days found that patients with very active inflammation saw significant reduction in levels of IL-8, IL-6, TNF alpha and edema, as well as improvement in signs and symptoms (Advances in Therapy. 2001;18:75-82).

The problem with these findings and others, he said, is that they don’t answer the question of whether the drugs have a direct anti-inflammatory effect or whether this effect is due to the drugs’ antibacterial properties.

He reviewed other studies of macrolides in chronic sinusitis, and noted that researchers have said the maximal effect may not appear until patients are on the drugs for 12 weeks or longer. Larger studies are needed to determine what effect such longer treatment will have on underlying bacteriology and resistance.

Dr. Gotfried then reviewed a number of studies of macrolides in asthma. Studies have shown macrolides can reduce airway edema, decrease mucous secretion, reduce airway hyperreactivity, and improve some measures of quality of life. Patients PCR-positive for chlamydia or mycoplasma have shown significant improvements in FEV1 with clarithromycin. Again, he added, 10 to 12 weeks of treatment may be necessary to see an effect.

Macrolides may be worth trying on corticosteroid-dependent asthmatics for two months, given that they are followed very closely and are given all the other therapy they need, Dr. Gotfried said.

But large-scale studies of macrolides in COPD need to be done, he added, to truly pin down their effects.

Summary: Does this Relate to Bacterial Resistance?

William Bishai, MD, PhD, Associate Professor, Johns Hopkins Medical Center, Baltimore, concluded the presentation with some summary comments.

He first mentioned the “in vivo/in vitro paradox” seen with pneumococcal pneumonia: patients with isolates that show resistance to drugs in vitro are not failing treatment. “This has been shown not only for the beta lactams,” Dr. Bishai said. “There is some evidence for it in the macrolide-resistant pneumococci.”

Current breakpoints are between 0.5 and 0.25 mcg/ml for susceptible/intermediate-level resistance for erythromycin and clarithromycin and between 0.5 and 1 mcg/ml for intermediate/high-level resistance, he noted. But animal studies found that strains of pneumococcus that would be considered resistant on the basis of these numbers could actually be successfully treated by a macrolide, Dr. Bishai added.

In terms of human experience with macrolide-resistant Streptococcus pneumonia, one large study by Lonks and colleagues with over 1,000 patients conducted over 13 years failed to identify very many treatment failures with strains that were resistant at the 1 to 8 mcg/ml level.

“One might then think about the breakpoints as being inadequate predictors of treatment outcome in macrolide resistant pneumococcal community-acquired pneumonia,” he added. One possible explanation for the clinical success with low-level resistant strains could be increased tissue concentration of macrolides on epithelial surfaces and in the intracellular compartment, he noted, as well as the drugs’ immunomodulatory effects.

Current ATS guidelines on treating inpatient community-acquired pneumonia recommend dual therapy with a macrolide and a beta lactam or mono-therapy with a fluoroquinolone, Dr. Bishai noted, but new evidence suggests that mortality is lower for patients on the combination therapy. Several possible mechanisms could be responsible, he added. A double-attack mechanism may be more effective, or there may be underlying atypical organisms that haven’t been identified that respond to a macrolide. And the third possibility, he said, is that the immunomodulatory effects of macrolides are responsible.

Faculty Disclosures:

William Bishai, MD
Research Grants: Abbott, Aventis, Bayer, Pfizer, Merck; Speaker: Abbott, Aventis, Roche, Pfizer, Merck

Mark Gotfried, MD
Research Support: Abbott Laboratories, Pfizer; Speaker’s Bureau: Abbott Laboratories;
Off-Label Use: Clarithromycin; Azithromycin for Anti-inflammatory Effects

Michael Niederman, MD
Speaker/Consultant: Pfizer, Bayer, GlaxoSmithKline, Pharmacia, Elon, Merck, AstraZeneca

Bruce Rubin, MD
Research Grant: Abbott Laboratories, Zambon Pharma; Speaker: Abbott Laboratories
Off-Label Use: erythromycin, azithromycin, clarithromycin for immunomodulation

Jun Tamaoki, MD
Reported nothing to disclose