Study objective: Long-term administration of macrolide antibiotics reduces sputum production in patients with chronic airway diseases, probably by inhibiting airway inflammation. The objective of the present study was to determine the acute effects of a macrolide on airway chloride secretion and sputum production.
Methods: We first investigated the effect of erythromycin treatment on chloride diffusion potential difference (CPD) across tracheal mucosa in vivo. Next, we conducted a double-blind, parallel-group study examining the effect of 7 days of treatment with clarithromycin (400 mg/d), amoxicillin (1,500 mg/d), or cefaclor (750 mg/d) in patients with chronic bronchitis or bronchiectasis without apparent respiratory infection.
Results: IV administration of erythromycin decreased the CPD of rabbit tracheal mucosa in a dose-dependent manner. Treatment of patients with clarithromycin decreased sputum production, whereas amoxicillin and cefaclor treatment had no effect. The percentage of patients whose sputum decreased > 30% from baseline (responders) was 38% in the clarithromycin group, 7% in the amoxicillin group, and 0% in the cefaclor group. During treatment with clarithromycin, the sputum solid composition increased and chloride concentration decreased in responders, but these changes were not observed in nonresponders.
Conclusion: Short-term administration of 14-membered macrolide reduces chronic airway hypersecretion, presumably by inhibiting chloride secretion and the resultant water secretion across the airway mucosa.
Key words: airway secretion; bronchiectasis; chronic bronchitis; macrolide antibiotics
Abbreviations: CPD = chloride diffusion potential difference; K-H = Krebs-Henseleit; mV = millivolt; PD = potential difference; %SC = percentage of solid composition
Macrolide antibiotics are effective in the treatment of acute bronchitis and chronic airway diseases, including chronic bronchitis and diffuse panbronchiolitis. (1,2) The efficacy of macrolide antibiotics may be derived not only from their antimicrobial activities but also from immunomodulatory and anti-inflammatory actions. (3) In addition, we have previously shown that administration of macrolides for 6 weeks reduces sputum production in patients with chronic airway hypersecretion, but the mechanism remain uncertain. Although it is possible that the antisecretory effect of macrolides could be associated with the inhibition of airway epithelial chloride secretion and the resultant reduction in water transport across the airway mucosa toward the lumen, (4) this hypothesis is based on in vitro evidence and thus may not accurately reflect the drug actions in vivo. Therefore, in the present study, we first examined the effect of erythromycin on chloride diffusion potential difference (CPD) across the rabbit tracheal mucosa in vivo. If the macrolide causes a rapid inhibition of chloride secretion in this experiment, then it is expected that short-term administration of the drug might reduce airway secretion. We thus conducted a double-blind study looking at the effect of 7 days of treatment with clarithromycin on sputum production in patients with chronic bronchitis or bronchiectasis.
MATERIALS AND METHODS
Measurement of Potential Difference In Vivo: Measurement of tracheal mucosal potential difference (PD) in vivo has been described in detail previously. (5,6) Briefly, Japanese white rabbits weighing 1.8 to 2.5 kg were anesthetized with intraperitoneal [alpha]-chloralose (50 mg/kg) and urethane (500 mg/kg), and the trachea was exposed. A polyethylene tube was inserted into the trachea 5 mm above the carina, through which the respirator (model SN-480-7; Shinano; Tokyo, Japan) was connected and mechanical ventilation was performed (tidal volume, 10 mL/kg; respiratory rate, 60 breaths/min). The upper tracheal cartilage rings were then incised transaxially, and the surface of membranous portion was exposed (Fig 1).
[FIGURE 1 OMITTED]
The exploring bridge, constructed of a polyethylene tube 2.5 mm in diameter for in vivo measurement of airway epithelial PD, was placed on the surface of the posterior membrane above the carina. Contact with the tracheal mucosal surface was ensured by continuous perfusion (0.3 mL/min) through the bridge with Krebs-Henseleit (K-H) solution of the following composition: 118 mM NaCl, 5.9 mM KCl, 2.5 mM Ca[Cl.sub.2], 1.2 mM MgS[O.sub.4], 1.2 mM Na[H.sub.2]P[O.sub.4], 1.2 mM NaHC[O.sub.3], and 25.5 mM glucose, warmed at 37[degrees]C and adjusted pH to 7.4. The whole area of exposed mucosal surface of the trachea was perfused, and the perfusion reservoir was connected to the calomel electrode via a polyethylene tube (1.5 mm in diameter) filled with 3% agar in saline solution.
The reference bridge, a 21-gauge needle that contained 3% agar in saline solution, was inserted into the subcutaneous space of the right anterior chest wall, which was isoelectric with the adventitial surface of the trachea. Each bridge was connected by a calomel electrode to a high-impedance voltmeter (model CEZ-9100; Nihon Kohden; Tokyo, Japan). The electrical signal was filtered to remove 60-cycle interference, and the PD between the tracheal mucosal surface and subcutaneous space (transmembrane PD) was continuously recorded on a pen recorder (model SR 6335; Graphtec; Tokyo, Japan).
In Japan, clarithromycin has been used much more frequently than erythromycin for the treatment of chronic airway diseases. Although they are both 14-membered macrolides, clarithromycin is difficult to dissolve in K-H solution compared with erythromycin. We thus used erythromycin in the animal study.
Effect of Erythromycin on PD: After equilibration, the tracheal mucosa was continuously perfused with K-H solution containing amiloride ([10.sup.-4] mol/L), a sodium channel blocker. Under this circumstance, the remaining PD (CPD) is an index of epithelial cellular and paracellular paths available for chloride diffusion. (7) After the CPD became stable, erythromycin was administered via mucosal or submucosal routes. For mucosal application, erythromycin at a final concentration of [10.sup.-5] mol/L or [10.sup.-4] mol/L was added to the amiloride-containing superfusing solution. For submucosal application, erythromycin, 10 mg/kg/h, was administered via the jugular vein by a bolus injection. Our preliminary study showed that vehicle alone administered by either route had no effect on CPD.
To determine the dose-response relationship for IV-administered erythromycin, increasing doses (1, 3, 10, and 30 mg/kg/h) were administered, while CPD was continuously recorded. In each instance, the next higher dose was administered 5 min after the CPD response to the preceding dose reached a plateau. Since even high concentrations of erythromycin did not alter the CPD when administered by superfusion, no dose-response curves were generated for this route of administration.
Patients: Forty-five patients, 38 to 77 years of age, with chronic bronchitis or bronchiectasis who had been continuously expectorating > 20 g of sputum per day for at least 2 weeks prior to the study were recruited after their consent was obtained. All cases of chronic bronchitis conformed to the World Hearth Organization definition of the disease. (8) Bronchiectasis was confirmed by CT of the chest. There was no evidence of current respiratory infection, based on chest radiography and sputum bacteriology.
Study Design: The study was conducted in a double-blind, parallel-group fashion. A doctor not involved in the disease follow-up or data analysis was assigned the task of classifying the patients into three groups matched for clinical diagnosis (Table 1). Patients continued their usual medications, including oral [[beta].sub.2]-adrenergic agonists, oral theophylline, oral mucolytic agents, inhaled corticosteroids, and inhaled anticholinergic agents. In the first group (the clarithromycin group; n = 16), patients received oral clarithromycin, 100 mg bid for 7 days. In the second group (the amoxicillin group; n = 15), patients received oral amoxicillin, 500 mg, tid for 7 days. In the third group (the cefaclor group; n = 14), patients received oral cefaclor, 250 mg tid for 7 days. Patient compliance with the medication schedule was assessed from an individual record supplied by each patient at the end of the trial.
Analysis of Sputum: The patients were given preweighed, covered, plastic cups, and were asked to collect and weigh all sputum expectorated during every 24 h of the trial period at home. At the beginning and after 7 days of the trial, sputum samples collected in the morning (7 AM to 10 AM) were transported to the laboratory. After determination of the wet weight, they were dried in a microwave oven (500 W for 30 min) and reweighed. The percentage of solid composition (%SC) was then calculated from the ratio of wet to dry weight. (9) After centrifugation of sputum, the concentration of chloride in the supernatant was measured by a chloridometer.
All date are expressed as means [+ or -] SEM. Statistical analysis was performed by Student t test or the Kruskal-Walls one-way analysis of variance, and p < 0.05 was considered statistically significant.
Transepithelial PD of Rabbit Tracheal Mucosa In Vivo
The in vivo PD of rabbit tracheal mucosa became stable within 5 min after perfusion with K-H solution, and the baseline PD value was 19.4 [+ or -] 1.7 millivolts (mV) [n = 14], lumen negative. As shown in Figure 2, application of amiloride at [10.sup.-4] mol/L of the superfusing solution reduced the PD to 11.0 [+ or -] 1.2 mV (n = 14); this value is identified as the baseline CPD. Subsequent application of erythromycin at [10.sup.-5] mol/L or [10.sup.-4] mol/L to the superfusate did not significantly alter the CDPD. However, IV administration of erythromycin at 10 mg/kg/h decreased the CDPD from 10.5 [+ or -] 0.6 to 6.8 [+ or -] 0.4 mV (p < 0.001; n = 7) within 2 min, a value that remained stable at least for the following 5 min.
[FIGURE 2 OMITTED]
As demonstrated in Figure 3, IV erythromycin reduced the CPD in a dose-dependent fashion. The maximum decrease in CPD from the baseline value was 5.7 [+ or -] 0.8 mV (p < 0.001; n = 9), and the apparent dose required to produce a half-maximal effect was 2.7 mg/kg/h. Administration of the vehicle alone had no effect.
[FIGURE 3 OMITTED]
Sputum Production in Patients With Chronic Bronchitis or Bronchiectasis
During the trial, no apparent adverse effects were seen in either treatment group. As shown in Figure 4, treatment with clarithromycin for 7 days decreased sputum production from 34 [+ or -] 5 to 25 [+ or -] 5 g/d (p < 0.05; n = 16), whereas amoxicillin and cefaclor had no effect. In the clarithromycin group, 6 of 16 patients (38%) showed the reduction of sputum volume by > 30% of baseline, whereas only 1 of 15 patients (6.7%) in the amoxicillin group and no patients in the cefaclor group did so. We defined the patients whose sputum volume was decreased > 30% of baseline as responders, and the others as nonresponders, and analysis of the sputum was compared between these two groups.
[FIGURE 4 OMITTED]
In responders, administration of clarithromycin for 7 days increased %SC from 2.5 [+ or -] 0.3% to 3.2 [+ or -] 0.3%. In contrast, the values for %SC in nonresponders were higher than those of responders and did not change after the treatment with clarithromycin (Fig 5). The sputum chloride concentrations were higher in responders than in nonresponders before treatment, and administration of clarithromycin decreased chloride concentrations from 191 [+ or -] 20 to 157 [+ or -] 18 mM (p < 0.05; n = 6) in responders, whereas it was without effect in nonresponders (Fig 5).
[FIGURE 5 OMITTED]
In our present study, we first looked at the effect of erythromycin on transepithelial PD of rabbit trachea in vivo. Epithelial cells in the central airway absorb Na from and secrete chloride toward the lumen, and the net ion flux through epithelial cellular and paracellular paths generates a transepithelial PD, which concomitantly promotes water movement across the airway mucosa. (10) In this study, the PD determined after addition of amiloride that eliminates the Na component of transport is assumed to reflect most likely chloride secretion. This CPD was rapidly reduced by IV erythromycin in a dose-dependent manner, indicating that this drug inhibits airway epithelial secretion of chloride toward the respiratory lumen. Since inhibition of chloride secretion reduces subsequent movement of water toward the lumen, the observed effect of erythromycin on CPD strongly suggests that it may inhibit water secretion and, hence, decrease the amount of sputum production. However, possible contributions of electrolyte transport processes other than chloride, such as amiloride-insensitive Na, Na-glucose co-transport, and bicarbonate diffusion, cannot be ruled out.
Previous in vitro studies (11,12) showed that airway epithelial chloride secretion was inhibited by macrolide antibiotics but not by a group of penicillin, cephalosporin, aminoglycoside or tetracycline, indicating the effect is specific for macrolides. Additionally, the inhibition of chloride secretion by macrolide can be observed only when the drug is applied to the submucosal side of the epithelium, possibly due to the difference in distribution of the receptor. (12) In agreement with this finding, IV but not mucosal application of erythromycin caused a decrease in CPD in our in vivo experiment. The mechanism by which macrolides inhibit airway epithelial chloride secretion is unknown. Several airway epithelial functions are controlled by the autonomic neural pathway, which plays a role in chloride secretion and the maintenance of epithelial PD in vivo, (13) and it is known that erythromycin inhibits the release of acetylcholine from the airway cholinergic nerve terminals. (14) Therefore, although it is possible that macrolides have a direct inhibitory action on airway epithelial chloride channels, the inhibition of cholinergic neurotransmission could also be involved.
On the basis of the finding in the CPD experiment, we conducted a clinical study and found that short-term treatment with clarithromycin but not amoxicillin or cefaclor caused a significant decrease in the amount of sputum expectorated by patients with chronic bronchitis or bronchiectasis. Furthermore, in responders to clarithromycin, the decrease in sputum production was accompanied by the increase in %SC of the sputum, which is suggestive of less hydration. However, although not significant, the mean calculated values of total solids were decreased (1.03 g/d on day 0, and 0.70 g/d on day 7). This finding is compatible with the previous finding that macrolide inhibits mucus glycoprotein secretion from human isolated airways. (3) Thus, the reduction of sputum volume by clarithromycin may be largely due to its effect on water transport, presumably involving active ion transport processes in the airway epithelium.
There seemed to exist responders and nonresponders in our patients. We found that responders had higher chloride content and lower %SC in their sputum compared with nonresponders, and that these values changed after treatment with clarithromycin along with the decrease in sputum volume. These results suggest that macrolide therapy is effective in patients having "watery" sputum, and that the drug may exert its effect by inhibiting airway epithelial chloride secretion and the concomitant secretion of water toward the lumen. However, actual ion transport properties of airway epithelium in chronic bronchitis and bronchiectasis are unknown, and the mechanism of difference between responders and nonresponders warrants further studies.
ACKNOWLEDGMENT: The authors thank Masayuki Shino and Yoshimi Sugimura for technical assistance.
(1) Suez D, Szefler SJ. Excessive accumulation of mucus in children with asthma. J Allergy Clin Immunol 1986; 77: 330-334
(2) Willey RF, Gould JC, Grant WB. A comparison of ampicillin, erythromycin and erythromycin with sulphametopyyrazine in the treatment of infective exacerbations of chronic bronchitis. Br J Dis Chest 1978; 72:12-20
(3) Goswami SK, Kivity S, Marom Z. Erythromycin inhibits respiratory glycoconjugate secretion from human airways in vitro. Am Rev Respir Dis 1990; 141:72-78
(4) Tamaoki J, Konno K. Effect of erythromycin on bioelectric properties of tracheal epithelium in culture. Sarcoidosis 1992; 9:645-648
(5) Boucher RC, Bromberg PA, Gatzy JT. Airway transepithelial electric potential in vivo: species and regional differences. J Appl Physiol 1980; 48:169-176
(6) Takemura H, Tamaoki J, Tagaya E, et al. Isoproterenol increases Cl diffusion potential difference of rabbit trachea through nitric oxide generation. J Pharmacol Exp Ther 1995; 274:584-588
(7) Stutts MJ, Knomles MR, Chinet T, et al. Abnormal ion transport in cystic fibrosis airway epithelium. In: Farmer SG, Hay DWP, eds. The airway epithelium: lung biology in health and disease (vol 55). New York, NY: Marcel Dekker, 1990; 301-334
(8) World Health Organization. Epidemiology of chronic non-specific respiratory diseases. Bull World Health Organ 1975; 52:251-259
(9) Rubin BK, Ramirez O, Zayas JG, et al. Collection and analysis of respiratory mucus from subjects without lung disease. Am Rev Respir Dis 1990; 115:989-995
(10) Welsh MJ, Widdicombe JH, Nedel JA. Fluid transport across the canine tracheal epithelium. J Appl Physiol 1980; 49: 905-909
(11) Ikeda K, Wu D, Takasaka T. Inhibition of acetylcholine-evoked Cl currents by 14-membered macrolide antibiotics in isolated acinar cells of the guinea pig nasal gland. Am J Respir Cell Mol Biol 1995; 13:449-454
(12) Tamaoki J, Isono K, Sakai N, et al. Erythromycin inhibits Cl secretion across canine tracheal epithelial cells. Eur Respir J 1992; 5:234-238
(13) Tamaoki J, Chiyotani A, Tagaya E, et al. Cholinergic control of rabbit tracheal transepithelial potential difference in vivo. Eur Respir J 1996; 9:1632-1636
(14) Tamaoki J, Tagaya E, Sakai A, et al. Effect of macrolide antibiotics on neutrally mediated contraction of human isolated bronchus. J Allergy Clin Immunol 1995; 95:853-859
* From the First Department of Medicine, Tokyo Women's Medical University School of Medicine, Tokyo, Japan.
This work was supported in part by grant 12770305 from the ministry of Education, Science and Culture, Japan.
Manuscript received August 16, 2001; revision accepted January 7, 2002.
Correspondence to: Atsushi Nagai, MD, First Department of Medicine, Tokyo Women's Medical University School of Medicine, 8-1 Kawada-Cho, Shinjuku, Tokyo 162-8666, Japan; e-mail: firstname.lastname@example.org
COPYRIGHT 2002 American College of Chest Physicians
COPYRIGHT 2002 Gale Group