Whooping cough--aspects of pathogenesis and treatment

Pharmacotherapy for Bordetella pertussis infection. I. A synthesis of laboratory sciences

There is considerable history and practice experience both with laboratory susceptibility testing for Bordetella pertussis and clinical treatment. This two-part narrative review provides a synthesis of the laboratory and clinical sciences as they apply to this bacterium and the clinical consequences of treating infection. It is generally held that antibiotic susceptibility testing for B. pertussis is not sufficiently standardised, but there has not been an urgent need to consolidate the same given the lack global experience with major resistance profiles. Experience in China, however, has provided concern for high-level macrolide resistance. The nature of and frequency of such resistance has raised the bar for reconsideration of susceptibility testing given that first-line treatment may be regionally compromised. Disk diffusion and Etest susceptibility testing can be recommended for screening resistance among individual isolates of B. pertussis and on an ad hoc manner. Disk diffusion, Etest and/or critical agar dilution testing can be recommended for large-scale studies. Standards for inoculum, growth atmosphere, timing of interpretation, preferred testing media and controls can be extrapolated from the publications to date. Such methods should be able to detect high-level resistance to several antibiotics, but especially macrolides. Concern for intermediate-susceptible categories requires consideration as well as the correlation with bacteriological and clinical outcomes. Provisional standards can be applied at this time, and modification or fine-tuning of any such standards are open to future investigation.

Pharmacotherapy for Bordetella pertussis infection. II. A synthesis of clinical sciences

Despite the plethora of studies that have examined laboratory susceptibility testing for Bordetella pertussis, assessments of treatment have lagged far behind both in quality and quantity. Macrolides and trimethoprim/sulfamethoxazole historically served the needs of both treatment and prevention, albeit there is still controversy about the degree of protection measured both bacteriologically and clinically. As high-level macrolide resistance has emerged in some geographic regions and since macrolides have been the mainstay of therapy, alternative antibiotics need to be defined for pertussis. In vitro susceptibility testing suggests the potential for several alternatives to macrolides, including trimethoprim/sulfamethoxazole, specific β-lactam agents, chloramphenicol, some quinolones and possibly some tetracyclines. For the latter antibiotics, more clinical studies for treatment and prophylaxis are required in to order to establish bacteriological-clinical correlates for outcome. In the interim, if the clinical circumstances mandate the use of proposed interim alternatives to macrolides, outcomes should be assessed with test of cure by culture, since genetic amplification technologies do not discriminate bacterial viability. Whereas there may be debate in regard to using placebo or macrolides as the controls for alternative antibiotic therapy in geographies where most B. pertussis isolates are antibiotic-susceptible, both placebo and macrolide controls should be assessed along with alternative antibiotics in well-designed controlled studies in regions pressured by macrolide resistance. Outcomes of clinical response and epidemiological patterns of disease should continue to be monitored given the degree of macrolide resistance that is emerging.

Effect of erythromycin treatment on antibody responses in pertussis

The effect of erythromycin treatment on antibody responses to Bordetella pertussis filamentous haemagglutinin (FHA) and pertussis toxin (PT) was investigated in convalescent blood samples from 105 children with pertussis. Erythromycin had been given to 59 children, median age 3.2 years (range 0.3-9.9) on median day 7 (range 11-14) after onset of disease while the remaining 46 children, age 3.45 (0.6-8.1) were untreated. No significant differences in IgG antibody concentration were noted to FHA by ELISA and to PT, neither by NT nor by ELISA, but a tendency towards lower median titers was seen to both antigens and by both type of assays in the erythromycin-treated group. Similarly, early erythromycin treatment i.e. within 7 days of onset of symptoms, did not influence significantly of the development of the antibody responses but tended to lead to lower titer levels. A significant response to PT and FHA was found in about 90% of blood samples, irrespective of treatment. All 56 children with culture-confirmed B. pertussis infection had a significant response to PT. The study has thus shown a slight but not significant effect of erythromycin treatment on antibody responses in pertussis.

Comparison of erythromycin ethylsuccinate and co-trimoxazole for treatment of pertussis

Fifty-five ambulatory children with early culture-proven pertussis were treated for two weeks either with erythromycin ethylsuccinate (n = 28) (50-80 mg/kg/day in three doses during meals) or with co-trimoxazole (n = 27) (6-10 mg trimethoprim/kg/day in two doses after meals). After completion of treatment, all patients in the erythromycin group were culture-negative, while in the co-trimoxazole group one child was still culture-positive. In this case vomiting may have played a role. Both agents appear to be able to eradicate Bordetella pertussis from the nasopharynx of patients with early whooping cough.

Treatment and prevention of pertussis by antimicrobial agents (Part II)

Suitable antimicrobials given during the catarrhal stage of whooping cough can attenuate the course of the disease. The efficacy of antibiotics administered prophylactically during the incubation period remains controversial but appears to be beneficial. Currently, erythromycin given for two weeks is the antibiotic of choice for pertussis. No treatment failures were observed with erythromycin estolate. Erythromycin ethylsuccinate and stearate must be given at high dosages (50-60 mg/kg/day) in order to achieve sufficient concentrations in the respiratory secretions. With ampicillin and amoxicillin treatment failures have been observed. The role of josamycin and co-trimoxazole in pertussis remains open.

Bordetella pertussis infection: a cause of persistent cough in adults

An enzyme-linked immunosorbent assay (ELISA) for Bordetella pertussis-specific immunoglobulin (Ig)A antibody in serum was used to demonstrate B. pertussis infection. The upper limit of normal for the assay (mean + 3 SD) was established by testing sera that had been collected from a group of healthy blood donors. All 13 patients with clinical whooping cough from whom B. pertussis was isolated and 25.7% of 218 adults, who were aged 18-81 years, who were referred to a consultant physician for the investigation of a persistent cough, had elevated levels of IgA to B. pertussis. The study confirms the B. pertussis IgA ELISA as a sensitive test for the diagnosis of B. pertussis infections and that these infections cause respiratory illness, in particular, persistent cough in adults.