[Would ampicillin alone again be enough for treating community-acquired pneumonias?]

MORE THAN SEMANTICS: The complexity of the problem raised by this apparently simple question becomes apparent if one examines the meanings behind the words.

Globally, the frequency of pathogens known to cause community-acquired pneumonia has changed little over time: Streptococcus pneumoniae, Haemophilus influenzae, Mycoplasma pneumoniae, Chlamydia pneumoniae, Legionella pneumoniae. The real question is the emergence of S. Pneumoniae resistant strains. Currently, the frequency of penicillin or ampicillin resistant strains (less then 10%) is not associated with an increased rate of treatment failure or mortality. "High" doses (150-200 MU/kg/d for penicillin or 2 g t.i.d. for ampicillin) can still be expected to control suspected or certain resistant pneumococcal pneumonia.

Would other antimicrobials be more effective? In France, "consensual" use of amoxicillin (1 g t.i.d.) "covers" 98% of the situations as resistant strains are found in only 2% of the adult cases. Thus, the lower frequency of pneumococcal resistance to cefotaxime or ceftriaxone (less than 0.5%) would not warrant their use in most cases of community-acquired pneumococcal pneumonia. Inversely, combination regimens with ampicillin or amoxicillin and cefotaxime or ceftriaxone or the amoxicillin-clavulanic acid association at standard doses provide good therapeutic alternatives for empirical prescriptions in cases with a high risk of infection other than pneumococcal infection (H.influenzae, methicillin-sensitive Staphylococcus aureus). The remarkable anti-pneumococcal activity of imipenem, vancomycin and teicoplanin should, for the time being, be reserved for selected patients with cross allergic reactions to penicillins and cephalosporins.

In case of severe community-acquired pneumonia, a combination regimen should be the rule in order to "cover" as many bacteria as possible. In the "healthy" adult under 60 years of age, amoxicillin combined with a macrolide or a fluoroquinolone can be recommended.

Bacteriological efficacy is not synonymous with cure. Secondary inflammatory processes and subsequent septic shock remain a major challenge.