Doxycycline is a long-acting/second generation tetracycline antibiotic, and currently is one of the most commonly prescribed antibiotics in the world, used to treat a wide variety of infectious agents including susceptible intracellular/zoonotic pathogens [ 1–6].

Because doxycycline was introduced prior to an appreciation of pharmacodynamic concepts, the optimal dosing regimen has not been determined. This in vitro study was conducted to determine the optimal dosing regimen for doxycycline based upon pharmacodynamic data.

Doxycycline was tested against selected Gram-positive pathogens, e.g. Staphylococcus aureus and Streptococcus pneumoniae and Gram-negative pathogens, e.g. P. multocida and E. coli. Time-kill studies were performed with each of these organisms at various serum concentrations representing two, four, eight, and 16 times the MIC of each test organism. The growth of the organisms was assessed by colony counts at various time points during a 24 h period to determine if doxycycline kills susceptible organisms by concentration or time-dependent kinetics. Studies were also carried to determine the PAE of doxycycline.

Escherichia coli (ATCC 25922), Staphylococcus aureus (ATCC 25923), Pasteurella multocida, and Streptococcus pneumoniae (ATCC 49619) were used as the test bacteria.

Inocula were prepared using Mueller-Hinton broth for E. coli, and Mueller Hinton broth with 1% Fildes and 20 mm NaCl for S. aureus and P. multocida, and cation-adjusted Mueller-Hinton broth (CAMH) with 5% lysed horse blood (LHB) for S.pneumoniae. The inocula were incubated at 37 °C with shaking until turbid (∼3–4 h). The turbidity was adjusted to 0.5 McFarland standard (representing a concentration of ∼108 CFU/mL). A portion of the 0.5 McFarland standard was diluted 1 : 100 (∼106 CFU/mL) with broth. One milliliter of this dilution was added to each cuvette well containing varying concentrations of doxycycline (0.02–1.6 mg/L). One cuvette well containing no antibiotic was used as a control for organism viability. The cuvettes were placed in an agitator and incubated at 37 °C for 24 h. The lowest concentration of doxycycline that inhibited visible growth for 24 h was recorded as the minimum inhibitory concentration (MIC) of the organism. The MIC for E. coli (ATCC 25922) was 1.5 mg/L, for Staphylococcus aureus (25923) 0.28 mg/L, for P. multocida 0.09 mg/L, and for Streptococcus pneumoniae (ATCC 49619) 0.16 mg/L [ 7,8].

The post-antibiotic effect (PAE) was determined for doxycycline for each of the organisms mentioned above by the method described by Craig and Gudmundsson [ 9]. An overnight growth of E. coli, Staphylococcus aureus, P. multocida, and Streptococcus pneumoniae was diluted into fresh broth, appropriate for each organism, to 106 CFU/mL and then incubated on a shaker at 37 °C for 3–4 h until logarithmic growth phase was achieved. At the end of this period, the inoculum size was determined and each tube containing the test organisms was then exposed to doxycycline at twice the MIC, four times the MIC, and 16 times the MIC for 1 h at 37 °C in a shaker. A suspension of each organism that was not exposed to antibiotics was used as control and was subjected to the same procedures described. Antibiotics were introduced at time zero of antimicrobial exposure. At the end of the exposure period the supernatant was decanted after centrifugation at 1200 g for 10 min and the pellet was re-suspended in fresh broth. The same procedure was repeated and after removal all tubes were again incubated at 37 °C.

Counts of CFU/mL were performed on all cultures at time zero, before and after washing and every hour thereafter until 6 h and the counts of CFU/mL were graphed. The PAE was determined by the following equation: PAE = T - C where T is the time required for the counts of CFU/mL in the test culture to increased one 1 log10 above the count observed immediately after antibiotic removal and C is the time required for the count of CFU/mL in an untreated control culture to increase 1 log10 above the count observed immediately after completion of the same procedure used on the test culture for antibiotic removal [ 9,10]

The data indicate that at low concentrations of doxycycline, i.e. at 2 to 4 times the MIC, inhibition of the organisms tested occurs in a time-dependent fashion. However, at higher concentrations, i.e. 8 to 16 times the MICs of the organisms, doxycycline exhibits concentration-dependent killing ( Figure 1).


Figure 1. Doxycycline kill curves: (a) Staphylococcus aureus; (b) Streptococcus pneumoniae; (c) Pasteurella multocida; (d) Escherichia coli

Download figure to PowerPoint

The PAE of doxycycline was demonstrated for the Gram-positive and Gram-negative organisms tested. Doxycycline's post-antibiotic effect is concentration dependent. The post-antibiotic effect for the Gram-positive and Gram-negative organisms tested is approximately equal ( Figure 2).


Figure 2. Doxycycline post-antibiotic effects (PAEs) (a) Staphylococcus aureus; (b) Streptococcus pneumoniae; (c) Pasteurella maltophilia; (d) Escherichia coli

Download figure to PowerPoint

For serious infections such as Legionnaire's disease, doxycycline therapy should be initiated with a 72 h loading regimen because of its high lipid solubility. A dose of 200 mg intravenously every 12 h is the preferred regimen to provide rapid and high serum/tissue concentration of doxycycline. If doxycycline is administered as a 100 mg intravenous dose every 12 h then 4–5 days of therapy are needed before the patient is fully saturated and a therapeutic effect can be expected. As with other antibiotics, a period of four to five serum half-lives is required before steady-state kinetics are achieved. Since the serum half-life (t1/2) of doxycycline is 22 h, it should be apparent that an initial 72 h loading regimen is required if a rapid therapeutic effect is desired in seriously ill patients [ 1,6]

The post-antibiotic effect of doxycycline differs from other tetracyclines, but is not clinically relevant since therapeutic serum levels are present over the entire duration of the dosing period [ 7,12,13]. Our data suggest that doxycycline may be administered on a 12 h or 24 h dosing basis when a 400 mg (intravenous/oral) daily dose is used. Clinically, with moderate to severe infections, it is important to use a doxycycline loading regimen to rapidly achieve high blood and tissue concentrations. Non-loading dose regimens should be used for patients who are not seriously ill. In terms of pharmacoeconomics, it is most cost effective to administer doxycycline intravenously on a once-daily basis because there is a hospital charge for each intravenously administered dose. A 24 h-dosing regimen eliminates the additional administration cost of a 12 h-dosing regimen.

Doxycycline is the most commonly used tetracycline for the treatment of a wide variety of infectious diseases. Introduced decades before pharmacodynamic considerations were appreciated, the dosing of doxycycline has been based on pharmacokinetic parameters. This is the first pharmacodynamic study of doxycycline.

Doxycycline exhibits time-dependent killing at two to four times the MIC, but dose-dependent killing at eight to 16 times the MIC of the organisms tested. Optimal dose-dependent killing may be achieved using 200 mg (intravenous/oral) every 12 h or 400 mg (intravenous/oral) every 24 h. In mild to moderate infections, time-dependent killing is sufficiently effective and doxycycline may be dosed as 100 mg (intravenous/oral) every 12 h or 200 mg (intravenous/oral) every 24 h.

Doxycycline exerts a PAE which is dose dependent and varies between 2.1 and 4.2 h.

In conclusion, doxycycline kills by time-dependent kinetics at low serum concentrations, but optimally at high serum concentrations by concentration-dependent kinetics. Non-critically ill patients may be treated effectively using either 100 mg (intravenous/oral) every 12 h or 200 mg (intravenous/oral) every 24 h regimen. Based on our in vitro pharmacodynamic data, a high-dose doxycycline regimen, i.e. 200 mg every 12 h or 400 mg every 24 h regimen should provide for optimal concentration-dependent killing. The PAE of doxycycline is clinically unimportant since adequate serum concentrations are maintained for the duration of the dosing interval if given on a 12 h or 24 h basis.


  1. Top of page
  2. Acknowledgments
  3. References

The authors wish to thank Nazli Chaudhry for technical assistance.


  1. Top of page
  2. Acknowledgments
  3. References
  • 1
    Cunha BA. The pharmacokinetics of doxycycline. Postgrad Med Comm1979;1:43 50.
  • 2
    Cunha BA, Comer JB, Jonas M. The tetracyclines. Med Clin North Am1982;66:293.
  • 3
    Cunha BA. Clinical uses of tetracyclines. In:BlackwoodRK, HlavkaJJ, BoothJH, eds. The tetracyclines.Berlin:Springer-Verlag,1985:393.
  • 4
    Klein NC & Cunha BA. Tetracyclines. Med Clin North Am1995;79:789 801.
  • 5
    Cunha BA. The virtues of doxycycline and the evils of erythromycin. Adv Ther1997;14:172 180.
  • 6
    Cunha BA. Current concepts in the antibiotic therapy of legionnaires' disease. Drugs Today1997;33:213 220.
  • 7
    Bundtzen RW, Gerber AU, Cohn DL, Craig WA. Post-antibiotic suppression of bacterial growth. Rev Infect Dis1981;3:28 37.
  • 8
    Craig WA. Pharmacokinetic/pharmacodynamic parameters: rationale for antibacterial dosing of mice and men. Clin Infect Dis1998;26:1 12.
  • 9
    Craig WA & Gudmundsson S. Post-antibiotic effect. In:LorianV. ed.Antibiotics in laboratory medicine,4th edn.Baltimore:Williams and Wilkens,1996:296 329.
  • 10
    Cars O & Odenholt-Tornqvist I. The post-antibiotic sub-MIC effect in vitro and in vivo. J Antimicrob Chemother1993;31:159 166.
  • 11
    Craig WA. Interrelationship between pharmacokinetics and pharmacodynamics in determining dosage regimens for broad-spectrum cephalosporins. Diagn Microbiol Infect Dis1995;22:89 96.
  • 12
    Totsuka K & Shimizu K. In vitro and in vivo post-antibiotic effects (PAE) of CL 31, 928 (DMG-DMDOT), a new glycylcycline against Staphylococcus aureus. In:34th Interscience Conference on Antimicrobial Agents and Chemotherapy.Washington DC:American Society of Microbiology,1994;179:Abstract F114.
  • 13
    Walker R, Andes D, Ebert S, Conklin R, Craig WA. Pharmacodynamic comparison of 6 dimethyl 6-deoxytetracycline (DMDOT) and minocycline in an animal infection model. In:34th Interscience Conference on Antimicrobial Agents and Chemotherapy.Washington DC:American Society of Microbiology,1994;179:(Abstract) F116.