Blind Treatment of Bacterial Infection

This PatientPlus article is written for healthcare professionals so the language may be more technical than the condition leaflets. You may find the abbreviations list helpful.

If a bacterial infection is suspected, it is often impracticable to wait for test results before starting treatment. Selecting the 'best guess' antibiotic should be guided by the following principles:[1]

  • Use antibiotics responsibly, considering issues such as safety, resistance and cost.
  • Check that an antibiotic is really needed - history and examination may yield clues as to whether a condition is bacterial or viral, but this is not always easy. Consider delayed antibiotics. Some viral conditions may need prophylaxis to prevent secondary bacterial overgrowth - eg, acute necrotising ulcerative gingivitis secondary to herpes simplex infection.
  • Blind prescribing does not obviate the need to take samples for culture and sensitivity, before starting treatment, whenever appropriate.[2] Depending on the clinical picture, this may include skin or wound swabs, high vaginal swabs, endocervical swabs, urine, faeces, sputum, blood, aspirate. In the hospital environment, consider cerebrospinal fluid. Where clinically appropriate, consider FBC, ESR, CRP, U&Es, LFTs, clotting, atypical serology, malaria film, serum for virology, CXR, and arterial blood gas analysis. Perform urinalysis.
  • Blind antibiotic prescribing for pyrexia of unknown origin (PUO) in a relatively well and stable patient is rarely helpful.
  • Calculating dosage is not an exact science but consider factors affecting absorption or bioavailability, such as age, weight, hepatic function, renal function, severity of infection and other medication:
    • Underdosing may result in significant failure of treatment and bacterial resistance in serious infection.
    • An excessive dose may result in toxicity, particularly for antibiotics with a narrow margin between the toxic and therapeutic dose (eg, an aminoglycoside).
    • Consider drug plasma monitoring, although this is difficult in primary care and may be more appropriate in an intermediate care setting.
  • Route of administration - most patients in primary care will cope with oral antibiotics, although some patients have difficulty swallowing tablets and may need liquid or dispersible preparations. Serious infections may require intravenous (IV) administration. Avoid intramuscular (IM) antibiotics in children, as these are likely to be painful.
  • Duration depends on condition and severity. Chronic infections such as tuberculosis may require prolonged treatment.
  • Follow local policy and national guidelines.[3]
  • Consider any other factors relating to the patient which are likely to be relevant - eg, ethnicity, history of allergy, whether immunocompromised, severity of condition, and whether taking other medication.
  • If female:
    • Check whether pregnant, breast-feeding or taking an oral contraceptive.
    • In pregnancy avoid tetracyclines, aminoglycosides, quinolones, high-dose metronidazole.
    • Short-term use of trimethoprim (there is a theoretical risk in the first trimester in patients with a poor diet, as it is a folate antagonist) or of nitrofurantoin (at term, there is a theoretical risk of neonatal haemolysis) is unlikely to cause problems.
  • Prescribing antibiotics after a telephone consultation should be the exception rather than the rule.
  • Choose simple generics first-line unless there is a very good case for using newer more expensive antibiotics.
  • Avoid widespread use of topical antibiotics, especially those readily used in oral forms, as this may spread resistance.
  • Clarithromycin is an acceptable alternative in patients who get gastrointestinal side-effects with erythromycin.
  • If blind treatment fails and test results are not available, check with a microbiologist.

Choosing the right drug in the absence of sensitivity results is an inexact science at the best of times but should be guided by the following principles:

  • History:
    • A detailed history may reveal the source of infection.
    • Ask about respiratory, gastrointestinal or genitourinary symptoms.
    • Ask about recent travel or treatment or conditions which could compromise the immune system.
  • Examination - check vital signs: temperature, pulse, blood pressure, respiratory rate and capillary return, to assess the severity of illness and signs of septicaemia.
  • Treatment:
    • After 'best guessing' the source of infection, follow local guidelines.
    • If there are none, use the guidance from the Health Protection Agency (HPA).[3]
    • Be ready to change treatment once drug sensitivities are known.
    • Treatment of most infections should not exceed seven days.
    • In a hospital or intermediate care setting, IV antibiotic therapy is usually reviewed after 48 hours and changed to oral preparations when possible.
    • If in doubt, ask a microbiologist.

Blind Treatment of Infection

TonsillitisMost sore throats are viral, but if bacterial tonsillitis is suspected:
  • Phenoxymethylpenicillin 500 mg QDS or 1 g BD for 10 days.
  • If allergic to penicillin, clarithromycin 250 mg-500 mg BD QDS for 10 days.
Otitis media in childhoodMany are viral - 80% resolve without antibiotics. If clinically appropriate:
  • Amoxicillin first-line - 40 mg/kg/day in three divided doses.
    Maximum 1 g TDS for 5 days.
  • Co-amoxiclav second-line for resistant Haemophilus influenzae - 1-6 years 156 mg TDS; 6-12 years 312 mg TDS.
  • If allergic to penicillin:
    • Erythromycin first-line - under 2 years 125 mg QDS; 2-8 years 250 mg QDS; other 250-500 mg QDS.
    • Azithromycin second-line -15-25 kg: 200 mg OD; 26-35 kg: 300 mg OD.
  • Reserve antibiotics for severe, or where symptoms are lasting more than 10 days. 69% resolve without antibiotics, and 84% resolve with antibiotics.
  • First-line - amoxicillin 500 mg TDS for 7 days or doxycycline or 200 mg stat/100 mg OD for 7 days or clarithromycin 250 mg/500 mg BD for 7 days or phenoxymethylpenicillin 250 mg QDS/500 mg BD for 7days.
  • Second-line - co-amoxiclav 625 mg TDS for 7 days or ciprofloxacin 250-500 mg BD plus metronidazole 400 mg TDS for 7 days.
Acute bronchitis/ lower respiratory tract infectionMarginal benefits in otherwise healthy adults. Patient leaflets can reduce antibiotic use.
  • Amoxicillin 500 mg TDS or or doxycycline 200 mg stat/100 mg OD for 5 days.
  • If there is penicillin allergy and tetracycline is contra-indicated, use erythromycin.
Acute exacerbation of chronic obstructive pulmonary disease30% are viral, 30-50% bacterial, and the rest undetermined.
Use antibiotics if there is increased dyspnoea and increased purulence of sputum volume. In penicillin allergy use clarithromycin if doxycycline is contra-indicated.
First-line - amoxicillin 500 mg TDS or doxycycline 200 mg stat/100 mg OD for 5 days:
  • If allergic to penicillin use clarithromycin 500 mg BD for 5 days.
  • Second-line - co-amoxiclav 625 mg TDS.
Community-acquired pneumoniaStart antibiotics immediately.
  • First-line: amoxicillin 500 mg-1 g TDS or clarithromycin 500 mg BD or doxycycline 200 mg stat/100 mg OD for up to 10 days.
  • If there is no response in 48 hours, consider admission or add clarithromycin to amoxicillin to cover mycoplasma (rare over 65) or oxytetracycline 250-500 mg QDS for up to 10 days.
  • If severely ill, give parenteral benzylpenicillin prior to admission. 
  • Consider risk factors for Staphylococcus aureus and Legionella spp.
MeningitisAdmit to hospital immediately.
  • Benzylpenicillin prior to admission, unless there is history of anaphylaxis, (NOT allergy). Ideally, IV but IM if a vein cannot be found.
  • Adults and children 10 years and over: 1200 mg.
  • Children 1-9 years: 600 mg.
  • Children under 1 year: 300 mg.
Uncomplicated urinary tract infection  (UTI) - ie no fever or flank pain
  • Amoxicillin resistance is common; therefore, ONLY use it if culture confirms susceptibility.
  • In the elderly (>65 years), do not treat asymptomatic bacteriuria; it occurs in 25% of women and 10% of men and is not associated with increased morbidity.
  • In the presence of a catheter, antibiotics will not eradicate bacteriuria; only treat if systemically unwell or if pyelonephritis is likely.
  • Do MSU on all treatment failures - extended-spectrum beta-lactamase enzyme-producing organisms increasing multiple resistance but still sensitive to nitrofurantoin.
Uncomplicated UTI (no fever or flank pain):
  • Use urine dipstick to exclude UTI negative nitrite and leukocyte 95% negative predictive value.
  • First-line - trimethoprim 200 mg BD PO or nitrofurantoin 50-100 mg QDS for 3 days in women, and for 7 days in men.
  • Second-line - depends on susceptibility of the organism isolated - eg, nitrofurantoin, amoxicillin, cefalexin, co-amoxiclav, quinolone, pivmecillinam.
  • Extended-spectrum beta-lactamase enzymes are increasing but often remain sensitive to nitrofurantoin.
UTI in pregnancy:
  • Send MSU for culture. Short-term use of trimethoprim or nitrofurantoin in pregnancy is unlikely to cause problems to the fetus. Also give folate if trimethoprim is used in the first trimester.
  • First-line - nitrofurantoin 100 mg MR BD or amoxicillin 500 mg TDS (if organism susceptible) for 7 days.
  • Second line - trimethoprim 200 mg BD for 7 days.
  • Third-line - cefalexin 500 mg BD or amoxicillin 250 mg TDS for 7 days.
Children (see also separate article Childhood Urinary Tract Infection):
  • Lower UTI: trimethoprim or nitrofurantoin or amoxicillin (if susceptible); second-line - cefalexin.
  • Upper UTI: co-amoxiclav; second-line - cefixime.
Recurrent UTI in women for 3 years or more:
  • Postcoital prophylaxis is as effective as prophylaxis taken nightly.
  • For prophylaxis use nitrofurantoin 50 mg stat postcoital or trimethoprim 100 mg OD at night.
Skin/soft tissue infectionsImpetigo:
  • Topical/oral produce similar results - reserve topical for very localised lesions to obviate resistance.
  • Reserve mupirocin for meticillin-resistant S. aureus (MRSA).
  • First-line oral medication - flucloxacillin 500 mg or clarithromycin 250-500 mg BD for 7 days.
  • Topical - use fusidic acid TDS for 5 days.
  • Using antibiotics, or adding them to steroids, in eczema does not improve healing unless there are visible signs of infection.
  • Flucloxacillin 500 mg QDS for 7 days. Use clarithromycin or clindamycin if there is allergy to penicillin.
  • If febrile and ill, admit for IV treatment.
  • If there has been water exposure, consult a microbiologist.
  • In facial cellulitis, use co-amoxiclav 625 TDS 7-14 days.
Leg ulcers:
  • Bacteria will always be present. Antibiotics do not improve healing. Culture swabs and antibiotics are only indicated if there is evidence of clinical infection such as inflammation/redness/cellulitis; increased pain; purulent exudate; rapid deterioration of the ulcer or pyrexia.
  • If there is active infection, use flucloxacillin or clarithromycin as for cellulitis.
  • Review antibiotics after culture results. Refer to specialist if there is severe infection.
  • Diabetic leg ulcers - co-amoxiclav 625 mg TDS for 7 days; review, and refer to a specialist in severe cases.
Animal bites:
  • Surgical toilet most important.
  • Assess tetanus and rabies risk.
  • Antibiotics should be considered for:
    • A puncture wound.
    • A bite involving a hand, foot, face, joint, tendon, or ligament.
    • The immunocompromised, those with diabetes, the elderly, and those who are asplenic.
  • First-line is co-amoxiclav 375-625 mg TDS for 7 days.
  • If there is allergy to penicillin, use metronidazole 200 mg-400 mg TDS PLUS doxycycline 100 mg BD.
Human bites:
  • Antibiotic prophylaxis is advised.
  • Assess HIV/hepatitis B and C risk.
  • Co-amoxiclav as for animal bites. If allergic to penicillin, use metronidazole 200-400 mg TDS PLUS doxycycline 100 mg BD or clarithromycin 250-500 mg BD.

NB: doses are for adults unless otherwise stated - for further details see the BNF.

The table is a brief summary. Guidance changes from time to time depending on prevailing antibiotic sensitivities. Check the HPA information for the most up-to-date guidance and details of the management of specific conditions.[3]

Further reading & references

  1. Cunha BA; Fever of unknown origin: focused diagnostic approach based on clinical clues from the history, physical examination, and laboratory tests. Infect Dis Clin North Am. 2007 Dec;21(4):1137-87, xi.
  2. British National Formulary; 64th Edition (Sep 2012) British Medical Association and Royal Pharmaceutical Society of Great Britain, London
  3. Primary Care Guidance; Public Health England

Disclaimer: This article is for information only and should not be used for the diagnosis or treatment of medical conditions. EMIS has used all reasonable care in compiling the information but make no warranty as to its accuracy. Consult a doctor or other health care professional for diagnosis and treatment of medical conditions. For details see our conditions.

Original Author:
Dr Laurence Knott
Current Version:
Peer Reviewer:
Dr Adrian Bonsall
Document ID:
454 (v5)
Last Checked:
Next Review: