Intended for healthcare professionals

Practice Rational Testing

Investigation of “non-responding” presumed lower respiratory tract infection in primary care

BMJ 2011; 343 doi: (Published 13 October 2011) Cite this as: BMJ 2011;343:d5840
  1. James D Chalmers, clinical research fellow ,
  2. Adam T Hill, consultant physician
  1. 1Department of Respiratory Medicine, Royal Infirmary of Edinburgh, Edinburgh EH16 4SA, UK
  1. Correspondence to: J Chalmers jamesdchalmers{at}

Most lower respiratory tract infections in primary care are self limiting, and laboratory and other investigations should be reserved for patients with specific risk factors

Learning points

  • Among patients prescribed an antibiotic for lower respiratory tract infection in primary care, 20-25% will present again within four weeks

  • Symptoms of lower respiratory tract infection usually resolve without specific treatment or investigation in a median of three weeks from first consultation

  • In most patients laboratory and other investigations are unlikely to be helpful and should be guided by specific risk factors

  • Microbiology testing is of little value in most patients. Sputum culture is recommended for patients with chronic lung disease with sputum production and for patients with suspected pneumonia who are not responding to initial therapy

  • Chest radiography is most useful if an alternative diagnosis (such as malignancy) is being considered

A 49 year old woman presents to her general practitioner for the second time with a cough. She first presented a week ago with a four day history of cough with purulent sputum. She reported fever but had no haemoptysis, chest pain, or recent weight loss. She had never smoked. Clinical examination was unremarkable. She was prescribed a one week course of amoxicillin 500 mg three times daily. She has presented again, reporting that the cough has not improved.

What is the next investigation?

The problem

Lower respiratory tract infections managed in primary care comprise a spectrum of disorders including acute bronchitis, community acquired pneumonia, and exacerbations of airways disease. Most lower respiratory tract infections are self limiting, and several studies in the United Kingdom and internationally suggest that antibiotic treatment does not significantly reduce recovery time.1 2 Historically, sputum purulence was regarded as an indicator of bacterial infection and an indication for antibiotic treatment. Recent data from over 3000 patients with acute cough in primary care showed that general practitioners were much more likely to prescribe antibiotics to patients with purulent sputum (odds ratio 3.2, 95% confidence interval 2.1-5.0) but that antibiotic treatment was of no benefit in terms of symptomatic improvement, regardless of sputum colour.3 Current data suggest that 50-70% of patients presenting in primary care with acute cough will receive an antibiotic at first presentation.3 4

A major clinical problem is what to do when a patient presents again with symptoms of lower respiratory tract infections that are not responding to the initial antibiotic. Evidence suggests that 20-25% of patients will return within four weeks of the initial consultation4 and that 60% of such patients will receive a further course of antibiotics.4 In many cases the patient will expect to undergo investigation and receive a further antibiotic course.5

Symptoms of lower respiratory tract infections are often slow to resolve, with the elderly and patients with chronic lung diseases often having the slowest recovery. Methods of assessing symptom recovery vary, but most suggest that the symptoms can be expected to last for two to three weeks. In a recent randomised controlled trial the median time to clinical recovery was 22 days from first consultation.6 In a large observational study the median time to patients feeling recovered was 11 days and the median time to normalisation of symptom scores was 15 days.2 Questionnaire studies suggest that 90% of patients still have symptoms of cough at two weeks and that 80% still have some symptoms at three weeks after consultation.7 Antibiotic treatment has not been shown to reduce this duration of symptoms.1 2

It is therefore not surprising that our patient has persistent symptoms one week after her consultation.

Recommended investigations

The first priority in assessment is to exclude community acquired pneumonia requiring hospital admission. The most important assessments can be made without laboratory or radiological investigations. Clinical history plus physical examination, pulse oximetry, and blood pressure measurement can identify most of the “alarm” signs and symptoms that might need further investigation or hospital admission (box 1).

Box 1 “Alarm” signs and symptoms indicating need for further investigation or admission to hospital

British Thoracic Society CRB65 score8
  • Confusion

  • Respiratory rate ≥30 breaths/min

  • Hypotension (systolic blood pressure <90 mm Hg and/or diastolic blood pressure ≤60 mm Hg

  • Age ≥65*

In patients with pneumonia, one or more of the above features suggest the need for hospital admission8

Other markers of severity
  • Hypoxaemia (oxygen saturations <92% on air)

  • Inability to take oral medication, or vomiting

  • Dehydration

  • Presence of major comorbidities

Features suggestive of pneumonia89
  • Physical signs on chest examination (crackles, decreased breath sounds, or dullness to percussion)

  • A short duration of symptoms (<24 hours)

  • Fever >38°C

  • Tachycardia >100 beats/min

  • Tachypnoea >20 breaths/min

  • Absence of sore throat and rhinorrhoea

  • *Although age is not in itself an indication for hospital admission, complications and mortality increase substantially in patients aged 65 and over, therefore a lower threshold for investigation or admission to hospital is advised

The British Thoracic Society recommends reassessment of patients with suspected pneumonia managed in the community within 48 hours, with those showing evidence of deterioration being referred to hospital.8

Chest radiography

Chest radiography at a second consultation (after antibiotic treatment for lower respiratory tract infection in primary care) will show consolidation in 10-15% of patients.4 However, the presence of radiographic changes is not an indication for additional antibiotic treatment if the patient has already received an appropriate antibiotic. Radiographic changes tend to lag behind clinical and microbiological resolution and are not sensitive as a way of monitoring treatment response. The main indications for chest radiography are to exclude non-infective diagnoses in high risk groups (box 2).

Box 2 Common indications for investigations in patients presenting again with non-resolving lower respiratory tract infections in the community

Chest radiography
  • Signs and symptoms of pneumonia (box 1)8

  • Haemoptysis

  • Pleuritic chest pain

  • A history of cigarette smoking

  • Symptoms suggestive of underlying malignancy, such as weight loss

  • Clinical signs of pleural effusion

  • Immunosuppression or risk factors for tuberculosis

Sputum culture
  • Pneumonia not responding to initial antibiotic treatment8

  • Patients with chronic obstructive pulmonary disease, bronchiectasis, or immunodeficiency10 11

  • Patients with risk factors for tuberculosis*

Testing for atypical organisms or viruses
  • Most appropriate during outbreaks, such as influenza A (H1N1) virus, Mycoplasma pneumoniae8

Laboratory blood testing
  • Full blood count (rarely helpful)

  • Urea and electrolytes (in chronic renal disease or when clinical concern exists over dehydration)

  • Liver function tests (in chronic liver disease)

  • C reactive protein (can be used to monitor response to treatment but unnecessary in most patients in primary care)6

  • *For countries with a low incidence of tuberculosis. In regions or countries with a high incidence of tuberculosis, sputum examination for acid fast bacillus should be performed more frequently.


In our case, is laboratory investigation necessary? Data from observational studies conducted in primary care suggest that it is unusual to grow a bacterial pathogen that will require further antibiotic treatment in patients presenting again with a lower respiratory tract infection. In one study, when extensive investigations (including sputum induction with nebulised saline, urinary antigen testing, paired serology testing, and polymerase chain reaction testing on throat swabs) were used, only 4% of patients had a “typical” bacterial pathogen grown, and the most frequent isolates were respiratory viruses, which would not prompt a change in treatment.4

Most patients with a lower respiratory tract infection in the community can be managed without microbiology investigations. Little evidence exists to support the use of sputum cultures in primary care. Unless specimens can be rapidly transported to the laboratory, fastidious organisms such as Haemophilus influenzae may die or be overgrown by normal flora. National guidelines therefore recommend limiting the use of sputum culture to patients with suspected pneumonia who are not responding to initial antibiotic treatment.8 Airways of patients with chronic obstructive pulmonary disease or bronchiectasis are often colonised with pathogens such as H influenzae and Pseudomonas aeruginosa. These organisms may be resistant to first line treatment, and sputum culture is therefore recommended in these groups. In patients with chronic bacterial colonisation, these organisms may continue to be grown in sputum cultures, even when the patient is clinically stable. A positive sputum culture in the absence of symptoms of an ongoing exacerbation does not require additional antibiotic treatment.10 11

In an area with a low incidence of tuberculosis, sputum examination for acid fast bacillus is not needed but should be considered in patients with risk factors for the disease (such as coming from countries with a high incidence of tuberculosis; homelessness; alcoholism; being elderly or immunosuppressed), particularly if there are associated symptoms such as night sweats, weight loss, or haemoptysis.

Urine antigen testing for Streptococcus pneumoniae is increasingly available but is unlikely to change management in our case as antibiotics effective against S pneumoniae have been prescribed. Throat swabs taken for polymerase chain reaction to identify respiratory viruses or atypical pathogens (such as Mycoplasma pneumoniae) are not considered routine investigations but may be appropriate in the right context, such as testing for H1N1 influenza during the pandemic of 2009.

Blood tests

Full blood count and biochemistry tests are unlikely to be helpful in patients presenting again with a lower respiratory tract infection. A raised leukocyte count is non-specific and is not a reliable indicator of ongoing infection. Point of care testing for C reactive protein has been shown to be a useful adjunct to clinical decision making in primary care.6 However, these tests are not widely available in primary care and are not needed for most patients. Testing of renal or liver function are unnecessary in the absence of chronic organ dysfunction or dehydration.


After examining the patient again, the general practitioner advised the patient about the natural course of acute cough and said that a duration of two to three weeks for acute cough to resolve was not unusual. The general practitioner explained that for people with her symptoms and signs, antibiotics had about as much chance of helping her marginally as causing her some harm.12


Cite this as: BMJ 2011;343:d5840


  • This series of occasional articles provides an update on the best use of key diagnostic tests in the initial investigation of common or important clinical presentations. The series advisers are Steve Atkin, professor, head of department of academic endocrinology, diabetes, and metabolism, Hull York Medical School; and Eric Kilpatrick, honorary professor, department of clinical biochemistry, Hull Royal Infirmary, Hull York Medical School. To suggest a topic for this series, please email us at practice{at}

  • Contributors: Both authors contributed to drafting and revising the manuscript and approved the final version.

  • Competing interests: All authors have completed the ICMJE uniform disclosure form at (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

  • Provenance and peer review: Commissioned; externally peer reviewed.

  • Patient consent not required (patient anonymised, dead, or hypothetical).


View Abstract