Article Text
Abstract
Respiratory tract infections (RTI) are among the most common acute conditions leading to GP consultations and to antibiotic prescribing in primary care, even though 70% are viral, and many others are minor self-limiting bacterial infections.1-4 Between 0.5% and 1.1% of adults have community-acquired pneumonia every year in the UK, most of whom are managed in primary care.4,5 The decision to prescribe antibiotics for an acute RTI in primary care is often based on clinical symptoms, which have low sensitivity and specificity, and high inter-observer variability.2,4 In primary care, it is very difficult to differentiate between diagnoses without additional tests.6 Unnecessary antibiotic prescribing may not aid recovery, exposes patients to potential adverse effects, may encourage repeat attendance and contributes to antibiotic resistance.2,7 One strategy aiming to reduce antibiotic prescribing in primary care is the use of biomarkers (e.g. C-reactive protein [CRP]).2 In the correct clinical context (e.g. in previously healthy people, not those with chronic lung disease) and as an adjunct to clinical assessment, a biomarker may help in the management of an RTI.2 In order to be used during the consultation, the results of a biomarker test must be rapidly available (e.g. ‘point-of-care’ [POC] testing).4 POC testing for CRP has recently been recommended as part of a national clinical guideline on the diagnosis and management of pneumonia.4 Here, we review the rationale for POC CRP testing and its advantages and disadvantages.
Statistics from Altmetric.com
Background
Guidance from the National Institute for Health and Care Excellence (NICE) on the management of RTIs recommends:3
advising patients that antibiotics may have no impact on symptoms in suspected viral infections, and may have adverse effects;
offering delayed prescribing of antibiotics if symptoms do not settle or get significantly worse (providing a ‘safety net’ for the small proportion of patients who develop a complication); and
offering an immediate antibiotic prescription if patients have symptoms and signs suggestive of serious illness and/or complications (e.g. pneumonia).
In patients with suspected pneumonia, a CRP test can help identify patients most likely to benefit from antibiotics.2,6 Although CRP tests are available through local hospital pathology services, a POC test potentially provides quicker results in general practice at the time of the consultation.4
CRP testing for pneumonia
CRP is a non-specific marker that is released 4–6 hours after an inflammatory trigger, with levels peaking after 36 hours and with a half-life of 4–7 hours.8 The serum concentration of CRP is usually <3mg/L but can increase to around 40mg/L for rhinovirus and 100mg/L for influenza and adenoviral infections; values >100mg/L are frequently found in patients with pneumonia.9
What do the results mean?
In a primary care study involving 2,820 patients presenting with cough in winter, of whom 140 (5%) had pneumonia confirmed on x-ray, the CRP level (in a laboratory test) correlated with the likelihood of pneumonia. The proportion of patients with pneumonia was 3% when CRP was <20mg/L (55/2,094 people), 5% at 20–30mg/L (11/227 people), 7% at 30–50mg/L (16/246 people), 15% at 50–100mg/L (24/159 people) and 35% at >100mg/L (34/96 people).10 The negative predictive value of the test (i.e. the chance of not having pneumonia if the CRP level was <20mg/L) was 97.4%. Signs and symptoms were useful in correctly identifying patients with a ‘low’ (<2.5% risk) or ‘high’ (>20%) diagnostic risk in only 26% of patients. In the 74% of patients in whom diagnostic doubt remained (estimated risk 2.5–20%), measurement of CRP concentration helped to correctly exclude pneumonia.
What is the POC CRP test?
A small sample of blood is taken from the finger and combined with a reagent containing anti-CRP antibodies.8 The antibodies form a complex with serum CRP, and the resulting complexes can be measured using an automatic reader. Results are available within 2–10 minutes. POC CRP tests can accurately detect CRP levels >8mg/L.7
How does the POC CRP test perform in a typical primary care population?
In a nested case control study of adult patients presenting with acute cough in 12 European countries, venous blood samples from 100 patients with and 100 patients without pneumonia were tested with five different POC CRP tests and a laboratory analyser.11 All the POC CRP tests and the laboratory analysis produced similar results and were deemed to be equally suitable in predicting radiographic pneumonia as a single test and when added to signs and symptoms.
Does POC CRP help reduce antibiotic prescribing?
A Cochrane systematic review2 of POC biomarker tests to guide antibiotic prescribing in patients with acute RTIs in primary care found six randomised controlled trials (RCTs) of CRP tests,12-17 including a total of 3,284 participants (of whom 139 were children) with suspected upper or lower RTI. Overall, fewer antibiotics were prescribed in the CRP group (631/1,685 [37%]) compared with the standard-of-care group (785/1,599 [49%]).2 There was no significant difference between CRP and standard-of-care groups in clinical recovery (at least substantial improvement at day 7 and 28 or need for re-consultations at day 28) or hospitalisations.2
In a more recent study in the Netherlands involving 939 patients with acute cough, GPs provided care as usual (including diagnostic work-up [with possible POC CRP testing] and management).18 Of the patients for whom GPs requested a POC CRP test, 46% were at low risk, 13% at high risk and 41% intermediate risk of complicated lower RTI. Those at intermediate risk are the only ones for whom the Dutch Acute Cough guideline recommends testing (i.e. testing was often not conducted in accordance with the guidelines).18 In 31% of the patients with intermediate risk, the POC CRP test result prompted a change in prescribing decision, but this was in line with the guidelines in only 60% of patients. Prior to the test, GPs intended to prescribe antibiotics for 31% of patients overall; after the test, this reduced to 28% (not significantly different). The authors suggested that the prescribing rate in the study was low in comparison to international data, which may limit the potential of POC CRP testing to reduce antibiotic prescribing among GPs who have a low baseline prescribing rate.
It is worth noting that some of these studies were funded by the manufacturers of the POC testing devices.14,18
Is POC CRP testing cost effective?
One trial13,19 involving 431 patients with suspected lower RTIs in general practice included an economic evaluation over 28 days.20 The groups were: usual care, POC CRP test, GP communication skills training or both interventions.20 The total mean cost was €35.96 per patient in the usual care group (antibiotic prescribing rate of 68%); €25.61 per patient in the GP communication group (antibiotic prescribing rate 33%); €37.58 per patient with the CRP test (antibiotic prescribing rate of 39%); and €37.78 per patient in the combination of communication skills and POC test group (antibiotic prescribing rate of 23%). Reconsultation rates, symptom scores and patient-reported time to recovery were similar for all groups, despite lower prescribing rates in the intervention groups. The authors suggested that this study showed that training in communication skills and clinician use of CRP POC tests, as well as the combination of both interventions, are cost-effective strategies to reduce antibiotic prescribing for lower RTI in primary care.20
Potential advantages of the test
The advantage of POC testing is the immediate availability of the result, allowing its implications to be discussed during the consultation.4,21
In studies funded by the manufacturer, GPs perceived that the POC CRP test was useful in persuading patients to agree to a non-antibiotic approach.21,22
In a study (funded by the pharmaceutical industry) of 120 patients with upper or lower RTI in an Irish general practice, patients' expectation of receiving an antibiotic prescription was lower in the CRP group than the no CRP group (36.6% vs. 48.3%) with no reduction in patient satisfaction (85% in each group were satisfied or very satisfied).23
Potential disadvantages of the test
As POC testing for CRP is not widely used in UK primary care at present, its introduction would require a change in care pathways and clinical practice.4 As such, there would be significant costs associated with purchasing and maintaining equipment, training staff and regular quality assurance of equipment.
The ‘real-life’ experience of these tests outside the trial setting is limited, and it is unclear whether the benefits would translate as well across all practices and individual prescribers.4 In particular, the rate of antibiotic prescribing varies widely between practitioners, hence the reduction in overall antibiotic use (the main benefit of POC CRP testing) would also vary accordingly.
Other considerations include the impact on the length of the consultation, handling errors (some GPs had problems using the semi-automated device), false-positive values that may increase inappropriate antibiotic use, or false-negative values (e.g. if the illness was still in its early stages and the CRP might not be initially raised) that could result in a lack of necessary antibiotic treatment.2,9,21,24
Also, the POC CRP test does not help to educate patients about the appropriate use of antibiotics in general, or specifically address patient expectations for the management of self-limiting infections.22
Increased reliance on diagnostic tests may restrict holistic patient care and cause overuse for ‘reassurance’ of a normal CRP, or sequential testing following a normal or borderline result if symptoms were not resolving.21
Key national recommendations on use of near patient CRP test for LRTI
Based on three RCTs reported in four papers,13,14,16,19 comparing CRP with usual care and an economic analysis,5 the NICE guideline on pneumonia suggested that:4
using CRP to assist antibiotic prescribing decisions at index consultation for people presenting in primary care with LRTI can reduce antibiotic treatment;
antibiotic prescribing at 28 days of follow-up in the group using CRP may be substantially reduced compared with those receiving usual care;
no difference was found in hospital admission between patients who received CRP and those who received usual care; and
POC CRP testing was cost effective compared with clinical assessment alone (incremental cost effectiveness ratio was £15,763 per quality-adjusted life year [QALY] gained).
No evidence was found about differences between the CRP and usual care groups for the outcomes of mortality, reconsultation, symptom resolution and quality-of-life.4
The guideline, therefore, recommended that for people presenting with symptoms of lower RTI in primary care, a POC CRP test should be considered if after clinical assessment a diagnosis of pneumonia has not been made and it is not clear whether antibiotics should be prescribed.4 The results of the CRP test should be used to guide antibiotic prescribing in people without a clinical diagnosis of pneumonia as follows:
Do not routinely offer antibiotic therapy if the CRP concentration is <20mg/L (due to a low risk of confirmed pneumonia in this group).
Consider a delayed antibiotic prescription (a prescription for use at a later date if symptoms worsen) if the CRP concentration is between 20 and 100mg/L (intermediate risk of confirmed pneumonia).
Offer antibiotic therapy if the CRP concentration is >100mg/L (high risk of confirmed pneumonia).4
It should be noted that, if antibiotic therapy is thought to be indicated following clinical assessment alone, with comorbidity and other factors taken into consideration, then CRP testing would not be necessary.4 The CRP test should be considered only when there is doubt about the need for antibiotics after a clinical assessment (i.e. not for patients with a clear clinical diagnosis of pneumonia, who should receive antibiotic treatment as soon as possible; and not for patients with a clinical diagnosis of a self-limiting illness, who would not need antibiotic therapy).4
Adopting the test in practice
Successful adoption models in European countries showed a distinct pattern of a slow and long early adoption phase, followed by policy changes that then triggered large-scale adoption.21 However, the process was regarded as slow, ranging from 2 to 7 years between initial adoption by opinion leaders and the eventual large-scale implementation.
Issues for adoption in the UK
In studies funded by the manufacturers of the testing kits, a recent survey found that 61% of UK GPs would welcome access to POC CRP in their clinic,25 and studies have repeatedly demonstrated a high rate of patient acceptability towards POC testing strategies.21
Issues that require further consideration include:21
reimbursement and incentivisation for appropriate use of CRP testing;
quality control processes for maintenance of devices;
training programmes for staff;
effects on clinic flow and workload; and
there is little evidence on the application of test results to different populations, including children and patients with comorbidity.
Costs
The NICE guideline development group estimated the cost of a CRP test (including equipment) to be £13.50.4 The group noted that analysers for CRP testing cost £700 and may be needed by around 5,500 GP practices at an estimated cost of £3.8 million in England.5 Providing POC CRP tests cost less than £15.13, they were deemed to be cost effective at a threshold of £20,000 per QALY.4
NICE has published briefing reviews for two CRP POC tests.26,27 The reviews do not make a judgement on the value of the tests and do not provide formal guidance on their use.
Conclusion
Lower respiratory tract infections are common, and vary between mild, self-limiting viral conditions to bacterial infections carrying a risk of severe complications. These conditions are hard to discriminate on clinical grounds only, and standard tests such as x-rays or laboratory blood tests require time to complete and to receive results. C-reactive protein (CRP) is a biomarker that is raised in acute tissue injury, and is particularly high during bacterial infection.
The use of CRP testing may reduce unnecessary antibiotic prescribing (which carries a risk of adverse effects and the development of antibiotic resistance), while targeting antibiotic therapy to patients most likely to benefit from it. A CRP test should be considered only when there is doubt about the need for antibiotics after a clinical assessment. The test is not necessary when the patient's condition clearly dictates a course of action, nor is it appropriate to use it when the patient has significant comorbidity.
Point-of-care (POC) CRP tests have been shown to produce results similar to laboratory-based CRP tests when compared under controlled conditions. A POC CRP test may be advantageous when a rapid result is needed to guide treatment choice. Nevertheless, there are likely to be significant issues relating to the adoption of POC CRP testing in primary care, including funding of the testing equipment and consumables, user training and ongoing quality control. Rapid uptake of POC CRP testing in primary care seems unlikely in the absence of a funded implementation programme.
References
[R=randomised controlled trial; M=meta-analysis]