Scoring pneumonia severity
There are several scoring systems available to help define pneumonia severity.
“Medicine is a science of uncertainty and an art of probability”.
These were William Osler’s words over a century ago, and on face value many would agree with them today. Osler was a consummate clinician, and his teachings detail the value of the history and clinical examination in diagnosing disease. It is hard to know what he would have thought of the plethora of scoring systems that are finding their way into modern medicine.
Pneumonia is a condition that has been subjected to rigorous attempts to develop tools that predict various outcomes, such as 30-day hospital mortality and need for intensive care admission.
These tools include the Pneumonia Severity Index (or PSI), the CURB-65 score, and from Australia, the CORB and SMART-COP scores.
The simplest of them (the CORB score), takes into account four clinical features to stratify patients into low and high-risk for ICU admission.
The new edition of Therapeutic Guidelines – Antibiotic, continues to use pneumonia severity scoring to inform the choice of empiric antibiotic therapy. In previous editions the PSI was used, but now the simpler and more clinically based scores like CORB and SMART-COP are recommended.
Overall, the PSI is a more complicated scoring system developed in the US in the late 1990s. It takes into account 20 variables, including seven laboratory or radiographic variables (available at www.tg.org.au/etg_demo/etg-psitable.pdf).
This makes it a more challenging tool to use in the primary care setting. However, many hand-held or web-based programs are available to deliver a PSI score, so that if the investigations are at hand, it remains a useful tool.
The creators of the PSI have demonstrated that the PSI has a slight statistical advantage over the CURB-65 in predicting low-risk patients, thus avoiding hospital admission.
The CORB score can be used to stratify patients already hospitalised for pneumonia into moderate or severe categories. Need for intensive respiratory or vasopressor support (IRVS) is assessed by a composite of the following variables: Confusion, Oxygen saturation < 90% on room air, Respiratory rate > 30 breaths/min, systolic Blood pressure < 90mmHg or diastolic Blood pressure < 60mmHg.
Groups are stratified according to score. A score of 2 or more identifies patients with ‘severe’ pneumonia who may require treatment in the intensive care unit.
Finally, the SMART-COP tool, developed in Australia, has been validated only for hospital use in predicting the need for ICU admission and vasopressor support.
Neither CORB nor SMART-COP are validated in predicting need for hospital admission and may be less useful in a general practice setting. Neither should be used to categorise patients as ‘low-risk’ or suitable for outpatient management.
While acknowledging their limitations, scoring systems are useful for antimicrobial stewardship purposes and limiting the use of cephalosporins and moxifloxacin in first-line management of community acquired pneumonia (CAP).
For outpatient management of mild CAP, either amoxycillin or an oral macrolide or doxycycline can be used. All patients with moderate CAP requiring hospital admission should be commenced on a penicillin/macrolide dual therapy unless there is a clear contraindication.
There is still widespread use of ceftriaxone as first-line IV therapy, but in fact benzylpenicillin and ampicillin remain the best antibiotics for pneumococcal pneumonia.
On their own, the scoring systems are not a substitute for clinical judgement. The Therapeutic Guidelines authors have recognised this and emphasise the use of pneumonia severity scores only as a guide.
For example, none of the scores take into account the social background of a patient. Nor do the simpler scores, such as CURB-65 and CORB, incorporate relevant risk factors such as diabetes or alcohol use.
Osler might have given his opinion of these scores using another of his famous quotes: “It is much more important to know what sort of patient has a disease than what sort of disease a patient has.”
CURB-65 SCORE (1 point each)
CONFUSION (acute onset)
UREA > 7 mmol/L
RESPIRATORY RATE ≥ 30/min
BLOOD PRESSURE (SBP < 90 mmHg or DBP ≤ 60 mmHg)
Age > 65 years
Several different scoring mechanisms exist for assessing severity of community acquired pneumoniaEach score has been validated in different clinical settings, and some are not suitable for predicting ‘low-risk’ groups
Scoring systems must be interpreted only in light of other relevant findings on history and clinical examination
Scoring systems can be useful in anti-microbial stewardship and limiting the use of cephalosporins and fluoroquinolones in first line treatment of pneumonia.
Tags: , Pathology