This afternoon a patient on the ward asked me “So do I have pneumonia then?”.

Not a straightforward answer.

She has a productive cough, crackles in her left base, a WCC of 24, CRP of 140, and presented with a lactate of 3.4, and a bicarbonate of 14. And a normal CXR.

SIRS? Yes. Sepsis? Yes? Chest infection? Almost certainly. Pneumonia? Depends.

The BTS Pneumonia Guideline says that the diagnosis of pneumonia depends on where the patient is: CXR is required in secondary care, but in primary care the diagnosis is made on clinical grounds only.

The ATS/IDSA consensus statement on community acquired pneumonia tells us: “In addition to a constellation of suggestive clinical features, a demonstrable infiltrate by chest radiograph or other imaging technique, with or without supporting microbiological data, is required for the diagnosis of pneumonia.”

So in the UK my patient has pneumonia if she’s seen in primary care, but not in secondary care because of her normal CXR.

In the US she doesn’t have pneumonia based on her CXR, but if she went on to have a CT, she might do.

In Australia she would need CXR evidence of new consolidation, so she doesn’t have CAP.

Assessment of the severity of said pneumonia gets more interesting, depending on the longitude of your residence. The BTS suggest using the CURB-65 score. The ATS suggest the PSI. The Australians suggest either the CORB score, or SMART-COP.

I told her that she has an LRTI, with systemic upset, and a normal CXR on admission, but I’ll be treating her as though she does have pneumonia, as repeating her CXR until it become positive seems a bit of a waste of time, and a tad excessive.

We’ve been looking at COPD treatment options in our MCN. The current discussion is the risk of pneumonia in COPD patients on inhaled corticosteroids, particularly in reference to the newest, and most recently licensed ICS/LABA therapies. Sami Suissa’s paper in Thorax has made all of us think twice about our treatment algorithms. But it begs the question – how was pneumonia defined in this (and other) clinical trials? Primary care definition? Secondary care definition? CXR? CT? The seeming paradox of ICS/LABA reducing exacerbations in COPD, yet increasing pneumonia risk, but reducing (not quite statistically) mortality is a puzzle to all of us, I think. I’ve been asked “Should we take all out COPD patients off Seretide?” a number of times. I don’t think its all that simple. To be continued, I suspect.

My patient has a significant pulmonary infection resulting in systemic upset, and sepsis. But the variation in guidelines across the globe means whether she has pneumonia or not depends on where she lives, which must be an issue for international multi-centre trials, and treatment guidelines.

Incidentally, she had a CURB-65 of 1 when she came in. I’m keeping her in the ward until the weekend.