Intended for healthcare professionals


Tom Nolan’s research reviews—1 June 2023

BMJ 2023; 381 doi: (Published 01 June 2023) Cite this as: BMJ 2023;381:p1224
  1. Tom Nolan, clinical editor; sessional GP, Surrey
  1. The BMJ, London

Scanning for pulmonary embolism

We all know that feeling when a colleague comes up to you with a grave look on their face and says, “You remember that patient you saw last week with….” On a good day the story might go that you admitted the patient to hospital, it turns out that they had a pulmonary embolism, and now they’re better. And on a bad day? Well, you’d better hope you made some good notes and, better still, followed a validated pulmonary embolism clinical decision rule. These are intended to answer the age-old question: “How do you know it isn’t a pulmonary embolism?” without always having to request a computed tomography pulmonary angiogram (CTPA). But CTPA testing seems to be going up regardless.

A retrospective analysis of 26 emergency departments in six European countries found that, between 2015 and 2019, the rate of CTPA testing increased from 836 per 100 000 emergency department visits to 1112 per 100 000. The rate of diagnosis of pulmonary embolism went up too, but, as diagnosis went up, so did the proportion of low risk pulmonary embolisms, while the proportion requiring intensive care went down.

Ann Intern Med doi:10.7326/M22-3116

Inhaled corticosteroids in COPD

We’ve been telling patients for years that if they have chronic obstructive pulmonary disease (COPD) and are prescribed a combined inhaled corticosteroid plus long acting β agonist (ICS-LABA) inhaler they are at increased risk of being admitted to hospital with pneumonia. From the introduction to a cohort study in JAMA Internal Medicine, it seems that this link isn’t so cut and dried after all: there have been some conflicting findings from trials, and applicability of trials to clinical practice is undermined by younger, more male participants in the studies than we typically see and by run-in periods (when patients stop other inhalers before the study intervention period begins).

A new cohort study reviewed records of over 130 000 patients in the US between 2014 and 2019, and compared ICS-LABA users with people who took long acting muscarinic antagonist (LAMA) and LABA (LAMA-LABA) inhalers. The people taking LAMA-LABA had an 8% lower rate of first moderate or severe COPD exacerbation (hazard ratio (HR) 0.92; 95% CI 0.89 to 0.96) and a 20% lower rate of first pneumonia hospitalisation (HR 0.80; 0.75 to 0.86), confirming that we’ve been giving the right advice all along.

JAMA Intern Med doi:10.1001/jamainternmed.2023.1245

Anticoagulant timing after stroke

When do you start a direct oral anticoagulant (DOAC) after a stroke in someone with atrial fibrillation? The 1-3-6-12 day rule refers to the day to commence a DOAC: 1, 3, 6, or 12 days depending on whether the patient had a transient ischaemic attack or mild, moderate, or severe stroke respectively—because the bigger the area of infarction the higher the risk of haemorrhage.

A randomised control trial aimed to find out if earlier initiation of DOACs—within 48 hours after a minor or moderate stroke, or 6-7 days after a major stroke—might be better for patients. The trial used a composite primary outcome that included recurrent ischaemic stroke, systemic embolism, major extracranial bleeding, symptomatic intracranial haemorrhage, and vascular death. There was no significant difference in this outcome between early anticoagulation and standard care after 30 days (2.9% v 4.1%), leaving the authors to conclude that “early treatment initiation can therefore be supported if indicated or if desired.”

N Engl J Med doi:10.1056/NEJMoa2303048

Pneumonia on steroids

A randomised control trial of 800 patients in France with severe community acquired pneumonia has found that treatment with intravenous hydrocortisone led to better rates of survival at 28 days compared with placebo. However, another recent study, set in the US, did not find any survival benefit from intravenous steroids (methylprednisolone in this case). So what to do?

Possible explanations, discussed in an editorial alongside the French study, include sex differences in the participants of the two studies: in the US study, 96% of participants were men, whereas 69% of those in the French study were men, and a subgroup analysis found a greater benefit among women. Another factor may be the cause of pneumonia: the French study had a relatively high proportion of patients with Streptococcus pneumoniae infection and excluded those with influenza, based on findings from observational studies that those given glucocorticoids in influenza seem to have higher mortality rates.

N Engl J Med doi:10.1056/NEJMoa2215145, doi:10.1056/NEJMe2302544

Second line diabetes drugs cross the line together

Which of the usual options for a second drug treatment for type 2 diabetes—in addition to metformin—are better at preventing kidney disease? A randomised control trial recruited people with type 2 diabetes who resemble many of patients we see in practice: participants had an HbA1c level between 6.8% and 8.5%, were already taking metformin, and didn’t have a severe reduction in renal function, symptomatic heart failure, or a recent cardiovascular event. They were randomly assigned to receive sitagliptin, glimepiride, liraglutide, or insulin glargine on top of metformin to see whether any one of these led to better renal outcomes (dialysis, transplant, death due to kidney disease, or progression of albuminuria). The study began enrolling patients in 2013, so didn’t include an SGLT2 inhibitor arm. None of treatments stood out over the five year follow up, and only 12% of the 5046 participants reached the composite primary endpoint, made up almost entirely (98.4%) of progression to albuminuria.

JAMA Intern Med doi:10.1001/jamainternmed.2023.1487


  • Competing interests: None declared

  • Provenance and peer review: Not commissioned; not peer reviewed