Tom Nolan’s research reviews—26 January 2023

Battle of the loops

The recent Christmas issue of The BMJ highlighted examples of mis-spelt antimicrobials including cefouroxime (4× better), trimethoprime (comes with free delivery), and, my favourite, tazosin (prescribe in haste, repent at leisure). If we searched the most mis-spelt drugs of all, my bet would be on furosemide to top the charts. I can’t be the only person who still hasn’t fully come to terms with the inconvenient “u”, added when the UK switched from British Approved Names to International Non-Proprietary Names 20 years ago.

If only something better would come along, so we can consign furosemide to the past, and I can stop resenting that extra syllable. We’ll have to endure it for the foreseeable future, it seems, after an open-label randomised trial comparing furosemide with fellow loop diuretic torsemide in patients discharged after hospital admission with heart failure found that the more expensive torsemide was no better than old frusey. The authors had hypothesised a 20% reduction in mortality at 12 months in those allocated to torsemide but found no difference in this, the primary end point of the study.

JAMA doi:10.1001/jama.2022.23924

Interval training for colorectal cancer screening

The debate over whether colonoscopy screening for colorectal cancer is superior to other, less invasive, screening methods is still raging. For those settled on colonoscopy over sigmoidoscopy or stool tests, another question is how often to repeat screening. In Germany, they offer colonoscopy screening at the age of 50 years and happen to have the largest registry of screening colonoscopies in the world, running at over 10 million. A cross sectional study that made use of that big registry found that incidence of colorectal cancer at repeat colonoscopy after 10 years was rare (between 0.1% and 0.5% depending on age and sex). The authors conclude that the current 10 year interval commonly recommended is safe, and could even be extended in some lower risk groups.

JAMA Intern Med doi:10.1001/jamainternmed.2022.6215

Appendicitis recovery

One way to reduce the carbon footprint of healthcare is to reduce the length of intravenous antibiotic treatment. Even if current practice may often seem more rooted in tradition than evidence, we need high quality evidence to give clinicians the confidence to alter antibiotic protocols and switch from intravenous to oral treatment sooner or offer shorter courses of treatment altogether.

For adults recovering from complex appendicitis, intravenous antibiotics for two days was non-inferior to five days of intravenous antibiotics in a large open-label multicentre randomised trial in the Netherlands. The study looked at 90 day rates of infectious complications and mortality as the primary endpoint. It also found lower rates of antibiotic related side effects (9% v 22%) but higher rates of re-admission (12% v 6%) in the group given intravenous antibiotics for two days rather than five.

Lancet doi:10.1016/S0140-6736(22)02588-0

Aspirin back on the menu

The authors of a trial of aspirin versus low molecular weight heparin (LMWH) as thromboprophylaxis after orthopaedic surgery argue that patients would strongly prefer aspirin if clinical outcomes are similar. There’s probably an environmental argument here too—not mentioned in the study—for favouring oral aspirin over LMWH injections. The study randomised 12 211 patients over 18 years old who had undergone surgery after a fracture of an extremity to receive either aspirin or LMWH. They found no difference in death from any cause at 90 days (which, as you might hope, were few) and in secondary outcomes of pulmonary embolism and bleeding complications—although deep vein thromboses were more common among those taking aspirin (2.51% v 1.71%). An editorial says that it’s time for guidelines to be updated to include aspirin as an option in this context.

N Engl J Med doi:10.1056/NEJMoa2205973

Rise and shine, it’s phlebotomy time

Over a three year period at the Yale New Haven Hospital there were over nine million blood draws from nearly 100 000 patients. Of these, 38.9% were taken between 4 am and 7 am and only 20.7% between 7 am and midday. Although early morning phlebotomy can help inform clinical assessment on morning rounds, the authors of this research letter argue that this needs to be balanced against the risk that disturbing patients’ sleep to take blood may be affecting patient outcomes, and call for further research. The other question is what proportion of these tests were clinically necessary and actually helped in clinical decision making. Then there’s all that plastic—again, not mentioned.

JAMA doi:10.1001/jama.2022.21509