Ann Robinson’s journal review—3 July 2019

Ann Robinson reviews the latest research from the top medical journals

Annals of Internal Medicine

Adrenal incidentalomasdo they need follow up?

Are adrenal incidentalomas, which are found by chance on imaging, really harmless? In this paper, the authors looked at 32 studies, including 4121 patients with benign non-functioning adrenal tumours (NFATs) or adenomas that cause mild autonomous cortisol excess (MACE). Only 2.5% of the tumours grew to a clinically significant extent over a mean follow-up period of 50 months, and no one developed adrenal cancer. Of those patients with NFAT or MACE, 99.9% didn’t develop clinically significant hormone (cortisol) excess. This was a group (especially those with MACE) with a high prevalence of hypertension, diabetes, and obesity. This could be because adrenal adenomas promote cardiometabolic problems, or vice versa, or maybe this group with multimorbidities is more likely be investigated. Adrenal incidentalomas are already found in around 1 in 20 abdominal CT scans, and this rate is likely to increase as imaging improves. So it’s good news that this study supports existing recommendations, which say that follow-up imaging in the 90% of incidentalomas that are smaller than 4 cm diameter is unnecessary.  


Chest pain and myocardial infarction—a new risk assessment tool using troponin

The COMPASS-MI project developed and tested a risk assessment tool which used initial and serial high sensitivity troponin (I or T) levels to calculate the risk of being diagnosed with a myocardial infarction (MI) and subsequent MI or death at 30 days in patients presenting to emergency departments with chest pain. For example, if the initial high sensitivity troponin I concentrations were <6 ng/L, and the change in level between 45 and 120 minutes was <4 ng/L, there was a 99.5% chance that the person hadn’t had an MI and a 99.8% chance of being alive a month later. The risk assessment tool also allowed over half of the patients to be classified as low risk. One caveat is that the data came from 15 international cohorts, and the diagnosis of MI may not have been consistent across the board. Existing rapid triage algorithms are not perfect; they may not differentiate between MI and other forms of myocardial injury, the timing of the second sample varies, and the outcome for people with raised troponin but no MI isn’t clear. This tool addressed these questions and seems to be a useful addition to the armamentarium.

Brain activity in unresponsive patients with acute brain injury

Agonising decisions about the withdrawal of life-sustaining therapies after brain injury rest on whether clinically unresponsive patients stand a chance of recovery. Up to 14% of chronically unresponsive patients may still have signs of brain activity in response to a spoken command (cognitive-motor dissociation) months or years after injury. But is there a similar dissociation in the first few days after a brain injury, and does it predict the outcome? This study found that 15% of patients with acute brain injury who were clinically unresponsive had evidence of brain activation (on electroencephalography) in response to spoken commands, and this group were more likely to have later recovery. But this single centre, descriptive study wasn’t powered to detect differences in long term outcomes, and larger, multicentre studies are needed.


COPD and spirometry measurements

How accurate is the ratio of the forced expiratory volume in the first second to the forced vital capacity (FEV1:FVC) in predicting chronic obstructive pulmonary disease (COPD)-related hospitalization and mortality? This careful study found that a FEV1:FVC threshold of less than 0.70 can help to identify individuals at risk of clinically significant COPD and had better prognostic accuracy than a range of fixed thresholds (0.75 to 0.65) or the lower limit of normal (LLN), defined as the lowest 2.5th percentile of a healthy reference group. But are physiological biomarkers such as FEV1:FVC measured by spirometry useful in complex conditions such as COPD? Finding evidence of airflow limitation makes it likely that breathlessness can be attributed to lung disease, but an editorial reminds us that “as multimorbidity is the rule rather than an exception in patients with COPD, excluding other causes of breathlessness, in particular heart disease, is mandatory.”

JAMA Internal Medicine

PSA—to test or not

How should a man decide whether to have a PSA test? In the UK, routine testing isn’t offered, but men over 50 years old can request the test from their GP. Digital decision aids abound, but do they help men to reach a decision compared with usual care (discussing the pros and cons with a doctor)? This systematic review and meta-analysis of 19 randomised clinical trials comparing decision aids for prostate cancer screening in 12 781 men found that decision aids are probably associated with a small reduction in “decisional conflict,” possibly associated with an increase in knowledge and possibly not associated with whether physicians and patients discuss prostate cancer screening or what screening decision the man makes. PSA is an imperfect test, and no amount of shared decision making can get round that.


Breast cancer treatment: two similar trials with opposite conclusions

Two studies (PERSEPHONE and PHARE) asked the same question and came to different conclusions; is a six month course of adjuvant trastuzumab as effective and less toxic for women with HER2-positive early breast cancer as the standard 12 month course? PERSEPHONE found that the shorter course was non-inferior to the longer course, with four-year disease-free survival of 89%, and that, unsurprisingly, the shorter course caused few serious adverse events and cardiotoxicity. But PHARE came to the opposite conclusion: the researchers did not find non-inferiority between six months and 12 months of trastuzumab, and the risk-benefit analysis favoured the longer course. The main difference between the trials was the choice of non-inferiority margin (1·15 in PHARE v 1.29 in PERSEPHONE). Where to set the margin is inherently controversial, and the PHARE authors question the very feasibility of non-inferiority trials “especially in the context of oncology trials, where the primary outcome is survival and any additional deaths could be considered unacceptable.”

Ann Robinson is an NHS GP and health writer and broadcaster