Alex Nowbar’s journal review—18 September 2019

Alex Nowbar reviews the latest research from the top medical journals


Benralizumab for the prevention of COPD exacerbations

Exacerbations of chronic obstructive pulmonary disease (COPD) in people with moderate to severe COPD with high eosinophil counts were not reduced by benralizumab in either the GALATHEA or TERRANOVA randomised controlled trials. The trials were funded by AstraZeneca. The rationale for these studies, which included thousands of participants, was debatable as earlier studies had shown no reduction in COPD exacerbations with benralizumab. However, in the subgroup of people with raised eosinophil counts there was a “numerical, albeit non-significant, improvement.” The drug, which has supporting evidence for asthma, really needs shelving for COPD.

Roxadustat rocks anaemia world

Chinese researchers tested whether roxadustat increases haemoglobin in patients with chronic kidney disease in two double blind randomised controlled trials. Posed as an alternative to epoeitin, roxadustat is an oral drug designed to inhibit hypoxia-inducible factor (HIF) prolyl hydroxylase. Both trials showed improvements in anaemia, one in patients receiving dialysis, and one in patients with chronic kidney disease not on dialysis. I expect regulators and patients will be pleased with this, but it is also our duty to expect evidence that roxadustat improves patients’ symptoms without increasing the risk of death or cardiovascular disease. There was also an unexplained higher rate of hyperkalaemia in those receiving roduxastat.


Uveitis treatment

An international randomised controlled trial of methotrexate versus mycophenolate has been inconclusive in showing which is more effective in controlling uveitis inflammation. Rathinam et al randomised 216 people at nine eye centres across the globe and assessed them at six months for “treatment success.” This endpoint was defined as control of inflammation in both eyes, 7.5 mg prednisone daily, and 2 drops of prednisolone acetate 1%, and no treatment failure due to safety or intolerability. I like this endpoint because it encompasses all the factors that make up treatment success, certainly from the clinician’s perspective. Being off steroids is critical in inflammatory conditions due to the serious adverse effects. It would have been good to know whether methotrexate or mycophenolate is better, so I commend this study for answering an important question, even if the answer was that mycophenolate was no better.

JAMA Internal Medicine

Heartening dialysis patterns

What happened to rates of peritoneal dialysis initiation after a programme to expand its use in an “integrated care delivery model” in Northern California? This observational study of 13,500 patients found that the proportion of new peritoneal dialysis initiation in patients starting dialysis rose from 15% in 2008, to 34% in 2018. 331 patients (84%) who started peritoneal dialysis in 2017 were still continuing with it at one year. There was no change in mortality over the study period. A system wide approach which included patient and caregiver education, education and support tools for health care professionals, streamlined system-level processes, monitoring, and continuous quality improvement could have been responsible for this improvement. But as this wasn’t a randomised controlled study, the improvement cannot be confidently attributed to the intervention and comparative costs are unknown. 

Disheartening kidney transplant patterns

Gander et al compared transplant rates between people receiving dialysis at for-profit facilities compared to nonprofit facilities in the US. I was not surprised to read that people were much less likely to be put on the waiting list and less likely to receive either a live or deceased donor kidney in for-profit facilities. Is there an unethical preference for for-profit facilities to perpetuate the need for dialysis instead of facilitating transplantation? Or are there systematic differences in the people who attend a for-profit facility versus a nonprofit facility like age, comorbidities, patient and staff cultures, financial status? Because this is an observational study, it is not possible to identify a reason for this discordance. Appropriateness of referral for transplant was also not assessed so it is unclear if nonprofit facilities are over-offering or if for-profits are under-offering.

Annals of Internal Medicine

Chronic disease geography in the US

Have you ever spared a thought for adults on low incomes living in affluent areas? It is typical to think about geography and income separately but Polyakova et al measured chronic disease prevalence in 6,363,097 adults with low income and aged over 65 using Medicare data stratified by income and local area. Their main findings were better health in low-income adults in affluent areas and high prevalence of chronic conditions in affluent areas. As is always the way with large cross-sectional studies, disease definitions may not represent actual morbidity because they do not include severity of conditions. For example, almost half of the people in their sample had “rheumatoid arthritis/ osteoarthritis”with ischaemic heart disease close behind in highest prevalence. Actually the three most common conditions were hypertension, hyperlipidaemia and anaemia, followed by arthritis and ischaemic heart disease. The size and comprehensiveness of this analysis are its strengths, but its findings are fairly unexciting and difficult to translate into action for improvement.


Surgical options for heavy menstrual bleeding

The HEALTH trial randomised 660 women aged <50 years with heavy menstrual bleeding to either endometrial ablation or laparoscopic supracervical hysterectomy. Hysterectomy may sound like a rather radical approach to heavy menstrual bleeding, but participants were required to have been referred to a gynaecologist for surgical treatment of the conditionso, by definition, it was severe enough to warrant treatment and the women had to have no desire for (further) children. The carefully chosen co-primary endpoint was patient satisfaction and menorrhagia-specific quality of life at 15 months. On this basis laparoscopic supracervical hysterectomy was superior to ablation, with 97% of women satisfied versus 87% of women in the ablation group, and 69% with the best menorrhagia quality of life score versus 54%. Complication rates were similar in both groups, but ablation was quicker to perform and had shorter recovery. Infection rates were similar, but voiding dysfunction was slightly more common in the laparoscopic supracervical hysterectomy group. The only caveat would be that it was open-label with no objective endpoints to assess its efficacy. That said, double-blind trials of surgical procedures are no mean feat.

Alleviating peanut allergy

A randomised controlled trial has shown the efficacy of oral immunotherapy in desensitising people with peanut allergy. The study randomly allocated 120 adults and children to three arms: two years build-up oral immunotherapy followed by withdrawal, two years build-up followed by continuation at a lower dose, or placebo (oat powder) throughout, with use of nose clips in an attempt to blind to smell.  But what counts as efficacy for an allergy therapy? Ideally one would want a safe “cure” without the need for further treatment, but oral immunotherapy isn’t a golden bullet. This study’s primary endpoint was passing a double blind, placebo controlled food challenge with peanut protein at 2 years, and at 2 years and 3 months. I would say this study length is a reasonable balance between long enough to claim the response is “sustained” and short enough to be a practical study with minimal dropout. This study found 21 of the 60 people in the build-up and withdraw arm passed both challenges, compared with only one of 25 of those in the placebo group. However, the build-up and continue arm were much less likely to become desensitised after passing the initial food challenge.

Adverse events (particularly gastrointestinal and skin disorders) were common with the oral immunotherapy. Allergists and patients with peanut allergy are probably best placed to decide if going through this therapy is “worth it”. I’m sure for many it would be, particularly as peanut exposure can be so hard to avoid. The last interesting detail is that biomarkers were identified that could predict responsiveness to the treatment. Since response rates were less than half, I suspect people would want to know their chance of success so that they could decide whether to embark on a lengthy course of an essentially low-dose of allergen.

Alex Nowbar is a clinical research fellow at Imperial College London.