Zackary Berger’s journal reviews—8 October 2018

Zackary Berger reviews the latest research from the top medical journals


Surgery versus physical therapy for patients with nonobstructive meniscal tears

Despite several trials showing that partial removal of the knee meniscus does no better than physical therapy in improving pain, the practice persists. Thus, researchers in the Netherlands conducted a noninferiority trial comparing surgery (arthroscopic partial meniscectomy) and physical therapy, for the outcome of patient-reported knee function. The trial was well done, with some limitations. The interventions were not blinded, for instance, and the exclusion criteria makes the result inapplicable to most of my patientsI can’t remember the last time I saw someone with symptomatic meniscal tear *without* osteoarthritis.

The noninferiority? It was achieved. So physical therapy is as good as meniscectomy. Did we know that already? I guess we did, but based on trials with slightly different designs and outcomes. In the introduction, the authors implicitly justify this study with the fact that previous studies haven’t changed practice. Unfortunately, exclusion criteria and lack of long-term follow up always provide ready-made excuses (“these patients aren’t my patients,” as I just said myself). Change is hard. Individual studies don’t make it happen. But they can change individual minds all the same.

Risk of hypertension in black American adults vs white American adults

Why do more African-Americans have high blood pressure than white Americans? This study looked at a cohort of white and African-American patients in the US who were recruited without high blood pressure, followed them up for a median of 9 years, determined what proportion of each group developed hypertension, and then analyzed which factors explained the disparity in hypertension between the groups.

It’s important to realize what was measured and what wasn’t. The study headline is that the Southern diet (fried food), was more widely eaten and eaten in greater quantities by African Americans compared with white people. This explained the largest proportion of the disparity in blood pressure among the variables recorded. But, keep in mind that this is derived from self-reported dietary habits, a data source afflicted by recall bias.

But the key phrase in the findings is, “Among the variables recorded.” Because what makes the African American experience unique is not genetic composition or physiologic differences, but the social fact of being black in America, a social variable that was unmeasured in this study. The variables of income and self-reported stress that were included in the study are not enough to capture the widespread effects of inequity. Do black Americans have equal access to healthy food and the economic resources needed to obtain it? Is income and education the only way to represent social disparities?

I know this study will be used to tell patients to eat healthier. Can we tell our societies to change the way they construct race?


Rivaroxaban in patients with heart failure, sinus rhythm, and coronary disease

Indication creep is not the title of a horror movie. When a medication is approved for one condition (like the anticoagulant rivaroxaban, for deep venous thrombosis and for the prevention of stroke in patients with atrial fibrillation), it becomes natural, or at least economically advantageous, to seek out other situations in which the medication can be supported by new evidence. Rivaroxaban is a useful tool in the antithrombotic arsenal for my patients with atrial fibrillation who can afford it. But I did cock an eyebrow at this trial, conducted, paid for, and partially written by the pharmaceutical company Janssen, asking whether rivaroxaban can reduce the rate of a composite outcome in patients with heart failure, sinus rhythm, and coronary artery disease.

Composite outcomes are commonly used to improve power, even if patients, in my experience, don’t understand what preventing-A-B-and-C means to them in real life. This trial, however, shows no such improvement. Back to the drawing board, or to devising new outcomes for which the medication can discover new marketing horizons.

Acceleration of BMI in early childhood and risk of sustained obesity

Researchers conducted a retrospective study in a cohort of German children to trace the relationships between these children’s birth weight, body mass index (BMI) as children and their BMI as adolescents. The trajectory of BMI from childhood to adolescence is not entirely straight as there is a period when BMI can decrease before again increasing. The most interesting finding of this paper is not that obesity in adolescence and obesity in childhood are associated (that seems obvious), but that there seems to be a particular time period, from 3 to 6 years old, after which progression to obesity is only rarely reversed in later years.

But, as with any study, the assumptions are worthy of examination. And the assumption, here, is that obesity is solely and forever a bad thing, a risk factor for diseaseor even a disease itselfa life with obesity is to be avoided, and the way obesity develops is through the actions of individuals. Perhaps we can consider the breadth of the literature, and the complexity of social context, in which obesity is related to genetics, weight loss is hard, environment helps shape behaviour, and different body types exist across the spectrum of health.

Annals of Internal Medicine

Treatments for primary basal cell carcinoma of the skin

Basal cell carcinoma (BCC) is the most common cancer in the United States. Though its mortality rate is lower than other cancers, those with BCC can experience significant distress and disability. Researchers conducted a systematic review of the evidence of treatments for BCC. They also conducted network meta-analyses, which enable indirect comparisons of treatments which have not been compared directly in the literature. Most individual comparisons were represented by three studies or fewer, which means that many relevant comparisons are not yet directly addressed in the literature.

The good news is that a number of treatmentssurgical excision, Mohs surgery, curettage and diathermy, and external-beam radiationappeared to be roughly equally effective in reducing recurrence. Cosmetic outcomes were better for photodynamic therapy than for excision or cryotherapy. The authors note that external beam radiation is becoming more and more popular, but not on the basis of substantial evidence when compared to other more common treatments, and it is significantly more expensive.

Moreover, there is no commonly used set of outcomes shared among studies evaluating BCC treatments, nor do the most widely cited networks of cancer surveillance (e.g. the US Center for Disease Control’s SEER network) include BCC. Thus, as with so many other conditions, we are in an evidence-poor zone. Patients already have skin in the game; researchers, clinicians, and funders need to meet them where they are.

High-cost status among dual-eligible Medicare and Medicaid beneficiaries

11 million Americans are “dual eligibles,” meaning that they benefit from Medicaid and Medicare. These people tend to be poorer, with more chronic conditions, sicker, and more disabled. Much attention has been paid to reducing health system costs by focusing on the “highest utilizers”—a somewhat cold and dehumanizing term for those who need to make frequent or intensive use of healthcare. Who are those highest users among beneficiaries of both Medicaid and Medicare?

Is a high user generally consistently to be found in that group, or can the level of use change from year to year?

Using administrative and claims databases, researchers calculated costs due to each patient’s Medicare and Medicaid use. They standardised costs, since prices vary wildly. They then cross-indexed these costs, using another database, against preventable hospitalizations. They only included patients who remained alive during all years covered by their sample (2008-2010), which comprised nearly 2 million beneficiaries of both Medicare and Medicaid.

In 2008, nearly 200,000 of the patients were categorized as “high-cost,” which the authors defined as the top 10% of users. Of those patients 55% were high-cost in all 3 years; conversely, this implies that 45% were not high-cost in all three years.

The surprising finding from this study? Persistently high-cost patients were different from transiently high-cost patients: the former had substantially higher costs for long term care (on average 69% of their total costs compared to 43%). Transiently high-cost patients had higher spending in the inpatient setting than persistently high cost patients. Thus, for persistently high-cost patients benefiting from both Medicare and Medicaid, a policy that incentivizes in-home rather than institutionalized long term care might be not just more humane (as seems clear), but cost saving.

The Lancet

Onchocerca volvulus infection in Ghana, Liberia, and the Democratic Republic of the Congo

Onchocerciasis is the second-most common infectious cause of blindness worldwide, and the suffering caused by the helminth Onchocerca volvulus also includes painful itching and bumps under the skin. (If you search the disease online in the US, Google tells you helpfully “US cases are rare,” which is ike saying that tiger attacks are infrequent in Manhattan.) Four countries in the Americas have been declared free of river blindness, and 99% of cases are now in Africa. Ivermectin is widely used, but its effectiveness might be decreasing. Thus this trial sought to determine whether moxidectin, a similar medication (though less subject to metabolization, and with a longer half-life), already used by veterinarians, is superior to ivermectin. Moxidectin has been tested in a phase 2 trial and studies in healthy volunteers.

The study found that the density of skin microfilariae was lower after moxidectin treatment than after ivermectin treatment, thus raising hopes that this alternative medication could reduce parasite transmission. As with all public health interventions, prioritization questions are uppermost. Though this is not my field, I wonder how vector control as a strategy compares to the investment and dissemination needed for this medication.

JAMA Internal Medicine

Increased daily water intake in premenopausal women with recurrent urinary tract infections

An extremely non-systematic reviewi.e. a Google searchshows approximately a zillion studies of cranberry juice or extract as a prophylactic against urinary tract infections, and not that many of water. So I was excited to see this trial, which randomized 140 pre-menopausal women who reported drinking less than 1.5L of water daily, to either a control group, or an intervention group, which was instructed to drink an additional three 500-ml bottles of Evian daily. I have no Evian-related disclosures, but the study was funded by Danone, which produces Evian. The women were on average 39 years old and had 3 UTIs a year.

This trial isn’t perfect (it doesn’t look like analysis was blinded, and I don’t know if patients in Sofia, Bulgaria, are comparable to mine in Baltimore, Maryland). But it made me happy to see that a low-cost, bare-bones intervention reduced the number of UTIs by an average of 1.5 per year, and in a secondary outcome reduced the number of antibiotic regimens used. Bottoms up!

Zackary Berger, is an associate professor at Johns Hopkins School of Medicine in the division of general internal medicine, and core faculty in the Johns Hopkins Berman Institute of Bioethics, both in Baltimore, Maryland. Zackary’s research focuses on shared decision making, patient-centered care, and health justice. He is also a practising physician at a free health clinic for undocumented, predominantly Spanish-speaking, immigrants.

Competing interests: None declared