US Highlights

Here are items from recent BMJ issues that may be of particular interest to American readers. Happy reading. Comments welcome.

Douglas Kamerow
US editor

US editor�s choice

June 24 

Carpal tunnel syndrome is common and can be disabling. When splinting and medication fail, surgery is often recommended. Open surgery can result in a long, painful rehabilitation, so Isam Atroshi and colleagues randomized 128 patients with carpal tunnel syndrome to receive either open or endoscopic surgery. They found slightly less postoperative pain in the endoscopic group but no difference in other outcomes, including time to work return and functional status. In a related editorial, Brent Graham points out that the diagnosis of carpal tunnel syndrome may be questionable. Often, doctors rely on electrodiagnostic testing to make the diagnosis, when in fact these tests are best interpreted in the context of clinical findings. He calls for better diagnostic criteria for this common problem.

Two drug trials in this issue disprove conventional wisdom. The DIPOM trial evaluated use of perioperative metoprolol in diabetic patients undergoing major non-cardiac surgery in a randomized controlled trial. They found no benefit on mortality or cardiac morbidity. Etelka Moll et al compared clomifene alone with clomifene plus metformin to induce ovulation in women with polycystic ovary syndrome. Their randomized placebo controlled trial found no difference between the two groups.

Finally, Steven Deeks provides an update on the ever changing world of antiretroviral treatment for HIV disease in adults. He reviews when to start drug therapy, based on the CD4 T cell count, and which three-drug combinations to use. He also discusses when, if ever, to discontinue drug therapy.

June 17

There are multiple guidelines on the use of statins to prevent heart disease, which originate from many different medical societies in different countries. Douglas Manuel and colleagues modeled the effects of six different guidelines on a population of almost 7000 Canadians. They sought to determine which guidelines were most effective (preventing the most deaths) and most efficient (exposing the fewest number of people to statins). Not surprisingly, they found that guidelines that focused their recommendations on those with the highest risk of heart disease were the most successful: using New Zealand, Australian, and British guidelines would have prevented the most deaths in the Canadian population as well as treated the fewest number of patients to do it. Following the American guidelines� �optional� recommendations, on the other hand, resulted in almost twice as many people receiving statin treatment with almost no corresponding decrease in deaths. 

In an accompanying commentary, Vishnu Madhok and Tom Fahey warn that risk estimation is not an exact science; different populations correlate differently with the gold standard Framingham risk scores. They urge more study of applying risk scores to different populations. But the population perspective of this analysis makes for sobering reading for those who have been aggressively treating patients according to the American guidelines. 

Low back pain is a common problem in primary care. B W Koes et al review its diagnosis and treatment. Unlike statin treatment, back pain recommendations across different guidelines are very consistent. They emphasize low intensity strategies, both in diagnostic testing and treatment. Imaging is generally recommended only if there is at least one �red flag� symptom or sign. The focus in treatment is to keep the patient active, use pain or anti-inflammatory medication, and avoid surgery if at all possible. Chronic low back pain is best prevented, as it is difficult to treat successfully. 

Much has been made of potential biases of the authors of medical journal articles. Joel Lexchin and Donald Light discuss commercial influences on the editors and publishers of journals and the biases that can result from these influences. Many medical journals do not require that editors declare personal conflicts of interest, and there have been documented cases of obvious bias related to these conflicts. Most journals receive the bulk of their revenue from advertising and reprint sales, and this income can create conflicts for editors and publishers. The authors give examples of these problems and advocate stricter rules to decrease the appearance and reality of commercial influence on medical journal content.

June 10

The mini-theme of this issue is climate change and its effects on the environment and health. Robin Stott explains the necessity for people in the developed world to conserve fossil fuel. By using less �carbon� we can improve the environment (as well as fitness and health), and by trading carbon credits we can transfer wealth to developing world. Mayer Hillman describes how to calculate a personal carbon footprint, and Anna Coote gives examples of what health services can do.

Rohan Ganguli and Martin Strassnig discuss the use of antipsychotic medications for patients with schizophrenia in light of the findings of the recently concluded CATIE trial. Newer, �atypical� antipyschotics were not found uniformly superior to older agents. The authors emphasize the need to balance effectiveness with side effects and to include patients and their caregivers in decision making.

Recent trials have shown that antibiotics can be discontinued after three days in children with pneumonia. What about adults? Rachida el Moussaoui and colleagues performed a double blinded randomized controlled trial comparing three versus eight days of amoxicillin therapy in 119 adults with mild to moderate to severe community acquired pneumonia who improved after three days of treatment. They found no significant differences in clinical and radiological outcomes. In a commentary on this article, John Paul points out that one of the strengths of this study is that it was done in nine hospitals.

Finally, Sharif Al-Ruzzeh and associates randomized 168 patients in England who were undergoing primary coronary artery bypass grafting to receive surgery with or without use of a cardiopulmonary bypass pump. One surgeon performed all of the operations. Those who had less invasive off-pump surgery had significantly better clinical outcomes and neurocognitive function as well as shorter hospital stays than the pump patients, with equally good graft patency.

June 3

In 2002, the BBC sponsored a randomized, controlled (but not blinded) trial of weight loss that was monitored in centers throughout the United Kingdom. A few of the participants were featured on television. Helen Truby et al report on the entire trial cohort. Over a six month period, the 231 patients lost an average of about 13 pounds each. The amount lost didn�t vary whether they were enrolled in supported programs such as Weight Watchers, home food substitution programs such as Slim Fast, or do-it-yourself programs instructed by books, such as the Atkins diet. The take-home message is that you can lose weight if you follow one of these programs. However, most of the weight was regained by the end of one year among the large proportion who stopped the diets after six months. In an accompanying editorial, David Arterburn states that there are numerous ways to safely lose weight in the short run. The challenge to patients, researchers, public health officials, and doctors is to find ways to maintain participation in programs that work.

Two studies this week have valuable lessons to impart in healthcare quality improvement. Peter Brampton and associates report their work in surveillance of patients with Barrett�s esophagus. Although they were working with a rather rare and exotic disease, their results are generalizable: publishing clinical guidelines and performing audits didn�t change clinical practice. Institution of a program with managers responsible for surveillance, however, led to dramatic improvements in guideline compliance.

Similarly, the much more ambitious 100 000 lives campaign in the US is well on its way to improving care for six hospital based conditions and procedures, report Joseph McCannon and colleagues. The lessons learned included choosing evidence based interventions, ensuring leadership commitment, setting clear aims, executing a plan, and measuring progress.

Finally, Uffe Ravnskov et al attack aggressive use of statin medications to lower cholesterol in high risk patients. They cite data that 85% of Norwegian men over age 40 would be categorized as high risk by the US National Cholesterol Education Program. To lower their LDL cholesterol levels to 70 mg/dl or below would require high doses of statins, with a correspondingly high level of side effects: myalgias and rhabdomyolysis, mental and neurological symptoms, even increased cancer. The authors argue that the dramatic increase in side effects brought on by high dose statins would outweigh any cardiac benefits.

Archive of US Highlights