Highlights of this issue

CMAJ 2000;162:1397

In 1998 an outbreak of 39 cases of Escherichia coli O157:H7 in southern Ontario resulted in the national recall of Genoa salami. A case–control study identified that this product, produced by fermenting and drying raw meat, was the likely source (odds ratio 8, 95% confidence interval 2–35). Samples obtained from the most commonly identified supplier tested positive for E. coli O157:H7. Only two-thirds of the people infected recalled eating salami in the week preceding their illness, but 95% said that they had consumed a sliced deli product, implicating the slicer as a possible source of contamination. A similar outbreak reported in the United States in 1994 led to more stringent US production standards. According to editorialist Susan Tamblyn, meaningful changes to Canada's food safety standards were not initiated until 1999.

Lung cancer tends to be a symptom-prompted diagnosis, discovered at an advanced stage, with an overall death rate of 90%. Currently agencies in Canada and the United States recommend against screening. When baseline data from the Early Lung Cancer Action Project (ELCAP) demonstrated that CT scanning is greatly superior to traditional radiography at detecting asymptomatic lung cancer, the National Cancer Institute developed plans for a long-term randomized controlled trial (RCT) to measure the life-saving effectiveness of screening for lung cancer.

Such an RCT would be redundant according to Olli Miettinen, who illustrates that the cost-effectiveness of screening for lung cancer can already be deduced. In the ELCAP study, 23 (70%) of the 31 cases of lung cancer detected were stage IA, which has a curability rate of 70%. Miettinen therefore argues that the curability rate among cases diagnosed through screening would be about 50% (0.7 × 0.7 = 0.49). The cost (C) of one CT scan is about $200. By deducing a case-detection rate of 0.5%, a reduction in overall mortality of 40 percentage points (from 90% to 50%) and perhaps a 10-year gain in the patient's life expectancy, Miettinen calculates that the gain in life expectancy (E) from a single repeat screen would be 0.02 years (0.005 × 0.4 × 10). The cost per life-year gained (C/E) would be $10 000 ($200/0.02). We don't need an RCT to prove that suitably specified CT screening for lung cancer would be cost-effective — we already know it. Ex Canada lux?

A recent RCT identified an increased risk of gallbladder disease among postmenopausal women given estrogen replacement therapy. Estrogen may prime inflammatory and nociceptive pathways and thus increase the risk of certain surgical procedures. Using administrative data for 800 000 female residents of Ontario aged 65 years and older, Muhammad Mamdani and colleagues have compared the incidence of cholecystectomy and appendectomy among women recently prescribed estrogen replacement therapy, levothyroxine or a dihydropyridine calcium-channel antagonist (DCCA). Compared with women taking a DCCA, those taking estrogen were significantly more likely to undergo cholecystectomy (age-adjusted risk-ratio [aRR] 1.9) and appendectomy (aRR 1.8). No significant difference in either outcome measure was found between the levothyroxine users and the DCCA users.

David Pelz and Stephen Lownie review some of the evidence and the controversies surrounding carotid angioplasty and stenting. Studies of selected patients who underwent this procedure have reported technical success rates of 83% to 99% and combined stroke and death rates of 1.4% to 9%. Issues such as the use of down-stream ballooning, the need for stenting after angioplasty and the true incidence of embolic events associated with carotid angioplasty and stenting need to be resolved. Until the results of large randomized trials such as the recently funded Carotid Revascularization Endarterectomy versus Stent Trial become available, the authors advise that the procedure be reserved for patients with significant contraindications to surgery.

© 2000 Canadian Medical Association or its licensors