By Louis Kuritzky, MD

Screening for Lung Cancer with Low-dose CT

SOURCE: Gould MK, et al. N Engl J Med 2014; 371:1813-1820.

The United States Preventive Services Task Force (USPSTF) gave a level B recommendation in support of annual low-dose computed tomography (LDCT) to screen for lung cancer in appropriate risk groups. The USPSTF decision was largely based on the National Lung Screening Trial (NLST), a mega-trial (n = 53,454) in the United States that randomized subjects to annual LDCT or chest X-ray. The primary endpoint of the study was lung cancer mortality, and all-cause mortality was a secondary endpoint. Inclusion criteria included at least a 30-pack/year history of smoking (if stopped within 15 years), ability and willingness to complete follow-up for abnormal findings, and absence of problematic comorbidities that might otherwise compromise long-term survival.

The good news is that LDCT was associated with a 20% relative risk reduction in lung cancer mortality and a 7% reduction in all-cause mortality, both of which were statistically significant. Should we end the discussion there?

Perhaps not. The NLST has several stark limitations. First, literally 95% of “positive” findings on LDCT were false-positive, and harms to patients during the follow-up evaluations were substantial, including deaths. Second, a not-inconsiderable number of “incidentalomas” were also detected, and follow-up data on whether these findings favorably (or unfavorably) affected study subjects’ lives has not yet been published.

Finally, an issue about the absolute magnitude of benefit. Although the 20% relative reduction in lung cancer mortality sounds impressive, the absolute risk reduction was very small; In the LDCT group, 356 of 26,309 died (1.3%) vs 443 of 26,035 in the chest X-ray group (1.7%), for an absolute risk reduction of 0.348%.

Although most major organizations have endorsed USPSTF recommendations, the American Academy of Family Physicians (AAFP) has issued a note of caution, based on lack of replication of these data in a community setting. Instead of universal screening, AAFP suggest a “shared decision-making” approach reminiscent of its advice about prostate cancer screening in the recent past.

Rethinking Acetaminophen for Acute Low Back Pain

SOURCE: Williams CM. Lancet 2014;384:1586-1596.

The natural history of acute low back pain indicates that somewhere between 60-70% of episodes have spontaneously resolved by 3 weeks and 80-90% by 3 months. We would hope that the goals of clinicians in their choice of pharmacotherapy and activities (physical therapy, exercise) are to shorten time to recovery, improve functional status during recovery, and provide symptom relief. A Cochrane Database analysis has confirmed the efficacy of NSAIDs for acute low back pain. What about acetaminophen? (Note: for readers who choose to review the original reference on this article, the word “paracetamol” is used in the original title, because that is the preferred term in the United Kingdom and Australia for what we call “acetaminophen” in the United States).

In this double-blind, placebo-controlled study conducted in Sydney, Australia, patients with acute low back pain (n = 1096) were randomized to treatment with PRN acetaminophen (up to 4000 mg/d) or placebo and followed for 3 months. The primary outcome was acute low back pain recovery, defined as a score of ? 1 on a 1-10 pain scale for at least 7 consecutive days.

No differences were found in time to recovery between groups. The authors suggest that although replication of their data with another clinical trial would make these conclusions more definitive, clinicians should be circumspect about use of acetaminophen in acute low back pain.

Non-obstructive Coronary Artery Disease: Not So Innocent After All

SOURCE: Maddox TM, et al. JAMA 2014;312:1754-1763.

It has become abundantly clear that coronary events (i.e., myocardial infarction [MI]) are not simply a result of “clogged pipes.” To the contrary, it has been documented that the majority of plaque ruptures occur within coronary arteries that have been atherosclerotically categorized as “non-obstructive.” In this report, obstructive coronary disease was defined as ? 50% stenosis of the left main coronary artery or ? 70% stenosis in other coronary arteries. Non-obstructive coronary disease was defined as 20-49% stenosis, and < 20% stenosis was categorized as “no apparent (coronary artery disease) CAD.”

The authors of this report studied all patients who underwent coronary arteriography in Veterans Adminstration hospitals from 2007-2012 (n = 37,684). Within this cohort, 8384 patients were reported to have non-obstructive coronary disease. They tracked the rate of admission for acute MI in the year following arteriography.

Compared to persons with no apparent CAD, the hazard ratio for MI at 1 year for persons with non-obstructive coronary disease was 2.0-4.5 (dependent upon the number of vessels involved); for persons with obstructive CAD, the hazard ratio was 9.0-19.5 (dependent on the number of vessels involved).