Search tips
Search criteria 


Logo of bmjThis ArticleThe BMJ
BMJ. 2008 January 5; 336(7634): 16–17.
PMCID: PMC2174765

All you need to read in the other general journals

Take care with p53

N Engl J Med 2007;357:2552-61 [PubMed]

The protein p53—so called because it is a phosphoprotein with a molecular mass of 53 kD—could be one of the most important molecules in biology, writes an expert in cancer genetics (p 2539). p53 helps control angiogenesis, apoptosis, and cell cycling, DNA repair, and regulation of oxidative stress, all of which are key processes in the growth and development of cancers. Abnormalities in TP53—the gene that encodes p53—are common in many cancers and are often associated with prognosis. In one recent study, disruptive mutations of TP53 within squamous cell cancers of the head and neck were associated with significantly shorter survival (hazard ratio for death 1.7, 95% CI 1.2 to 2.4)2.4).

figure scut0501.f1

Exploiting what we already know about this protein and its gene won’t be easy. The functions of p53 are so diverse that manipulating them—with gene therapy, for example—could have widespread and unpredictable effects. Overexpression of TP53 causes premature ageing in mice, while at least one polymorphism is associated with longer life in humans, despite a small increase in the risk of cancer. p53 seems to be “master and commander” of the cellular processes that control growth, life, and death. We should be careful what we do with it.

One vertebral fracture leads to another, for at least 15 years

JAMA 2007;298:2761-7 [PubMed]

Once a woman reaches 65, her absolute risk of a vertebral fracture depends firstly on whether or not she already has one and secondly on her bone mineral density (BMD). In the longest cohort study so far, women with a previous fracture and established osteoporosis had a greater than 50% chance of a new vertebral fracture over the next 15 years, compared with less than 10% for women with neither. The two risk factors operated independently of each otherother.

figure scut0501.f2

History of fracture was a better predictor of future fractures (odds ratio 4.21, 95% CI 3.33 to 5.34) than BMD (odds ratio per standard deviation decrease in BMD at the hip 1.78, 1.58 to 2.00), and the authors suggest that women with a vertebral fracture should be treated for osteoporosis regardless of the results of x ray absorptiometry. They also say that doctors should try harder to identify women with these kinds of fractures. Up to two thirds remain undiagnosed and untreated. BMD at the lumbar spine was a better predictor than at other sites. The authors estimate that each standard deviation decrease in BMD ages a woman’s bones by around five years.

Vitamin D still an unproved treatment for chronic kidney disease

Ann Intern Med 2007;147:840-53 [PubMed]

Most patients with severe chronic kidney disease take vitamin D to help normalise their disrupted bone metabolism. Renal doctors have prescribed it for years, guidelines recommend it, and third party payers willingly fund the treatment. All that may have to change, writes an editorial (p 880), after a recent meta analysis found very little robust evidence to support giving any kind of vitamin D to patients with chronic renal failurefailure.

figure scut0501.f3

Traditional compounds such as calcitriol and alfacalcidol had inconsistent effects on serum concentrations of parathyroid hormone (the main target for treatment), but increased serum concentrations of both calcium and phosphate. Newer treatments significantly reduced parathyroid hormone concentrations, but neither group of drugs had any discernible effect on bone pain, vascular calcification, risk of fractures, need for parathyroidectomy, or survival. Few trials bothered to look, relying instead on surrogate markers of bone metabolism. The authors and the linked editorial agree that this oversight leaves a large hole in the evidence. Vitamin D might have a beneficial effect on clinical risk factors, but it is also possible that vitamin D could be useless or even harmful in the long run. Large placebo controlled trials with hard clinical end points are still justified and must be done. Doctors should consider changing their practice until the results are in.

Bacteria invade cells during urinary tract infections

PLoS Med 2007;4:e329 doi: 10.1371/journal.pmed.0040329 [PubMed]

Many women treated for a urinary tract infection get another one within six months. They may be reinfecting themselves with faecal Eschericia coli, but researchers from the US have found an alternative explanation—communities of bacteria that live within bladder cells and periodically break out to cause symptomatic reinfectionreinfection.

figure scut0501.f4

They looked for the intracellular bacteria in 100 volunteers by examining exfoliated bladder cells in specimens of midstream urine. A combination of light microscopy, immunofluorescence, and electron microscopy found intracellular communities in 14 of 80 specimens from women with acute urinary tract infections. All the colonies were of E coli. They looked slightly different from regular E coli, however, and seemed to form long filaments when released. Researchers found the filaments in nearly half of the infected specimens. Women with intracellular bacteria or filaments in their urine had significantly worse cystitis than women without them. No intracellular bacteria or filaments were seen in 20 samples from women without active infections.

These researchers have already characterised an intracellular cycle of bacteria that reinfects urine in laboratory mice. They suspect the same thing could be happening in some humans. Intracellular bacteria are safe from many antibiotics and largely protected from host immunity.

CT angiography is as good as ventilation perfusion scanning at excluding pulmonary emboli

JAMA 2007;298:2743-53 [PubMed]

For years, the diagnostic test of choice for patients with suspected pulmonary embolism was a ventilation perfusion scan. Doctors in the developed world now have an alternative—computed tomography of the pulmonary vessels—which they are turning to in increasing numbers. Evidence is lagging behind practice, but at least one head to head trial has been done and suggests that the new technology is just as good as the old at ruling out clinically important emboli.

Two of 561 (0.4%) adults cleared of pulmonary embolism using a standard diagnostic pathway that included computed tomography (CT) angiography developed clinically important venous thromboembolism within three months. This compared with six of 611 (1%) adults cleared by the same pathway that used a ventilation perfusion scan instead of CT angiography (difference -0.6%, 95% CI -16% to 0.3%).

This was a non-inferiority trial, so we still have no evidence that CT angiography is “better” or more accurate than ventilation perfusion scanning, says an editorial (p 2788). CT angiography picked up significantly more emboli in adults with suspected disease (133/701 (19%) v 101/716 (14%); P=0.01). But we don’t know how many were false positives or clinically unimportant. Ventilation perfusion scanning isn’t dead yet, say the study’s authors.

US task force rules against routine screening for carotid artery stenosis

Ann Intern Med 2007;147:860-70 [PubMed]

Ann Intern Med 2007;147:854-9 [PubMed]

In 1996 the US Preventive Services Task Force was unable to rule decisively on whether or not routine screening for carotid artery stenosis would prevent strokes. Several large randomised controlled trials later, the task force has decided that screening followed by a carotid endarterectomy is likely to do more harm than good and should be discouraged. A careful (but not completely systematic) review of all the data published in English concluded that available screening tests weren’t accurate enough and the surgical treatment for stenosis wasn’t safe enough to justify screening people without symptoms.

No randomised trials have directly looked at the effects of screening on later risk of stroke. In randomised trials of carotid endarterectomy, surgery reduced the five year risk of stroke from around 11% to around 6% in people with substantial silent stenosis. But the patients and surgeons were all carefully selected, and these benefits are unlikely to translate well to primary care populations, say the review’s authors. Between 3% and 5% of patients die or have a stroke within a month of surgery, and any benefits don’t materialise for at least 18 months. The risk of a perioperative heart attack was unaccounted for in trial results. So were the possible harms of screening tests, which include false positive results from duplex ultrasonography and strokes after cerebral angiography.

Inconsistent results leave doubts over mechanical bowel preparation

Lancet 2007;370:2112-7 [PubMed]

Cleaning out the bowel before an elective colorectal operation makes sense in theory, but does it help prevent anastomotic leaks in practice? One randomised trial reported an absolute difference of only 0.6% in the risk of a leak between groups of patients who did and did not have mechanical bowel preparation before surgery (4.8% (32/670) with preparation v 5.4% (37/684) without; P=0.69) and the researchers conclude that the practice is unnecessary for elective patients.

The authors of an accompanying comment are more cautious, however (p 2073). Patients who did not have the bowel preparation and ate normally the day before surgery were significantly more likely to get an intra-abdominal abscess (4.7% (32/684) v 2.2% (15/670); P=0.02), and it is unclear how many had septic complications that were too minor to need intervention but could still cause strictures or poor bowel function in the long run. Surgeons should assess the need for a mechanically prepared bowel on a case by case basis, they say. It is too early to give it up entirely, particularly for patients with low anastomoses who are more likely to leak than those having surgery higher up. Patients in this trial were prepared using oral polyethylene glycol with bisacodyl, or sodium phosphate.

Articles from The BMJ are provided here courtesy of BMJ Publishing Group