But the recent UK PREMET Study of 900 pregnant women has strongly suggested that metronidazole not only doesn’t reduce premature births, but may also actually contribute to the very condition it is prescribed to treat (Br J Obstet Gynaecol, 2006; 113: 65-74).

Indeed, according to Andrew Shennan, the author of the report, and professor of obstetrics for Tommy’s, the baby charity, almost 1000 babies are born prematurely in the UK each year due to the use of metronidazole.

But these findings are nothing new. It has been well known for some years that the antibiotic doesn’t prevent premature births in high-risk pregnant women (N Engl J Med, 2000; 342: 534-40).

Given this evidence, it is somewhat alarming that ineffective and dangerous drugs are still being regularly prescribed to such a vulnerable patient population. What’s equally alarming is the extent to which the serious side-effects of metronidazole have been known.

In fact, more than 20 years ago, a link between cancer and this drug emerged after three patients with Crohn’s disease were treated with large doses of this antibiotic. All three developed cancer (Am J Gastroenterol, 1985; 80: 978-82). In this report, the association was worrying enough for the authors to suggest that it warranted concern.

Indeed, the drug’s manufacturer - Baxter Healthcare Corporation - has itself issued a number of warnings related to the use of metronidazole. For example, the drug has been shown to cause cancer and tumours in rats and mice, and to be toxic to the fetuses in pregnant mice when administered directly into the abdominal cavity. Although findings in animals do not necessarily apply to humans, there have been no proper well-controlled studies of the use of the drug in pregnant women and, according to Baxter’s own drug information sheet, metronidazole “should be used during pregnancy only if clearly needed”.

But cancer and preterm births in at-risk women aren’t the only causes of concern with metronidazole. The drug has been associated with inflammation of the pancreas, albeit rarely (J Pancreas [online], 2004; 5: 516-9; Ann Pharmacother, 2000; 34: 1152-5), as well as vomiting, diarrhoea, headaches and seizures.

More serious, however, is the effect of metronidazole on the central nervous system (see box, page 16). In a 1997 report, a 48-year-old Nigerian man with a liver abscess due to amoebic infection developed an encephalopathy (degenerative brain disease) three days after taking oral metronidazole (Afr J Med Med Sci, 1997; 26: 97-8).

Such brain damage was also found in a study carried out in Korea. The findings also suggested that metronidazole can induce a whole host of other neurological adverse effects - from seizures to peripheral neuropathy (numbness, tingling, burning or weakness in the hands and/or feet) (Korean J Gastroenterol, 2005; 45: 195-200).

Furthermore, there has been a case report of a 25-year-old Mexican woman who had taken metronidazole (500 mg three times a day) for three days before developing two ulcers of the mid-oesophagus. In her case, the ulcers began to heal a week after she stopped taking the drug (Rev Gastroenterol Mex, 1998; 63: 106-7).

The case above suggests that the antibiotic should not be prescribed to those with gastroenterological problems - conditions for which it has been prescribed for decades. Yet, despite such adverse effects with metronidazole, it is still widely prescribed for various conditions and infections such as Crohn’s disease and Helicobacter pylori, and for the treatment of anaerobic infections and liver abscesses.

More research into the safety of this drug has been firmly recommended by a number of studies, including, in particular, investigations into the connection between large doses (around 500 mg/kg/day) of metronidazole and cancer.

Amanda Diamond