Herbal suppositories for bacterial vaginosis?

Bacterial vaginosis is a common condition which is more than a triviality. It can have serious consequences including pelvic inflammatory disease, endometritis, postoperative vaginal cuff infections, preterm labor, premature rupture of membranes, and chorioamnionitis. Therefore, it is important to treat it adequately with antibiotics. But are there other options as well?

There are plenty of alternative or ‘natural’ treatments on offer. But do they work?

This trial was conducted on 127 women with bacterial vaginosis to compare a vaginal suppository of metronidazole with Forzejeh, a vaginal suppository of herbal Persian medicine combination of

• Tribulus terrestris,
• Myrtus commuis,
• Foeniculum vulgare,
• Tamarindus indica.

The patients (63 in metronidazole group and 64 in Forzejeh group) received the medications for 1 week. Their symptoms including the amount and odour of discharge and cervical pain were assessed using a questionnaire. Cervical inflammation and Amsel criteria (pH of vaginal discharge, whiff test, presence of clue cells and Gram staining) were investigated at the beginning of the study and 14 days after treatment.

The amount and odour of discharge, Amsel criteria, pelvic pain and cervical inflammation significantly decreased in Forzejeh and metronidazole groups (p = <.001). There was no statistically significant difference between the metronidazole and Fozejeh groups with respect to any of the clinical symptoms or the laboratory assessments.

The authors concluded that Forzejeh … has a therapeutic effect the same as metronidazole in bacterial vaginosis.

The plants in Fozejeh are assumed to have antimicrobial activities. Forzejeh has been used in folk medicine for many years but was only recently standardised. According to the authors, this study shows that the therapeutic effects of Forzejeh on bacterial vaginosis is similar to metronidazole.

Yet, I am far less convinced than these Iranian researchers. As this trial compared two active treatments, it was an equivalence study. As such, it requires a different statistical approach and a much larger sample size. The absence of a difference between the two groups is most likely due to the fact that the study was too small to show it.

If I am correct, the present investigation only demonstrates yet again that, with flawed study-designs, it is easy to produce false-positive results.