Discussion Questions

• ■
  

  How does the study design affect the researchers’ results?

  
  
• ■
  

  How might you explain the superior predictivity of the Nugent score?

  
  
• ■
  

  Do you agree with the researchers’ stated study limitations?

  
  
• ■
  

  Why might bacterial vaginosis be associated with STIs?

  
  
• ■
  

  How will this study influence the way you practice?

Bacterial vaginosis (BV) is a common condition associated with a shift in the normal flora of the vagina. Sometimes this shift is accompanied by a symptomatic vaginal discharge that prompts the patient to seek medical care. Additionally, several studies have linked BV with an increased incidence of sexually transmitted infections (STIs). Notable new research by Allsworth and Peipert analyzes different diagnostic strategies for the assessment of BV in an effort to identify patients at increased risk for STIs. The participants in our Journal Club were eager to determine whether categorization of BV severity would provide a marker that enhanced detection of previously undiagnosed STIs?

There are 2 methods for classifying BV in patients: the Nugent score, which is often used in research settings, and Amsel’s criteria, which is a useful clinical tool. Nugent scoring is based on stratifying and estimating the prevalence of certain types of Gram-positive and Gram-variable rods from a vaginal fluid Gram stain. With Amsel’s criteria, 3 of 4—a positive whiff test, the presence of thin white/gray discharge, ph >4.5, or more than 20% clue cells per HPF—indicate BV. The Nugent scores are often reserved for research trials because they have been shown to be more sensitive and specific than Amsel’s criteria.

In this study, both the Amsel’s and Nugent criteria were used to assess BV severity. A statistically significant correlation between BV severity and incident STI was only identified in patients with high Nugent scores (≥9). However, Journal Club participants agreed that, although the Nugent score resulted in a more accurate diagnosis of BV, it is too cumbersome a test for the average clinician. A Gram stain slide must be sent to a certified reader, often a pathologist, for grading of microflora in the sample.

This article used data from Project PROTECT, a randomized controlled trial that compared 2 methods of promoting dual-method contraception among women prone to high-risk sexual activity. In the original study, all patients used a computer-based multimedia program. Upon completion, women in the intervention arm received individualized feedback based on the transtheoretical model of behavior change; those in the control arm received general information on contraception. Primary outcomes of this trial were self-reported use of dual methods of contraception and the biological outcomes of an unintended pregnancy or an incident or recurrent sexually transmitted disease.

The new study was designed as a secondary analysis of the prospective data obtained in Project Protect. Journal Club members were concerned about how the randomization and effects of the intervention in the primary trial shaped the data in the secondary analysis. First, it is unclear what type of contraception women were using throughout the trial. Certain forms of contraception, particularly barrier methods and those that alter cervical mucus, can reduce the risk of STI transmission and lower the incidence of pelvic inflammatory disease (PID). Second, the use of condoms prevents the vagina from being exposed to semen. Because semen is an alkalinizing agent, it can spur alteration of the vaginal flora, potentially predisposing women to a higher incidence of BV. If the women included in this trial were using these contraceptive methods, the results would be different from those obtained in a group of women who were not using contraception; particularly condoms.

Additionally, this research trial was designed to be a prospective study from baseline diagnosis of BV until the first positive STI result. The authors only report assessing BV status at the onset of the study and do not describe whether or not they assessed BV status at any other time during the 24-month follow-up period. Furthermore, the authors do not address whether or not women diagnosed with BV were treated at the time of initial diagnosis or at any other time during the course of the study. It was also unclear how often women were screened or if they underwent routine STI screening during the study. These issues make it difficult to ascertain whether having continuous, recurrent, or 1 unique episode of BV would make women more predisposed to STI acquisition. At the time of the incident STI, the authors make no comment as to the presence or severity of BV. This study does suggest that the most extreme cases of BV, as denoted by Nugent score, are associated with higher occurrence of STI.

After discussing this study, Journal Club participants took away an increased awareness and appreciation for the link between high-risk behaviors, BV, and STIs. A metaanalysis of 43 studies found new or multiple sex partners—male or female—and inconsistent condom use put women at higher risk of BV, but it did not identify BV as an STI. Nonetheless, BV infection has also been associated with an increased risk of transmission of herpes simplex virus (HSV) and the human immunodeficiency virus (HIV). Finally, previous research has demonstrated that treating recurrent BV can decrease the incidence of Chlamydia trachomatis infections. Therefore, we agreed with the authors’ hypothesis that an increased incidence of BV would be associated with an increased incidence of STIs. However, we questioned whether complex scoring systems were more sensitive than the use of historical risk factors alone. An assessment of the effectiveness of historical risk factors in comparison to BV scoring criteria would have enhanced the overall value of this study.

The finding that the presence of BV and the severity of BV each increase the risk of STIs was interesting to us because it confirmed previous findings, but we did not feel that it would alter clinical practice. Use of Nugent scoring in clinical practice would neither be more efficient than use of Amsel’s criteria nor more cost-effective for global implementation. In routine practice, it would be unfeasible to send a Gram stain to a certified reader any time a patient was suspected of having BV. But we do encourage providers to assess high-risk sexual habits and understand their associations with STIs.