Definitions for the level of evidence (I-IV) and grade of recommendation (A-C) are provided at the end of the "Major Recommendations" field.
Recommended Tests
A variety of tests, which reflect the changes in vaginal ecology, have been used to diagnose bacterial vaginosis (BV). Isolation of the bacteria associated with this condition, such as G. vaginalis, has a poor specificity (these bacteria being present in a proportion of normal women, albeit in smaller numbers) and is discouraged. Other tests that detect the biochemical changes associated with BV are more useful for studies on pathogenesis rather than for clinical diagnosis and include the detection of sialidase and proline aminopeptidase. Two diagnostic methods for BV have been used extensively in genitourinary medicine clinics and remain the tests of choice. Both require interpretation within the given clinical scenario.
- Amsel's Criteria
The presence of three or more of the composite criteria is considered consistent with BV, and this has been used as the gold standard for many years.
- Typical appearance of discharge at vaginal examination
- Vaginal discharge pH > 4.5
- Positive 'whiff test' following the addition of potassium hydroxide to a sample of discharge
- Clue-cells on dark-ground microscopy of a saline wet mount preparation
The criteria are simple to perform, particularly in a clinic setting and require minimal material with the exception of a microscope. However, the disadvantages are that the patient must undergo a vaginal examination and the recognition of the vaginal discharge and the fishy 'smell' has a subjective endpoint. In the majority of United Kingdom (UK) clinics the 'whiff' test is no longer performed because of the caustic nature of the potassium hydroxide, hence invalidating the method, which is dependent on measurement of all four criteria to achieve a high sensitivity for the diagnosis of BV.
- Appearance of Gram-Stained Vaginal Smear
The grading or scoring of Gram-stained smears offers an alternative to use of the composite criteria; it has the advantage of a more objective endpoint, and allows for a common approach that can be audited. A microscope is required. The original method divided patients into two groups, with or without BV (normal), but subsequent methods have included an intermediate category, believed to be a transition between normal and BV. The disadvantage is that multiple methods have been described and are in use resulting in a lack of consistency in diagnosis and reporting. One method which is widely used particularly for research studies requires counting of bacteria; this is time consuming and not feasible in a busy genitourinary medicine (GUM) clinic. A number of simplified schemes have been described but the grading of vaginal flora described by Ison and Hay allows a method of assessment that gives a good correlation with Amsel's criteria for the diagnosis of BV and correlates well with other scoring methods. This latter method has been endorsed by the Bacterial Special Interest Group of the British Association for Sexual Health and HIV (BASHH).
Recommended Diagnostic Test: Appearance of Gram-Stained Smear According to Modified Ison-Hay Scoring System
Modified Ison-Hay suggests five grades of flora:
Grade 0 - Epithelial cells with no bacteria
Grade 1 - Normal vaginal flora (lactobacillus morphotypes alone)
Grade II - Reduced numbers of lactobacillus morphotypes with a mixed bacterial flora
Grade III - Mixed bacterial flora only, few or absent lactobacillus morphotypes
Grade IV - Gram positive cocci only
Grades 0, I and IV are found in women without BV
Grade II is intermediate and not found in women with BV as defined by Amsel's criteria.
Grade III is consistent with BV as diagnosed by Amsel's criteria.
Thus, only Grade III flora is indicative of BV. There is some evidence to suggest that Grade II flora responds to oral, but not vaginal clindamycin in pregnant women. There is insufficient evidence on the clinical significance of grades 0, II and IV in the non-pregnant population and their response to standard treatment regimens for BV.
Diagnostic Methods for BV
| Amsel' s Criteria |
Gram-Stain Ison/Hay |
Convenient to Perform |
Yes |
Yes |
Microscope Required |
Yes |
Yes |
Caustic Material Required |
Yes |
No |
Reproducible |
No |
Yes |
Screening Should Take Place in the Following Patient Groups
- Women presenting with vaginal discharge, an offensive odour or any genital symptom. (Evidence Level Ia; Grade of Recommendation A)
- Women found to have a copious discharge at examination. (Grade of Recommendation A)
- Pregnant women with a history of previous pre-term labour may be offered screening. (Evidence Level Ia; Grade of Recommendation A)
- To date there is insufficient evidence to support routine screening of asymptomatic pregnant women. (Evidence Level Ia; Grade of Recommendation A)
- There is some evidence to support screening and treating BV prior to termination of pregnancy to reduce subsequent endometritis and pelvic inflammatory disease (PID) (Evidence Level Ib; Grade of Recommendation B)
- There is a complete lack of evidence to inform any decision on screening asymptomatic non-pregnant women as regards PID outcomes. (Evidence Level IV; Grade of Recommendation C)
Sites for Testing
Vaginal wall smear following the insertion of a speculum is recommended.
Pre-pubertal women and those declining speculum examination
- Posterior vaginal wall sample -blind
Pregnant women
Sex workers
Men
Recommendation for Test of Cure
There is no available evidence to support or refute need for a test of cure. (Evidence Level IV; Grade of Recommendation C)
Definitions:
Levels of Evidence
Ia: Evidence obtained from meta-analysis of randomised controlled trials
Ib: Evidence obtained from at least one randomised controlled trial
IIa: Evidence obtained from at least one well designed controlled study without randomisation
IIb: Evidence obtained from at least one other type of well designed quasi-experimental study
III: Evidence obtained from well designed non-experimental descriptive studies
IV: Evidence obtained from expert committee reports or opinions and/or clinical experience of respected authorities
Grading of Recommendations
- Evidence at level Ia or Ib
- Evidence at level IIa, IIb, or III
- Evidence at level IV