|
.gif)
|
|
The Nugent Gram stain scoring system
for evaluation of vaginal smears for the presence of bacterial vaginosis should be adjusted to
compensate for differences among the image areas covered by different microscopes.
Diagnosis of bacterial vaginosis: need for validation of microscopic
image area used for scoring bacterial morphotypes.
Larsson P-G, Carlsson B, Fåhraeus L, Jakobsson T, Forsum U.
Sex Transm Infect 2004;80:63-67.
Summary:
Question
How does the area covered by the microscopic field influence the score of a Gram-stained vaginal smear
evaluated for bacterial vaginosis (BV) using the Nugent method?
Design
Nugent scores, which are assigned to Gram-stained vaginal smears according to the number of specific
bacterial morphotypes seen per microscopic 1000X field, were determined using two microscopes with
different size fields. The Nugent scores obtained by one microscope were recalculated based on the
relative area covered by the field of that microscope compared to the area covered by a microscope used
as the reference for Nugent scoring.
Participants
Slides from a treatment study for BV (n=913) and from a trial for screening and treatment of women with
BV during pregnancy (n=8985) were used in the evaluation. All slides were Gram stained and scored
according to the Nugent method by one investigator. At least 4 fields per slide were evaluated.
Description of Tests and Diagnostic Standard
The image areas covered by 6 different microscopic set-ups were measured using a stage micrometer with
a 0.01 mm interval scale. The image area of one microscope, a Zeiss FL30 from the early 1990s with an
area of 0.0165 mm2, which had been used in an international validation of Nugent scoring, was used as
the reference for Nugent scoring in this study. Graphs and tables were constructed that adjusted the
number of bacteria used as the cut-off points for the Nugent scores according to the increase in area of
the microscopic field. The adjusted numbers of bacteria were used to recalculate new scores after
compensation for the image area of the microscope. The adjusted Nugent scores from one microscope
were compared to the Nugent scores obtained using the reference microscope. Most of the slides with a
Nugent score above 3 were re-evaluated by one investigator alone or together with another
investigator using the Hay/Ison classification.
Main Outcome Measures
Weighted kappa indexes were calculated to investigate the agreement between the adjusted Nugent
scores and the Nugent scores obtained from the reference microscope and the Hay/Ison scores.
Main Results
The image area of the 6 microscope set-ups varied between 0.0165 and 0.049 mm2. The Nugent scoring
cut-offs for the different microscope set-ups were adjusted using the 0.0165 mm2 area as the 100%
reference (table). When the Nugent scores for slides from two patient studies, which were determined
on a microscope with a relative field area 297% larger than the reference, were recalibrated and
compared to the original Nugent scores, slides assigned to the normal group and the BV group did not
significantly change Nugent scores. However, in one study, among the 232 slides originally assigned to
the intermediate group, 36 were reclassified as normal, 178 as intermediate, and 18 as BV. Among the
1176 slides originally assigned to the intermediate group in the other study, 701 were still classified as
intermediate, 49 were reclassified as BV, and 426 as normal. When the recalibrated scores were
compared to the scores obtained on the reference microscope, the weighted kappa values were 0.94 for
one study and 0.88 for the other. A comparison of slides with Hay/Ison scoring and recalibrated Nugent
scores for the two studies gave kappa values of 0.88 and 0.90.
Assignment of points for numbers of bacterial morphotypes in Nugent scoring using different
microscope set-ups
|
Nugent score
|
Number of points by relative size of microscopic field
|
|
100%
|
210%
|
297%
|
| 0 |
>30 |
>50 |
>90 |
| 1 |
5-29 |
10-49 |
15-89 |
| 2 |
1-4 |
2-8 |
3-12 |
| 3 |
0-1 |
0-2 |
0-3 |
| 4 |
0 |
0 |
0 |
Authors' Conclusions
The Nugent Gram stain scoring system should be adjusted to compensate for differences among the
image areas covered by different microscopes.
Source of funding: None given
For correspondence: P-G Larsson, Department of Obstetrics and Gynecology, Kärnsjukhuset, 541 85 Skövde, Sweden. E-mail address: p-g.larsson@vgregion.se.
|
