York Hospital AFFIRM Poster CVS 2010 44x92_template

Report
Comparison of Primary Care and Clinical Laboratory Methods for
the Diagnosis of Bacterial Vaginosis and Yeast Vaginitis
Arthur E. Crist, Jr.1, David. Bankert1, Roslyn V. Mallory2, Elliot L. Rank3, Michelle L. Althouse3
Clinical Microbiology Laboratory1 and Department of Obstetrics and Gynecology2, York Hospital, York, PA and BD Diagnostics3, Sparks, MD
BACKGROUND
Traditional diagnosis of bacterial vaginosis (BV) or yeast
vaginitis (YV) relies on patient symptomatology, clinical
findings during vaginal examination, and analysis of
vaginal fluid. Once a diagnosis is established, antimicrobial
chemotherapy is initiated. Alternatively, vaginal specimens
may be sent to a clinical or hospital based laboratory where
microscopy (Nugent Scored Gram Stain), culture, and
molecular methods, e.g. BD AFFIRM™ VPIII Test, may be
utilized. Failure to make the correct diagnoses may delay
appropriate antimicrobial chemotherapy leading to
possible sequelae in the case of BV, whereas patients
misdiagnosed or treated empirically are subjected to
unnecessary antimicrobial agents.
OBJECTIVE
To compare primary care based and clinical
laboratory based test methods to patient
diagnosis and treatment for bacterial vaginosis
and yeast vaginitis
REFERENCES
1. Amsel, R., P.A. Totten, C.A. Spiegel, et al. 1983. Non-specific
vaginitis: diagnostic criteria and microbial and epidemiologic
associations. Am. J. Med. 74: 14.
2. Brown, H.L., D.D. Fuller, T.E. Davis, et al. 2001. Evaluation of the
Affirm Ambient Temperature Transport System for the detection
and identification of Trichomonas vaginalis, Gardnerella vaginalis,
and Candida species from vaginal fluid specimens. J. Clin.
Microbiol. 39: 3197-3199.
3. Nugent, R.P., M.A. Krohn, and S.L. Hillier. 1991. Reliability of
diagnosing bacterial vaginosis is improved by a standardized
method of Gram stain interpretation. J. Clin. Microbiol. 29: 297301.
printed by
www.postersession.com
MATERIALS AND METHODS
Institutional Review Board approval and patient consent were
obtained prior to specimen collection. There were 84 patients
enrolled in the study from three primary care sites. Each site utilized
their routine methods for diagnosis of vulvovaginitis which for BV
includes: (i) the presence of a homogenous vaginal discharge: (ii)
vaginal pH>4.5 (iii) a positive amine “Whiff” test; and (iv)
microscopic presence of clue cells on a saline wet mount. Three out of
4 of these criteria, when present, are indicative of BV according to
Amsel Criteria (AC) (1) . For YV each patient was assessed for (i) the
presence of abnormal vaginal discharge; (ii) vaginal pH 4.0-4.5; (iii) a
negative amine test; and (iv) microscopic presence of yeast cells or
pseudohyphae on a KOH wet mount (WM). An additional 3 swabs of
the vaginal discharge were collected and sent to the Microbiology
Laboratory at York Hospital. One specimen was collected using a
BBL CultureSwab and tested for BV using the Nugent Scored Gram
Stain (NGS) as previously described (3). A second specimen was
collected with a ESwab transport device which contains a flocked
swab in 1 mL of liquid Amies transport medium (ATM). A 10 and 100
µL sample of ATM were inoculated onto Inhibitory Mold Agar
w/Gentamicin and CHROMagar Candida for yeast isolation. Yeast
identification was performed using conventional methods. A third
specimen was collected with the Affirm Ambient Temperature
Transport System for the DNA probe based AFFIRM™ VPIII Test
(AFF) for BV and YV. The AFF was performed according to the
manufacturers instructions and has been previously described (2).
Patient medical records were reviewed by one of us (RVM) who was
blinded to the study results. Data were analyzed by comparing
diagnosis and treatment to a consensus of testing methods (CTM).
ACKNOWLEDGEMENTS
RESULTS
TABLE 1. Comparison of Methods in Patients Diagnosed with Bacterial Vaginosis (n=36)
Method
Sensitivity
Specificity
PPV
NPV
Amsel Criteria
AFFIRM
Nugent GSa
a
c
95 (21/22)
100 (22/22)
57 (8/14)
71 (10/14)
78 (21/27)
85 (22/26)
89 (8/9)
100 (10/10)
50 (11/11)
100 (14/14)
100 (11/11)
56 (14/25)
92 (22/24)
96 (23/24)
58 (7/12)
75 (9/12)
81 (22/27)
88 (23/26)
78 (7/9)
90 (9/10)
75 (18/24)
100 (12/12)
100 (18/18)
67 (12/18)
Positive Nugent Score of >6
Amsel Criteria
AFFIRM
Nugent GSb
b
Positive Nugent Score of >3
c
Percentage with fractions in parentheses to show how each was determined
TABLE 2. Comparison of Methods in Patients Diagnosed with Yeast Vaginitis (n=30)
Method
Sensitivity
Specificity
PPV
NPV
Wet Mount
AFFIRM
Yeast Culture
a
a
100 (21/21)
86 (18/21)
100 (21/21)
11 (1/9)
100 (9/9)
100 (9/9)
72 (21/29)
100 (18/18)
100 (21/21)
100 (1/1)
75 (9/12)
100 (9/9)
Percentage with fractions in parentheses to show how each was determined
CONCLUSIONS
TABLE 3. Patients Not Diagnosed with Bacterial Vaginosis or Yeast Vaginitis
Consensus of Methods Positive for:
No.
Treatment*
MET
CD
FLUC
NYS
FLUC & NYS
Bacterial Vaginosis (BV)
5
1
2
Yeast (Y)
1
1a
BV and Y
1
Negative for BV and Y
17
1
5
2
* Treatment: Metronidazole (MET); Clindamycin (CD); Fluconazole (FLUC); Nystatin (NYS)
We would like to thank BD Diagnostic for providing financial
support and York Women’s Health Centers of Queen Street, East
York, and Shrewsbury for participating in the study. We would also
like to thank Heather Challenger and Kimberly Young for providing
technical support.
There were 36 patients diagnosed and treated for BV. Twenty
two were positive and 14 negative based on a CTM. The
sensitivity and specificity for AC, NGS>6, and AFF were 95%
and 57%, 50% and 100%, and 100% and 71%, respectively
(TABLE 1). Using a NGS score of >3 as the criteria for BV
improved the sensitivity of the NGS to 75% without affecting
specificity. Using NGS>3 as part of the CTM did not
significantly effect the sensitivity or specificity of the AC or
AFF. There were 30 patients diagnosed and treated for YV. Out
of these, 21 were positive and 14 negative based on a CTM. The
sensitivity and specificity for WM, YC, and AFF were 100%
and 11%, 100% and 100%, and 86% and 100%, respectively
(TABLE 2). Out of the 43 patients positive for BV or YV, 13.9%
(6/43) were positive for both.
There were 24 patients that had a diagnosis of
something other than BV or YV. Five were positive for BV, 1
for YV, and 1 for both based on a CTM. Only 1 of the 7
patients was treated with an appropriate antimicrobial agent.
Conversely, out of the 17 negative patients, 8 were treated with
an antifungal agent (TABLE 3).
a
Patient diagnosed with Trichomonas but negative for Trichomonas on wet mount, culture, and AFFIRM
Other
2
None
1
1
4
5
1. Methods utilized and results generated at primary
care sites can lead to under or over utilization of
antimicrobial agents.
2. The AFF provides a sensitive and specific method
for identifying patients with BV and YV, provides
faster turnaround time than culture, and
eliminates the subjectivity associated with other
methods
3. Diagnosis of BV and YV remains a challenge for
both the clinician and clinical laboratory.
4. Additional studies may be warranted due to the
small sample volume

similar documents