One in 5 patients with symptoms of vaginitis have 2 causes of vaginitis
In a recent study, 1,471 patients with a symptom of vaginitis (abnormal vaginal discharge, itching or irritation, or odor) self-collected a vaginal swab and had a vaginal swab collected by a clinician.12 The swabs were placed in buffer and the samples were tested by NAAT using the BD Max system (Franklin Lakes, New Jersey) for the presence of nucleic acid sequences of the microorganisms responsible for the most common causes of vaginitis. In this cohort, using the clinician collected vaginal swabs for NAAT, the investigators reported the following pattern of detection of nucleic acid sequences: 36.1%, bacterial vaginosis pattern; 16.2%, Candida spp.; 1.6%, T vaginalis; 0.7%, Candida glabrata; and 0.1%, Candida krusei. Nucleic acid sequences of multiple organisms were detected in 21.7% of patients, including 13.9% with bacterial vaginosis pattern plus Candida spp., 4.9% with bacterial vaginosis pattern plus T vaginalis, 0.3% with Candida spp. plus T vaginalis, 0.2% with Candida spp. plus Candida glabrata, 0.2% with bacterial vaginosis pattern plus Candida glabrata, and 2.2% with all 3 organisms. A total of 23.8% of the women had no detectable nucleic acid sequences associated with organisms known to cause vaginitis.
In another study of 1,491 patients with a symptom of vaginitis, clinician-collected vaginal swabs were tested by NAAT using the Aptima BV and Aptima Candida/Trichomonas systems (Hologic, Marlborough, Massachusetts) for the presence of nucleic acid sequences of microorganisms responsible for most cases of vaginitis.13 The investigators reported the following pattern of detection of nucleic acid sequences: 28.6%, bacterial vaginosis pattern; 14.2%, Candida spp.; 3%, T vaginalis; 1.9%, Candida glabrata.13 Nucleic acid sequences from multiple organisms were detected in 23.3% of patients. Nucleic acid sequences suggesting the presence of two different causes of vaginitis were detected among 20.8% of patients, including bacterial vaginosis plus Candida spp., 11.1%; bacterial vaginosis plus T vaginalis, 7.2%; Candida spp. plus T vaginalis, 1.0%; Candida spp. plus Candida glabrata, 0.9%; bacterial vaginosis plus Candida spp., 0.5%; Candida glabrata plus T vaginalis, 0.1%. Nucleic acid sequences suggesting the presence of 3 different causes of vaginitis were detected in 2.4% of patients, the most common being the combination of bacterial vaginosis plus Candida spp. plus T vaginalis, 1.7% and bacterial vaginosis plus Candida spp. plus Candida glabrata, 0.5%. Nucleic acid sequences suggesting the presence of 4 different causes of vaginitis were detected in 0.1% of patients. A total of 28.8% of the women had no detectable nucleic acid sequences associated with organisms known to cause vaginitis.13
In clinical practice it is uncommon to see the diagnosis of multiple causes of vaginitis recorded in the medical record of a patient. This suggests that we are not effectively identifying the 20% of patients with multiple causes of vaginitis.
When multiple organisms that cause vaginitis are present, NAAT is superior to clinical evaluation for diagnosis
In a study of 1,264 patients with symptoms of vaginitis who had an identified microbial cause, more than 20% had multiple organisms detected by NAAT.10 The reference methods for diagnosis in this study were Nugent scoring with Amsel criteria to resolve intermediate Nugent scores for bacterial vaginosis, culture for Candida, and culture for T vaginalis. Compared with the reference method for diagnosis, the sensitivities for NAAT and clinician detection of cases of bacterial vaginosis plus Candida were 74% and 18%, respectively (P <.0001). Compared with the reference method for diagnosis, the sensitivities for NAAT and clinician detection of cases of bacterial vaginosis plus T vaginalis were 72% and 21%, respectively (P <.0001). Compared with the reference method for diagnosis, the sensitivities for NAAT and clinician detection of cases of bacterial vaginosis plus Candida plus T vaginalis were 80% and 10%, respectively (P <.0005).10 Based on this one study, it appears that clinicians are not very effective at diagnosing a case of vaginitis caused by multiple different microorganisms.
Patient collection of a vaginal swab for NAAT
Multiple studies have reported that collection of a vaginal swab for NAAT by the patient or a clinician results in similar excellent test performance.4,12,13 This observation might catalyze the development of clinical protocols where patients with vaginitis could collect the swab for NAAT analysis, without needing to have a speculum examination by a clinician.
When collecting a vaginal specimen for NAAT it is important that no vaginal lubricants or creams contaminate the collection swab. Vaginal lubricants and creams may inhibit the polymerase chain reaction enzymes resulting in a false negative. The swab may be directly inserted into the vagina to collect the specimen or a speculum without a lubricant, except water can be used to facilitate specimen collection. To collect a specimen without a speculum the swab is inserted 2 inches into the vagina and rotated for 10 to 15 seconds.
What should clinicians do while waiting for a NAAT result?
A major problem with NAAT testing for vaginitis is that the results are not available at the initial patient visit, impacting the ability to make an immediate diagnosis and provide targeted antibiotic treatment. Given that bacterial vaginosis and Candida species are the most common causes of infectious vaginitis in many populations of gynecology patients, one approach is to initiate treatment with one dose of an oral antifungal agent and a multiday course of vaginal metronidazole. Once the NAAT test results are available, the treatment can be refined to specific infectious agents identified by the test, or the antibiotics can be discontinued if no relevant microorganisms are detected. Another approach would be to wait until the NAAT test is completed and then prescribe the appropriate antibiotic. My sense is that most patients would not favor this wait and see approach.
Barriers to the use of NAAT for vaginitis
A barrier to the use of NAAT for the diagnosis of vaginitis is that leading organizations do not currently recommend NAAT as a primary approach to diagnosis, favoring microscopy and measurement of vaginal pH.9 In addition, clinicians and patients may be rightfully concerned about the cost of NAAT, which can be substantial.
Vaginitis, especially when it is recurrent, can be stressful14 and have an impact on a patient’s quality of life15,16 and sexual health.17 Arguably, our current practice algorithms for diagnosing the cause of vaginitis are not optimized.18 Our failure to accurately diagnose the cause of vaginitis contributes to inappropriate antibiotic treatment and return visits because of inadequate initial treatment.18 We can improve and simplify our approach to the diagnosis of vaginitis by prioritizing the use of NAAT.19 In turn, reliably making the right diagnosis will result in the optimization of treatment. ●