|Year : 2017 | Volume
| Issue : 1 | Page : 43-48
Comparison of visual, clinical, and microbiological diagnosis of symptomatic vaginal discharge in the reproductive age group
Abdelsattar M Farhan, Elsayed A Eldesouky, Essam A Gaballah, Mohammed E Soltan
Department of Obestetrics and Gynecology and Department of Microbiology, Faculty of Medicine, Alazhar University, Cairo, Egypt
|Date of Submission||30-Oct-2016|
|Date of Acceptance||14-Dec-2016|
|Date of Web Publication||24-May-2017|
Mohammed E Soltan
Menoufiya, Berket Elsabee, Gamal Abd Elnaser Street, Cairo, Postal Code 31111
Source of Support: None, Conflict of Interest: None
Symptomatic vaginal discharge is the most frequent symptom in women of reproductive age group. Discharge may be physiological or pathological, for which bacterial vaginosis is the most common cause, followed by candidiasis and trichomonas. Multiple infections can also coexist.
This study aimed to determine the microbiological and clinical profile of symptomatic vaginal discharge and its utility in the management of genital tract infection.
Participants and methods
This was a prospective study of 200 women with vaginal discharge referred for clinical examination and pelvic examination through a speculum examination according to the criteria of vaginal discharge. To determine the cause, high vaginal swab by Ayre’s spatula was obtained for the Microbiology Department through transport media. The collected data were tabulated and analyzed.
Bacterial vaginosis is the most common cause of vaginal discharge, followed by candidiasis and trichomonas in our setup. For the etiological diagnosis of symptomatic vaginal discharge, the microbiological diagnostic approach is the best. A specific diagnosis was made in 83.5% of cases, whereas an etiology was not found in 16.5% of cases.
Treatment on the basis of a clinical diagnosis is not accurate. Thus, the most ideal approach in this study is the microbiological diagnostic approach for the etiological diagnosis of symptomatic vaginal discharge.
Keywords: bacterial vaginosis, candidiasis, microbiological diagnosis, trichomonas, vaginal discharge
|How to cite this article:|
Farhan AM, Eldesouky EA, Gaballah EA, Soltan ME. Comparison of visual, clinical, and microbiological diagnosis of symptomatic vaginal discharge in the reproductive age group. Benha Med J 2017;34:43-8
|How to cite this URL:|
Farhan AM, Eldesouky EA, Gaballah EA, Soltan ME. Comparison of visual, clinical, and microbiological diagnosis of symptomatic vaginal discharge in the reproductive age group. Benha Med J [serial online] 2017 [cited 2018 Apr 25];34:43-8. Available from: http://www.bmfj.eg.net/text.asp?2017/34/1/43/206900
| Introduction|| |
Vaginal discharge in the reproductive age group is the most common complaint encountered everyday worldwide .
Physiological vaginal discharge is made up of a combination of mucoid secretions from the endocervical cells, sloughed epithelial cells, vaginal transudate, and products from the normal flora of the vagina, for example, lactobacilli . This discharge is characteristically white or clear and has minimal odor; it varies both in quantity and consistency among women, during pregnancy, and with the stage of the menstrual cycle. The amount of discharge, although variable, is usually from 1 to 4 ml over 24 h; during pregnancy, vaginal discharge is thicker and white-cream in color .
In women of reproductive age, discharge is usually caused by infection and causative organisms may or may not be sexually transmitted presents by vaginal discharge . Symptomatic vaginal discharge is caused by inflammation because of infection of the vaginal mucosa. This occurs in 1–14% of all women in the reproductive age group and is responsible for 5–10 million outpatient per day visits per year throughout the world; the most common cause of symptomatic vaginal discharge is bacterial vaginosis, followed by candidiasis and trichomoniasis .
| Aim|| |
The aim of this study is to compare the clinical diagnosis made on per speculum examination and microbiological diagnosis by a simple microscopic examination to determine the etiology of symptomatic vaginal discharge in the reproductive age group.
| Participants and methods|| |
The study was approved by the Ethics Committee of Al-Azhar University.
This is a prospective comparative study that included 200 patients with vaginal discharge. Etiology was identified by a visual and clinical examination, then wet swabs were obtained for aerobic, anaerobic culture, and microscopic examination. Patients were recruited from the outpatient clinic of gynecology at Sayed Galal Hospital (Al-Azhar University Hospitals).
Women between 15 and 45 years with vaginal discharge were included in the study.
Pregnant women, postmenopausal women, women in whom per speculum and pelvic examination was not possible, menstruating women, those who had received antimicrobials/antifungals (topical/oral) in the previous 1 month, posthysterectomy status, patients who had delivered or undergone an abortion in the past 6 weeks, and prepubertal females were excluded from the study.
All women were subjected to a complete assessment of history: complaints, present and previous history; obstetric history, pelvic operation; menstrual history; family planning history, diabetes mellitus.
Complaints and present history included vaginal discharge (quality/quantity/odor/color), itching, pururitis, dyspareunia, urine upsets as frequency and burning sensation, abdominal pain, new sexual partners, use of soaps or detergents, douching.
Speculum examination for criteria suggestive for a etiology as white, clumpy, curdy discharge, erythema and edema of vagina and vulva in candidiasis, off-white or yellow, frothy discharge, erythema of vulva and cervix (strawberry cervix) in trichomoniasis, and white or gray, thin, copious discharge in bacterial vaginosis diagnosis by history and clinical examination by applying of each criteria has be tabulated.
High vaginal swab and an endocervical swab by Ayre’s spatula were obtained for the Microbiological Department at the Faculty of Medicine Al-Azhar University through transport media for aerobic and anaerobic culture, in addition to a simple microscopic examination.
Bacterial vaginosis was diagnosed by the presence of three out of four Amsel criteria: thin, homogenous vaginal discharge, vaginal pH greater than 4.5, and a positive whiff test (fishy amine odor when 10% potassium hydroxide solution is added), presence of clue cells (vaginal epithelial cells with borders obscured by adherent coccobacilli on wet-mount preparation or Gram stain) .
Wet-mount preparation was influenced by the number of parasites in the vaginal fluid sample of motile trichomonads; motile trichomonads with flagella slightly larger than a leukocyte may be seen .
Vaginal pH is usually normal (4.0–4.5). Microscopic examination indicated visible budding yeast .
| Results|| |
[Table 1] shows that 3% of the patients were negative by clinical diagnosis, whereas 16.5% were negative by microbiological diagnosis; bacterial vaginosis was 44% by clinical and 34% by microbiological diagnosis. Candida was found in 29.5% of patients by clinical examination and in 22.5% of patients by microbiology; finally, trichomonas was found in 2% of patients by clinical examination and in 1% of patients by microbiological diagnosis.
[Table 2] shows that there was a statistically significance difference between the two methods [P<0.001 (highly significant), Fisher’s exact test=273.3].
|Table 2 Differences in the diagnosis of cases by clinical and microbiological tests|
Click here to view
[Table 3] shows that there were highly statistically significant agreements between microbiological diagnosis and clinical diagnosis of candida (κ=0.68, P<0.001), with a degree of agreement of 84%.
|Table 3 Degree of agreement between clinical and microbiological diagnosis of candida|
Click here to view
[Table 4] shows that there were highly statistically significant agreements between microbiological diagnosis and clinical diagnosis of bacterial vaginosis (κ=0.65, P<0.001), with a degree of agreement of 82%.
|Table 4 Degree of agreement between clinical and microbiological diagnosis of bacterial vaginosis|
Click here to view
[Table 5] shows that there were highly statistically significant agreements between microbiological diagnosis and clinical diagnosis of trichomonas (κ=0.24, P<0.001), with a degree of agreement of 94%.
|Table 5 Degree of agreement between clinical and microbiological diagnosis of trichomonas|
Click here to view
[Table 6] shows that clinical diagnosis had a sensitivity of 100%, a specificity of 76.1%, and an accuracy of 84% in the diagnosis of candida, a sensitivity of 100%, a specificity of 67.6%, and an accuracy of 82% in the diagnosis of bacterial vaginosis, and a sensitivity of 100%, a specificity of 93.3%, and an accuracy of 94% in the diagnosis of trichomonas.
|Table 6 Validity of clinical diagnosis of candida, bacterial vaginosis, and trichomonas in comparison with microbiology as the gold standard|
Click here to view
| Discussion|| |
This prospective study included 200 women in whom criteria were assessed to compare the diagnosis of symptomatic vaginal discharge by clinical and microbiological means. The mean age of the patients was 32±6 years (range: 20–45 years).
In our study, we observed an increase in the prevalence of bacterial vaginosis: 88 (44%) patients by clinical diagnosis and 68 (34%) by microbiological diagnosis, followed by candidiasis: 59 (29.5%) patients by clinical diagnosis and 45 (22.5%) by microbiological diagnosis. Mixed candidiasis and bacterial vaginosis was found in 33 (16.5%) patients by clinical diagnosis and 21 (10.5%) patients by microbiological diagnosis; isolated Trichomonas vaginalis was diagnosed clinically in four (2%) patients and in two (1%) patients by a microbiological fresh film for trichomonas.
Etiology was found in 83.5% of cases and not found in 33 (16.5%) in microbiological diagnosis while only six (3%) diagnosis cant reached in clinical diagnosis.
In the study by Rekha and Jyothi , the etiological diagnosis was made in 146 (72%) patients included; in the remaining 28% of the patients, a diagnosis could not be made using the microbiological diagnostic approach.
In the study by Vijayalakshmi et al. , 200 women presented with vaginal discharge; a diagnosis was made for 161 (80%) of these women and in 39 (20%) women, the etiological diagnosis could not be made.
In the study by Sowjanya and colleagues, bacterial vaginosis was diagnosed in 48%, candidiasis in 24%, and trichomoniasis in 3.3% of the cases by clinical examination. Microbiological diagnosis indicated pathological organisms in 62% of cases, whereas in 38% of the cases, the discharge was physiological. Among the pathological organisms isolated, the common isolates were bacterial vaginosis, found in 26.6%, candidiasis, found in 20%, and trichomoniasis, found in 5.33%. In all, 10% of the cases had mixed infections .
Donder et al.  suggested that the group of patients in whom the etiology is not clear and not found clinically, the diagnosis is not made microbiologically, and there is no growth on culture may probably have normal physiological discharge or less frequently viral vaginitis, aerobic vaginitis, or vaginal lactobacillosis, which are not routinely detected.
In our study: 88 (44%) of 200 patients were diagnosed with an isolated bacterial vaginosis by clinical examination; 63 (71.6%) of these patients were positive for bacterial vaginosis by microbiological diagnosis. Escherichia coli and Staphylococcus spp. were found in eight (9.1%) and four (4.5%) patients, respectively; in 13 (14.8%) of these patients, no etiology was identified.
Clinical diagnosis of bacterial vaginosis, compared with microbiological diagnosis, had a sensitivity of 100%, a specificity of 67.6%, and an accuracy of 82% in the diagnosis of bacterial vaginosis [P<0.001 (highly significant), positive predictive value (PPV)=100% and negative predictive value (NPV)=71.2%, area under the curve 0.88 (very good)].
In the study by Vijayalakshmi and colleagues, the prevalence of bacterial vaginosis was identified more by the clinical diagnostic approach than the microbiological approach. Out of 200 cases, clinically, bacterial vaginosis was found to be positive in 108 cases and microscopically in 106 cases; 96 cases were both clinically and microscopically positive for bacterial vaginosis .
In the study by Sowjanya and colleagues, on comparison of visual and clinical methods with the microbiological method, the sensitivity of visual diagnosis for bacterial vaginosis was 81.48% and specificity was 45.8%. The PPV of visual diagnosis was 45.83% and the NPV was 81.48%. The sensitivity of clinical diagnosis for bacterial vaginosis was 94.4% and specificity was 66.6%. The PPV of clinical diagnosis was 61.44% and the NPV was 91.04%. If WHO blanket therapy is used for treatment, bacterial vaginosis would be overtreated .
In our study, 59 (29.5%) patients were diagnosed clinically with an isolated candidiasis; 39 (66.1%) of these patients were positive for candidiasis in microbiological diagnosis. E. coli and Staphylococcus spp. were found in two (6.3%) and six (10.2%) patients, respectively. The etiology was not identified in 12 (20.3%) patients diagnosed clinically with candidiasis.
Clinical diagnosis was compared with microbiological diagnosis, and it was noted that clinical diagnosis had a sensitivity of 100%, a specificity of 76.1%, and an accuracy of 84% in the diagnosis of candida [P<0.001 (highly significant), PPV=67.3%, NPV=100%, and AUC 0.88 (very good)].
In the study by Vijayalakshmi and colleagues, the prevalence of candida was identified more by the microbiological approach than the clinical approach. Candidiasis was found to be positive in 45 cases and microscopically in 51 cases; 40 cases were both clinically and microscopically positive for candidiasis .
In the study by Sowjanya and colleagues, the sensitivity of visual diagnosis for candidiasis was 52.6% and specificity was 80.3%. The PPV of visual diagnosis was 47.61% and the NPV was 83.33%. The sensitivity of clinical diagnosis for candidiasis was 63.1% and specificity was 85.7%. The PPV of clinical diagnosis was 60% and the NPV was 87.27%. If visual and clinical methods are used, candidiasis would be overtreated .
In our study: four (2%) of 200 patients were diagnosed clinically with an isolated T. vaginalis; two (50%) of these patients were positive for trichomonas in microbiological diagnosis. In one (25%) of these patients, the etiology was not found and the other patient (25%) was diagnosed with E. coli.
Clinical diagnosis was compared with microbiological diagnosis, and it was noted that clinical diagnosis had a sensitivity of 100%, a specificity of 93.3%, and an accuracy of 94% in the diagnosis of trichomonas [P=0.022 (significant), PPV=14.3%, NPV=100%, and AUC 0.97 (excellent)].
In the study by Vijayalakshmi and colleagues, the prevalence of trichomoniasis was identified more by the microbiological approach than the clinical approach. Clinically, trichomoniasis was found to be positive in eight cases and microscopically in six cases; five cases were both clinically and microscopically positive for trichomoniasis .
In the study by Sowjanya and colleagues, the sensitivity of visual diagnosis for trichomoniasis was 35.2% and specificity was 98.4%. The PPV of visual diagnosis was 75% and the NPV was 92.25%. The sensitivity of clinical diagnosis for trichomoniasis was 63.1% and specificity was 85.7%. The PPV of clinical diagnosis was 60% and the NPV was 87.27%. Trichomoniasis would be undertreated if identified by visual and clinical methods .
In our study, we can conclude that bacterial vaginosis has high sensitivity and PPV, and low specificity, moderate NPV, if we give emprical treatment as identified by clinical diagnosis some cases will recieve treatment unnecessarily. Candidiasis has high sensitivity and low specificity, and a moderate PPV, with a high NPV, if we give emprical treatment as identified by clinical diagnosis some cases will recieve treatment unnecess. Trichomonas has very high specificity and sensitivity, but a low PPV and a high NPV; so emprical treatment of cases as identified by clincal diagnosis some cases would be missed.
In the study by Vijayalakshmi and colleagues, clinical diagnosis had higher sensitivity (90.6%) for diagnosing bacterial vaginosis and moderate sensitivity (83.3%) for trichomoniasis and (78.4%) for candidiasis; clinical diagnosis had a higher specificity (98.4%) for trichomoniasis and 96.6% for candidiasis, and moderate specificity for bacterial vaginosis. Trichomoniasis has a NPV of 99.5%.
This implies that if the clinical approach is used to diagnose discharge, bacterial vaginosis and trichomoniasis would be overtreated, whereas candidiasis would be undertreated .
In our microbiological examination, 32% of patients had no trichomonal vaginitis, bacterial vaginosis, or candidiasis by the microbiological diagnostic approach; the visual, clinical approach diagnosed 20% of those cases with candidiasis only. If we apply WHO blanket therapy, these patients will receive unnecessary treatment. Also, in the swabs 8.5% of patients, E. coli was found; 7% had Staphylococcus albus that was not diagnosed clinically.
Not only does overdiagnosis place a financial burden on the health system but it also carries the risk of possible social consequences in the community. Also, adverse effects of drugs may occur.
Oral metronidazole is associated with anorexia, nausea, vomiting, and skin rashes. It crosses the blood–brain barrier and may cause dizziness, convulsions, and peripheral neuropathy. Transient leukopenia and disulfiram-like action have been observed with the use of metronidazole. Overall, 10–30% of patients treated with the drug developed candidiasis 3 weeks later. Antifungal therapy is associated with renal and hepatic complications, hypersensitivity reactions, nausea and vomiting, flatulence, and rarely angioedema. Food and Drug Administration does not recommend blanket therapy and combination therapy for the treatment of vaginal infections .
Our study is similar to that of Vijayalakshmi and colleagues and Sowjanya and colleagues; they reported that microbiological diagnosis is the best method.
In contrast with WHO recommends that all women with abnormal vaginal discharge be treated empirically with metronidazole and when candida is identified, treatment for candidiasis should also be initiated .
| Conclusion|| |
Vaginal discharge is the most common complaint in gynecological outpatients; the most common pathogen is bacterial vaginosis, 44%. Clinical diagnosis has high sensitivity and low specificity; thus, a simple microscopic evaluation has also been determined to be accurate. Microbiological examination is recommended as the best means to avoid overtreatment and undertreatment of cases.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Thulkar J, Kriplani A, Agarwal N, Vishnubhatla S. Aetiology & risk factors of recurrent vaginitis & its association with various contraceptive methods. Indian J Med Res 2010; 131:83–87.
Mylonas I, Bergauer F. Diagnosis of vaginal discharge by wet mount microscopy: a simple and underrated method. Obstet Gynecol Surv 2011; 66:359–368.
Khamees SS. Characterization of vaginal discharge among women complaining of genital tract infection Int J Pharm Life Sci 2012; 3:0976–7126.
Sowjanya R, Prathyusha V, Sai Sree Sudha R. Comparative study of visual, clinical and microbiological diagnosis of white discharge. IOSR J Dent Med Sci 2015; 14:24–27.
Amsel R, Patricia A, Spiegel CA, Kirk CS, David E, King KH. Nonspecific vaginitis. Diagnostic criteria and microbial and epidemiologic associations. Am J Med 1983; 74:14–22.
Wiese W, Patel SR, Patel SC, Ohl CA, Estrada CA. A meta-analysis of the Papanicolaou smear and wet mount for the diagnosis of vaginal trichomoniasis. Am J Med 2000; 108:301–308.
Van Schalkwyk J, Yudin MH, Infectious Disease Committee, Allen V, Bouchard C, Boucher M, Boucoiran I et al.
Vulvovaginitis: screening for and management of trichomoniasis, vulvovaginal candidiasis, and bacterial vaginosis. J Obstet Gynaecol Can 2015; 37:266–274.
Rekha S, Jyothi S. Comparison of visual, clinical and microbiological diagnosis of symptomatic vaginal discharge in the reproductive age group. Int J Pharm Biomed Res 2010; 1:144–148.
Vijayalakshmi D, Patil SS, Sambarey PW. Clinical and microscopic correlation of vaginal discharge. Int J Contemp Med Res 2016; 3:1328–1331.
Donder GG, Vereecken A, Bosmans E, Dekeersmaecker A, Salembier G, Spitz B. Definition of a type of abnormal vaginal flora that is distinct from bacterial vaginosis: aerobic vaginitis. BJOG 2002; 109:34–43.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]