Comparison of baseline characteristics
Before treatment, no statistically significant differences were observed between the two groups in age, disease duration, body mass index (BMI), menstrual status, number of HPV-infected genotypes, or other baseline characteristics (P > 0.05; Table 1).
| Observation group (n=57) | Control group (n=68) | Statistic | P value |
| Age (years) | 41.37±8.21 | 42.07±8.75 | t=-0.461 | 0.645 |
| Disease duration (months) | 14.82±5.76 | 15.26±6.14 | t=-0.410 | 0.682 |
| BMI (kg/m²) | 22.67±2.58 | 22.94±2.64 | t=-0.573 | 0.567 |
| Regular menstruation | 42 (73.68) | 50 (73.53) | χ2=0.120 | 0.942 |
| Irregular menstruation | 10 (17.54) | 13 (19.12) | | |
| Menopause | 5 (8.77) | 5 (7.35) | | |
| Single genotype infection | 36 (63.16) | 42 (61.76) | χ2=0.026 | 0.873 |
| Multiple genotype infection | 21 (36.84) | 26 (38.24) | | |
| HPV16/18 positive | 20 (35.09) | 25 (36.76) | χ2=0.038 | 0.846 |
| Number of HPV-positive genotypes | 1.39±0.53 | 1.41±0.58 | t=-0.259 | 0.796 |
| TCT NILM | 42 (73.68) | 50 (73.53) | χ2=0.000 | 1 |
| TCT ASC-US | 10 (17.54) | 12 (17.65) | | |
| TCT LSIL | 5 (8.77) | 6 (8.82) | | |
| Baseline vaginal pH | 5.18±0.43 | 5.22±0.45 | t=-0.489 | 0.626 |
| Baseline CD4+/CD8+ | 1.17±0.25 | 1.15±0.27 | t=0.441 | 0.66 |
| Baseline total TCM syndrome score (points) | 18.46±4.09 | 18.22±4.36 | t=0.309 | 0.758 |
Table 1: Baseline characteristics of the two groups. Baseline demographic and clinical characteristics of the observation and control groups, including age, disease duration, body mass index, menstrual status, HPV infection characteristics, cervical cytology findings, vaginal pH, CD4+/CD8+ ratio, and total traditional Chinese medicine syndrome score.
Comparison of clinical efficacy and HPV outcomes
After treatment, the total effective rate was higher in the observation group than in the control group (89.47% vs. 72.06%; χ2 = 5.878, P = 0.015), and the association remained after adjustment for measured baseline covariates (adjusted RR = 1.23, 95% CI 1.08–1.40, P = 0.002). Complete conversion to negative occurred in 66.67% and 45.59% of patients, respectively; however, the overall four-category HPV outcome distribution did not reach statistical significance (χ2 = 7.406, P = 0.060). The time to first HPV conversion was shorter in the observation group (P = 0.028), whereas the HPV16/18-specific conversion rate did not differ significantly between groups (70.00% vs. 48.00%, P = 0.138; Table 2). In exploratory analyses stratified by baseline HPV16/18 status, complete HR-HPV conversion occurred in 1/20 (5.00%) versus 0/25 (0.00%) HPV16/18-positive patients (Fisher’s exact P = 0.444) and in 37/37 (100.00%) versus 31/43 (72.09%) HPV16/18-negative patients (Fisher’s exact P < 0.001) in the observation and control groups, respectively. Most HPV16/18-positive infections were multiple-type infections at baseline (17/20 and 24/25, respectively; Supplementary Table 1).
| Observation group (n=57) | Control group (n=68) | Statistic | P value |
| Total effective rate | 51 (89.47) | 49 (72.06) | χ2=5.878 | 0.015 |
| Complete conversion to negative | 38 (66.67) | 31 (45.59) | χ2=7.406 | 0.06 |
| Partial conversion to negative | 13 (22.81) | 18 (26.47) |
| Persistent positivity | 5 (8.77) | 15 (22.06) |
| Newly positive HPV genotype | 1 (1.75) | 4 (5.88) |
| HPV16/18 conversion to negative | 14/20 (70.00) | 12/25 (48.00) | χ2=2.204 | 0.138 |
| Time to first HPV conversion (months, effective cases) | 2.25±0.74 | 2.59±0.76 | t=-2.238 | 0.028 |
Table 2: Clinical efficacy and HPV outcomes after treatment. Post-treatment clinical efficacy and HPV outcomes in the observation and control groups, including complete conversion to negative, partial conversion to negative, persistent positivity, newly positive HPV genotype, total effective rate, HPV16/18-specific conversion, and time to first HPV conversion. The overall comparison across the four HPV outcome categories was χ2 = 7.406, P = 0.060.
Comparison of physicochemical and enzymatic indicators of vaginal microecology
Before treatment, no differences were observed in vaginal microecological examination results between the two groups (P > 0.05). After treatment, vaginal pH was lower in the observation group than in the control group (P < 0.001), and the reduction in vaginal pH was greater in the observation group (P = 0.004). In addition, the rate of normal hydrogen peroxide status and the proportion of vaginal cleanliness grade I–II were higher in the observation group than in the control group (P = 0.015 and P = 0.007, respectively), whereas leukocyte esterase positivity and amine-test positivity were lower (P = 0.022 and P = 0.026, respectively). The difference in sialidase positivity between the two groups was not statistically significant (P = 0.055; Table 3).
| Observation group (n=57) | Control group (n=68) | Statistic | P value |
| Vaginal pH (before treatment) | 5.18±0.43 | 5.22±0.45 | t=-0.489 | 0.626 |
| Vaginal pH (after treatment) | 4.54±0.31 | 4.78±0.34 | t=-4.150 | <0.001 |
| Reduction in vaginal pH | 0.63±0.36 | 0.43±0.41 | t=2.906 | 0.004 |
| Normal hydrogen peroxide (before treatment) | 19 (33.33) | 21 (30.88) | χ2=0.086 | 0.77 |
| Normal hydrogen peroxide (after treatment) | 43 (75.44) | 37 (54.41) | χ2=5.950 | 0.015 |
| Leukocyte esterase positive (before treatment) | 41 (71.93) | 50 (73.53) | χ2=0.040 | 0.841 |
| Leukocyte esterase positive (after treatment) | 17 (29.82) | 34 (50.00) | χ2=5.226 | 0.022 |
| Sialidase positive (before treatment) | 31 (54.39) | 39 (57.35) | χ2=0.111 | 0.739 |
| Sialidase positive (after treatment) | 12 (21.05) | 25 (36.76) | χ2=3.673 | 0.055 |
| Amine test positive (before treatment) | 29 (50.88) | 35 (51.47) | χ2=0.004 | 0.947 |
| Amine test positive (after treatment) | 10 (17.54) | 24 (35.29) | χ2=4.934 | 0.026 |
| Vaginal cleanliness grade I-II (before treatment) | 18 (31.58) | 20 (29.41) | χ2=0.069 | 0.793 |
| Vaginal cleanliness grade I-II (after treatment) | 45 (78.95) | 38 (55.88) | χ2=7.394 | 0.007 |
Table 3: Vaginal physicochemical and enzymatic indicators before and after treatment. Comparison of vaginal pH, hydrogen peroxide status, leukocyte esterase positivity, sialidase positivity, amine-test positivity, and vaginal cleanliness grade before and after treatment in the observation and control groups.
Comparison of vaginal flora and pathogenic microorganism detection
After treatment, Lactobacillus predominance was more frequent in the observation group than in the control group (P = 0.008), whereas fungal detection, bacterial vaginosis (BV)-related abnormalities, and mixed infection were less frequent (P < 0.05). Trichomonas detection did not differ between groups (P = 0.500; Table 4). Across all patients, normal post-treatment hydrogen peroxide status was more frequent when Lactobacillus predominance was present than when it was absent (62/81 [76.54%] vs. 18/44 [40.91%]; χ2 = 15.714, P < 0.001). The same direction was observed in the observation group (84.09% vs. 46.15%) and the control group (67.57% vs. 38.71%).
| Observation group (n=57) | Control group (n=68) | Statistic | P value |
| Lactobacillus predominance (before treatment) | 20 (35.09) | 22 (32.35) | χ2=0.104 | 0.747 |
| Lactobacillus predominance (after treatment) | 44 (77.19) | 37 (54.41) | χ2=7.055 | 0.008 |
| Fungal detection (before treatment) | 7 (12.28) | 9 (13.24) | χ2=0.025 | 0.874 |
| Fungal detection (after treatment) | 3 (5.26) | 12 (17.65) | χ2=4.503 | 0.034 |
| Trichomonas detection (before treatment) | 2 (3.51) | 3 (4.41) | Fisher | 1 |
| Trichomonas detection (after treatment) | 0 (0.00) | 2 (2.94) | Fisher | 0.5 |
| BV-related abnormalities (before treatment) | 30 (52.63) | 36 (52.94) | χ2=0.001 | 0.972 |
| BV-related abnormalities (after treatment) | 9 (15.79) | 23 (33.82) | χ2=5.295 | 0.021 |
| Mixed infection (before treatment) | 13 (22.81) | 16 (23.53) | χ2=0.009 | 0.924 |
| Mixed infection (after treatment) | 4 (7.02) | 14 (20.59) | χ2=4.633 | 0.031 |
Table 4: Vaginal flora and microorganism detection before and after treatment. Comparison of Lactobacillus predominance, fungal detection, Trichomonas detection, bacterial vaginosis-related abnormalities, and mixed infection before and after treatment in the observation and control groups.
Comparison of peripheral T-cell subset indicators
Before treatment, no between-group differences were observed in T-cell subset measures (P > 0.05). After treatment, the CD4+ T-cell percentage was higher in the observation group than in the control group (P = 0.006), the CD8+ T-cell percentage was lower (P = 0.024), and the CD4+/CD8+ ratio was higher (P < 0.001). The increase in CD4+ T cells, the decrease in CD8+ T cells, and the increase in the CD4+/CD8+ ratio were also greater in the observation group (P < 0.001, P = 0.004, and P < 0.001, respectively; Figure 1).

Figure 1: Peripheral T-cell subset measures before and after treatment. (A) CD4+ T-cell percentages before and after treatment and the corresponding increase. (B) CD8+ T-cell percentages before and after treatment and the corresponding decrease. (C) CD4+/CD8+ ratios before and after treatment and the corresponding increase. Changes in CD4+ and CD8+ T cells are expressed as percentage points, whereas changes in the CD4+/CD8+ ratio are unitless. #P < 0.05 versus the corresponding pretreatment value; &P < 0.05 versus the observation group. Please click here to view a larger version of this figure.
Comparison of local cervical signs and gynecological examination findings
After treatment, the incidences of abnormal cervical color and contact bleeding were lower in the observation group than in the control group (P = 0.015 and P = 0.038, respectively). No between-group differences were observed in cervical columnar epithelial ectropion, cervical hypertrophy, or cervical polyps/neoplasm-like changes (P > 0.05; Table 5).
| Observation group (n=57) | Control group (n=68) | Statistic | P value |
| Cervical columnar epithelial ectropion (before treatment) | 34 (59.65) | 42 (61.76) | χ2=0.058 | 0.809 |
| Cervical columnar epithelial ectropion (after treatment) | 15 (26.32) | 27 (39.71) | χ2=2.492 | 0.114 |
| Abnormal cervical color (before treatment) | 46 (80.70) | 55 (80.88) | χ2=0.001 | 0.98 |
| Abnormal cervical color (after treatment) | 14 (24.56) | 31 (45.59) | χ2=5.950 | 0.015 |
| Cervical hypertrophy (before treatment) | 22 (38.60) | 25 (36.76) | χ2=0.044 | 0.833 |
| Cervical hypertrophy (after treatment) | 9 (15.79) | 19 (27.94) | χ2=2.634 | 0.105 |
| Cervical polyps/neoplasm-like changes (before treatment) | 5 (8.77) | 7 (10.29) | χ2=0.083 | 0.774 |
| Cervical polyps/neoplasm-like changes (after treatment) | 3 (5.26) | 5 (7.35) | Fisher | 0.726 |
| Contact bleeding (before treatment) | 26 (45.61) | 32 (47.06) | χ2=0.026 | 0.872 |
| Contact bleeding (after treatment) | 6 (10.53) | 17 (25.00) | χ2=4.326 | 0.038 |
Table 5: Local cervical examination findings before and after treatment. Comparison of cervical columnar epithelial ectropion, abnormal cervical color, cervical hypertrophy, cervical polyps or neoplasm-like changes, and contact bleeding before and after treatment in the observation and control groups.
Comparison of improvement magnitude in major outcomes
The reductions in Nugent score, Donders score, and total TCM syndrome score were greater in the observation group than in the control group (all P < 0.001). The recovery rate of Lactobacillus predominance and the rate of microecological normalization were also higher in the observation group (81.08% vs. 58.70%, P = 0.029; 70.18% vs. 50.00%, P = 0.022; Table 6).
| Observation group (n=57) | Control group (n=68) | Statistic | P value |
| Reduction in Nugent score (points) | 3.21±1.68 | 2.16±1.55 | t=3.628 | <0.001 |
| Reduction in Donders score (points) | 3.05±1.44 | 2.03±1.37 | t=4.056 | <0.001 |
| Reduction in total TCM syndrome score (points) | 11.74±3.91 | 8.38±3.78 | t=4.865 | <0.001 |
| Recovery of Lactobacillus predominance | 30/37 (81.08) | 27/46 (58.70) | χ2=4.777 | 0.029 |
| Microecological normalization | 40 (70.18) | 34 (50.00) | χ2=5.226 | 0.022 |
Table 6: Additional change-score and microecological normalization outcomes. Comparison of reductions in Nugent score, Donders score, and total traditional Chinese medicine syndrome score, together with recovery of Lactobacillus predominance and microecological normalization, between the observation and control groups.
Safety and adverse reactions
During treatment, recorded adverse events occurred in 15/57 (26.32%) patients in the observation group and 29/68 (42.65%) patients in the control group; the between-group difference did not reach statistical significance (P = 0.057). Increased vaginal discharge, vulvovaginal candidiasis, and bacterial vaginosis were recorded less frequently in the observation group (P = 0.040, P = 0.034, and P = 0.049, respectively), whereas gastrointestinal discomfort was numerically more frequent (12.28% vs. 2.94%; Fisher’s exact P = 0.078). These individual comparisons should not be interpreted as evidence of an overall safety advantage (Table 7).
| Adverse reaction/safety indicator | Observation group (n=57) | Control group (n=68) | Statistic | P value |
| Vaginal itching | 6 (10.53) | 14 (20.59) | χ2=2.336 | 0.126 |
| Vaginal burning sensation | 4 (7.02) | 11 (16.18) | χ2=2.463 | 0.117 |
| Lower abdominal distension | 5 (8.77) | 12 (17.65) | χ2=2.079 | 0.149 |
| Increased vaginal discharge | 8 (14.04) | 20 (29.41) | χ2=4.218 | 0.04 |
| Abnormal vaginal bleeding | 1 (1.75) | 3 (4.41) | Fisher | 0.625 |
| Vulvovaginal candidiasis | 3 (5.26) | 12 (17.65) | χ2=4.503 | 0.034 |
| Bacterial vaginosis | 4 (7.02) | 13 (19.12) | χ2=3.864 | 0.049 |
| Gastrointestinal discomfort | 7 (12.28) | 2 (2.94) | Fisher | 0.078 |
| Treatment discontinuation | 1 (1.75) | 4 (5.88) | Fisher | 0.375 |
| Total adverse reactions | 15 (26.32) | 29 (42.65) | χ2=3.626 | 0.057 |
Table 7: Recorded adverse events during treatment. Comparison of recorded adverse events in the observation and control groups, including vaginal itching, vaginal burning sensation, lower abdominal distension, increased vaginal discharge, abnormal vaginal bleeding, vulvovaginal candidiasis, bacterial vaginosis, gastrointestinal discomfort, treatment discontinuation, and total adverse reactions.
Data Availability
The deidentified participant-level raw data supporting the conclusions of this study are provided in Supplementary File 1.
Supplementary Table 1: Scoring criteria for the traditional Chinese medicine syndrome score.
Scoring criteria for the eight items included in the total traditional Chinese medicine syndrome score. Each item is scored from 0–3 based on symptom severity, with higher scores indicating greater severity. Please click here to download this file.
Supplementary Table 2: Exploratory HPV outcomes stratified by baseline HPV16/18 status.
Exploratory comparison of HPV outcomes according to baseline HPV16/18 status. HPV16/18-negative patients had HR-HPV infection involving one or more of the other HR-HPV genotypes included in the assay. Individual non-16/18 genotype labels were not retained in the deidentified analytic dataset; therefore, type-specific clearance estimates for individual non-16/18 genotypes were not reconstructed. These analyses were exploratory. Please click here to download this file.
Supplementary File 1: The deidentified participant-level raw data for the protocol. Please click here to download this file.