Effect of Alprostadil Plus Ginkgo Biloba Extract on Hearing and Quality of Life in Sudden Sensorineural Hearing Loss: Retrospective Comparative Study

Summary

Using a single-center retrospective comparative design, this study showed that alprostadil combined with Ginkgo biloba extract improved PTA-4 recovery, hearing-loss distribution, SF-36 domains, and HHIA scores in sudden sensorineural hearing loss patients compared with alprostadil alone, without increasing adverse reactions, suggesting the short-term potential of early combination therapy.

Abstract

Sudden sensorineural hearing loss is associated with inner-ear microcirculatory disturbance and oxidative stress, which can damage auditory cells and cause permanent hearing impairment without timely treatment. Current management targets blood supply and inflammation, but single-agent efficacy remains unstable. This retrospective study explored whether alprostadil combined with Ginkgo biloba extract is associated with improved audiological outcomes and quality of life. Patients treated from January 2024 to December 2024 at a single center were included. A total of 100 patients who met the eligibility criteria were identified and divided into two groups based on their actual treatment: 50 who received alprostadil alone (control group) and 50 who received alprostadil plus Ginkgo biloba extract (combination therapy group). Primary outcomes were changes in affected-ear pure-tone average threshold [four-frequency pure-tone average (PTA-4)] and quality-of-life scales [36-Item Short Form Health Survey (SF-36); Hearing Handicap Inventory for Adults (HHIA)]. Assessments were performed before treatment and at follow-up as documented in the medical records. Statistical analyses included repeated-measures analysis of variance and independent-samples t-tests, with Bonferroni correction when required (α = 0.05/k), and adverse events (AE) were compared using the χ² test or Fisher’s exact test. The PTA-4 group × time interaction was significant (P = 0.002). On day 90, PTA-4 improvement was greater in the combination therapy group than in the control group (P = 0.002). After 90 days, the proportion of mild hearing loss differed between groups (P = 0.040), whereas the proportions for other grades did not. Significant group × time interactions were observed for all SF-36 dimensions (all P < 0.05) and HHIA scores (P = 0.007). On days 10, 30, and 90, HHIA improved more in the combination therapy group; SF-36 scores were also higher at follow-up. All included patients had completed treatment and 90-day follow-up per medical records and were included in the efficacy and safety analyses. Alprostadil plus Ginkgo biloba extract improved short-term audiological and quality-of-life outcomes without increasing adverse reactions, supporting the short-term potential of early combination therapy.

Introduction

Sudden sensorineural hearing loss is a hearing impairment that develops within a short period of time; common clinical symptoms include tinnitus, aural fullness, and hearing loss. Some patients can recover spontaneously, while a considerable proportion may develop irreversible hearing loss, thereby affecting social interaction and emotional stability^1,2. Current therapy mainly focuses on improving microcirculation in the inner ear and alleviating inflammatory damage, with commonly used drugs including glucocorticoids, vasodilators, and neurotrophic agents; however, efficacy is variable, and relapses are not uncommon^3,4. In recent years, several studies have reported the potential value of alprostadil in protecting vascular endothelium and regulating hemorheology⁵; there is also literature concerning the antioxidant and vasodilatory effects of Ginkgo biloba extract⁶, but evidence for alprostadil in sudden sensorineural hearing loss remains limited, and evidence for Ginkgo biloba extract is likewise heterogeneous, particularly regarding audiological recovery, quality-of-life benefit, and follow-up durability. In clinical practice, the effect of a single intervention in some patients is not ideal, especially when the onset time is prolonged, or vascular risk factors are superimposed; ischemia and oxidative stress in the inner ear often form a vicious cycle, hindering effective restoration of auditory function⁷.

A small number of publications have explored the positive effects of alprostadil or Ginkgo biloba extract used alone, but most evaluated monotherapy rather than add-on treatment, and many lacked rigorous controlled designs or adequate follow-up, making it difficult for healthcare personnel to obtain a complete intervention strategy or to determine whether adding Ginkgo biloba extract to alprostadil provides additional short-term clinical benefit without increasing adverse reactions^8,9. Therefore, based on a retrospective comparative analysis of real-world clinical data, the present study compares alprostadil monotherapy with alprostadil plus Ginkgo biloba extract in patients with sudden sensorineural hearing loss to investigate the comprehensive benefits in terms of the magnitude of hearing recovery and quality-of-life improvement, and to obtain more reliable information regarding safety and compliance. This study assessed the extent of inner-ear lesions through changes in pure-tone thresholds and hearing grading and used the SF-36 and HHIA scales to evaluate the multidimensional effects of the intervention on patients’ physical, psychological, and social functioning. This study aims to provide additional real-world evidence for the use of combination therapy in sudden sensorineural hearing loss, evaluate its effects on audiological outcomes and patient-reported quality of life, and offer further evidence to support comprehensive treatment strategies.

Protocol

The retrospective study was reviewed and approved by the Medical Ethics Committee of Tianjin First Central Hospital (approval No. (2025) No. 49). The medical records used for this retrospective analysis were derived from patients treated between January 2024 and December 2024. The present retrospective comparative analysis was conducted on de-identified routinely collected clinical data. All subjects had provided written informed consent for the use of their clinical data for research purposes at the time of their initial treatment, and the ethics committee approved the use of these data for this retrospective study.

Study design

This was a retrospective comparative study from a single center. Patients with sudden sensorineural hearing loss who were treated between January 2024 and December 2024 were identified. Based on their actual treatment regimen, they were divided into two groups: those who received alprostadil alone (control group) and those who received alprostadil combined with Ginkgo biloba extract (combination therapy group). The assignment was not randomized but reflected clinical practice decisions. Outcome assessors who performed audiometric and questionnaire evaluations were blinded to the study grouping at the time of data extraction and scoring. The follow-up endpoint was defined as day 90 after the first administration. The primary endpoint was the change in affected-ear four-frequency pure-tone average (PTA-4) at day 90. Secondary endpoints included hearing loss grading at day 90, Hearing Handicap Inventory for Adults (HHIA) scores at days 10, 30, and 90, and 36-Item Short Form Health Survey (SF-36) scores at days 30 and 90. Safety endpoints included adverse reactions recorded from the first administration through day 90. No exploratory outcomes were predefined for this retrospective analysis.

Inclusion and exclusion criteria for study subjects

The study subjects were patients with unilateral sudden sensorineural hearing loss. The diagnostic criteria for sudden sensorineural hearing loss were sudden-onset sensorineural hearing decline progressing within 72 h, with pure-tone audiometry showing a hearing-threshold increase of ≥30 decibels hearing level (dB HL) at any 3 consecutive frequencies compared with the previous hearing baseline or the contralateral healthy ear, and a time from onset to presentation of ≤14 days. All patients had undergone medical history collection, otoscopy, acoustic immittance testing, bone-conduction, and air-conduction pure-tone audiometry to confirm sensorineural hearing loss and to exclude conductive hearing impairment due to external or middle-ear disease.

The inclusion criteria were: age 18–70 years; unilateral onset; meeting the above diagnostic criteria for sudden sensorineural hearing loss; completing air-conduction threshold testing at 0.5, 1, 2, and 4 kHz in the affected ear; no treatment with alprostadil, Ginkgo biloba extract, batroxobin, vinpocetine, betahistine, hyperbaric oxygen, or intratympanic injection before presentation; completing follow-up on day 10, day 30, and day 90.

The exclusion criteria were: bilateral sudden sensorineural hearing loss; hearing impairment caused by previous external ear or middle ear diseases; hearing impairment caused by previous ear surgery; Meniere's disease; otosclerosis; noise-induced hearing loss; definite intracranial space-occupying lesion or auditory nerve lesion; severe cardiovascular disease; liver dysfunction; renal dysfunction; hematological disease; coagulation dysfunction; current anticoagulant or antiplatelet treatment; known allergy to alprostadil or Ginkgo biloba extract; pregnancy; lactation; inability to cooperate with audiological examinations; inability to complete study follow-up according to the protocol.

Baseline assessment process

Baseline data were extracted from medical records, including age, sex, time from onset to presentation, affected-ear side, history of hypertension, and baseline PTA-4. The time from onset to presentation was recorded in days, with the date the patient first developed definite hearing decline as the start point and the date baseline data were first extracted at the study center as the end point. A history of hypertension was determined according to previous definite diagnosis records, records of long-term antihypertensive drug use, or medical records before inclusion.

Sample size estimation

The sample size was determined by the number of eligible patients identified during the study period (January 2024 to December 2024). A total of 100 patients (50 per group) were included. Based on the observed difference in PTA-4 at day 90 (3.67 dB HL) and pooled standard deviation (5.88 dB HL), a post-hoc power calculation indicated that this sample size provided >80% power at a two-sided α = 0.05 for detecting a difference of this magnitude, which was consistent with the effect size assumed from previous studies.

Group assignment and blinding

The group assignment was based on the actual treatment received, as documented in the medical records. Patients were divided into the control group (alprostadil alone) or the combination therapy group (alprostadil plus Ginkgo biloba extract) according to the treatment plan determined by the attending physicians in routine practice. No randomization was involved. Audiological examiners, personnel assessing the SF-36 and HHIA scales, data entry and verification personnel, and statistical analysts were all blinded to group information during data extraction and analysis. Group codes were used in the data collection forms to record the treatment regimens, and unblinding was performed after completion of the statistical analysis.

Treatment regimen and administration procedures

All subjects had started treatment within 24 h after diagnosis and received continuous treatment for 10 days, as documented in the treatment records. All administrations were completed in the designated treatment area of the Department of Otolaryngology by nurses trained in accordance with standardized protocols. Before each infusion, the subject's name, patient identification number, drug name, dose, solvent, administration date, and administration time were checked. All drugs were prepared aseptically immediately before use, and the infusion was completed within 2 h after preparation. During the infusion period, the start and end times, infusion reaction, and completion status of the treatment for that day were recorded.

The control group received an alprostadil injection. For each administration, 10 µg of alprostadil was added to 100 mL of 5% glucose injection and infused via a peripheral vein, with the infusion time set to 30–40 min once daily for 10 consecutive days.

The combination therapy group received Ginkgo biloba extract injection in addition to the same alprostadil treatment as the control group. The dose, solvent, infusion time, and treatment course of alprostadil were identical to those in the control group. Each ampoule of Ginkgo biloba extract injection was 5 mL and contained 17.5 mg of Ginkgo biloba extract; 2 ampoules were used each time, with a total dose of 35 mg, added to 100 mL of 5% glucose injection, and infused through a peripheral vein, with the infusion time controlled at 30–40 min, once daily for 10 consecutive days. The two drugs were prepared separately and infused separately. After completion of the alprostadil infusion, the line was flushed with 5% glucose injection, and then the Ginkgo biloba extract injection was infused.

Both groups received identical basic otologic management, including noise avoidance, discontinuation of headphone use, sleep management, tinnitus education, monitoring for medication reactions, and scheduled follow-up appointments. During the study treatment and follow-up period, systemic glucocorticoids, intratympanic glucocorticoids, oral Ginkgo biloba extract, batroxobin, vinpocetine, betahistine, hyperbaric oxygen, and acupuncture were not initiated. When subjects required the above treatments due to changes in their condition, they were managed according to clinical needs, and this was recorded as a protocol deviation, while safety data were retained. This sequential administration approach was consistent with the labeled or routine intravenous use of both agents and was intended to avoid direct admixture. Because direct compatibility evidence for this specific sequential regimen is limited, the two drugs were prepared and infused separately, the line was flushed between infusions, and predefined monitoring and stopping criteria were used to minimize potential infusion-related risk.

Peripheral venous access, infusion administration, and clinically indicated emergency management were performed in accordance with institutional standard operating procedures; used needles/sharps and all drug-related medical waste generated during administration were disposed of in designated sharps or medical-waste containers according to hospital policy and local regulations.

Audiological assessment procedures

Pure-tone audiometry was performed by designated audiological examiners in a soundproof booth. Before the database lock, the examiners were unaware of the subjects' group information. The pure-tone audiometer underwent professional calibration before study initiation, and device self-testing and biological calibration were performed before each testing day. Subjects avoided exposure to intense noise before testing and sat quietly for 5 min before the test. During testing, ear canal patency was first confirmed, and then headphones were worn for air-conduction threshold measurement; when necessary, the nature of hearing loss was confirmed alongside bone-conduction thresholds and acoustic immittance results. Hearing-threshold measurement used a 5 dB step method, and the lowest sound intensity at which the subject gave repeated responses at the same intensity level that met the criterion was taken as the threshold at that frequency.

PTA-4 was the primary audiological indicator, calculated as the arithmetic mean of the air-conduction thresholds at 0.5, 1, 2, and 4 kHz in the affected ear, expressed in dB HL. The formula was:

PTA-4 = (0.5 kHz threshold + 1 kHz threshold + 2 kHz threshold + 4 kHz threshold) / 4.

The time points for PTA-4 assessment were baseline, day 10, day 30, and day 90. The day 10 assessment was performed within 24 h of completion of the 10th treatment; the allowed window for day 30 was ± 3 days, and for day 90, ± 7 days. The same assessment procedure, calculation method, and recording format were used at all time points.

Hearing loss grading was determined according to PTA-4¹⁰, with assessment time points at baseline and day 90. A PTA-4 of 26–40 dB HL was defined as mild hearing loss, 41–60 dB HL as moderate hearing loss, 61–80 dB HL as severe hearing loss, and >80 dB HL as profound hearing loss. Hearing loss grading was independently assessed by two researchers who were unaware of group information; when results were inconsistent, the final grade was determined after recalculating PTA-4 using the original pure-tone audiometry records.

Quality-of-life scale assessment procedures

Quality of life was evaluated using the SF-36 and HHIA scales. The questionnaires used in this study followed the standard instruments and scoring methods described in the corresponding references. Scale assessments were completed by researchers who had received standardized training and were unaware of the group information. The assessments were conducted in a quiet room, and subjects completed the questionnaires independently. When reading difficulty was present, the assessors read the items and fixed response options aloud one by one, without explaining the items or providing answers. Missing items were checked immediately upon questionnaire collection, and if any were found, the subjects completed them on the spot. Therefore, no missing-item imputation was required for the scale analyses. Before entry of the scale results, two researchers checked the subject number, assessment date, and raw scores.

The SF-36 scale¹¹ was used to evaluate general health-related quality of life and included 8 dimensions: physical functioning, role physical, bodily pain, general health, vitality, social functioning, role emotional, and mental health. Each dimension was converted to a score of 0–100 according to the standard scoring rules described for the instrument, and a higher score indicated a better health status in that dimension. The assessment time points for SF-36 were baseline, day 30, and day 90.

The HHIA scale¹² was used to evaluate the effects of hearing impairment on social functioning and emotional status and contained 25 items. Each item was scored as 4 points for "Yes", 2 points for "Sometimes", and 0 points for "No", with a total score range of 0–100; a higher score indicated more severe social and emotional effects caused by hearing impairment. The assessment time points for HHIA were baseline, day 10, day 30, and day 90.

Safety monitoring and determination of adverse reactions

Safety monitoring covered the period from the first administration to the end of follow-up on day 90. During treatment, blood pressure and pulse had been measured daily before infusion, as per routine clinical practice. Infusion reactions were monitored during infusion, and observation continued for 30 min after the end of infusion, as documented in the nursing records. During follow-up on days 10, 30, and 90, new symptoms occurring after treatment, aggravation of original symptoms, and medical management were recorded in the medical records.

Adverse reactions were recorded by category: headache, dizziness, nausea, rash, blood pressure fluctuations, and serious adverse events (SAEs). Headache, dizziness, nausea, and rash were documented as events that newly occurred after the first administration or worsened relative to baseline and were recorded by the investigators. Blood pressure fluctuation was defined as a change of ≥20 mmHg in systolic blood pressure or ≥10 mmHg in diastolic blood pressure relative to baseline, or symptomatic blood pressure abnormalities requiring clinical management. Serious adverse events were defined as events resulting in death, life-threatening conditions, requiring hospitalization, prolongation of hospital stay, or causing persistent functional impairment. The incidence of each type of adverse reaction was calculated as the number of subjects who experienced that event at least once; repeated occurrences of the same type of event in the same subject were counted as 1 case only.

When mild adverse reactions occurred, observation was continued and recording was completed; when persistent blood pressure abnormalities, generalized rash, marked gastrointestinal reactions, or conditions judged by the investigators to preclude continuation of infusion occurred, the infusion on that day was suspended and clinical management was provided; when serious adverse events occurred, the study medication was stopped immediately, medical management and ethics reporting were completed, and the event was recorded as a safety endpoint event.

Follow-up and adherence control

The study follow-up time points were days 10, 30, and 90. The day 10 follow-up was completed simultaneously with the end-of-treatment assessment, and on days 30 and 90, follow-up data were extracted from the medical records for patients who returned for re-examination on the scheduled dates. Follow-up content included pure-tone audiometry, scale assessment, inquiry about adverse reactions, and verification of concomitant treatment. At each follow-up, it was verified whether the subjects had received treatment outside the study protocol during the observation period, and the drug name, treatment date, and reason for treatment were recorded.

Adherence was determined by treatment and follow-up completion. Patients who had completed 10 prescribed intravenous treatments and had completed the audiological and scale assessments on day 10, day 30, and day 90, per medical records, were classified as having good adherence. Missed use of study medication, failure to complete the primary outcome assessment within the allowed window, initiation of prohibited treatment during the observation period, or loss to follow-up were recorded as protocol deviations. All protocol deviations were jointly verified by the principal investigator and the statistical personnel before database lock.

Data management and quality control

All study data were extracted and recorded in uniform data collection forms. The raw data included screening records, consent records for clinical data use, group assignment records, administration records, original pure-tone audiometry records, original SF-36 questionnaires, original HHIA questionnaires, adverse reaction records, and follow-up records. Data entry was performed independently by two persons, with item-by-item verification. When inconsistencies were found between entered values and the raw records, the raw records prevailed, and traces of modification were retained. Logical checks were completed before the database lock. The checking content included continuity of patient identification numbers, number of treatment days, follow-up time windows, calculated PTA-4 values, hearing loss grading, total scale scores, and adverse reaction classification.

Statistical analysis

Statistical analysis was performed using statistical analysis software. Since this was a retrospective study, all analyses were conducted on the available complete data, and no missing-value imputation was needed because all patients had complete records. Continuous variables were first tested for normality and homogeneity of variance. Continuous variables conforming to a normal distribution and homogeneity of variance were expressed as mean ± standard deviation, and between-group comparisons were performed using the independent-samples t-test; continuous variables not conforming to a normal distribution were expressed as median and interquartile range, and between-group comparisons were performed using the Mann-Whitney U test. Categorical variables were expressed as counts and percentages, and between-group comparisons were performed using the χ² test; when the expected frequencies did not meet the conditions for the χ² test, Fisher's exact test was used.

The longitudinal changes in PTA-4 scores, SF-36 dimension scores, and HHIA scores were analyzed using repeated-measures analysis of variance (ANOVA). The model included group, time, and group × time interaction terms. When the sphericity assumption was violated, the Greenhouse-Geisser correction was applied. After the group × time interaction term was statistically significant, further between-group comparisons were performed at each time point, and the Bonferroni method was used to correct the significance level. PTA-4 and HHIA both included 4 time points, and the corrected significance level was 0.0125; SF-36 included 3 time points, and the corrected significance level was 0.0167. Effect size was expressed as Cohen's d, and the calculation formula was the difference in means between the two groups divided by the pooled standard deviation.

As a secondary audiological outcome, hearing loss grading was compared between the two groups at baseline and on day 90, with respect to the proportions of patients with mild, moderate, severe, and profound hearing loss. This analysis was used to describe changes in the composition of hearing-loss severity after treatment, and no multiple corrections were applied. For key categorical comparisons, the effect size was expressed as a risk difference with 95% confidence interval (CI). The incidence of adverse reactions was compared using the χ² test or Fisher's exact test. All statistical tests were two-sided tests, and for analyses without Bonferroni correction, P < 0.05 was considered statistically significant.

Results

General data and baseline characteristics

A total of 149 patients were assessed for eligibility from the medical records, of whom 49 were excluded. Ultimately, 100 patients were included in the analysis, with 50 in each group. All patients were included in the final efficacy and safety analyses (Figure 1). There were no statistically significant differences between the control group and the combination therapy group in age, sex, time from onset to visit, baseline PTA-4, side of the affected ear, or history of hypertension (all P > 0.05; Table 1), indicating that the baseline characteristics of the two groups were balanced and comparable.

Audiological evaluation results

PTA-4 improved in both groups over time. On day 90 after treatment, the degree of threshold improvement in the combination therapy group was significantly better than that in the control group (P = 0.002); the mean PTA-4 decreased by 17.09 dB HL in the control group and 21.37 dB HL in the combination therapy group, with day 90 values of 52.36 dB HL ± 6.07 dB HL and 48.79 dB HL ± 5.66 dB HL, respectively (Table 2). After 90 days of treatment, the proportion of patients with mild hearing loss was higher in the combination therapy group than in the control group (30.0% vs. 16.0%; risk difference, 14.00%; 95% CI, −2.59%–29.74%), whereas the proportions of moderate, severe, and profound hearing loss did not differ significantly between groups. These hearing-grade comparisons were supportive, categorical analyses, and should be interpreted with caution (Table 3).

Quality-of-life assessment results

The combination therapy group showed better SF-36 outcomes during follow-up. On day 30, the combination therapy group had significantly higher PF, BP, VT, SF, RE, RP, GH, and MH scores than the control group (all P < 0.01, Cohen’s d range: 0.532–0.749); on day 90, each dimension score was likewise significantly higher than that of the control group (all P < 0.01, Cohen’s d range: 0.694–0.894) (Table 4). HHIA scores in the combination therapy group were 4.28, 5.18, and 6.82 points lower than those in the control group on day 10, day 30, and day 90, respectively (all P < 0.01, Cohen’s d range: 0.624–0.925) (Table 5).

Adverse reactions and safety

The differences in the incidence of headache, dizziness, nausea, rash, blood pressure fluctuation, and other serious adverse events between the two groups were small, and the corresponding risk differences had 95% CIs that crossed 0; these safety comparisons were descriptive/supportive analyses and should be interpreted cautiously (Table 6). One other serious adverse event was recorded in the control group, whereas none was recorded in the combination therapy group.

DATA AVAILABILITY:

All de-identified patient-level data, audiometry datasets, SF-36/HHIA scoring data, adverse-event records, statistical analysis files, and source data for all figures and tables are provided in the Supplementary File 1.

Figure 1: Flowchart of patient inclusion. Control group, alprostadil alone; combination therapy group, alprostadil plus Ginkgo biloba extract. PTA-4, four-frequency pure-tone average; HHIA, Hearing Handicap Inventory for Adults; SF-36, 36-Item Short Form Health Survey. All included patients (n = 100; 50 per group) had received the assigned intervention and completed follow-up per medical records, and were included in the efficacy and safety analyses. Please click here to view a larger version of this figure.

Indicator	Control group (n = 50)	Combined treatment group (n = 50)	Statistic value	p value
Age (years)	52.38 ± 5.84	53.16 ± 5.02	t = 0.695	0.488
Sex (male/female, n, %)	27 (54 %) / 23 (46 %)	25 (50 %) / 25 (50 %)	χ² = 0.160	0.689
Time from onset to visit (days)	4.28 ± 0.53	4.42 ± 0.51	t = 1.027	0.307
Baseline PTA-4 (dB HL)	69.45 ± 7.64	70.16 ± 7.32	t = 0.466	0.642
Side of affected ear (left/right, n, %)	26 (52 %) / 24 (48 %)	23 (46 %) / 27 (54 %)	χ² = 0.360	0.548
History of hypertension (n, %)	11 (22 %)	9 (18 %)	χ² = 0.491	0.484

Table 1: Comparison of baseline demographic and clinical characteristics between the two groups. Control group, alprostadil alone; combination therapy group, alprostadil plus Ginkgo biloba extract. The baseline analysis included all patients (n = 100; 50 per group). Continuous variables are presented as mean ± SD and categorical variables as n (%). Age, time from onset to presentation, and baseline PTA-4 were compared using independent-samples t tests; sex, side of affected ear, and history of hypertension were compared using χ² tests. PTA-4 was defined as the arithmetic mean of air-conduction thresholds at 0.5, 1, 2, and 4 kHz. The reported P values were unadjusted. A two-sided P < 0.05 was considered statistically significant. Please click here to download this Table.

Time point	Control group	Combined treatment group	Mean difference (95% CI)	t value	p value
Baseline (before treatment)	69.45 ± 7.64	70.16 ± 7.32	0.71 (−2.26 to 3.68)	0.466	0.642
Day 10	63.21 ± 7.02	60.85 ± 6.28	−2.31 (−4.97–0.35)	1.662	0.101
Day 30	57.44 ± 6.51	55.12 ± 5.94	−2.37 (−4.59 to −0.15)	2.251	0.027
Day 90	52.36 ± 6.07	48.79 ± 5.66	−3.67 (−5.98 to −1.36)	3.192	0.002
	Group × time interaction: F = 4.381, p = 0.014

Table 2: Comparison of PTA-4 (average pure-tone threshold) between the two groups at different time points (mean ± s). Data are presented as mean ± SD. Control group, alprostadil alone; combination therapy group, alprostadil plus Ginkgo biloba extract. All patients completed the PTA-4 assessments and were included in the efficacy analysis (n = 100; 50 per group). Mean difference was calculated as the combination therapy group minus the control group. PTA-4 was defined as the arithmetic mean of air-conduction thresholds at 0.5, 1, 2, and 4 kHz, expressed in dB HL; lower values indicate better hearing recovery. Between-group comparisons at each time point were performed using independent-samples t tests, and the group × time interaction was tested by repeated-measures ANOVA. The P values shown in the table are unadjusted for between-group comparisons at individual time points; because four time-point comparisons were performed, the Bonferroni-corrected significance threshold was α = 0.0125. Please click here to download this Table.

Control group n (%)	Combined treatment group n (%)	RD（Combination group –Control group，%，95%CI）	χ² value	p value
3 (6.0 %)	2 (4.0 %)	−2.00% (−12.62%–8.25%)	0.163	0.686
8 (16.0 %)	15 (30.0 %)	14.00% (−2.59%–29.74%)	4.217	0.04
12 (24.0 %)	9 (18.0 %)	−6.00% (−21.74%–10.06%)	0.577	0.447
21 (42.0 %)	23 (46.0 %)	4.00% (−14.96%–22.56%)	0.205	0.651
21 (42.0 %)	26 (52.0 %)	10.00% (-9.27%–28.27%)	1.154	0.283
15 (30.0 %)	9 (18.0 %)	−12.00% (−28.03%– 4.81%)	2.308	0.129
14 (28.0 %)	13 (26.0 %)	−2.00% (−19.01%– 15.16%)	0.051	0.824
6 (12.0 %)	3 (6.0 %)	−6.00% (−18.44%– 6.05%)	1.071	0.301

Table 3: Distribution of hearing-grade categories in the two groups. Data are presented as n (%). Control group, alprostadil alone; combination therapy group, alprostadil plus Ginkgo biloba extract. All patients had evaluable baseline and day 90 hearing-grade data and were included in the efficacy analysis (n = 100; 50 per group). Risk difference was calculated as the combination therapy group minus the control group. Percentages were calculated using the total number of patients in each group (n = 50) as the denominator. Hearing-loss severity was classified according to World Health Organization criteria based on PTA-4: mild, 26–40 dB HL; moderate, 41–60 dB HL; severe, 61–80 dB HL; and profound, >80 dB HL. Between-group comparisons for each hearing-loss category at baseline and day 90 were performed using χ² tests, with Fisher’s exact test used when expected cell counts were small. These were category-specific between-group comparisons rather than an overall distribution test, and the reported P values were unadjusted. A two-sided P < 0.05 was considered statistically significant. Please click here to download this Table.

Dimension	Time	Control group	Combined treatment group	Mean difference (95% CI)	t value	p value	Cohen’s d
PF	Baseline	55.27 ± 4.95	54.62 ± 5.14	−0.65 (−2.69 to 1.39)	0.573	0.568	0.115
	Day 30	62.13 ± 5.11	66.59 ± 5.43	4.46 (2.10 to 6.82)	3.78	<0.001	0.749
	Day 90	68.45 ± 5.77	74.51 ± 6.09	6.06 (3.36 to 8.76)	4.551	<0.001	0.894
		Group × time interaction: F = 6.128, p = 0.015
RP	Baseline	50.96 ± 6.28	51.44 ± 6.01	0.48 (−1.94 to 2.90)	0.376	0.708	0.075
	Day 30	58.27 ± 5.94	60.69 ± 6.23	2.42 (−0.04 to 4.88)	1.952	0.054	0.39
	Day 90	63.41 ± 6.32	68.15 ± 5.98	4.74 (2.01 to 7.47)	3.502	0.001	0.694
		Group × time interaction: F = 4.927, p = 0.028
BP	Baseline	57.83 ± 4.89	58.64 ± 5.27	0.81 (−1.19 to 2.81)	0.747	0.457	0.148
	Day 30	63.25 ± 5.24	66.16 ± 5.48	2.91 (0.73 to 5.09)	2.668	0.009	0.532
	Day 90	68.42 ± 5.93	72.57 ± 5.86	4.15 (1.80 to 6.50)	3.512	0.001	0.696
		Group × time interaction: F = 5.306, p = 0.023
GH	Baseline	52.41 ± 5.01	53.26 ± 5.57	0.85 (-1.24 to 2.94)	0.72	0.473	0.143
	Day 30	59.07 ± 5.18	61.42 ± 5.44	2.35 (0.13 to 4.57)	2.066	0.042	0.421
	Day 90	64.69 ± 5.35	69.45 ± 5.87	4.76 (2.40 to 7.12)	4.068	<0.001	0.803
		Group × time interaction: F = 5.491, p = 0.021
VT	Baseline	54.39 ± 4.96	53.72 ± 4.64	−0.67 (−2.58 to 1.24)	0.67	0.505	0.134
	Day 30	60.11 ± 5.45	64.38 ± 5.22	4.27 (1.85 to 6.69)	3.541	0.001	0.702
	Day 90	66.24 ± 5.67	71.56 ± 6.18	5.32 (2.81 to 7.83)	4.246	<0.001	0.836
		Group × time interaction: F = 6.002, p = 0.017
SF	Baseline	58.61 ± 5.22	59.38 ± 4.88	0.77 (−1.26 to 2.80)	0.699	0.486	0.139
	Day 30	63.72 ± 5.35	67.21 ± 5.43	3.49 (1.14 to 5.84)	2.95	0.004	0.588
	Day 90	69.08 ± 5.74	73.48 ± 6.03	4.40 (2.04 to 6.76)	3.728	<0.001	0.739
		Group × time interaction: F = 5.775, p = 0.019
RE	Baseline	55.02 ± 5.41	55.67 ± 5.84	0.65 (−1.55 to 2.85)	0.573	0.568	0.115
	Day 30	61.35 ± 5.52	64.83 ± 5.95	3.48 (1.06 to 5.90)	2.841	0.006	0.566
	Day 90	66.78 ± 5.64	71.84 ± 5.32	5.06 (2.75 to 7.37)	4.445	<0.001	0.873
		Group × time interaction: F = 5.986, p = 0.017
MH	Baseline	56.18 ± 5.06	55.72 ± 5.23	−0.46 (−2.50 to 1.58)	0.42	0.676	0.084
	Day 30	62.09 ± 5.51	65.36 ± 5.88	3.27 (0.91 to 5.63)	2.743	0.007	0.547
	Day 90	67.42 ± 5.81	72.39 ± 6.02	4.97 (2.53 to 7.41)	4.087	<0.001	0.805
		Group × time interaction: F = 6.115, p = 0.015

Table 4: Comparison of SF-36 dimension scores between the two groups (mean ± s). Data are presented as mean ± SD. Control group, alprostadil alone; combination therapy group, alprostadil plus Ginkgo biloba extract. All patients completed the SF-36 assessments and were included in the efficacy analysis (n = 100; 50 per group). Scores for each SF-36 domain range from 0 to 100, with higher scores indicating better health-related quality of life. PF, physical functioning; RP, role-physical; BP, bodily pain; GH, general health; VT, vitality; SF, social functioning; RE, role-emotional; MH, mental health. Between-group comparisons at each time point were performed using independent-samples t tests, and the group × time interaction for each domain was analyzed using repeated-measures ANOVA. The P values shown in the table are unadjusted for the between-group comparisons at individual time points; because three time-point comparisons were performed for each domain, the Bonferroni-corrected significance threshold was α = 0.0167. Cohen’s d denotes the standardized effect size for between-group differences. Please click here to download this Table.

Time point	Control group	Combined treatment group	Mean difference (95% CI)	t value	p value	Cohen’s d
Baseline	72.63 ± 6.85	71.84 ± 6.42	−0.72 (−3.41 to 1.97)	0.582	0.562	0.116
Day 10	64.58 ± 6.01	60.23 ± 6.14	−4.28 (−6.90 to−1.66)	3.156	0.002	0.624
Day 30	56.42 ± 5.58	51.37 ± 5.44	−5.18 (−7.70 to−2.66)	4.114	<0.001	0.767
Day 90	48.06 ± 5.12	41.12 ± 4.68	−6.82 (−9.77 to −3.87)	5.327	<0.001	0.925
	Group × time interaction: F = 7.532, p = 0.007

Table 5: Comparison of HHIA scores between the two groups (mean ± s). Data are presented as mean ± SD. Control group, alprostadil alone; combination therapy group, alprostadil plus Ginkgo biloba extract. All patients completed the HHIA assessments and were included in the efficacy analysis (n = 100; 50 per group). HHIA refers to the Hearing Handicap Inventory for Adults, with a total score ranging from 0 to 100; higher scores indicate a greater negative impact of hearing impairment on social and emotional functioning. Between-group comparisons at each time point were performed using independent-samples t tests, and the group × time interaction was analyzed using repeated-measures ANOVA. The P values shown in the table are unadjusted for between-group comparisons at individual time points; because four time-point comparisons were performed, the Bonferroni-corrected significance threshold was α = 0.0125. Cohen’s d denotes the standardized effect size for between-group differences. Please click here to download this Table.

Adverse reaction type	Control group (n = 50)	Combined treatment group (n = 50)	RD（Combination group–Control group，%，95%CI）	Statistic value	p value
Headache	6 (12.0 %)	4 (8.0 %)	−4.00% (−16.76%–8.58%)	χ²=0.417	0.519
Dizziness	4 (8.0 %)	5 (10.0 %)	2.00% (−10.22%–14.35%)	Fisher	0.736
Nausea	5 (10.0 %)	3 (6.0 %)	−4.00% (−16.02%–7.68%)	Fisher	0.459
Rash	2 (4.0 %)	1 (2.0 %)	−2.00% (−11.60%–6.98%)	Fisher	0.556
Blood pressure fluctuation	3 (6.0 %)	1 (2.0 %)	−4.00% (−14.35%–5.36%)	Fisher	0.302
Other serious adverse events	1 (2.0 %)	0 (0.0 %)	−2.00% (−10.50%–5.32%)	Fisher	1

Table 6: Incidence of adverse reactions in the two groups. Data are presented as n (%), and percentages were calculated using the total number of patients in each group (n = 50) as the denominator. The safety analysis included all patients who received at least one dose of study medication; in this study, this comprised all patients (n = 100; 50 per group). The control group received alprostadil alone, and the combination therapy group received alprostadil plus Ginkgo biloba extract. Risk difference was calculated as the combination therapy group minus the control group. Adverse events were monitored from the first administration through day 90. Between-group comparisons for headache, dizziness, nausea, rash, and blood pressure fluctuation were performed using χ² tests; other serious adverse events were analyzed using Fisher’s exact test because of sparse counts. The reported P values were unadjusted. A two-sided P < 0.05 was considered statistically significant. Please click here to download this Table.

Supplementary File 1: Raw data used in the study. All de-identified patient-level data, audiometry datasets, SF-36/HHIA scoring data, adverse-event records, statistical analysis files, and source data for all figures and tables.Please click here to download this file.

Discussion

This study evaluated the auditory outcomes and quality of life of patients with sudden sensorineural hearing loss, and the results showed that patients who received a combination of alprostadil and Ginkgo biloba extract showed more significant improvements in pure-tone thresholds, especially at 90 days after treatment, with a more pronounced decrease in average hearing threshold and a markedly higher proportion of mild hearing loss than with alprostadil alone. By day 90, PTA-4 decreased from 70.16 ± 7.32 to 48.79 ± 5.66 dB HL in the combination group and from 69.45 ± 7.64 to 52.36 ± 6.07 dB HL in the control group, corresponding to mean improvements of 21.37 and 17.09 dB HL, respectively. Together with the higher proportion of patients with mild hearing loss at day 90 in the combination group (30.0% vs. 16.0%), this finding suggests a shift toward less severe hearing impairment and may be associated with less hearing-related difficulty in daily life, indicating that the observed benefit was not only statistically significant but also clinically interpretable in terms of reduced hearing-loss severity, suggesting that the combined regimen can better protect or repair inner-ear function, reduce hearing damage, and promote functional recovery¹³. However, as a secondary audiological outcome, this categorical finding should be interpreted with caution because only the mild hearing-loss category differed significantly at day 90, whereas the other categories did not, and no multiple-comparison correction was applied to the hearing-grade analysis. In terms of quality of life, the combination therapy group exhibited better performance in physical functioning, mental health, and subjective hearing handicap during the follow-up period; more effective hearing recovery enabled patients to achieve greater improvements in the social, psychological, and emotional levels¹⁴. The reduction in HHIA indicates a lower self-perceived hearing handicap, whereas the improvement across multiple SF-36 domains suggests broader gains in physical, social, and emotional functioning. The concurrent improvement in HHIA and multiple SF-36 domains further supports the clinical relevance of these changes with respect to patient-perceived burden and daily functioning. One other serious adverse event was recorded in the control group, but there were no significant differences in the types or incidence of adverse events between the two groups, indicating that the combined regimen is well tolerated and safe, with potential short-term clinical applicability. This safety finding should be interpreted in the context of the study protocol, because the two agents were not mixed or co-infused but were administered sequentially with line flushing and predefined monitoring and stopping criteria to reduce potential compatibility and infusion-related risk. Direct evidence regarding the safety of this specific sequential intravenous regimen remains limited. This study adds to the evidence for expanding treatment strategies for sudden sensorineural hearing loss and suggests that combined intervention may yield short-term clinical benefits during the 90-day observation period. Considering that adverse reactions did not increase significantly after treatment and that multiple quality-of-life dimensions showed a positive trend of improvement, these findings support further evaluation of this regimen in larger, multicenter studies. However, formal minimal clinically important difference thresholds for PTA-4, HHIA, and SF-36 were not available for this population; therefore, the clinical importance of the observed changes should be interpreted with appropriate caution. Future work should clarify potential mechanisms through longer follow-up and monitoring of biochemical indices to provide more comprehensive evidence to improve the treatment system for sudden sensorineural hearing loss.

Alprostadil mainly acts by dilating blood vessels, inhibiting platelet aggregation, and reducing blood viscosity, thereby providing more adequate oxygen and blood supply to the inner-ear microcirculation, which is particularly critical in the acute phase of sudden sensorineural hearing loss¹⁵. The organ of Corti and auditory nerve cells in the inner ear are highly dependent on the blood supply; once oxygen delivery declines due to hypoperfusion or vascular spasm, hearing function often drops rapidly and may even result in permanent damage¹⁶. In this study, alprostadil significantly improved the mean hearing threshold in patients with sudden sensorineural hearing loss, and this finding may be partly explained by mechanisms proposed in previous studies, including maintenance of inner-ear hair-cell activity and attenuation of oxidative stress induced by microcirculatory disorders. Ginkgo biloba extract, rich in flavonoids, terpenes, and other active components, has been reported to produce multi-target synergistic benefits by promoting vasodilation, lowering blood viscosity, and scavenging oxygen-free radicals¹⁷. Microcirculatory disturbance and cellular oxidative stress are the most common pathological bases of sudden sensorineural hearing loss; relying on a single drug may not fully address the multiple pathways underlying hemodynamic impairment and free-radical accumulation. The synergistic action of alprostadil and Ginkgo biloba extract may help intervene early in the inner-ear blood supply and cellular metabolic disorders and potentially lessen the ongoing damage to auditory structures caused by inflammatory responses and harmful metabolic products¹⁸. In the combination therapy group of this study, in addition to the greater improvement in mean hearing threshold, patients’ scores for mood and social interaction were also higher. Once hearing function is restored, it has positive implications for patients’ social activities, mental health, and subjective experience of handicap. Timely restoration of inner-ear perfusion and suppression of oxidative damage may help reduce irreversible structural degeneration within the auditory pathway, creating favorable conditions for subsequent neural and vascular repair. Because this regimen did not increase the risk of serious adverse events and was associated with significant improvements in clinical outcomes, the present findings suggest the short-term feasibility of this intervention, although blood-flow optimization and neuroprotection were not directly measured in this trial.

This study, compared with previous literature, more systematically evaluated the audiological and quality-of-life effects of alprostadil combined with Ginkgo biloba extract on sudden sensorineural hearing loss, aiming to provide additional real-world evidence for current clinical practice. Previous studies mostly focused on single-drug interventions, such as the effectiveness of alprostadil or Ginkgo biloba extract in some patients with sudden sensorineural hearing loss, but because of the influence of disease course, individual constitution, and complications, the actual effects often showed instability¹⁹. In this retrospective comparative analysis, the study found that the two drugs used in combination could achieve multiple synergies in stabilizing blood flow and inhibiting oxidative stress, and, by comparing intervention effects at different time points, verified the persistence of this potentiation. Unlike previous reports that merely focused on hearing-threshold improvement, this study placed greater emphasis on the comprehensive evaluation of quality-of-life indicators and used multidimensional scales to quantify patients’ psychological, social, and emotional states²⁰. The results showed that combined therapy could more effectively alleviate subjective hearing handicap and related negative emotions, demonstrating the value of the intervention for patients’ overall functioning and mental health. For patients with sudden sensorineural hearing loss, hearing recovery is not limited to the improvement of threshold values, but also involves negative chain reactions such as social handicaps and decreased self-confidence caused by auditory impairment²¹; during the follow-up period of this study, these negative factors could be effectively relieved through active intervention, reflecting the overall improving trend of quality of life under the intervention regimen. In addition, the continuous observation up to 90 days showed that the early hearing improvement produced by the combination therapy could be consolidated in the subsequent stage, which is consistent with other literature viewpoints indicating that “early detection and early treatment can significantly reduce the risk of later functional impairment”²², and also provides a reference value for the time window in subsequent studies. Considering the complex and diverse etiological mechanisms of sudden sensorineural hearing loss, it is crucial to determine the appropriate timing of intervention early and adopt a targeted, comprehensive regimen. With a relatively long follow-up period and multidimensional indicators, this study further suggested that the combined treatment regimen may have greater clinical feasibility and a potential for lasting benefit compared with single-drug interventions, and also laid a preliminary foundation for future validation studies with larger sample sizes.

This study adopted a retrospective comparative design using real-world clinical data and conducted multidimensional assessments of audiological outcomes and quality of life, including pure-tone thresholds, hearing grades, the SF-36, and HHIA scales. The results verified the feasibility and safety of the combined use of alprostadil and a Ginkgo biloba extract for improving sudden sensorineural hearing loss, without a marked increase in adverse events. However, because this was a retrospective study without randomization, performance bias and reporting bias cannot be excluded for patient-reported outcomes such as the SF-36 and HHIA, as well as for adverse events, although the risk of bias for objective audiological outcomes was reduced by assessor blinding. The retrospective nature inherently limits causal inference. Group assignment was not randomized, and unmeasured confounding may exist. Furthermore, because this study was single-center and the sample size was limited, the extrapolation of the results to larger populations still needs further verification in multicenter studies with large samples; meanwhile, follow-up lasted only to 90 days, so longer-term therapeutic benefits, relapse conditions, and long-term safety could not be evaluated; in addition, inflammatory factors and oxidative-stress indicators were not deeply examined, and subsequent studies should elucidate the specific mechanisms of the combined therapy at the molecular level, so as to achieve wider application in the comprehensive prevention and treatment of sudden sensorineural hearing loss.

As observed in this retrospective cohort, alprostadil combined with Ginkgo biloba extract may represent a promising early therapeutic option for sudden sensorineural hearing loss, with potential to improve short-term hearing recovery and patient-perceived hearing handicap, physical functioning, and psychosocial well-being without an apparent increase in safety burden. Its clinical significance should still be interpreted with caution in light of the single-center design, moderate sample size, relatively short follow-up period, lack of randomization, and the absence of minimal clinically important difference thresholds. At present, this regimen is best regarded as a potentially applicable treatment strategy that warrants further confirmation in larger, multicenter studies with longer follow-up to establish its robustness, generalizability, and longer-term benefit.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
1. Study drugs and infusion solutions
Alprostadil Injection	Beijing Tide Pharmaceutical Co., Ltd.	NMPA Approval No. H10980024	2 mL:10 μg; study drug for intravenous infusion.
Extract of Ginkgo Biloba Leaves Injection	Zhong Hao International Co., Ltd. (marketing authorization holder); Chi Sheng Pharma & Biotech Co., Ltd. (manufacturer)	NMPA Approval No. HC20181022	5 mL:17.5 mg; injection; 10 ampoules/box; add-on study drug.
Glucose Injection, 5%	Chimin Health Management Co., Ltd.	NMPA Approval No. H19994045	100 mL:5 g; used as the intravenous infusion diluent and line-flushing solution.
2. Infusion administration and safety supplies
Disposable sterile intravenous infusion set for gravity infusion	Hospital standard consumable	GB 8368-2018; catalog number N/A	Used for peripheral venous infusion of separately prepared study drugs.
Peripheral intravenous access supplies	Hospital standard consumables	N/A	Infusion needle or indwelling catheter, tourniquet, skin disinfectant, and sterile dressing; used to establish and maintain peripheral venous access.
Medical electronic blood pressure monitor	OMRON Healthcare Co., Ltd.	HBP-1300	Upper-arm medical electronic blood pressure monitoring device; used for blood pressure and pulse monitoring before infusion.
Emergency management supplies for infusion reactions	Hospital emergency cart/standard emergency medicines	N/A	Available in the treatment area for clinically indicated management of infusion reactions or serious adverse events.
3. Audiological assessment equipment and environment
Diagnostic audiometer	Interacoustics A/S	AD629	Used for air-conduction threshold testing and PTA-4 calculation.
Soundproof booth/audiometry room	Department of Otolaryngology, Tianjin First Central Hospital	N/A	Used to provide a controlled testing environment for pure-tone audiometry.
Audiometer calibration record and biological calibration checklist	Study site/certified calibration service	N/A	Used to document professional calibration before study initiation and daily device self-testing/biological calibration.
Original pure-tone audiometry record sheet	Department of Otolaryngology, Tianjin First Central Hospital	N/A	Used to record air-conduction and bone-conduction thresholds and to support PTA-4 calculation and hearing-grade classification.
4. Questionnaires and scoring materials
36-Item Short Form Health Survey (SF-36)	RAND	RAND 36-Item Health Survey 1.0	A 36-item health-related quality-of-life questionnaire covering 8 domains; source instrument and scoring instructions are provided by RAND; scored according to the RAND 36-Item Health Survey 1.0 standard scoring instructions.
SF-36 scoring and verification sheet	Study team, Department of Otolaryngology, Tianjin First Central Hospital	N/A	Used to convert raw SF-36 responses to 0–100 domain scores and verify subject number, assessment date, and raw scores.
Hearing Handicap Inventory for Adults (HHIA)	Original instrument: Newman et al.	N/A	A 25-item self-assessment scale for social and emotional effects of hearing impairment; source instrument originally described by Newman et al. (1990) and psychometrically reevaluated by Cassarly et al. (2020); scored Yes/Sometimes/No = 4/2/0.
HHIA scoring and verification sheet	Study team, Department of Otolaryngology, Tianjin First Central Hospital	N/A	Used to calculate total HHIA scores and verify subject number, assessment date, and raw scores.
5. Study documentation and trial management materials
Case report form (CRF)	Study team, Department of Otolaryngology, Tianjin First Central Hospital	N/A	Used to record screening, informed consent, randomization number, baseline variables, treatment, follow-up, outcomes, and concomitant treatment.
Infusion administration record form	Study team, Department of Otolaryngology, Tianjin First Central Hospital	N/A	Used to record subject ID, drug name, dose, solvent, administration date/time, infusion start/end time, infusion reaction, and completion status.
Follow-up and concomitant treatment record form	Study team, Department of Otolaryngology, Tianjin First Central Hospital	N/A	Used at day 10, day 30, and day 90 to record follow-up completion, audiometry, scale assessment, adverse events, and concomitant or prohibited treatment.
Adverse event record form	Study team, Department of Otolaryngology, Tianjin First Central Hospital	N/A	Used to record adverse reactions, serious adverse events, onset time, management, outcome, and relationship assessment when applicable.
Basic otologic management checklist	Study team, Department of Otolaryngology, Tianjin First Central Hospital	N/A	Used to standardize noise avoidance, discontinuation of headphone use, sleep management, tinnitus explanation, medication-reaction observation, and follow-up reminders; routine vitamin B therapy was not used.
6. Statistical software
IBM SPSS Statistics 26.0	IBM	26	Statistical analysis software.