Intraoperative Use of Platelet-Rich Plasma in Tympanoplasty: Effects on Graft Success and Audiological Outcomes - A Prospective Controlled Trial
Intraoperative Use of Platelet-Rich Plasma in Tympanoplasty: Effects on Graft Success and Audiological Outcomes - A Prospective Controlled Trial
This single-center controlled trial evaluated whether intraoperative application of autologous platelet-rich plasma (PRP)-impregnated absorbable gelatin sponge improves graft take rates and audiological outcomes in adults undergoing tympanoplasty for chronic otitis media with tympanic membrane perforation. Forty-eight patients were allocated by systematic alternation to a PRP group (PRP-impregnated Spongostan placed in the middle ear and lateral to the graft) or a control group (dry Spongostan). The primary outcome was graft integrity at two months; the secondary outcome was improvement in the air-bone gap.
Study design and ethics This prospective, single-center, randomized controlled trial was conducted at the Otolaryngology-Head and Neck Surgery Clinic of Sisli Hamidiye Etfal Training and Research Hospital, Istanbul, between April 1 and June 30, 2023. The protocol received ethics committee approval (decision no. 3857, April 4, 2023). All participants provided written informed consent before enrolment. The trial was conducted in accordance with the Declaration of Helsinki.
Participants Adults aged 18-65 years were eligible if they had a central tympanic membrane perforation present for at least six months without active discharge in the preceding three months, were able to undergo and cooperate with preoperative pure-tone audiometry, and were available for follow-up. Exclusion criteria were: planned concurrent mastoidectomy, cholesteatoma on preoperative imaging or intraoperative inspection, a history of ventilation tube insertion on the affected side, revision surgery, uncontrolled hypertension or diabetes mellitus, coagulation disorder or antiplatelet/anticoagulant therapy, anaemia (Hgb <12 g/dL) or thrombocytopaenia (platelets <150,000/μL), active malignancy, and serious psychiatric illness.
Randomization Patients who met inclusion criteria were allocated by simple alternation: the first eligible patient was assigned to the PRP group, the second to the control group, the third to the PRP group, and so on. Patients were not informed of their group assignment (single-blind design). Surgeons were necessarily aware of group allocation.
Sample size Sample size was calculated with G*Power version 3.1.6. Based on an expected large effect size (Cohen's d = 0.62) derived from comparable prospective cohort data, a two-sided α of 0.05, and a power of 80%, the required sample was 48 patients (24 per group).
PRP preparation PRP was prepared manually at the start of each operation in the PRP group. After intravenous access was established, 8.5 mL of venous blood was drawn into a glass tube containing 1.5 mL of acid citrate dextrose (ACD-A solution; Becton Dickinson, Plymouth, UK) using an 18-gauge needle. The tube was centrifuged at 419 × g (2,500 rpm) for 8 minutes in a Tomos Centrifuge 802B (Tomos Life Science Group, Shanghai, China), balanced against a tube of equal weight containing normal saline. The platelet-rich buffy coat layer and the overlying plasma were carefully aspirated using a spinal cannula into a sterile, anticoagulant- free 10-mL tube. This tube was then centrifuged at 822 × g (3,500 rpm) for 8 minutes. The upper two-thirds of the supernatant were discarded, and the remaining platelet concentrate was gently resuspended by inverting the tube several times. This final product was used to saturate the spongostans immediately before placement.
Surgical technique The surgeries were performed under general anesthesia by one of six experienced ENT specialists. The approach (transcanal, endoauricular, or retroauricular) was determined preoperatively based on the location and size of the perforation, the anatomy of the external auditory canal, and the surgeon's preference. After standard preparation and local infiltration with 20 mg/mL lidocaine plus 0.0125 mg/mL epinephrine, the perforation edges were de- epithelialized with an instrument. The ossicular chain was examined and reconstructed as needed. In the PRP group, PRP-impregnated spongy bone was placed in the middle ear cavity and next to the graft. In the control group, dry spongy bone was placed in the same positions. The graft material (cartilage, temporal fascia, or fat) was selected according to the perforation characteristics and the surgeon's preference. An under- or over-under placement technique was used as needed. The external auditory canal was filled with antibiotic-impregnated gauze strips (bacitracin 500 IU/g + neomycin 5 mg/g ointment). All patients received 0.3% ciprofloxacin ear drops three times a day for one week after surgery.
Outcome measures and follow-up Patients were reviewed at two weeks (suture removal and packing removal) and at two months. At the two-month visit, the tympanic membrane was examined under microscopy and graft integrity was recorded (binary outcome: intact or perforated). Pure-tone audiometry was performed in a sound-treated booth using an Interacoustics AC 40 audiometer in accordance with ISO standards and American Speech-Language-Hearing Association criteria. Thresholds were measured at 250, 500, 1000, 2000, and 4000 Hz by air and bone conduction. The ABG was calculated as the mean air-bone gap across 500, 1000, 2000, and 4000 Hz. Hearing gain was defined as the difference between preoperative and postoperative ABG. Perforation area was measured intraoperatively using a calibrated 2-mm Rosen elevator and diamond burrs of known diameter; it was recorded for descriptive purposes only and was not a selection criterion.
Statistical analysis Analyses were performed with SPSS 15.0 for Windows. Categorical variables were compared with Chi-square tests; continuous variables were compared with independent- samples t-tests when normally distributed (Shapiro-Wilk test, p>0.05) or Mann-Whitney U tests otherwise. Within-group pre- to postoperative changes were assessed with paired t-tests or Wilcoxon signed-rank tests as appropriate. Bonferroni correction was applied for multiple comparisons. Statistical significance was defined as p<0.05 (two-tailed).
Inclusion Criteria:
Exclusion Criteria: