Study design
We hypothesized that PRP injection administered in combination with TELD has clinical benefits and could improve AF remodeling. This prospective, open-label, randomized, controlled pilot trial included patients who underwent TELD with PRP injection between July 2018 and January 2019. This study was approved by the Institutional Review Board of Beijing Haidian Hospital (2018002) and registered on the Chinese Clinical Trial Registry (ChiCTR 1800017228). All patients provided written informed consent.
Power analysis was performed to calculate a sample size that could achieve > 80% power based on the visual analog scale (VAS) score difference in the pre-experiment. With an effect size of 0.77, we estimated that at least 28 patients would be needed in each group (PRP and control groups).
Sixty patients were enrolled and randomly assigned to receive either TELD with PRP injection or TELD only, with a ratio of 1:1. Inclusion criteria were as follows: symptoms of low back pain and leg pain associated with herniated nucleus pulposus based on magnetic resonance imaging (MRI) data, single-level involvement without calcification, no lumbar surgery history, failed conservative treatment after 8 weeks, and platelet count > 150 × 109/L. Patients who were pregnant, were receiving medications affecting platelet function, had coagulation disorders, had cauda equina syndrome, or were aged < 18 years or > 80 years were excluded.
PRP preparation
PRP injection was performed using a sterile WEGO PRP preparation kit (Wego New Life Medical Devices Co., Ltd., Shandong, China). First, an anticoagulant was extracted using a 50 mL syringe, which sufficiently lubricated the inner wall of the syringe. Whole blood was collected from the median cubital vein. Subsequently, the blood was centrifuged at 2500 rpm for 10 min to separate the whole blood and form a buffy coat layer containing platelets and white blood cells. Then, a 20 mL syringe was used to remove the excess red blood cells at 1 mm below the cone. Finally, a second centrifugation was conducted at 2750 rpm for 10 min to further clear the buffy coat layer. The supernatant was removed with a 20 mL syringe. The pellet contained the PRP (Fig. 1) and was freshly prepared before use. The PRP volume for injection was 4 mL.
Surgical procedure
All procedures were performed under local anesthesia administered by the same experienced orthopedic surgeon using the technique introduced by Hoogland 15. Patients were placed in the prone position. An 18G needle was introduced from a skin entry point to the superior articular process of the lower involved vertebrae of the herniated disc under fluoroscopy monitoring. After the needle reached the target, a guide wire was inserted. A series of dilators were introduced, and a reamer was subsequently inserted through the cannula for foraminoplasty. While the neuroforamen was large enough for the working channel, the endoscope was introduced to observe the relationship of the nerve root and herniated nucleus pulposus under continuous irrigation. After the removal of the extruded or sequestrated fragment, the bulging annulus underneath the nerve root was also removed in cases where it was necessary. After draining out the fluids, the fresh PRP was mixed with 0.4 mL thrombin solution (1:10) to activate the platelets which simultaneously yielded a gel. This PRP gel mix was injected into the local site around the annuloplasty of patients in the PRP group (Fig. 2). In the control group, only discectomy was performed. The incision was closed without drainage in both groups.
MRI technique
Imaging of lumbar spines was obtained by a 1.5-Tesla MRI machine (HD-xt, GE Healthcare, USA). The postoperative MRI parameters were turbo echo T2-weighted (T2W) with 3000 ms repetition time and 75–85 ms echo time. Axial imaging was set with a slice thickness of 4 mm and an interslice gap of 0.4 mm, 180 × 180 mm field of view, and 400 × 288 matrix. Sagittal imaging was performed with a slice thickness of 3 mm without a gap, 280 × 280 mm field of view, and 400 × 312 matrix.
Measurements
For the clinical evaluation, each patient completed a questionnaire consisting of standardized outcome assessments at baseline and at 3 days, 6 months, and 1 year after surgery; all patients returned for a clinical follow-up. Data including a VAS score of back pain and leg pain and Oswestry disability index (ODI) were obtained.
Postoperative MRI parameters of the patients at 3 days and 1 year after surgery were assessed by one spine surgeon and one radiologist and compared between the two groups. Disc height (DH) was computed as the mean of the anterior and posterior DH from MRI data. Spine cross-sectional area (SCSA) was calculated on the axial cut of the T2W image using the miPlatform 3.0 software (Hinacom Software and Technology, Beijing, China). The SCSA was measured in the lower endplate plane, midline plane, and upper endplate plane, and the average was obtained.
Statistical analysis
Measurement data were expressed as mean ± standard deviation. Categorical variables were presented as frequencies and percentages. Normally distributed data such as age, platelet count, DH, and SCSA were analyzed by the independent sample t-test, and the Mann-Whitney U test was used to analyze the difference in VAS and ODI. The significance of the differences in sex, surgical levels, and annular defect type was analyzed using the chi-square test. Interobserver reliability was calculated using the intraclass correlation coefficient (ICC). Analyses were performed using SPSS version 19.0 (IBM Corp. Armonk, NY). A P < 0.05 was considered statistically significant.