- Baseline Data
The baseline features of all cases are summarized in Table 1. Patient characteristics were not significantly different in the three groups. Grading of the extent of fibrovascular proliferation was defined as follows[15]: Grade 1, focal adhesions only; Grade 2, broad adhesion≥1 sites or vitreous-retinal adhesion at disk, macula or arcade; Grade 3, vitreous-retinal attachment extending to the periphery.
- Intraoperative findings
Surgical findings and additional surgical procedures are shown in Table 2.
The incidence of iatrogenic retinal breaks was comparable in the three groups respectively, and it was statistically significantly less in Group B and Group C than in Group A (P=0.015 and 0.028, respectively, Chi-square Test). Intraoperative bleeding was classified into 3 grades[15]: Grade 1, minor bleeding that stopped either spontaneously or by transient bottle elevation; Grade 2, moderate bleeding requiring endodiathermy or with formation of broadsheets of clots extending away from the bleeding site; Grade 3, thick clot formation covering at least half of the posterior pole or interfering with the surgical plane. There was a significant difference in the severity of intraoperative bleeding among the three groups (P=0.019, Chi-square Test). The incidence of severe intraoperative bleeding was statistically significantly less in Group B (P=0.020) and Group C (P=0.011) than in Group A. There was no significant difference between the Group B and Group C (P=0.975). The type of internal tamponading agent was comparable in the three groups. The use of long-term internal tamponade agents (C3F8 and silicon oil) were statistically significant in three groups (P=0.003, Chi-square Test), which is less in Group B and Group C than in Group A (P=0.028 and 0.000, respectively). Additionally, the data showed that the mean surgical time was statistically significantly shorter in Group B and Group C than in Group A (P=0.000 and 0.000, respectively, ANOVA and LSD test).
3.Visual Acuity
BCVA was converted from Snellen chart values to the logarithm of minimum angle of resolution (logMAR) equivalent units for statistical calculations. As shown in Table 3, at baseline examination, the mean BCVA was 1.567±0.340 logMAR, 1.555±0.343 logMAR, and 1.587±0.340 logMAR respectively in groups A, B, and C. There was no statistically significant difference between the three groups (P=0.893, ANOVA and LSD test). At the 1-month follow-up, the mean BCVA had increased to 1.250±0.398 logMAR in group A, 0.946±0.365 logMAR in Group B, and 0.872±0.416 logMAR in Group C. The postoperative mean BCVA was statistically significantly better in Groups B (P=0.002) and Group C (P=0.000) than in Group A respectively at 1 month after surgery. There was no significant difference between Group B and Group C (P=0.371). The 3-month postoperative BCVA was not significantly different between Groups A and Group B (P=0.083). However, Group C had better visual acuity than Group A (P=0.000) and Group B (P=0.044) respectively. 6 months after surgery, visual acuities were not significantly different among the three groups (P=0.155).
- Postoperative adverse events
The primary postoperative adverse events, recorded in Table 4, were high IOP and the incidence of early recurrent VH.
A transient (<1 week) increase in IOP (>25 mmHg) occurred in 4 patients, 5 patients and 17 patients in Group A, B, C respectively. After one month of surgery, the patients with high IOP decreased to 2, 1 and 3 respectively. Analysis of these data showed a statistically significant increase in Group C compared with Group A and B at 1 week after surgery (P=0.045, Chi-square Test). There was no difference at 1 month (P=0.719).
VH was scaled according to the Diabetic Retinopathy Vitrectomy Study grading system 1 week and 1 month after surgery. Grade 0: No VH; Grade 1: Mild VH with visible fundus details; Grade 2: Moderate VH with no visible fundus details but with an orange fundus reflex; Grade 3: Severe VH with no retinal details and no orange fundus reflex. The incidence of VH was significantly higher in Group A than that in Group B (P=0.040, Chi-square Test) and Group C (P=0.019) within 1 month after the operation. No significant difference was observed between Group B and Group C (P=0.947).
- Postoperative exudative retinal detachment and choroidal detachment
Exudative retinal detachment and choroidal detachment are relatively rare postoperative complications for the patients with PDR, mostly due to excessive laser energy [16]. We observed the effect of IVTA on exudative reticular detachment and choroidal detachment to clarify whether TA had the effect of reducing these complications. However, there were no statistically significant differences in the incidence of exudative retinal detachment(P=0.659, Chi-square Test) and choroidal detachment (P=0.261) among the three groups (Table 5).