Table 4.
The pooling results of the incidence, therapy and prognosis of CMVR-related RD in aIDS patients.
| Category | No. of studies | Pooled incidence | 95% CI | P-value | I 2 | Sensitivity analysis | Selected model | |
|---|---|---|---|---|---|---|---|---|
| Incidence | ||||||||
| General RD | 10 | 24% | 18% | 29% | P<0.01 | 83% | Negative | Random-effect model |
| RD at baseline/enrollment | 40 | 13% | 10% | 16% | P<0.01 | 92% | Negative | Random-effect model |
| New RD after anti-CMV therapy | 48 | 12% | 10% | 14% | P<0.01 | 69% | Negative | Random-effect model |
| Therapy choices for RD | ||||||||
| PPV with SO or gas tamponade | 24 | 52% | 32% | 73% | P<0.01 | 97% | Negative | Random-effect model |
| Laser | 3 | 25% | 4% | 54% | P<0.01 | 94% | Negative | Random-effect model |
| Scleral buckle | 2 | 19% | 9% | 30% | P=0.28 | 14% | Negative | Fixed-effect model |
| No therapy a | 4 | 46% | 32% | 60% | P=0.14 | 50% | Negative | Random-effect model |
| Anatomic success rate | ||||||||
| PPV with SO or gas tamponade | 12 | 89% | 85% | 93% | P=0.07 | 40% | Negative | Fixed-effect model |
| Laser | 3 | 65% | 43% | 84% | P=0.13 | 52% | Negative | Random-effect model |
| SO-dependent eyes | 3 | 26% | 0% | 92% | P<0.01 | 98% | Negative | Random-effect model |
a
Patients with CMVR-related RD who did not receive any therapy because the retinal reattachment was inoperable or the patients were systemically unwell to undergo surgery.
AIDS, acquired immunodeficiency syndrome; CI, confidence interval; CMV, cytomegalovirus; CMVR, cytomegalovirus retinitis; PPV, pars plana vitrectomy; RD, retinal detachment; SO, silicone oil.