Anti-VEGF for prevention of postoperative vitreous cavity haemorrhage after vitrectomy for proliferative diabetic retinopathy

Review question
Does anti-vascular endothelial growth factor (anti-VEGF) reduce the occurrence of posterior vitreous cavity haemorrhage (POVCH) after vitrectomy surgery for proliferative diabetic retinopathy?

POVCH is a significant complication following vitrectomy (removal of the vitreous gel from the posterior chamber of the eye) for the treatment of proliferative retinopathy (the growth of abnormal blood vessels from the retina, a layer of tissue at the back of the eye), occurring in approximately 30% of cases. POVCH has two main forms: early, when haemorrhage (bleeding) is present in the first few postoperative days, and late, when haemorrhage occurs a number of months after surgery. The presence of POVCH delays visual recovery, can lead to elevated pressure within the eye, and can make further treatment for diabetic retinopathy difficult. Ten per cent of patients require revision surgery, which has significant implications for resources, time, and cost. The use of anti-VEGF before surgery (preoperatively) has been proposed as an intervention to reduce the incidence of POVCH.

Search date
The evidence is up to date to May 2015.

Key results
The electronic database searches identified 12 randomised controlled trials that met the inclusion criteria. We performed a number of analyses that suggest that pre- or intraoperative anti-VEGF may reduce the incidence of early POVCH. The effect on late POVCH was unclear. We are currently unable to comment on the effect of anti-VEGF treatment on postoperative visual acuity due to significant differences in the studies' design and outcomes.

The risk of adverse events when using preoperative anti-VEGF appears small.

Quality of evidence
We are reasonably certain that anti-VEGF reduces the incidence of early POVCH (high-quality evidence) but less certain about its effects on late POVCH and risk of adverse effects.

Authors' conclusions: 

The use of pre- or intraoperative bevacizumab lowers the incidence of early POVCH. The reported complications from its use appear to be low. Futher randomised studies that look at other anti-VEGF medications are ongoing and will strengthen the current review findings, giving both surgeons and patients evidence to guide treatment choices in the management of proliferative retinopathy.

Read the full abstract...

Postoperative vitreous cavity haemorrhage (POVCH) is a significant complication following vitrectomy for proliferative diabetic retinopathy (PDR). It delays visual recovery and can make further treatment difficult if the view of the fundus is significantly obscured. A number of interventions to reduce the incidence of POVCH have been proposed, including the perioperative use of anti-vascular endothelial growth factor (anti-VEGF). Anti-VEGFs reduce vascular proliferation and the vascularity of neovascular tissue, which is often the source of bleeding following vitrectomy.


This updated review aimed to summarise the effects of anti-VEGF use to reduce the occurrence of POVCH after vitrectomy surgery for PDR.

Search strategy: 

We searched CENTRAL (which contains the Cochrane Eyes and Vision Group Trials Register) (2015, Issue 4), Ovid MEDLINE, Ovid MEDLINE In-Process and Other Non-Indexed Citations, Ovid MEDLINE Daily, Ovid OLDMEDLINE (January 1946 to May 2015), PubMed (January 1966 to May 2015), EMBASE (January 1980 to May 2015), Latin American and Caribbean Health Sciences (LILACS) (January 1982 to May 2015), the ISRCTN registry (, (, and the the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) ( We did not use any date or language restrictions in the electronic searches for trials. We last searched the electronic databases on 26 May 2015.

Selection criteria: 

We included all randomised controlled trials (RCTs) and quasi-RCTs that looked at the use of anti-VEGFs and the incidence of POVCH in people undergoing vitrectomy for PDR.

Data collection and analysis: 

Both review authors independently assessed and extracted the data. We used standard methodological procedures expected by Cochrane.

The primary outcomes of the review were the incidence of early and late POVCH following perioperative anti-VEGF administration. Secondary outcomes included best-corrected visual acuity at six months following surgery, the incidence of vitreous cavity washout or revision vitrectomy at six months, adverse effects of intervention (cataract, iris rubeosis and rubeotic glaucoma, retinal detachment, increased inflammation and systemic side effects), quality of life measures performed at least six months following vitrectomy, and density of POVCH.

Main results: 

The current review included 12 RCTs that looked at the pre- or intraoperative use of intravitreal bevacizumab to prevent postoperative vitreous haemorrhage during pars plana vitrectomy for complications of PDR. The studies were conducted in a variety of countries (three from Iran, two from Italy, two from Egypt, and the remaining from South Korea, USA, Mexico, Pakistan, and Japan). The inclusion criteria for entry into the studies were standard complications of proliferative retinopathy: non-clearing vitreous haemorrhage, tractional retinal detachment involving the macula, or combined tractional rhegmatogenous detachment. The included studies randomised a total of 654 eyes. The average age of the participants was 54 years.

We identified methodological issues in all included studies. Risk of bias was highest for masking of participants and investigators (four studies were an 'open label' design), and a number of studies were unclear when describing randomisation methods and sequence allocation.

Participants receiving intravitreal bevacizumab in addition to pars plana vitrectomy were less likely to experience early POVCH (grade 2) compared to people undergoing pars plana vitrectomy alone (risk ratio (RR) 0.28, 95% confidence interval (CI) 0.08 to 0.96, 2 studies, 144 eyes, high-quality evidence). This corresponds to an absolute effect of 130 fewer people (95% CI 167 fewer to 7 fewer) with early POVCH per 1000 people when treated with intravitreal bevacizumab. We saw similar results for all grades of POVCH (RR 0.35, 95% CI 0.23 to 0.53, 9 studies, 512 eyes) and when excluding cases where assessment of outcome was impossible due to presence of silicone oil (RR 0.34, 95% CI 0.19 to 0.60, 6 studies, 302 eyes).

The effect of pre- or intraoperative intravitreal bevacizumab on the incidence of late postoperative haemorrhage was uncertain (RR 0.72, 95% CI 0.30 to 1.72, 3 studies, 196 eyes, low-quality evidence). The absolute effect was 55 fewer people (95% CI 138 fewer to 143 more) with late POVCH per 1000 people when treated with intravitreal bevacizumab. This outcome was rarer and was only reported in a few studies. We are currently unable to provide an estimate of the effect of intravitreal bevacizumab on postoperative visual acuity due to significant study heterogeneity.

No local or systemic complications of intravitreal bevacizumab were reported by the RCTs. The risk of postoperative retinal detachment was lower in the participants treated with pre- or intraoperative bevacizumab (RR 0.46, 95% CI 0.19 to 1.08, 7 studies, 372 participants, low-quality evidence); the absolute effect was 49 fewer people (95% CI:73 fewer to 8 more) with postoperative retinal detachment per 1000 people when treated with intravitreal bevacizumab.