This study aimed to perform an updated meta-analysis examining the efficacy of rt-PA intravenous thrombolysis in patients with CRAO. In this meta-analysis, data from 121 patients with CRAO receiving the rt-PA intravenous thrombolysis among seven studies were analyzed. 62 of 121 patients with CRAO showed improved VA. Compared with the conservative treatment (13.0%), the improvement rate was significantly higher in patients treated with rt-PA intravenous thrombolysis (40.4%). Overall, the pooled evidence suggested that rt-PA intravenous thrombolysis was effective in treating CRAO. Furthermore, rt-PA intravenous thrombolysis in CRAO was a safe therapeutic option. Complications following rt-PA intravenous thrombolysis were detected in only 11 patients. The reported major complication was hemorrhage with no associated mortality.
rt-PA is firstly approved for acute ischemic stroke by the U.S. Food and Drug Administration in 1996. A prior Cochrane systematic review concluded that rt-PA treatment for acute ischemic stroke was safe and effective [15]. CRAO is a stroke of the eye and analogous to ischemic cerebral stroke. As a thrombolytic agent, rt-PA can be administered intravenously or intra-arterially for the treatment of CRAO [10,16]. Intravenous administration has the advantage of easier access with the nonspecialized physicians in the emergency room and reduced risk of haemorrhagic complications [17]. However, intra-arterial thrombolysis needs to be performed by an interventional radiologist in the catheter room following sufficient preoperative preparation, thus, delaying the optimal therapeutic time window, which is currently recognized as a limiting factor for the effective treatment of CRAO. Moreover, a RCT study suggested that intra-arterial rtPA could not be recommended for the management of acute CRAO due to similar VA outcomes and the higher rate of adverse events compared with the conservative treatment [18].
Based on the above mentioned shreds of evidence, intravenous rt-PA treatment was inclined to be used for the treatment of CRAO. Until 2011, the first RCT on intravenous rt-PA treatment in CRAO was performed with negative conclusion [6]. The major reasons for this negative finding of the study included the small sample size (8 patients in the treatment group) and prolonged therapeutic time window from symptom onset to treatment with the average time of 14.4h. Yet another RCT concluded that 16 of 24 patients (66.7%) had exhibited significant visual improvement [10]. In general, the improvement rate of VA following intravenous rt-PA treatment ranged from 0% to 83.3% [6–12]. However, the pooled effect of intravenous rt-PA treatment in CRAO remains unclear. Thus, we performed this meta-analysis.
Unlike an ischemic cerebral stroke, wherein there are guideline-based recommendations for treatment with rt-PA within 4.5 hours of symptom onset [19], the same robust data are lacking in CRAO. Besides, there is a paucity of data on the optimal time window for effective intervention. The mean time to treatment in the RCT performed by Chen and their colleagues was 14.4 h, longest time window in this meta-analysis [8]. However, the research result of VA improvement was negative. Although the time to treatment was shortest with the average of 3.05h in the prospective study conducted by Schultheiss and the co-workers, the visual improvement was not satisfactory with the improvement rate of 25% [12]. Thus, a considerable association between time to treatment and visual improvement was concluded [7]. This prospective study indicated that time to treatment ≤6.5 hours was significantly associated with a better gain of lines of vision. 53% of patients receiving thrombolytic treatment within 6.5 hours after the onset of symptoms achieved an improvement of three or more Snellen lines, whereas no patient treated after >6.5 hours exhibited a comparable visual outcome. Supposedly, the sooner the intravenous rt-PA therapy was applied for CRAO, the better the visual improvement was achieved. A patients-level meta-analysis showed that intravenous fibrinolysis of rt-PA in CRAO was beneficial at 4.5hours or earlier after symptom onset [1]. This time window of efficacy was in accord with the ischemic cerebral stroke.
This meta-analysis has some inevitable limitations that must be acknowledged. First, a potential source of heterogeneity from the study design and time to treatment exists in the present meta-analysis. Second, there is publication bias because of the relatively small number of patients included in this analysis. Therefore, the results should be interpreted with caution prior the clinical implication of rt-PA intravenous thrombolytic therapy.